Update store.py

Co-authored-by: silversub <silversub@gmail.com>

.dockerignore

@@ -1,18 +1,63 @@
-.git
-.github
-changedetectionio/processors/__pycache__
-changedetectionio/api/__pycache__
-changedetectionio/model/__pycache__
-changedetectionio/blueprint/price_data_follower/__pycache__
-changedetectionio/blueprint/tags/__pycache__
-changedetectionio/blueprint/__pycache__
-changedetectionio/blueprint/browser_steps/__pycache__
-changedetectionio/fetchers/__pycache__
-changedetectionio/tests/visualselector/__pycache__
-changedetectionio/tests/restock/__pycache__
-changedetectionio/tests/__pycache__
-changedetectionio/tests/fetchers/__pycache__
-changedetectionio/tests/unit/__pycache__
-changedetectionio/tests/proxy_list/__pycache__
-changedetectionio/__pycache__
+# Git
+.git/
+.gitignore
 
+# GitHub
+.github/
+
+# Byte-compiled / optimized / DLL files
+**/__pycache__
+**/*.py[cod]
+
+# Caches
+.mypy_cache/
+.pytest_cache/
+.ruff_cache/
+
+# Distribution / packaging
+build/
+dist/
+*.egg-info*
+
+# Virtual environment
+.env
+.venv/
+venv/
+
+# IntelliJ IDEA
+.idea/
+
+# Visual Studio
+.vscode/
+
+# Test and development files
+test-datastore/
+tests/
+docs/
+*.md
+!README.md
+
+# Temporary and log files
+*.log
+*.tmp
+tmp/
+temp/
+
+# Training data and large files
+train-data/
+works-data/
+
+# Container files
+Dockerfile*
+docker-compose*.yml
+.dockerignore
+
+# Development certificates and keys
+*.pem
+*.key
+*.crt
+profile_output.prof
+
+# Large binary files that shouldn't be in container
+*.pdf
+chrome.json

.github/ISSUE_TEMPLATE/bug_report.md

@@ -27,6 +27,10 @@ A clear and concise description of what the bug is.
 **Version**
 *Exact version* in the top right area: 0....
 
+**How did you install?**
+
+Docker, Pip, from source directly etc
+
 **To Reproduce**
 
 Steps to reproduce the behavior:

.github/test/Dockerfile-alpine

@@ -2,32 +2,33 @@
 # Test that we can still build on Alpine (musl modified libc https://musl.libc.org/)
 # Some packages wont install via pypi because they dont have a wheel available under this architecture.
 
-FROM ghcr.io/linuxserver/baseimage-alpine:3.18
+FROM ghcr.io/linuxserver/baseimage-alpine:3.21
 ENV PYTHONUNBUFFERED=1
 
 COPY requirements.txt /requirements.txt
 
 RUN \
-  apk add --update --no-cache --virtual=build-dependencies \
+ apk add --update --no-cache --virtual=build-dependencies \
+    build-base \
     cargo \
-    g++ \
-    gcc \
+    git \
     jpeg-dev \
     libc-dev \
     libffi-dev \
-    libjpeg \
     libxslt-dev \
-    make \
     openssl-dev \
-    py3-wheel \
     python3-dev \
+    zip \
     zlib-dev && \
   apk add --update --no-cache \
+    libjpeg \
     libxslt \
-    python3 \
-    py3-pip && \
+    nodejs \
+    poppler-utils \
+    python3 && \
   echo "**** pip3 install test of changedetection.io ****" && \
-  pip3 install -U pip wheel setuptools && \
-  pip3 install -U --no-cache-dir --find-links https://wheel-index.linuxserver.io/alpine-3.18/ -r /requirements.txt && \
+  python3 -m venv /lsiopy  && \
+  pip install -U pip wheel setuptools && \
+  pip install -U --no-cache-dir --find-links https://wheel-index.linuxserver.io/alpine-3.21/ -r /requirements.txt && \
   apk del --purge \
     build-dependencies

.github/workflows/containers.yml

@@ -95,7 +95,7 @@ jobs:
           push: true
           tags: |
             ${{ secrets.DOCKER_HUB_USERNAME }}/changedetection.io:dev,ghcr.io/${{ github.repository }}:dev
-          platforms: linux/amd64,linux/arm64,linux/arm/v6,linux/arm/v7,linux/arm/v8
+          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8,linux/arm64/v8
           cache-from: type=gha
           cache-to: type=gha,mode=max
 
@@ -103,6 +103,27 @@ jobs:
 #          provenance: false
 
       # A new tagged release is required, which builds :tag and :latest
+      - name: Debug release info
+        if: github.event_name == 'release' && startsWith(github.event.release.tag_name, '0.')
+        run: |
+          echo "Release tag: ${{ github.event.release.tag_name }}"
+          echo "Github ref: ${{ github.ref }}"
+          echo "Github ref name: ${{ github.ref_name }}"
+          
+      - name: Docker meta :tag
+        if: github.event_name == 'release' && startsWith(github.event.release.tag_name, '0.')
+        uses: docker/metadata-action@v5
+        id: meta
+        with:
+            images: |
+                ${{ secrets.DOCKER_HUB_USERNAME }}/changedetection.io
+                ghcr.io/dgtlmoon/changedetection.io
+            tags: |
+                type=semver,pattern={{version}},value=${{ github.event.release.tag_name }}
+                type=semver,pattern={{major}}.{{minor}},value=${{ github.event.release.tag_name }}
+                type=semver,pattern={{major}},value=${{ github.event.release.tag_name }}
+                type=raw,value=latest
+
       - name: Build and push :tag
         id: docker_build_tag_release
         if: github.event_name == 'release' && startsWith(github.event.release.tag_name, '0.')
@@ -111,12 +132,8 @@ jobs:
           context: ./
           file: ./Dockerfile
           push: true
-          tags: |
-            ${{ secrets.DOCKER_HUB_USERNAME }}/changedetection.io:${{ github.event.release.tag_name }}
-            ghcr.io/dgtlmoon/changedetection.io:${{ github.event.release.tag_name }}
-            ${{ secrets.DOCKER_HUB_USERNAME }}/changedetection.io:latest
-            ghcr.io/dgtlmoon/changedetection.io:latest
-          platforms: linux/amd64,linux/arm64,linux/arm/v6,linux/arm/v7,linux/arm/v8
+          tags: ${{ steps.meta.outputs.tags }}
+          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8,linux/arm64/v8
           cache-from: type=gha
           cache-to: type=gha,mode=max
 # Looks like this was disabled

.github/workflows/pypi-release.yml

@@ -45,9 +45,12 @@ jobs:
     - name: Test that the basic pip built package runs without error
       run: |
         set -ex
-        sudo pip3 install --upgrade pip 
-        pip3 install dist/changedetection.io*.whl
+        ls -alR 
+        
+        # Find and install the first .whl file
+        find dist -type f -name "*.whl" -exec pip3 install {} \; -quit
         changedetection.io -d /tmp -p 10000 &
+        
         sleep 3
         curl --retry-connrefused --retry 6 http://127.0.0.1:10000/static/styles/pure-min.css >/dev/null
         curl --retry-connrefused --retry 6 http://127.0.0.1:10000/ >/dev/null

.github/workflows/test-container-build.yml

@@ -23,8 +23,28 @@ on:
   # Changes to requirements.txt packages and Dockerfile may or may not always be compatible with arm etc, so worth testing
   # @todo: some kind of path filter for requirements.txt and Dockerfile
 jobs:
-  test-container-build:
+  builder:
+    name: Build ${{ matrix.platform }} (${{ matrix.dockerfile == './Dockerfile' && 'main' || 'alpine' }})
     runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        include:
+          # Main Dockerfile platforms
+          - platform: linux/amd64
+            dockerfile: ./Dockerfile
+          - platform: linux/arm64
+            dockerfile: ./Dockerfile
+          - platform: linux/arm/v7
+            dockerfile: ./Dockerfile
+          - platform: linux/arm/v8
+            dockerfile: ./Dockerfile
+          - platform: linux/arm64/v8
+            dockerfile: ./Dockerfile
+          # Alpine Dockerfile platforms (musl via alpine check)
+          - platform: linux/amd64
+            dockerfile: ./.github/test/Dockerfile-alpine
+          - platform: linux/arm64
+            dockerfile: ./.github/test/Dockerfile-alpine
     steps:
         - uses: actions/checkout@v4
         - name: Set up Python 3.11
@@ -47,24 +67,14 @@ jobs:
             version: latest
             driver-opts: image=moby/buildkit:master
 
-        # https://github.com/dgtlmoon/changedetection.io/pull/1067
-        # Check we can still build under alpine/musl
-        - name: Test that the docker containers can build (musl via alpine check)
-          id: docker_build_musl
-          uses: docker/build-push-action@v6
-          with:
-            context: ./
-            file: ./.github/test/Dockerfile-alpine
-            platforms: linux/amd64,linux/arm64
-
-        - name: Test that the docker containers can build
+        - name: Test that the docker containers can build (${{ matrix.platform }} - ${{ matrix.dockerfile }})
           id: docker_build
           uses: docker/build-push-action@v6
           # https://github.com/docker/build-push-action#customizing
           with:
             context: ./
-            file: ./Dockerfile
-            platforms: linux/amd64,linux/arm64,linux/arm/v6,linux/arm/v7,linux/arm/v8
-            cache-from: type=local,src=/tmp/.buildx-cache
-            cache-to: type=local,dest=/tmp/.buildx-cache
+            file: ${{ matrix.dockerfile }}
+            platforms: ${{ matrix.platform }}
+            cache-from: type=gha
+            cache-to: type=gha,mode=max
 

.github/workflows/test-only.yml

@@ -8,13 +8,13 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - name: Lint with flake8
+      - name: Lint with Ruff
         run: |
-          pip3 install flake8
-          # stop the build if there are Python syntax errors or undefined names
-          flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
-          # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
-          flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
+          pip install ruff
+          # Check for syntax errors and undefined names
+          ruff check . --select E9,F63,F7,F82
+          # Complete check with errors treated as warnings
+          ruff check . --exit-zero
 
   test-application-3-10:
     needs: lint-code
@@ -28,7 +28,6 @@ jobs:
     uses: ./.github/workflows/test-stack-reusable-workflow.yml
     with:
       python-version: '3.11'
-      skip-pypuppeteer: true
 
   test-application-3-12:
     needs: lint-code
@@ -37,3 +36,9 @@ jobs:
       python-version: '3.12'
       skip-pypuppeteer: true
 
+  test-application-3-13:
+    needs: lint-code
+    uses: ./.github/workflows/test-stack-reusable-workflow.yml
+    with:
+      python-version: '3.13'
+      skip-pypuppeteer: true

.github/workflows/test-stack-reusable-workflow.yml

@@ -7,7 +7,7 @@ on:
         description: 'Python version to use'
         required: true
         type: string
-        default: '3.10'
+        default: '3.11'
       skip-pypuppeteer:
         description: 'Skip PyPuppeteer (not supported in 3.11/3.12)'
         required: false
@@ -64,14 +64,16 @@ jobs:
           echo "Running processes in docker..."
           docker ps
 
-      - name: Test built container with Pytest (generally as requests/plaintext fetching)
+      - name: Run Unit Tests
         run: |
           # Unit tests
-          echo "run test with unittest"
           docker run test-changedetectionio  bash -c 'python3 -m unittest changedetectionio.tests.unit.test_notification_diff'
           docker run test-changedetectionio  bash -c 'python3 -m unittest changedetectionio.tests.unit.test_watch_model'
           docker run test-changedetectionio  bash -c 'python3 -m unittest changedetectionio.tests.unit.test_jinja2_security'
-          
+          docker run test-changedetectionio  bash -c 'python3 -m unittest changedetectionio.tests.unit.test_semver'
+
+      - name: Test built container with Pytest (generally as requests/plaintext fetching)
+        run: |
           # All tests
           echo "run test with pytest"
           # The default pytest logger_level is TRACE
@@ -84,16 +86,16 @@ jobs:
         run: |
           # Playwright via Sockpuppetbrowser fetch
           # tests/visualselector/test_fetch_data.py will do browser steps  
-          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_content.py'
-          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/test_errorhandling.py'
-          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/visualselector/test_fetch_data.py'
-          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_custom_js_before_content.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest  -vv --capture=tee-sys --showlocals --tb=long --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_content.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest  -vv --capture=tee-sys --showlocals --tb=long --live-server-host=0.0.0.0 --live-server-port=5004 tests/test_errorhandling.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest  -vv --capture=tee-sys --showlocals --tb=long --live-server-host=0.0.0.0 --live-server-port=5004 tests/visualselector/test_fetch_data.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest  -vv --capture=tee-sys --showlocals --tb=long --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_custom_js_before_content.py'
 
 
       - name: Playwright and SocketPuppetBrowser - Headers and requests
         run: |       
           # Settings headers playwright tests - Call back in from Sockpuppetbrowser, check headers
-          docker run --name "changedet" --hostname changedet --rm -e "FLASK_SERVER_NAME=changedet" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000?dumpio=true" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio; pytest --live-server-host=0.0.0.0  --live-server-port=5004 tests/test_request.py'
+          docker run --name "changedet" --hostname changedet --rm -e "FLASK_SERVER_NAME=changedet" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000?dumpio=true" --network changedet-network test-changedetectionio  bash -c 'find .; cd changedetectionio; pytest --live-server-host=0.0.0.0  --live-server-port=5004 tests/test_request.py; pwd;find .'
 
       - name: Playwright and SocketPuppetBrowser - Restock detection
         run: |                            
@@ -170,13 +172,33 @@ jobs:
           curl --retry-connrefused --retry 6  -s -g -6 "http://[::1]:5556"|grep -q checkbox-uuid
 
           # Check whether TRACE log is enabled.
-          # Also, check whether TRACE is came from STDERR
-          docker logs test-changedetectionio 2>&1 1>/dev/null | grep 'TRACE log is enabled' || exit 1
+          # Also, check whether TRACE came from STDOUT
+          docker logs test-changedetectionio 2>/dev/null | grep 'TRACE log is enabled' || exit 1
           # Check whether DEBUG is came from STDOUT
           docker logs test-changedetectionio 2>/dev/null | grep 'DEBUG' || exit 1
 
           docker kill test-changedetectionio
 
+      - name: Test HTTPS SSL mode
+        run: |
+          openssl req -x509 -newkey rsa:4096 -keyout privkey.pem -out cert.pem -days 365 -nodes -subj "/CN=localhost"
+          docker run --name test-changedetectionio-ssl --rm -e SSL_CERT_FILE=cert.pem -e SSL_PRIVKEY_FILE=privkey.pem -p 5000:5000 -v ./cert.pem:/app/cert.pem -v ./privkey.pem:/app/privkey.pem -d test-changedetectionio
+          sleep 3
+          # Should return 0 (no error) when grep finds it
+          # -k because its self-signed
+          curl --retry-connrefused --retry 6 -k https://localhost:5000 -v|grep -q checkbox-uuid
+      
+          docker kill test-changedetectionio-ssl
+
+      - name: Test IPv6 Mode
+        run: |
+          # IPv6 - :: bind to all interfaces inside container (like 0.0.0.0), ::1 would be localhost only
+          docker run --name test-changedetectionio-ipv6 --rm -p 5000:5000 -e LISTEN_HOST=:: -d test-changedetectionio
+          sleep 3
+          # Should return 0 (no error) when grep finds it on localhost
+          curl --retry-connrefused --retry 6 http://[::1]:5000 -v|grep -q checkbox-uuid
+          docker kill test-changedetectionio-ipv6
+
       - name: Test changedetection.io SIGTERM and SIGINT signal shutdown
         run: |
           
@@ -231,9 +253,9 @@ jobs:
           docker logs test-cdio-basic-tests > output-logs/test-cdio-basic-tests-stdout-${{ env.PYTHON_VERSION }}.txt
           docker logs test-cdio-basic-tests 2> output-logs/test-cdio-basic-tests-stderr-${{ env.PYTHON_VERSION }}.txt
 
-      - name: Store container log
+      - name: Store everything including test-datastore
         if: always()
         uses: actions/upload-artifact@v4
         with:
           name: test-cdio-basic-tests-output-py${{ env.PYTHON_VERSION }}
-          path: output-logs
+          path: .

.gitignore

@@ -1,14 +1,30 @@
-__pycache__
-.idea
-*.pyc
-datastore/url-watches.json
-datastore/*
-__pycache__
-.pytest_cache
-build
-dist
-venv
-test-datastore/*
-test-datastore
+# Byte-compiled / optimized / DLL files
+**/__pycache__
+**/*.py[cod]
+
+# Caches
+.mypy_cache/
+.pytest_cache/
+.ruff_cache/
+
+# Distribution / packaging
+build/
+dist/
 *.egg-info*
+
+# Virtual environment
+.env
+.venv/
+venv/
+.python-version
+
+# IDEs
+.idea
 .vscode/settings.json
+
+# Datastore files
+datastore/
+test-datastore/
+
+# Memory consumption log
+test-memory.log

.pre-commit-config.yaml

@@ -0,0 +1,9 @@
+repos:
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.11.2
+    hooks:
+      # Lint (and apply safe fixes)
+      - id: ruff
+        args: [--fix]
+      # Fomrat
+      - id: ruff-format

.ruff.toml

@@ -0,0 +1,48 @@
+# Minimum supported version
+target-version = "py310"
+
+# Formatting options
+line-length = 100
+indent-width = 4
+
+exclude = [
+    "__pycache__",
+    ".eggs",
+    ".git",
+    ".tox",
+    ".venv",
+    "*.egg-info",
+    "*.pyc",
+]
+
+[lint]
+# https://docs.astral.sh/ruff/rules/
+select = [
+    "B", # flake8-bugbear
+    "B9",
+    "C", 
+    "E", # pycodestyle
+    "F", # Pyflakes
+    "I", # isort
+    "N", # pep8-naming
+    "UP", # pyupgrade
+    "W", # pycodestyle
+]
+ignore = [
+    "B007", # unused-loop-control-variable
+    "B909", # loop-iterator-mutation
+    "E203", # whitespace-before-punctuation
+    "E266", # multiple-leading-hashes-for-block-comment
+    "E501", # redundant-backslash
+    "F403", # undefined-local-with-import-star
+    "N802", # invalid-function-name
+    "N806", # non-lowercase-variable-in-function
+    "N815", # mixed-case-variable-in-class-scope
+]
+
+[lint.mccabe]
+max-complexity = 12
+
+[format]
+indent-style = "space"
+quote-style = "preserve"

COMMERCIAL_LICENCE.md

@@ -0,0 +1,54 @@
+# Generally
+
+In any commercial activity involving 'Hosting' (as defined herein), whether in part or in full, this license must be executed and adhered to.
+
+# Commercial License Agreement
+
+This Commercial License Agreement ("Agreement") is entered into by and between Web Technologies s.r.o. here-in ("Licensor") and (your company or personal name) _____________ ("Licensee"). This Agreement sets forth the terms and conditions under which Licensor provides its software ("Software") and services to Licensee for the purpose of reselling the software either in part or full, as part of any commercial activity where the activity involves a third party.
+
+### Definition of Hosting
+
+For the purposes of this Agreement, "hosting" means making the functionality of the Program or modified version available to third parties as a service. This includes, without limitation:
+- Enabling third parties to interact with the functionality of the Program or modified version remotely through a computer network.
+- Offering a service the value of which entirely or primarily derives from the value of the Program or modified version.
+- Offering a service that accomplishes for users the primary purpose of the Program or modified version.
+
+## 1. Grant of License
+Subject to the terms and conditions of this Agreement, Licensor grants Licensee a non-exclusive, non-transferable license to install, use, and resell the Software. Licensee may:
+- Resell the Software as part of a service offering or as a standalone product.
+- Host the Software on a server and provide it as a hosted service (e.g., Software as a Service - SaaS).
+- Integrate the Software into a larger product or service that is then sold or provided for commercial purposes, where the software is used either in part or full.
+
+## 2. License Fees
+Licensee agrees to pay Licensor the license fees specified in the ordering document. License fees are due and payable as specified in the ordering document. The fees may include initial licensing costs and recurring fees based on the number of end users, instances of the Software resold, or revenue generated from the resale activities.
+
+## 3. Resale Conditions
+Licensee must comply with the following conditions when reselling the Software, whether the software is resold in part or full:
+- Provide end users with access to the source code under the same open-source license conditions as provided by Licensor.
+- Clearly state in all marketing and sales materials that the Software is provided under a commercial license from Licensor, and provide a link back to https://changedetection.io.
+- Ensure end users are aware of and agree to the terms of the commercial license prior to resale.
+- Do not sublicense or transfer the Software to third parties except as part of an authorized resale activity.
+
+## 4. Hosting and Provision of Services
+Licensee may host the Software (either in part or full) on its servers and provide it as a hosted service to end users. The following conditions apply:
+- Licensee must ensure that all hosted versions of the Software comply with the terms of this Agreement.
+- Licensee must provide Licensor with regular reports detailing the number of end users and instances of the hosted service.
+- Any modifications to the Software made by Licensee for hosting purposes must be made available to end users under the same open-source license conditions, unless agreed otherwise.
+
+## 5. Services
+Licensor will provide support and maintenance services as described in the support policy referenced in the ordering document should such an agreement be signed by all parties. Additional fees may apply for support services provided to end users resold by Licensee.
+
+## 6. Reporting and Audits
+Licensee agrees to provide Licensor with regular reports detailing the number of instances, end users, and revenue generated from the resale of the Software. Licensor reserves the right to audit Licensee’s records to ensure compliance with this Agreement.
+
+## 7. Term and Termination
+This Agreement shall commence on the effective date and continue for the period set forth in the ordering document unless terminated earlier in accordance with this Agreement. Either party may terminate this Agreement if the other party breaches any material term and fails to cure such breach within thirty (30) days after receipt of written notice.
+
+## 8. Limitation of Liability and Disclaimer of Warranty
+Executing this commercial license does not waive the Limitation of Liability or Disclaimer of Warranty as stated in the open-source LICENSE provided with the Software. The Software is provided "as is," without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages, or other liability, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the Software or the use or other dealings in the Software.
+
+## 9. Governing Law
+This Agreement shall be governed by and construed in accordance with the laws of the Czech Republic.
+
+## Contact Information
+For commercial licensing inquiries, please contact contact@changedetection.io and dgtlmoon@gmail.com.

Dockerfile

@@ -1,8 +1,5 @@
 # pip dependencies install stage
 
-# @NOTE! I would love to move to 3.11 but it breaks the async handler in changedetectionio/content_fetchers/puppeteer.py
-#        If you know how to fix it, please do! and test it for both 3.10 and 3.11
-
 ARG PYTHON_VERSION=3.11
 
 FROM python:${PYTHON_VERSION}-slim-bookworm AS builder
@@ -26,20 +23,34 @@ WORKDIR /install
 
 COPY requirements.txt /requirements.txt
 
-# --extra-index-url https://www.piwheels.org/simple  is for cryptography module to be prebuilt (or rustc etc needs to be installed)
-RUN pip install --extra-index-url https://www.piwheels.org/simple  --target=/dependencies -r /requirements.txt
+# Use cache mounts and multiple wheel sources for faster ARM builds
+ENV PIP_CACHE_DIR=/tmp/pip-cache
+RUN --mount=type=cache,target=/tmp/pip-cache \
+    pip install \
+    --extra-index-url https://www.piwheels.org/simple \
+    --extra-index-url https://pypi.anaconda.org/ARM-software/simple \
+    --cache-dir=/tmp/pip-cache \
+    --target=/dependencies \
+    -r /requirements.txt
 
 # Playwright is an alternative to Selenium
 # Excluded this package from requirements.txt to prevent arm/v6 and arm/v7 builds from failing
 # https://github.com/dgtlmoon/changedetection.io/pull/1067 also musl/alpine (not supported)
-RUN pip install --target=/dependencies playwright~=1.41.2 \
+RUN --mount=type=cache,target=/tmp/pip-cache \
+    pip install \
+    --cache-dir=/tmp/pip-cache \
+    --target=/dependencies \
+    playwright~=1.48.0 \
     || echo "WARN: Failed to install Playwright. The application can still run, but the Playwright option will be disabled."
 
 # Final image stage
 FROM python:${PYTHON_VERSION}-slim-bookworm
+LABEL org.opencontainers.image.source="https://github.com/dgtlmoon/changedetection.io"
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
     libxslt1.1 \
+    # For presenting price amounts correctly in the restock/price detection overview
+    locales \
     # For pdftohtml
     poppler-utils \
     zlib1g \
@@ -68,7 +79,7 @@ COPY changedetection.py /app/changedetection.py
 # Github Action test purpose(test-only.yml).
 # On production, it is effectively LOGGER_LEVEL=''.
 ARG LOGGER_LEVEL=''
-ENV LOGGER_LEVEL "$LOGGER_LEVEL"
+ENV LOGGER_LEVEL="$LOGGER_LEVEL"
 
 WORKDIR /app
 CMD ["python", "./changedetection.py", "-d", "/datastore"]

MANIFEST.in

@@ -1,8 +1,11 @@
 recursive-include changedetectionio/api *
 recursive-include changedetectionio/blueprint *
 recursive-include changedetectionio/content_fetchers *
+recursive-include changedetectionio/conditions *
 recursive-include changedetectionio/model *
+recursive-include changedetectionio/notification *
 recursive-include changedetectionio/processors *
+recursive-include changedetectionio/realtime *
 recursive-include changedetectionio/static *
 recursive-include changedetectionio/templates *
 recursive-include changedetectionio/tests *

README.md

@@ -41,6 +41,20 @@ Using the **Browser Steps** configuration, add basic steps before performing cha
 After **Browser Steps** have been run, then visit the **Visual Selector** tab to refine the content you're interested in.
 Requires Playwright to be enabled.
 
+### Awesome restock and price change notifications
+
+Enable the _"Re-stock & Price detection for single product pages"_ option to activate the best way to monitor product pricing, this will extract any meta-data in the HTML page and give you many options to follow the pricing of the product.
+
+Easily organise and monitor prices for products from the dashboard, get alerts and notifications when the price of a product changes or comes back in stock again!
+
+[<img src="docs/restock-overview.png" style="max-width:100%;" alt="Easily keep an eye on product price changes directly from the UI"  title="Easily keep an eye on product price changes directly from the UI" />](https://changedetection.io?src=github)
+
+Set price change notification parameters, upper and lower price, price change percentage and more.
+Always know when a product for sale drops in price.
+
+[<img src="docs/restock-settings.png" style="max-width:100%;" alt="Set upper lower and percentage price change notification values"  title="Set upper lower and percentage price change notification values" />](https://changedetection.io?src=github)
+
+
 
 ### Example use cases
 
@@ -75,7 +89,7 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
 #### Key Features
 
 - Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
-- Target elements with xPath(1.0) and CSS Selectors, Easily monitor complex JSON with JSONPath or jq
+- Target elements with xPath 1 and xPath 2, CSS Selectors, Easily monitor complex JSON with JSONPath or jq
 - Switch between fast non-JS and Chrome JS based "fetchers"
 - Track changes in PDF files (Monitor text changed in the PDF, Also monitor PDF filesize and checksums)
 - Easily specify how often a site should be checked
@@ -91,13 +105,28 @@ We [recommend and use Bright Data](https://brightdata.grsm.io/n0r16zf7eivq) glob
 
 Please :star: star :star: this project and help it grow! https://github.com/dgtlmoon/changedetection.io/
 
+### Conditional web page changes
+
+Easily [configure conditional actions](https://changedetection.io/tutorial/conditional-actions-web-page-changes), for example, only trigger when a price is above or below a preset amount, or [when a web page includes (or does not include) a keyword](https://changedetection.io/tutorial/how-monitor-keywords-any-website)
+
+<img src="./docs/web-page-change-conditions.png" style="max-width:80%;" alt="Conditional web page changes"  title="Conditional web page changes"  />
+
+### Schedule web page watches in any timezone, limit by day of week and time.
+
+Easily set a re-check schedule, for example you could limit the web page change detection to only operate during business hours.
+Or perhaps based on a foreign timezone (for example, you want to check for the latest news-headlines in a foreign country at 0900 AM),
+
+<img src="./docs/scheduler.png" style="max-width:80%;" alt="How to monitor web page changes according to a schedule"  title="How to monitor web page changes according to a schedule"  />
+
+Includes quick short-cut buttons to setup a schedule for **business hours only**, or **weekends**.
+
 ### We have a Chrome extension!
 
 Easily add the current web page to your changedetection.io tool, simply install the extension and click "Sync" to connect it to your existing changedetection.io install.
 
 [<img src="./docs/chrome-extension-screenshot.png" style="max-width:80%;" alt="Chrome Extension to easily add the current web-page to detect a change."  title="Chrome Extension to easily add the current web-page to detect a change."  />](https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop)
 
-[Goto the Chrome Webstore to download the extension.](https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop)
+[Goto the Chrome Webstore to download the extension.](https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop) ( Or check out the [GitHub repo](https://github.com/dgtlmoon/changedetection.io-browser-extension) ) 
 
 ## Installation
 
@@ -272,6 +301,10 @@ I offer commercial support, this software is depended on by network security, ae
 [release-link]: https://github.com/dgtlmoon/changedetection.io/releases
 [docker-link]: https://hub.docker.com/r/dgtlmoon/changedetection.io
 
+## Commercial Licencing
+
+If you are reselling this software either in part or full as part of any commercial arrangement, you must abide by our COMMERCIAL_LICENCE.md found in our code repository, please contact dgtlmoon@gmail.com and contact@changedetection.io .
+
 ## Third-party licenses
 
 changedetectionio.html_tools.elementpath_tostring: Copyright (c), 2018-2021, SISSA (Scuola Internazionale Superiore di Studi Avanzati), Licensed under [MIT license](https://github.com/sissaschool/elementpath/blob/master/LICENSE)

changedetection.py

@@ -1,6 +1,8 @@
-#!/usr/bin/python3
+#!/usr/bin/env python3
 
 # Only exists for direct CLI usage
 
 import changedetectionio
-changedetectionio.main()
+
+if __name__ == '__main__':
+    changedetectionio.main()

changedetectionio/PLUGIN_README.md

@@ -0,0 +1,98 @@
+# Creating Plugins for changedetection.io
+
+This document describes how to create plugins for changedetection.io. Plugins can be used to extend the functionality of the application in various ways.
+
+## Plugin Types
+
+### UI Stats Tab Plugins
+
+These plugins can add content to the Stats tab in the Edit page. This is useful for adding custom statistics or visualizations about a watch.
+
+#### Creating a UI Stats Tab Plugin
+
+1. Create a Python file in a directory that will be loaded by the plugin system.
+
+2. Use the `global_hookimpl` decorator to implement the `ui_edit_stats_extras` hook:
+
+```python
+import pluggy
+from loguru import logger
+
+global_hookimpl = pluggy.HookimplMarker("changedetectionio")
+
+@global_hookimpl
+def ui_edit_stats_extras(watch):
+    """Add custom content to the stats tab"""
+    # Calculate or retrieve your stats
+    my_stat = calculate_something(watch)
+    
+    # Return HTML content as a string
+    html = f"""
+    <div class="my-plugin-stats">
+        <h4>My Plugin Statistics</h4>
+        <p>My statistic: {my_stat}</p>
+    </div>
+    """
+    return html
+```
+
+3. The HTML you return will be included in the Stats tab.
+
+## Plugin Loading
+
+Plugins can be loaded from:
+
+1. Built-in plugin directories in the codebase
+2. External packages using setuptools entry points
+
+To add a new plugin directory, modify the `plugin_dirs` dictionary in `pluggy_interface.py`.
+
+## Example Plugin
+
+Here's a simple example of a plugin that adds a word count statistic to the Stats tab:
+
+```python
+import pluggy
+from loguru import logger
+
+global_hookimpl = pluggy.HookimplMarker("changedetectionio")
+
+def count_words_in_history(watch):
+    """Count words in the latest snapshot"""
+    try:
+        if not watch.history.keys():
+            return 0
+            
+        latest_key = list(watch.history.keys())[-1]
+        latest_content = watch.get_history_snapshot(latest_key)
+        return len(latest_content.split())
+    except Exception as e:
+        logger.error(f"Error counting words: {str(e)}")
+        return 0
+
+@global_hookimpl
+def ui_edit_stats_extras(watch):
+    """Add word count to the Stats tab"""
+    word_count = count_words_in_history(watch)
+    
+    html = f"""
+    <div class="word-count-stats">
+        <h4>Content Analysis</h4>
+        <table class="pure-table">
+            <tbody>
+                <tr>
+                    <td>Word count (latest snapshot)</td>
+                    <td>{word_count}</td>
+                </tr>
+            </tbody>
+        </table>
+    </div>
+    """
+    return html
+```
+
+## Testing Your Plugin
+
+1. Place your plugin in one of the directories scanned by the plugin system
+2. Restart changedetection.io
+3. Go to the Edit page of a watch and check the Stats tab to see your content

changedetectionio/__init__.py

@@ -1,41 +1,62 @@
-#!/usr/bin/python3
+#!/usr/bin/env python3
 
 # Read more https://github.com/dgtlmoon/changedetection.io/wiki
 
-__version__ = '0.45.26'
+__version__ = '0.50.3'
 
 from changedetectionio.strtobool import strtobool
 from json.decoder import JSONDecodeError
 import os
-os.environ['EVENTLET_NO_GREENDNS'] = 'yes'
-import eventlet
-import eventlet.wsgi
 import getopt
+import platform
 import signal
-import socket
+
 import sys
 
+# Eventlet completely removed - using threading mode for SocketIO
+# This provides better Python 3.12+ compatibility and eliminates eventlet/asyncio conflicts
 from changedetectionio import store
 from changedetectionio.flask_app import changedetection_app
 from loguru import logger
 
-
 # Only global so we can access it in the signal handler
 app = None
 datastore = None
 
+def get_version():
+    return __version__
+
 # Parent wrapper or OS sends us a SIGTERM/SIGINT, do everything required for a clean shutdown
 def sigshutdown_handler(_signo, _stack_frame):
-    global app
-    global datastore
     name = signal.Signals(_signo).name
-    logger.critical(f'Shutdown: Got Signal - {name} ({_signo}), Saving DB to disk and calling shutdown')
-    datastore.sync_to_json()
-    logger.success('Sync JSON to disk complete.')
-    # This will throw a SystemExit exception, because eventlet.wsgi.server doesn't know how to deal with it.
-    # Solution: move to gevent or other server in the future (#2014)
-    datastore.stop_thread = True
+    logger.critical(f'Shutdown: Got Signal - {name} ({_signo}), Fast shutdown initiated')
+    
+    # Set exit flag immediately to stop all loops
     app.config.exit.set()
+    datastore.stop_thread = True
+    
+    # Shutdown workers immediately
+    try:
+        from changedetectionio import worker_handler
+        worker_handler.shutdown_workers()
+    except Exception as e:
+        logger.error(f"Error shutting down workers: {str(e)}")
+    
+    # Shutdown socketio server fast
+    from changedetectionio.flask_app import socketio_server
+    if socketio_server and hasattr(socketio_server, 'shutdown'):
+        try:
+            socketio_server.shutdown()
+        except Exception as e:
+            logger.error(f"Error shutting down Socket.IO server: {str(e)}")
+    
+    # Save data quickly
+    try:
+        datastore.sync_to_json()
+        logger.success('Fast sync to disk complete.')
+    except Exception as e:
+        logger.error(f"Error syncing to disk: {str(e)}")
+    
     sys.exit()
 
 def main():
@@ -44,9 +65,8 @@ def main():
 
     datastore_path = None
     do_cleanup = False
-    host = ''
-    ipv6_enabled = False
-    port = os.environ.get('PORT') or 5000
+    host = os.environ.get("LISTEN_HOST", "0.0.0.0").strip()
+    port = int(os.environ.get('PORT', 5000))
     ssl_mode = False
 
     # On Windows, create and use a default path.
@@ -87,10 +107,6 @@ def main():
         if opt == '-d':
             datastore_path = arg
 
-        if opt == '-6':
-            logger.success("Enabling IPv6 listen support")
-            ipv6_enabled = True
-
         # Cleanup (remove text files that arent in the index)
         if opt == '-c':
             do_cleanup = True
@@ -102,10 +118,24 @@ def main():
         if opt == '-l':
             logger_level = int(arg) if arg.isdigit() else arg.upper()
 
+
+    logger.success(f"changedetection.io version {get_version()} starting.")
+    # Launch using SocketIO run method for proper integration (if enabled)
+    ssl_cert_file = os.getenv("SSL_CERT_FILE", 'cert.pem')
+    ssl_privkey_file = os.getenv("SSL_PRIVKEY_FILE", 'privkey.pem')
+    if os.getenv("SSL_CERT_FILE") and os.getenv("SSL_PRIVKEY_FILE"):
+        ssl_mode = True
+
+    # SSL mode could have been set by -s too, therefor fallback to default values
+    if ssl_mode:
+        if not os.path.isfile(ssl_cert_file) or not os.path.isfile(ssl_privkey_file):
+            logger.critical(f"Cannot start SSL/HTTPS mode, Please be sure that {ssl_cert_file}' and '{ssl_privkey_file}' exist in in {os.getcwd()}")
+            os._exit(2)
+
     # Without this, a logger will be duplicated
     logger.remove()
     try:
-        log_level_for_stdout = { 'DEBUG', 'SUCCESS' }
+        log_level_for_stdout = { 'TRACE', 'DEBUG', 'INFO', 'SUCCESS' }
         logger.configure(handlers=[
             {"sink": sys.stdout, "level": logger_level,
              "filter" : lambda record: record['level'].name in log_level_for_stdout},
@@ -142,8 +172,26 @@ def main():
 
     app = changedetection_app(app_config, datastore)
 
+    # Get the SocketIO instance from the Flask app (created in flask_app.py)
+    from changedetectionio.flask_app import socketio_server
+    global socketio
+    socketio = socketio_server
+
     signal.signal(signal.SIGTERM, sigshutdown_handler)
     signal.signal(signal.SIGINT, sigshutdown_handler)
+    
+    # Custom signal handler for memory cleanup
+    def sigusr_clean_handler(_signo, _stack_frame):
+        from changedetectionio.gc_cleanup import memory_cleanup
+        logger.info('SIGUSR1 received: Running memory cleanup')
+        return memory_cleanup(app)
+
+    # Register the SIGUSR1 signal handler
+    # Only register the signal handler if running on Linux
+    if platform.system() == "Linux":
+        signal.signal(signal.SIGUSR1, sigusr_clean_handler)
+    else:
+        logger.info("SIGUSR1 handler only registered on Linux, skipped.")
 
     # Go into cleanup mode
     if do_cleanup:
@@ -153,18 +201,18 @@ def main():
 
 
     @app.context_processor
-    def inject_version():
+    def inject_template_globals():
         return dict(right_sticky="v{}".format(datastore.data['version_tag']),
                     new_version_available=app.config['NEW_VERSION_AVAILABLE'],
-                    has_password=datastore.data['settings']['application']['password'] != False
+                    has_password=datastore.data['settings']['application']['password'] != False,
+                    socket_io_enabled=datastore.data['settings']['application']['ui'].get('socket_io_enabled', True)
                     )
 
     # Monitored websites will not receive a Referer header when a user clicks on an outgoing link.
-    # @Note: Incompatible with password login (and maybe other features) for now, submit a PR!
     @app.after_request
     def hide_referrer(response):
         if strtobool(os.getenv("HIDE_REFERER", 'false')):
-            response.headers["Referrer-Policy"] = "no-referrer"
+            response.headers["Referrer-Policy"] = "same-origin"
 
         return response
 
@@ -181,15 +229,21 @@ def main():
         from werkzeug.middleware.proxy_fix import ProxyFix
         app.wsgi_app = ProxyFix(app.wsgi_app, x_prefix=1, x_host=1)
 
-    s_type = socket.AF_INET6 if ipv6_enabled else socket.AF_INET
-
-    if ssl_mode:
-        # @todo finalise SSL config, but this should get you in the right direction if you need it.
-        eventlet.wsgi.server(eventlet.wrap_ssl(eventlet.listen((host, port), s_type),
-                                               certfile='cert.pem',
-                                               keyfile='privkey.pem',
-                                               server_side=True), app)
 
+    # SocketIO instance is already initialized in flask_app.py
+    if socketio_server:
+        if ssl_mode:
+            logger.success(f"SSL mode enabled, attempting to start with '{ssl_cert_file}' and '{ssl_privkey_file}' in {os.getcwd()}")
+            socketio.run(app, host=host, port=int(port), debug=False,
+                         ssl_context=(ssl_cert_file, ssl_privkey_file), allow_unsafe_werkzeug=True)
+        else:
+            socketio.run(app, host=host, port=int(port), debug=False, allow_unsafe_werkzeug=True)
     else:
-        eventlet.wsgi.server(eventlet.listen((host, int(port)), s_type), app)
-
+        # Run Flask app without Socket.IO if disabled
+        logger.info("Starting Flask app without Socket.IO server")
+        if ssl_mode:
+            logger.success(f"SSL mode enabled, attempting to start with '{ssl_cert_file}' and '{ssl_privkey_file}' in {os.getcwd()}")
+            app.run(host=host, port=int(port), debug=False,
+                    ssl_context=(ssl_cert_file, ssl_privkey_file))
+        else:
+            app.run(host=host, port=int(port), debug=False)

changedetectionio/api/Import.py

@@ -0,0 +1,62 @@
+import os
+from changedetectionio.strtobool import strtobool
+from flask_restful import abort, Resource
+from flask import request
+import validators
+from . import auth
+
+
+class Import(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    @auth.check_token
+    def post(self):
+        """
+        @api {post} /api/v1/import Import a list of watched URLs
+        @apiDescription Accepts a line-feed separated list of URLs to import, additionally with ?tag_uuids=(tag  id), ?tag=(name), ?proxy={key}, ?dedupe=true (default true) one URL per line.
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/import --data-binary @list-of-sites.txt -H"x-api-key:8a111a21bc2f8f1dd9b9353bbd46049a"
+        @apiName Import
+        @apiGroup Watch
+        @apiSuccess (200) {List} OK List of watch UUIDs added
+        @apiSuccess (500) {String} ERR Some other error
+        """
+
+        extras = {}
+
+        if request.args.get('proxy'):
+            plist = self.datastore.proxy_list
+            if not request.args.get('proxy') in plist:
+                return "Invalid proxy choice, currently supported proxies are '{}'".format(', '.join(plist)), 400
+            else:
+                extras['proxy'] = request.args.get('proxy')
+
+        dedupe = strtobool(request.args.get('dedupe', 'true'))
+
+        tags = request.args.get('tag')
+        tag_uuids = request.args.get('tag_uuids')
+
+        if tag_uuids:
+            tag_uuids = tag_uuids.split(',')
+
+        urls = request.get_data().decode('utf8').splitlines()
+        added = []
+        allow_simplehost = not strtobool(os.getenv('BLOCK_SIMPLEHOSTS', 'False'))
+        for url in urls:
+            url = url.strip()
+            if not len(url):
+                continue
+
+            # If hosts that only contain alphanumerics are allowed ("localhost" for example)
+            if not validators.url(url, simple_host=allow_simplehost):
+                return f"Invalid or unsupported URL - {url}", 400
+
+            if dedupe and self.datastore.url_exists(url):
+                continue
+
+            new_uuid = self.datastore.add_watch(url=url, extras=extras, tag=tags, tag_uuids=tag_uuids)
+            added.append(new_uuid)
+
+        return added

changedetectionio/api/Notifications.py

@@ -0,0 +1,145 @@
+from flask_expects_json import expects_json
+from flask_restful import Resource
+from . import auth
+from flask_restful import abort, Resource
+from flask import request
+from . import auth
+from . import schema_create_notification_urls, schema_delete_notification_urls
+
+class Notifications(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    @auth.check_token
+    def get(self):
+        """
+        @api {get} /api/v1/notifications Return Notification URL List
+        @apiDescription Return the Notification URL List from the configuration
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/notifications -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            HTTP/1.0 200
+            {
+                'notification_urls': ["notification-urls-list"]
+            }
+        @apiName Get
+        @apiGroup Notifications
+        """
+
+        notification_urls = self.datastore.data.get('settings', {}).get('application', {}).get('notification_urls', [])        
+
+        return {
+                'notification_urls': notification_urls,
+               }, 200
+    
+    @auth.check_token
+    @expects_json(schema_create_notification_urls)
+    def post(self):
+        """
+        @api {post} /api/v1/notifications Create Notification URLs
+        @apiDescription Add one or more notification URLs from the configuration
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/notifications/batch -H"x-api-key:813031b16330fe25e3780cf0325daa45" -H "Content-Type: application/json" -d '{"notification_urls": ["url1", "url2"]}'
+        @apiName CreateBatch
+        @apiGroup Notifications
+        @apiSuccess (201) {Object[]} notification_urls List of added notification URLs
+        @apiError (400) {String} Invalid input
+        """
+
+        json_data = request.get_json()
+        notification_urls = json_data.get("notification_urls", [])
+
+        from wtforms import ValidationError
+        try:
+            validate_notification_urls(notification_urls)
+        except ValidationError as e:
+            return str(e), 400
+
+        added_urls = []
+
+        for url in notification_urls:
+            clean_url = url.strip()
+            added_url = self.datastore.add_notification_url(clean_url)
+            if added_url:
+                added_urls.append(added_url)
+
+        if not added_urls:
+            return "No valid notification URLs were added", 400
+
+        return {'notification_urls': added_urls}, 201
+    
+    @auth.check_token
+    @expects_json(schema_create_notification_urls)
+    def put(self):
+        """
+        @api {put} /api/v1/notifications Replace Notification URLs
+        @apiDescription Replace all notification URLs with the provided list (can be empty)
+        @apiExample {curl} Example usage:
+            curl -X PUT http://localhost:5000/api/v1/notifications -H"x-api-key:813031b16330fe25e3780cf0325daa45" -H "Content-Type: application/json" -d '{"notification_urls": ["url1", "url2"]}'
+        @apiName Replace
+        @apiGroup Notifications
+        @apiSuccess (200) {Object[]} notification_urls List of current notification URLs
+        @apiError (400) {String} Invalid input
+        """
+        json_data = request.get_json()
+        notification_urls = json_data.get("notification_urls", [])
+
+        from wtforms import ValidationError
+        try:
+            validate_notification_urls(notification_urls)
+        except ValidationError as e:
+            return str(e), 400
+        
+        if not isinstance(notification_urls, list):
+            return "Invalid input format", 400
+
+        clean_urls = [url.strip() for url in notification_urls if isinstance(url, str)]
+        self.datastore.data['settings']['application']['notification_urls'] = clean_urls
+        self.datastore.needs_write = True
+
+        return {'notification_urls': clean_urls}, 200
+        
+    @auth.check_token
+    @expects_json(schema_delete_notification_urls)
+    def delete(self):
+        """
+        @api {delete} /api/v1/notifications Delete Notification URLs
+        @apiDescription Deletes one or more notification URLs from the configuration
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/notifications -X DELETE -H"x-api-key:813031b16330fe25e3780cf0325daa45" -H "Content-Type: application/json" -d '{"notification_urls": ["url1", "url2"]}'
+        @apiParam {String[]} notification_urls The notification URLs to delete.
+        @apiName Delete
+        @apiGroup Notifications
+        @apiSuccess (204) {String} OK Deleted
+        @apiError (400) {String} No matching notification URLs found.
+        """
+
+        json_data = request.get_json()
+        urls_to_delete = json_data.get("notification_urls", [])
+        if not isinstance(urls_to_delete, list):
+            abort(400, message="Expected a list of notification URLs.")
+
+        notification_urls = self.datastore.data['settings']['application'].get('notification_urls', [])
+        deleted = []
+
+        for url in urls_to_delete:
+            clean_url = url.strip()
+            if clean_url in notification_urls:
+                notification_urls.remove(clean_url)
+                deleted.append(clean_url)
+
+        if not deleted:
+            abort(400, message="No matching notification URLs found.")
+
+        self.datastore.data['settings']['application']['notification_urls'] = notification_urls
+        self.datastore.needs_write = True
+
+        return 'OK', 204
+    
+def validate_notification_urls(notification_urls):
+    from changedetectionio.forms import ValidateAppRiseServers
+    validator = ValidateAppRiseServers()
+    class DummyForm: pass
+    dummy_form = DummyForm()
+    field = type("Field", (object,), {"data": notification_urls, "gettext": lambda self, x: x})()
+    validator(dummy_form, field)

changedetectionio/api/Search.py

@@ -0,0 +1,51 @@
+from flask_restful import Resource, abort
+from flask import request
+from . import auth
+
+class Search(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    @auth.check_token
+    def get(self):
+        """
+        @api {get} /api/v1/search Search for watches
+        @apiDescription Search watches by URL or title text
+        @apiExample {curl} Example usage:
+            curl "http://localhost:5000/api/v1/search?q=https://example.com/page1" -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            curl "http://localhost:5000/api/v1/search?q=https://example.com/page1?tag=Favourites" -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            curl "http://localhost:5000/api/v1/search?q=https://example.com?partial=true" -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+        @apiName Search
+        @apiGroup Watch Management
+        @apiQuery {String} q Search query to match against watch URLs and titles
+        @apiQuery {String} [tag] Optional name of tag to limit results (name not UUID)
+        @apiQuery {String} [partial] Allow partial matching of URL query
+        @apiSuccess (200) {Object} JSON Object containing matched watches
+        """
+        query = request.args.get('q', '').strip()
+        tag_limit = request.args.get('tag', '').strip()
+        from changedetectionio.strtobool import strtobool
+        partial = bool(strtobool(request.args.get('partial', '0'))) if 'partial' in request.args else False
+
+        # Require a search query
+        if not query:
+            abort(400, message="Search query 'q' parameter is required")
+
+        # Use the search function from the datastore
+        matching_uuids = self.datastore.search_watches_for_url(query=query, tag_limit=tag_limit, partial=partial)
+
+        # Build the response with watch details
+        results = {}
+        for uuid in matching_uuids:
+            watch = self.datastore.data['watching'].get(uuid)
+            results[uuid] = {
+                'last_changed': watch.last_changed,
+                'last_checked': watch['last_checked'],
+                'last_error': watch['last_error'],
+                'title': watch['title'],
+                'url': watch['url'],
+                'viewed': watch.viewed
+            }
+
+        return results, 200

changedetectionio/api/SystemInfo.py

@@ -0,0 +1,54 @@
+from flask_restful import Resource
+from . import auth
+
+
+class SystemInfo(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+        self.update_q = kwargs['update_q']
+
+    @auth.check_token
+    def get(self):
+        """
+        @api {get} /api/v1/systeminfo Return system info
+        @apiDescription Return some info about the current system state
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/systeminfo -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            HTTP/1.0 200
+            {
+                'queue_size': 10 ,
+                'overdue_watches': ["watch-uuid-list"],
+                'uptime': 38344.55,
+                'watch_count': 800,
+                'version': "0.40.1"
+            }
+        @apiName Get Info
+        @apiGroup System Information
+        """
+        import time
+        overdue_watches = []
+
+        # Check all watches and report which have not been checked but should have been
+
+        for uuid, watch in self.datastore.data.get('watching', {}).items():
+            # see if now - last_checked is greater than the time that should have been
+            # this is not super accurate (maybe they just edited it) but better than nothing
+            t = watch.threshold_seconds()
+            if not t:
+                # Use the system wide default
+                t = self.datastore.threshold_seconds
+
+            time_since_check = time.time() - watch.get('last_checked')
+
+            # Allow 5 minutes of grace time before we decide it's overdue
+            if time_since_check - (5 * 60) > t:
+                overdue_watches.append(uuid)
+        from changedetectionio import __version__ as main_version
+        return {
+                   'queue_size': self.update_q.qsize(),
+                   'overdue_watches': overdue_watches,
+                   'uptime': round(time.time() - self.datastore.start_time, 2),
+                   'watch_count': len(self.datastore.data.get('watching', {})),
+                   'version': main_version
+               }, 200

changedetectionio/api/Tags.py

@@ -0,0 +1,156 @@
+from flask_expects_json import expects_json
+from flask_restful import abort, Resource
+from flask import request
+from . import auth
+
+# Import schemas from __init__.py
+from . import schema_tag, schema_create_tag, schema_update_tag
+
+
+class Tag(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    # Get information about a single tag
+    # curl http://localhost:5000/api/v1/tag/<string:uuid>
+    @auth.check_token
+    def get(self, uuid):
+        """
+        @api {get} /api/v1/tag/:uuid Single tag - get data or toggle notification muting.
+        @apiDescription Retrieve tag information and set notification_muted status
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/tag/cc0cfffa-f449-477b-83ea-0caafd1dc091 -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            curl "http://localhost:5000/api/v1/tag/cc0cfffa-f449-477b-83ea-0caafd1dc091?muted=muted" -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+        @apiName Tag
+        @apiGroup Tag
+        @apiParam {uuid} uuid Tag unique ID.
+        @apiQuery {String} [muted] =`muted` or =`unmuted` , Sets the MUTE NOTIFICATIONS state
+        @apiSuccess (200) {String} OK When muted operation OR full JSON object of the tag
+        @apiSuccess (200) {JSON} TagJSON JSON Full JSON object of the tag
+        """
+        from copy import deepcopy
+        tag = deepcopy(self.datastore.data['settings']['application']['tags'].get(uuid))
+        if not tag:
+            abort(404, message=f'No tag exists with the UUID of {uuid}')
+
+        if request.args.get('muted', '') == 'muted':
+            self.datastore.data['settings']['application']['tags'][uuid]['notification_muted'] = True
+            return "OK", 200
+        elif request.args.get('muted', '') == 'unmuted':
+            self.datastore.data['settings']['application']['tags'][uuid]['notification_muted'] = False
+            return "OK", 200
+
+        return tag
+
+    @auth.check_token
+    def delete(self, uuid):
+        """
+        @api {delete} /api/v1/tag/:uuid Delete a tag and remove it from all watches
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/tag/cc0cfffa-f449-477b-83ea-0caafd1dc091 -X DELETE -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+        @apiParam {uuid} uuid Tag unique ID.
+        @apiName DeleteTag
+        @apiGroup Tag
+        @apiSuccess (200) {String} OK Was deleted
+        """
+        if not self.datastore.data['settings']['application']['tags'].get(uuid):
+            abort(400, message='No tag exists with the UUID of {}'.format(uuid))
+
+        # Delete the tag, and any tag reference
+        del self.datastore.data['settings']['application']['tags'][uuid]
+        
+        # Remove tag from all watches
+        for watch_uuid, watch in self.datastore.data['watching'].items():
+            if watch.get('tags') and uuid in watch['tags']:
+                watch['tags'].remove(uuid)
+
+        return 'OK', 204
+
+    @auth.check_token
+    @expects_json(schema_update_tag)
+    def put(self, uuid):
+        """
+        @api {put} /api/v1/tag/:uuid Update tag information
+        @apiExample {curl} Example usage:
+            Update (PUT)
+            curl http://localhost:5000/api/v1/tag/cc0cfffa-f449-477b-83ea-0caafd1dc091 -X PUT -H"x-api-key:813031b16330fe25e3780cf0325daa45" -H "Content-Type: application/json" -d '{"title": "New Tag Title"}'
+
+        @apiDescription Updates an existing tag using JSON
+        @apiParam {uuid} uuid Tag unique ID.
+        @apiName UpdateTag
+        @apiGroup Tag
+        @apiSuccess (200) {String} OK Was updated
+        @apiSuccess (500) {String} ERR Some other error
+        """
+        tag = self.datastore.data['settings']['application']['tags'].get(uuid)
+        if not tag:
+            abort(404, message='No tag exists with the UUID of {}'.format(uuid))
+
+        tag.update(request.json)
+        self.datastore.needs_write_urgent = True
+
+        return "OK", 200
+
+
+    @auth.check_token
+    # Only cares for {'title': 'xxxx'}
+    def post(self):
+        """
+        @api {post} /api/v1/watch Create a single tag
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/watch -H"x-api-key:813031b16330fe25e3780cf0325daa45" -H "Content-Type: application/json" -d '{"name": "Work related"}'
+        @apiName Create
+        @apiGroup Tag
+        @apiSuccess (200) {String} OK Was created
+        @apiSuccess (500) {String} ERR Some other error
+        """
+
+        json_data = request.get_json()
+        title = json_data.get("title",'').strip()
+
+
+        new_uuid = self.datastore.add_tag(title=title)
+        if new_uuid:
+            return {'uuid': new_uuid}, 201
+        else:
+            return "Invalid or unsupported tag", 400
+
+class Tags(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    @auth.check_token
+    def get(self):
+        """
+        @api {get} /api/v1/tags List tags
+        @apiDescription Return list of available tags
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/tags -H"x-api-key:813031b16330fe25e3780cf0325daa45"
+            {
+                "cc0cfffa-f449-477b-83ea-0caafd1dc091": {
+                    "title": "Tech News",
+                    "notification_muted": false,
+                    "date_created": 1677103794
+                },
+                "e6f5fd5c-dbfe-468b-b8f3-f9d6ff5ad69b": {
+                    "title": "Shopping",
+                    "notification_muted": true,
+                    "date_created": 1676662819
+                }
+            }
+        @apiName ListTags
+        @apiGroup Tag Management
+        @apiSuccess (200) {String} OK JSON dict
+        """
+        result = {}
+        for uuid, tag in self.datastore.data['settings']['application']['tags'].items():
+            result[uuid] = {
+                'date_created': tag.get('date_created', 0),
+                'notification_muted': tag.get('notification_muted', False),
+                'title': tag.get('title', ''),
+                'uuid': tag.get('uuid')
+            }
+
+        return result, 200

changedetectionio/api/Watch.py

@@ -3,25 +3,16 @@ from changedetectionio.strtobool import strtobool
 
 from flask_expects_json import expects_json
 from changedetectionio import queuedWatchMetaData
+from changedetectionio import worker_handler
 from flask_restful import abort, Resource
 from flask import request, make_response
 import validators
 from . import auth
 import copy
 
-# See docs/README.md for rebuilding the docs/apidoc information
+# Import schemas from __init__.py
+from . import schema, schema_create_watch, schema_update_watch
 
-from . import api_schema
-
-# Build a JSON Schema atleast partially based on our Watch model
-from changedetectionio.model.Watch import base_config as watch_base_config
-schema = api_schema.build_watch_json_schema(watch_base_config)
-
-schema_create_watch = copy.deepcopy(schema)
-schema_create_watch['required'] = ['url']
-
-schema_update_watch = copy.deepcopy(schema)
-schema_update_watch['additionalProperties'] = False
 
 class Watch(Resource):
     def __init__(self, **kwargs):
@@ -57,7 +48,7 @@ class Watch(Resource):
             abort(404, message='No watch exists with the UUID of {}'.format(uuid))
 
         if request.args.get('recheck'):
-            self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
+            worker_handler.queue_item_async_safe(self.update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
             return "OK", 200
         if request.args.get('paused', '') == 'paused':
             self.datastore.data['watching'].get(uuid).pause()
@@ -75,6 +66,7 @@ class Watch(Resource):
         # Return without history, get that via another API call
         # Properties are not returned as a JSON, so add the required props manually
         watch['history_n'] = watch.history_n
+        # attr .last_changed will check for the last written text snapshot on change
         watch['last_changed'] = watch.last_changed
         watch['viewed'] = watch.viewed
         return watch
@@ -245,7 +237,7 @@ class CreateWatch(Resource):
 
         new_uuid = self.datastore.add_watch(url=url, extras=extras, tag=tags)
         if new_uuid:
-            self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid, 'skip_when_checksum_same': True}))
+            worker_handler.queue_item_async_safe(self.update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid}))
             return {'uuid': new_uuid}, 201
         else:
             return "Invalid or unsupported URL", 400
@@ -283,8 +275,6 @@ class CreateWatch(Resource):
         list = {}
 
         tag_limit = request.args.get('tag', '').lower()
-
-
         for uuid, watch in self.datastore.data['watching'].items():
             # Watch tags by name (replace the other calls?)
             tags = self.datastore.get_all_tags_for_watch(uuid=uuid)
@@ -302,113 +292,7 @@ class CreateWatch(Resource):
 
         if request.args.get('recheck_all'):
             for uuid in self.datastore.data['watching'].keys():
-                self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
+                worker_handler.queue_item_async_safe(self.update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
             return {'status': "OK"}, 200
 
-        return list, 200
-
-class Import(Resource):
-    def __init__(self, **kwargs):
-        # datastore is a black box dependency
-        self.datastore = kwargs['datastore']
-
-    @auth.check_token
-    def post(self):
-        """
-        @api {post} /api/v1/import Import a list of watched URLs
-        @apiDescription Accepts a line-feed separated list of URLs to import, additionally with ?tag_uuids=(tag  id), ?tag=(name), ?proxy={key}, ?dedupe=true (default true) one URL per line.
-        @apiExample {curl} Example usage:
-            curl http://localhost:5000/api/v1/import --data-binary @list-of-sites.txt -H"x-api-key:8a111a21bc2f8f1dd9b9353bbd46049a"
-        @apiName Import
-        @apiGroup Watch
-        @apiSuccess (200) {List} OK List of watch UUIDs added
-        @apiSuccess (500) {String} ERR Some other error
-        """
-
-        extras = {}
-
-        if request.args.get('proxy'):
-            plist = self.datastore.proxy_list
-            if not request.args.get('proxy') in plist:
-                return "Invalid proxy choice, currently supported proxies are '{}'".format(', '.join(plist)), 400
-            else:
-                extras['proxy'] = request.args.get('proxy')
-
-        dedupe = strtobool(request.args.get('dedupe', 'true'))
-
-        tags = request.args.get('tag')
-        tag_uuids = request.args.get('tag_uuids')
-
-        if tag_uuids:
-            tag_uuids = tag_uuids.split(',')
-
-        urls = request.get_data().decode('utf8').splitlines()
-        added = []
-        allow_simplehost = not strtobool(os.getenv('BLOCK_SIMPLEHOSTS', 'False'))
-        for url in urls:
-            url = url.strip()
-            if not len(url):
-                continue
-
-            # If hosts that only contain alphanumerics are allowed ("localhost" for example)
-            if not validators.url(url, simple_host=allow_simplehost):
-                return f"Invalid or unsupported URL - {url}", 400
-
-            if dedupe and self.datastore.url_exists(url):
-                continue
-
-            new_uuid = self.datastore.add_watch(url=url, extras=extras, tag=tags, tag_uuids=tag_uuids)
-            added.append(new_uuid)
-
-        return added
-
-class SystemInfo(Resource):
-    def __init__(self, **kwargs):
-        # datastore is a black box dependency
-        self.datastore = kwargs['datastore']
-        self.update_q = kwargs['update_q']
-
-    @auth.check_token
-    def get(self):
-        """
-        @api {get} /api/v1/systeminfo Return system info
-        @apiDescription Return some info about the current system state
-        @apiExample {curl} Example usage:
-            curl http://localhost:5000/api/v1/systeminfo -H"x-api-key:813031b16330fe25e3780cf0325daa45"
-            HTTP/1.0 200
-            {
-                'queue_size': 10 ,
-                'overdue_watches': ["watch-uuid-list"],
-                'uptime': 38344.55,
-                'watch_count': 800,
-                'version': "0.40.1"
-            }
-        @apiName Get Info
-        @apiGroup System Information
-        """
-        import time
-        overdue_watches = []
-
-        # Check all watches and report which have not been checked but should have been
-
-        for uuid, watch in self.datastore.data.get('watching', {}).items():
-            # see if now - last_checked is greater than the time that should have been
-            # this is not super accurate (maybe they just edited it) but better than nothing
-            t = watch.threshold_seconds()
-            if not t:
-                # Use the system wide default
-                t = self.datastore.threshold_seconds
-
-            time_since_check = time.time() - watch.get('last_checked')
-
-            # Allow 5 minutes of grace time before we decide it's overdue
-            if time_since_check - (5 * 60) > t:
-                overdue_watches.append(uuid)
-        from changedetectionio import __version__ as main_version
-        return {
-                   'queue_size': self.update_q.qsize(),
-                   'overdue_watches': overdue_watches,
-                   'uptime': round(time.time() - self.datastore.start_time, 2),
-                   'watch_count': len(self.datastore.data.get('watching', {})),
-                   'version': main_version
-               }, 200
+        return list, 200

changedetectionio/api/__init__.py

@@ -0,0 +1,33 @@
+import copy
+from . import api_schema
+from ..model import watch_base
+
+# Build a JSON Schema atleast partially based on our Watch model
+watch_base_config = watch_base()
+schema = api_schema.build_watch_json_schema(watch_base_config)
+
+schema_create_watch = copy.deepcopy(schema)
+schema_create_watch['required'] = ['url']
+
+schema_update_watch = copy.deepcopy(schema)
+schema_update_watch['additionalProperties'] = False
+
+# Tag schema is also based on watch_base since Tag inherits from it
+schema_tag = copy.deepcopy(schema)
+schema_create_tag = copy.deepcopy(schema_tag)
+schema_create_tag['required'] = ['title']
+schema_update_tag = copy.deepcopy(schema_tag)
+schema_update_tag['additionalProperties'] = False
+
+schema_notification_urls = copy.deepcopy(schema)
+schema_create_notification_urls = copy.deepcopy(schema_notification_urls)
+schema_create_notification_urls['required'] = ['notification_urls']
+schema_delete_notification_urls = copy.deepcopy(schema_notification_urls)
+schema_delete_notification_urls['required'] = ['notification_urls']
+
+# Import all API resources
+from .Watch import Watch, WatchHistory, WatchSingleHistory, CreateWatch
+from .Tags import Tags, Tag
+from .Import import Import
+from .SystemInfo import SystemInfo
+from .Notifications import Notifications

changedetectionio/api/api_schema.py

@@ -1,5 +1,7 @@
 # Responsible for building the storage dict into a set of rules ("JSON Schema") acceptable via the API
 # Probably other ways to solve this when the backend switches to some ORM
+from changedetectionio.notification import valid_notification_formats
+
 
 def build_time_between_check_json_schema():
     # Setup time between check schema
@@ -98,8 +100,6 @@ def build_watch_json_schema(d):
         }
     }
 
-    from changedetectionio.notification import valid_notification_formats
-
     schema['properties']['notification_format'] = {'type': 'string',
                                                    'enum': list(valid_notification_formats.keys())
                                                    }
@@ -112,6 +112,35 @@ def build_watch_json_schema(d):
 
     schema['properties']['time_between_check'] = build_time_between_check_json_schema()
 
+    schema['properties']['browser_steps'] = {
+        "anyOf": [
+            {
+                "type": "array",
+                "items": {
+                    "type": "object",
+                    "properties": {
+                        "operation": {
+                            "type": ["string", "null"],
+                            "maxLength": 5000  # Allows null and any string up to 5000 chars (including "")
+                        },
+                        "selector": {
+                            "type": ["string", "null"],
+                            "maxLength": 5000
+                        },
+                        "optional_value": {
+                            "type": ["string", "null"],
+                            "maxLength": 5000
+                        }
+                    },
+                    "required": ["operation", "selector", "optional_value"],
+                    "additionalProperties": False  # No extra keys allowed
+                }
+            },
+            {"type": "null"},  # Allows null for `browser_steps`
+            {"type": "array", "maxItems": 0}  # Allows empty array []
+        ]
+    }
+
     # headers ?
     return schema
 

changedetectionio/api/auth.py

@@ -11,22 +11,14 @@ def check_token(f):
         datastore = args[0].datastore
 
         config_api_token_enabled = datastore.data['settings']['application'].get('api_access_token_enabled')
-        if not config_api_token_enabled:
-            return
-
-        try:
-            api_key_header = request.headers['x-api-key']
-        except KeyError:
-            return make_response(
-                jsonify("No authorization x-api-key header."), 403
-            )
-
         config_api_token = datastore.data['settings']['application'].get('api_access_token')
 
-        if api_key_header != config_api_token:
-            return make_response(
-                jsonify("Invalid access - API key invalid."), 403
-            )
+        # config_api_token_enabled - a UI option in settings if access should obey the key or not
+        if config_api_token_enabled:
+            if request.headers.get('x-api-key') != config_api_token:
+                return make_response(
+                    jsonify("Invalid access - API key invalid."), 403
+                )
 
         return f(*args, **kwargs)
 

changedetectionio/apprise_asset.py

@@ -1,11 +0,0 @@
-import apprise
-
-# Create our AppriseAsset and populate it with some of our new values:
-# https://github.com/caronc/apprise/wiki/Development_API#the-apprise-asset-object
-asset = apprise.AppriseAsset(
-   image_url_logo='https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/changedetectionio/static/images/avatar-256x256.png'
-)
-
-asset.app_id = "changedetection.io"
-asset.app_desc = "ChangeDetection.io best and simplest website monitoring and change detection"
-asset.app_url = "https://changedetection.io"

changedetectionio/async_update_worker.py

@@ -0,0 +1,449 @@
+from .processors.exceptions import ProcessorException
+import changedetectionio.content_fetchers.exceptions as content_fetchers_exceptions
+from changedetectionio.processors.text_json_diff.processor import FilterNotFoundInResponse
+from changedetectionio import html_tools
+from changedetectionio.flask_app import watch_check_update
+
+import asyncio
+import importlib
+import os
+import time
+
+from loguru import logger
+
+# Async version of update_worker
+# Processes jobs from AsyncSignalPriorityQueue instead of threaded queue
+
+async def async_update_worker(worker_id, q, notification_q, app, datastore):
+    """
+    Async worker function that processes watch check jobs from the queue.
+    
+    Args:
+        worker_id: Unique identifier for this worker
+        q: AsyncSignalPriorityQueue containing jobs to process
+        notification_q: Standard queue for notifications
+        app: Flask application instance
+        datastore: Application datastore
+    """
+    # Set a descriptive name for this task
+    task = asyncio.current_task()
+    if task:
+        task.set_name(f"async-worker-{worker_id}")
+    
+    logger.info(f"Starting async worker {worker_id}")
+    
+    while not app.config.exit.is_set():
+        update_handler = None
+        watch = None
+
+        try:
+            # Use asyncio wait_for to make queue.get() cancellable
+            queued_item_data = await asyncio.wait_for(q.get(), timeout=1.0)
+        except asyncio.TimeoutError:
+            # No jobs available, continue loop
+            continue
+        except Exception as e:
+            logger.error(f"Worker {worker_id} error getting queue item: {e}")
+            await asyncio.sleep(0.1)
+            continue
+        
+        uuid = queued_item_data.item.get('uuid')
+        fetch_start_time = round(time.time())
+        
+        # Mark this UUID as being processed
+        from changedetectionio import worker_handler
+        worker_handler.set_uuid_processing(uuid, processing=True)
+        
+        try:
+            if uuid in list(datastore.data['watching'].keys()) and datastore.data['watching'][uuid].get('url'):
+                changed_detected = False
+                contents = b''
+                process_changedetection_results = True
+                update_obj = {}
+
+                # Clear last errors
+                datastore.data['watching'][uuid]['browser_steps_last_error_step'] = None
+                datastore.data['watching'][uuid]['last_checked'] = fetch_start_time
+
+                watch = datastore.data['watching'].get(uuid)
+
+                logger.info(f"Worker {worker_id} processing watch UUID {uuid} Priority {queued_item_data.priority} URL {watch['url']}")
+
+                try:
+                    watch_check_update.send(watch_uuid=uuid)
+
+                    # Processor is what we are using for detecting the "Change"
+                    processor = watch.get('processor', 'text_json_diff')
+
+                    # Init a new 'difference_detection_processor'
+                    processor_module_name = f"changedetectionio.processors.{processor}.processor"
+                    try:
+                        processor_module = importlib.import_module(processor_module_name)
+                    except ModuleNotFoundError as e:
+                        print(f"Processor module '{processor}' not found.")
+                        raise e
+
+                    update_handler = processor_module.perform_site_check(datastore=datastore,
+                                                                         watch_uuid=uuid)
+
+                    # All fetchers are now async, so call directly
+                    await update_handler.call_browser()
+
+                    # Run change detection (this is synchronous)
+                    changed_detected, update_obj, contents = update_handler.run_changedetection(watch=watch)
+
+                except PermissionError as e:
+                    logger.critical(f"File permission error updating file, watch: {uuid}")
+                    logger.critical(str(e))
+                    process_changedetection_results = False
+
+                except ProcessorException as e:
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot)
+                    if e.xpath_data:
+                        watch.save_xpath_data(data=e.xpath_data)
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': e.message})
+                    process_changedetection_results = False
+
+                except content_fetchers_exceptions.ReplyWithContentButNoText as e:
+                    extra_help = ""
+                    if e.has_filters:
+                        has_img = html_tools.include_filters(include_filters='img',
+                                                             html_content=e.html_content)
+                        if has_img:
+                            extra_help = ", it's possible that the filters you have give an empty result or contain only an image."
+                        else:
+                            extra_help = ", it's possible that the filters were found, but contained no usable text."
+
+                    datastore.update_watch(uuid=uuid, update_obj={
+                        'last_error': f"Got HTML content but no text found (With {e.status_code} reply code){extra_help}"
+                    })
+
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot, as_error=True)
+
+                    if e.xpath_data:
+                        watch.save_xpath_data(data=e.xpath_data)
+                        
+                    process_changedetection_results = False
+
+                except content_fetchers_exceptions.Non200ErrorCodeReceived as e:
+                    if e.status_code == 403:
+                        err_text = "Error - 403 (Access denied) received"
+                    elif e.status_code == 404:
+                        err_text = "Error - 404 (Page not found) received"
+                    elif e.status_code == 407:
+                        err_text = "Error - 407 (Proxy authentication required) received, did you need a username and password for the proxy?"
+                    elif e.status_code == 500:
+                        err_text = "Error - 500 (Internal server error) received from the web site"
+                    else:
+                        extra = ' (Access denied or blocked)' if str(e.status_code).startswith('4') else ''
+                        err_text = f"Error - Request returned a HTTP error code {e.status_code}{extra}"
+
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot, as_error=True)
+                    if e.xpath_data:
+                        watch.save_xpath_data(data=e.xpath_data, as_error=True)
+                    if e.page_text:
+                        watch.save_error_text(contents=e.page_text)
+
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text})
+                    process_changedetection_results = False
+
+                except FilterNotFoundInResponse as e:
+                    if not datastore.data['watching'].get(uuid):
+                        continue
+
+                    err_text = "Warning, no filters were found, no change detection ran - Did the page change layout? update your Visual Filter if necessary."
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text})
+
+                    # Filter wasnt found, but we should still update the visual selector so that they can have a chance to set it up again
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot)
+
+                    if e.xpath_data:
+                        watch.save_xpath_data(data=e.xpath_data)
+
+                    # Only when enabled, send the notification
+                    if watch.get('filter_failure_notification_send', False):
+                        c = watch.get('consecutive_filter_failures', 0)
+                        c += 1
+                        # Send notification if we reached the threshold?
+                        threshold = datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts', 0)
+                        logger.debug(f"Filter for {uuid} not found, consecutive_filter_failures: {c} of threshold {threshold}")
+                        if c >= threshold:
+                            if not watch.get('notification_muted'):
+                                logger.debug(f"Sending filter failed notification for {uuid}")
+                                await send_filter_failure_notification(uuid, notification_q, datastore)
+                            c = 0
+                            logger.debug(f"Reset filter failure count back to zero")
+
+                        datastore.update_watch(uuid=uuid, update_obj={'consecutive_filter_failures': c})
+                    else:
+                        logger.trace(f"{uuid} - filter_failure_notification_send not enabled, skipping")
+
+                    process_changedetection_results = False
+
+                except content_fetchers_exceptions.checksumFromPreviousCheckWasTheSame as e:
+                    # Yes fine, so nothing todo, don't continue to process.
+                    process_changedetection_results = False
+                    changed_detected = False
+                    
+                except content_fetchers_exceptions.BrowserConnectError as e:
+                    datastore.update_watch(uuid=uuid,
+                                         update_obj={'last_error': e.msg})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.BrowserFetchTimedOut as e:
+                    datastore.update_watch(uuid=uuid,
+                                         update_obj={'last_error': e.msg})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.BrowserStepsStepException as e:
+                    if not datastore.data['watching'].get(uuid):
+                        continue
+
+                    error_step = e.step_n + 1
+                    from playwright._impl._errors import TimeoutError, Error
+
+                    # Generally enough info for TimeoutError (couldnt locate the element after default seconds)
+                    err_text = f"Browser step at position {error_step} could not run, check the watch, add a delay if necessary, view Browser Steps to see screenshot at that step."
+
+                    if e.original_e.name == "TimeoutError":
+                        # Just the first line is enough, the rest is the stack trace
+                        err_text += " Could not find the target."
+                    else:
+                        # Other Error, more info is good.
+                        err_text += " " + str(e.original_e).splitlines()[0]
+
+                    logger.debug(f"BrowserSteps exception at step {error_step} {str(e.original_e)}")
+
+                    datastore.update_watch(uuid=uuid,
+                                         update_obj={'last_error': err_text,
+                                                   'browser_steps_last_error_step': error_step})
+
+                    if watch.get('filter_failure_notification_send', False):
+                        c = watch.get('consecutive_filter_failures', 0)
+                        c += 1
+                        # Send notification if we reached the threshold?
+                        threshold = datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts', 0)
+                        logger.error(f"Step for {uuid} not found, consecutive_filter_failures: {c}")
+                        if threshold > 0 and c >= threshold:
+                            if not watch.get('notification_muted'):
+                                await send_step_failure_notification(watch_uuid=uuid, step_n=e.step_n, notification_q=notification_q, datastore=datastore)
+                            c = 0
+
+                        datastore.update_watch(uuid=uuid, update_obj={'consecutive_filter_failures': c})
+
+                    process_changedetection_results = False
+
+                except content_fetchers_exceptions.EmptyReply as e:
+                    # Some kind of custom to-str handler in the exception handler that does this?
+                    err_text = "EmptyReply - try increasing 'Wait seconds before extracting text', Status Code {}".format(e.status_code)
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text,
+                                                                'last_check_status': e.status_code})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.ScreenshotUnavailable as e:
+                    err_text = "Screenshot unavailable, page did not render fully in the expected time or page was too long - try increasing 'Wait seconds before extracting text'"
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text,
+                                                                'last_check_status': e.status_code})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.JSActionExceptions as e:
+                    err_text = "Error running JS Actions - Page request - "+e.message
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot, as_error=True)
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text,
+                                                                'last_check_status': e.status_code})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.PageUnloadable as e:
+                    err_text = "Page request from server didnt respond correctly"
+                    if e.message:
+                        err_text = "{} - {}".format(err_text, e.message)
+
+                    if e.screenshot:
+                        watch.save_screenshot(screenshot=e.screenshot, as_error=True)
+
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text,
+                                                                'last_check_status': e.status_code,
+                                                                'has_ldjson_price_data': None})
+                    process_changedetection_results = False
+                    
+                except content_fetchers_exceptions.BrowserStepsInUnsupportedFetcher as e:
+                    err_text = "This watch has Browser Steps configured and so it cannot run with the 'Basic fast Plaintext/HTTP Client', either remove the Browser Steps or select a Chrome fetcher."
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': err_text})
+                    process_changedetection_results = False
+                    logger.error(f"Exception (BrowserStepsInUnsupportedFetcher) reached processing watch UUID: {uuid}")
+
+                except Exception as e:
+                    logger.error(f"Worker {worker_id} exception processing watch UUID: {uuid}")
+                    logger.error(str(e))
+                    datastore.update_watch(uuid=uuid, update_obj={'last_error': "Exception: " + str(e)})
+                    process_changedetection_results = False
+
+                else:
+                    if not datastore.data['watching'].get(uuid):
+                        continue
+
+                    update_obj['content-type'] = update_handler.fetcher.get_all_headers().get('content-type', '').lower()
+
+                    if not watch.get('ignore_status_codes'):
+                        update_obj['consecutive_filter_failures'] = 0
+
+                    update_obj['last_error'] = False
+                    cleanup_error_artifacts(uuid, datastore)
+
+                if not datastore.data['watching'].get(uuid):
+                    continue
+
+                if process_changedetection_results:
+                    # Extract title if needed
+                    if datastore.data['settings']['application'].get('extract_title_as_title') or watch['extract_title_as_title']:
+                        if not watch['title'] or not len(watch['title']):
+                            try:
+                                update_obj['title'] = html_tools.extract_element(find='title', html_content=update_handler.fetcher.content)
+                                logger.info(f"UUID: {uuid} Extract <title> updated title to '{update_obj['title']}")
+                            except Exception as e:
+                                logger.warning(f"UUID: {uuid} Extract <title> as watch title was enabled, but couldn't find a <title>.")
+
+                    try:
+                        datastore.update_watch(uuid=uuid, update_obj=update_obj)
+
+                        if changed_detected or not watch.history_n:
+                            if update_handler.screenshot:
+                                watch.save_screenshot(screenshot=update_handler.screenshot)
+
+                            if update_handler.xpath_data:
+                                watch.save_xpath_data(data=update_handler.xpath_data)
+
+                            # Ensure unique timestamp for history
+                            if watch.newest_history_key and int(fetch_start_time) == int(watch.newest_history_key):
+                                logger.warning(f"Timestamp {fetch_start_time} already exists, waiting 1 seconds")
+                                fetch_start_time += 1
+                                await asyncio.sleep(1)
+
+                            watch.save_history_text(contents=contents,
+                                                    timestamp=int(fetch_start_time),
+                                                    snapshot_id=update_obj.get('previous_md5', 'none'))
+
+                            empty_pages_are_a_change = datastore.data['settings']['application'].get('empty_pages_are_a_change', False)
+                            if update_handler.fetcher.content or (not update_handler.fetcher.content and empty_pages_are_a_change):
+                                watch.save_last_fetched_html(contents=update_handler.fetcher.content, timestamp=int(fetch_start_time))
+
+                            # Send notifications on second+ check
+                            if watch.history_n >= 2:
+                                logger.info(f"Change detected in UUID {uuid} - {watch['url']}")
+                                if not watch.get('notification_muted'):
+                                    await send_content_changed_notification(uuid, notification_q, datastore)
+
+                    except Exception as e:
+                        logger.critical(f"Worker {worker_id} exception in process_changedetection_results")
+                        logger.critical(str(e))
+                        datastore.update_watch(uuid=uuid, update_obj={'last_error': str(e)})
+
+                # Always record attempt count
+                count = watch.get('check_count', 0) + 1
+
+                # Record server header
+                try:
+                    server_header = update_handler.fetcher.headers.get('server', '').strip().lower()[:255]
+                    datastore.update_watch(uuid=uuid, update_obj={'remote_server_reply': server_header})
+                except Exception as e:
+                    pass
+
+                datastore.update_watch(uuid=uuid, update_obj={'fetch_time': round(time.time() - fetch_start_time, 3),
+                                                               'check_count': count})
+
+        except Exception as e:
+            logger.error(f"Worker {worker_id} unexpected error processing {uuid}: {e}")
+            logger.error(f"Worker {worker_id} traceback:", exc_info=True)
+            
+            # Also update the watch with error information
+            if datastore and uuid in datastore.data['watching']:
+                datastore.update_watch(uuid=uuid, update_obj={'last_error': f"Worker error: {str(e)}"})
+        
+        finally:
+            # Always cleanup - this runs whether there was an exception or not
+            if uuid:
+                try:
+                    # Mark UUID as no longer being processed
+                    worker_handler.set_uuid_processing(uuid, processing=False)
+                    
+                    # Send completion signal
+                    if watch:
+                        #logger.info(f"Worker {worker_id} sending completion signal for UUID {watch['uuid']}")
+                        watch_check_update.send(watch_uuid=watch['uuid'])
+
+                    update_handler = None
+                    logger.debug(f"Worker {worker_id} completed watch {uuid} in {time.time()-fetch_start_time:.2f}s")
+                except Exception as cleanup_error:
+                    logger.error(f"Worker {worker_id} error during cleanup: {cleanup_error}")
+            
+            # Brief pause before continuing to avoid tight error loops (only on error)
+            if 'e' in locals():
+                await asyncio.sleep(1.0)
+            else:
+                # Small yield for normal completion
+                await asyncio.sleep(0.01)
+
+        # Check if we should exit
+        if app.config.exit.is_set():
+            break
+
+    # Check if we're in pytest environment - if so, be more gentle with logging
+    import sys
+    in_pytest = "pytest" in sys.modules or "PYTEST_CURRENT_TEST" in os.environ
+    
+    if not in_pytest:
+        logger.info(f"Worker {worker_id} shutting down")
+
+
+def cleanup_error_artifacts(uuid, datastore):
+    """Helper function to clean up error artifacts"""
+    cleanup_files = ["last-error-screenshot.png", "last-error.txt"]
+    for f in cleanup_files:
+        full_path = os.path.join(datastore.datastore_path, uuid, f)
+        if os.path.isfile(full_path):
+            os.unlink(full_path)
+
+
+
+async def send_content_changed_notification(watch_uuid, notification_q, datastore):
+    """Helper function to queue notifications using the new notification service"""
+    try:
+        from changedetectionio.notification_service import create_notification_service
+        
+        # Create notification service instance
+        notification_service = create_notification_service(datastore, notification_q)
+        
+        notification_service.send_content_changed_notification(watch_uuid)
+    except Exception as e:
+        logger.error(f"Error sending notification for {watch_uuid}: {e}")
+
+
+async def send_filter_failure_notification(watch_uuid, notification_q, datastore):
+    """Helper function to send filter failure notifications using the new notification service"""
+    try:
+        from changedetectionio.notification_service import create_notification_service
+        
+        # Create notification service instance
+        notification_service = create_notification_service(datastore, notification_q)
+        
+        notification_service.send_filter_failure_notification(watch_uuid)
+    except Exception as e:
+        logger.error(f"Error sending filter failure notification for {watch_uuid}: {e}")
+
+
+async def send_step_failure_notification(watch_uuid, step_n, notification_q, datastore):
+    """Helper function to send step failure notifications using the new notification service"""
+    try:
+        from changedetectionio.notification_service import create_notification_service
+        
+        # Create notification service instance
+        notification_service = create_notification_service(datastore, notification_q)
+        
+        notification_service.send_step_failure_notification(watch_uuid, step_n)
+    except Exception as e:
+        logger.error(f"Error sending step failure notification for {watch_uuid}: {e}")

changedetectionio/auth_decorator.py

@@ -0,0 +1,33 @@
+import os
+from functools import wraps
+from flask import current_app, redirect, request
+from loguru import logger
+
+def login_optionally_required(func):
+    """
+    If password authentication is enabled, verify the user is logged in.
+    To be used as a decorator for routes that should optionally require login.
+    This version is blueprint-friendly as it uses current_app instead of directly accessing app.
+    """
+    @wraps(func)
+    def decorated_view(*args, **kwargs):
+        from flask import current_app
+        import flask_login
+        from flask_login import current_user
+
+        # Access datastore through the app config
+        datastore = current_app.config['DATASTORE']
+        has_password_enabled = datastore.data['settings']['application'].get('password') or os.getenv("SALTED_PASS", False)
+
+        # Permitted
+        if request.endpoint and 'diff_history_page' in request.endpoint and datastore.data['settings']['application'].get('shared_diff_access'):
+            return func(*args, **kwargs)
+        elif request.method in flask_login.config.EXEMPT_METHODS:
+            return func(*args, **kwargs)
+        elif current_app.config.get('LOGIN_DISABLED'):
+            return func(*args, **kwargs)
+        elif has_password_enabled and not current_user.is_authenticated:
+            return current_app.login_manager.unauthorized()
+
+        return func(*args, **kwargs)
+    return decorated_view

changedetectionio/blueprint/backups/__init__.py

@@ -0,0 +1,164 @@
+import datetime
+import glob
+import threading
+
+from flask import Blueprint, render_template, send_from_directory, flash, url_for, redirect, abort
+import os
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.flask_app import login_optionally_required
+from loguru import logger
+
+BACKUP_FILENAME_FORMAT = "changedetection-backup-{}.zip"
+
+
+def create_backup(datastore_path, watches: dict):
+    logger.debug("Creating backup...")
+    import zipfile
+    from pathlib import Path
+
+    # create a ZipFile object
+    timestamp = datetime.datetime.now().strftime("%Y%m%d%H%M%S")
+    backupname = BACKUP_FILENAME_FORMAT.format(timestamp)
+    backup_filepath = os.path.join(datastore_path, backupname)
+
+    with zipfile.ZipFile(backup_filepath.replace('.zip', '.tmp'), "w",
+                         compression=zipfile.ZIP_DEFLATED,
+                         compresslevel=8) as zipObj:
+
+        # Add the index
+        zipObj.write(os.path.join(datastore_path, "url-watches.json"), arcname="url-watches.json")
+
+        # Add the flask app secret
+        zipObj.write(os.path.join(datastore_path, "secret.txt"), arcname="secret.txt")
+
+        # Add any data in the watch data directory.
+        for uuid, w in watches.items():
+            for f in Path(w.watch_data_dir).glob('*'):
+                zipObj.write(f,
+                             # Use the full path to access the file, but make the file 'relative' in the Zip.
+                             arcname=os.path.join(f.parts[-2], f.parts[-1]),
+                             compress_type=zipfile.ZIP_DEFLATED,
+                             compresslevel=8)
+
+        # Create a list file with just the URLs, so it's easier to port somewhere else in the future
+        list_file = "url-list.txt"
+        with open(os.path.join(datastore_path, list_file), "w") as f:
+            for uuid in watches:
+                url = watches[uuid]["url"]
+                f.write("{}\r\n".format(url))
+        list_with_tags_file = "url-list-with-tags.txt"
+        with open(
+                os.path.join(datastore_path, list_with_tags_file), "w"
+        ) as f:
+            for uuid in watches:
+                url = watches[uuid].get('url')
+                tag = watches[uuid].get('tags', {})
+                f.write("{} {}\r\n".format(url, tag))
+
+        # Add it to the Zip
+        zipObj.write(
+            os.path.join(datastore_path, list_file),
+            arcname=list_file,
+            compress_type=zipfile.ZIP_DEFLATED,
+            compresslevel=8,
+        )
+        zipObj.write(
+            os.path.join(datastore_path, list_with_tags_file),
+            arcname=list_with_tags_file,
+            compress_type=zipfile.ZIP_DEFLATED,
+            compresslevel=8,
+        )
+
+    # Now it's done, rename it so it shows up finally and its completed being written.
+    os.rename(backup_filepath.replace('.zip', '.tmp'), backup_filepath.replace('.tmp', '.zip'))
+
+
+def construct_blueprint(datastore: ChangeDetectionStore):
+    backups_blueprint = Blueprint('backups', __name__, template_folder="templates")
+    backup_threads = []
+
+    @login_optionally_required
+    @backups_blueprint.route("/request-backup", methods=['GET'])
+    def request_backup():
+        if any(thread.is_alive() for thread in backup_threads):
+            flash("A backup is already running, check back in a few minutes", "error")
+            return redirect(url_for('backups.index'))
+
+        if len(find_backups()) > int(os.getenv("MAX_NUMBER_BACKUPS", 100)):
+            flash("Maximum number of backups reached, please remove some", "error")
+            return redirect(url_for('backups.index'))
+
+        # Be sure we're written fresh
+        datastore.sync_to_json()
+        zip_thread = threading.Thread(target=create_backup, args=(datastore.datastore_path, datastore.data.get("watching")))
+        zip_thread.start()
+        backup_threads.append(zip_thread)
+        flash("Backup building in background, check back in a few minutes.")
+
+        return redirect(url_for('backups.index'))
+
+    def find_backups():
+        backup_filepath = os.path.join(datastore.datastore_path, BACKUP_FILENAME_FORMAT.format("*"))
+        backups = glob.glob(backup_filepath)
+        backup_info = []
+
+        for backup in backups:
+            size = os.path.getsize(backup) / (1024 * 1024)
+            creation_time = os.path.getctime(backup)
+            backup_info.append({
+                'filename': os.path.basename(backup),
+                'filesize': f"{size:.2f}",
+                'creation_time': creation_time
+            })
+
+        backup_info.sort(key=lambda x: x['creation_time'], reverse=True)
+
+        return backup_info
+
+    @login_optionally_required
+    @backups_blueprint.route("/download/<string:filename>", methods=['GET'])
+    def download_backup(filename):
+        import re
+        filename = filename.strip()
+        backup_filename_regex = BACKUP_FILENAME_FORMAT.format("\d+")
+
+        full_path = os.path.join(os.path.abspath(datastore.datastore_path), filename)
+        if not full_path.startswith(os.path.abspath(datastore.datastore_path)):
+            abort(404)
+
+        if filename == 'latest':
+            backups = find_backups()
+            filename = backups[0]['filename']
+
+        if not re.match(r"^" + backup_filename_regex + "$", filename):
+            abort(400)  # Bad Request if the filename doesn't match the pattern
+
+        logger.debug(f"Backup download request for '{full_path}'")
+        return send_from_directory(os.path.abspath(datastore.datastore_path), filename, as_attachment=True)
+
+    @login_optionally_required
+    @backups_blueprint.route("", methods=['GET'])
+    def index():
+        backups = find_backups()
+        output = render_template("overview.html",
+                                 available_backups=backups,
+                                 backup_running=any(thread.is_alive() for thread in backup_threads)
+                                 )
+
+        return output
+
+    @login_optionally_required
+    @backups_blueprint.route("/remove-backups", methods=['GET'])
+    def remove_backups():
+
+        backup_filepath = os.path.join(datastore.datastore_path, BACKUP_FILENAME_FORMAT.format("*"))
+        backups = glob.glob(backup_filepath)
+        for backup in backups:
+            os.unlink(backup)
+
+        flash("Backups were deleted.")
+
+        return redirect(url_for('backups.index'))
+
+    return backups_blueprint

changedetectionio/blueprint/backups/templates/overview.html

@@ -0,0 +1,36 @@
+{% extends 'base.html' %}
+{% block content %}
+    {% from '_helpers.html' import render_simple_field, render_field %}
+    <div class="edit-form">
+        <div class="box-wrap inner">
+            <h4>Backups</h4>
+            {% if backup_running %}
+                <p>
+                    <strong>A backup is running!</strong>
+                </p>
+            {% endif %}
+            <p>
+                Here you can download and request a new backup, when a backup is completed you will see it listed below.
+            </p>
+            <br>
+                {% if available_backups %}
+                    <ul>
+                    {% for backup in available_backups %}
+                        <li><a href="{{ url_for('backups.download_backup', filename=backup["filename"]) }}">{{ backup["filename"] }}</a> {{  backup["filesize"] }} Mb</li>
+                    {% endfor %}
+                    </ul>
+                {% else %}
+                    <p>
+                    <strong>No backups found.</strong>
+                    </p>
+                {% endif %}
+
+            <a class="pure-button pure-button-primary" href="{{ url_for('backups.request_backup') }}">Create backup</a>
+            {% if available_backups %}
+                <a class="pure-button button-small button-error " href="{{ url_for('backups.remove_backups') }}">Remove backups</a>
+            {% endif %}
+        </div>
+    </div>
+
+
+{% endblock %}

changedetectionio/blueprint/browser_steps/__init__.py

@@ -22,45 +22,56 @@ from loguru import logger
 
 browsersteps_sessions = {}
 io_interface_context = None
+import json
+import hashlib
+from flask import Response
+import asyncio
+import threading
 
+def run_async_in_browser_loop(coro):
+    """Run async coroutine using the existing async worker event loop"""
+    from changedetectionio import worker_handler
+    
+    # Use the existing async worker event loop instead of creating a new one
+    if worker_handler.USE_ASYNC_WORKERS and worker_handler.async_loop and not worker_handler.async_loop.is_closed():
+        logger.debug("Browser steps using existing async worker event loop")
+        future = asyncio.run_coroutine_threadsafe(coro, worker_handler.async_loop)
+        return future.result()
+    else:
+        # Fallback: create a new event loop (for sync workers or if async loop not available)
+        logger.debug("Browser steps creating temporary event loop")
+        loop = asyncio.new_event_loop()
+        asyncio.set_event_loop(loop)
+        try:
+            return loop.run_until_complete(coro)
+        finally:
+            loop.close()
 
 def construct_blueprint(datastore: ChangeDetectionStore):
     browser_steps_blueprint = Blueprint('browser_steps', __name__, template_folder="templates")
 
-    def start_browsersteps_session(watch_uuid):
-        from . import nonContext
+    async def start_browsersteps_session(watch_uuid):
         from . import browser_steps
         import time
-        global browsersteps_sessions
-        global io_interface_context
-
+        from playwright.async_api import async_playwright
 
         # We keep the playwright session open for many minutes
         keepalive_seconds = int(os.getenv('BROWSERSTEPS_MINUTES_KEEPALIVE', 10)) * 60
 
         browsersteps_start_session = {'start_time': time.time()}
 
-        # You can only have one of these running
-        # This should be very fine to leave running for the life of the application
-        # @idea - Make it global so the pool of watch fetchers can use it also
-        if not io_interface_context:
-            io_interface_context = nonContext.c_sync_playwright()
-            # Start the Playwright context, which is actually a nodejs sub-process and communicates over STDIN/STDOUT pipes
-            io_interface_context = io_interface_context.start()
+        # Create a new async playwright instance for browser steps
+        playwright_instance = async_playwright()
+        playwright_context = await playwright_instance.start()
 
         keepalive_ms = ((keepalive_seconds + 3) * 1000)
         base_url = os.getenv('PLAYWRIGHT_DRIVER_URL', '').strip('"')
         a = "?" if not '?' in base_url else '&'
         base_url += a + f"timeout={keepalive_ms}"
 
-        try:
-            browsersteps_start_session['browser'] = io_interface_context.chromium.connect_over_cdp(base_url)
-        except Exception as e:
-            if 'ECONNREFUSED' in str(e):
-                return make_response('Unable to start the Playwright Browser session, is it running?', 401)
-            else:
-                # Other errors, bad URL syntax, bad reply etc
-                return make_response(str(e), 401)
+        browser = await playwright_context.chromium.connect_over_cdp(base_url, timeout=keepalive_ms)
+        browsersteps_start_session['browser'] = browser
+        browsersteps_start_session['playwright_context'] = playwright_context
 
         proxy_id = datastore.get_preferred_proxy_for_watch(uuid=watch_uuid)
         proxy = None
@@ -82,14 +93,20 @@ def construct_blueprint(datastore: ChangeDetectionStore):
                 logger.debug(f"Browser Steps: UUID {watch_uuid} selected proxy {proxy_url}")
 
         # Tell Playwright to connect to Chrome and setup a new session via our stepper interface
-        browsersteps_start_session['browserstepper'] = browser_steps.browsersteps_live_ui(
-            playwright_browser=browsersteps_start_session['browser'],
+        browserstepper = browser_steps.browsersteps_live_ui(
+            playwright_browser=browser,
             proxy=proxy,
-            start_url=datastore.data['watching'][watch_uuid].get('url')
+            start_url=datastore.data['watching'][watch_uuid].link,
+            headers=datastore.data['watching'][watch_uuid].get('headers')
         )
+        
+        # Initialize the async connection
+        await browserstepper.connect(proxy=proxy)
+        
+        browsersteps_start_session['browserstepper'] = browserstepper
 
         # For test
-        #browsersteps_start_session['browserstepper'].action_goto_url(value="http://example.com?time="+str(time.time()))
+        #await browsersteps_start_session['browserstepper'].action_goto_url(value="http://example.com?time="+str(time.time()))
 
         return browsersteps_start_session
 
@@ -98,10 +115,8 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     @browser_steps_blueprint.route("/browsersteps_start_session", methods=['GET'])
     def browsersteps_start_session():
         # A new session was requested, return sessionID
-
+        import asyncio
         import uuid
-        global browsersteps_sessions
-
         browsersteps_session_id = str(uuid.uuid4())
         watch_uuid = request.args.get('uuid')
 
@@ -110,7 +125,19 @@ def construct_blueprint(datastore: ChangeDetectionStore):
 
         logger.debug("Starting connection with playwright")
         logger.debug("browser_steps.py connecting")
-        browsersteps_sessions[browsersteps_session_id] = start_browsersteps_session(watch_uuid)
+
+        try:
+            # Run the async function in the dedicated browser steps event loop
+            browsersteps_sessions[browsersteps_session_id] = run_async_in_browser_loop(
+                start_browsersteps_session(watch_uuid)
+            )
+        except Exception as e:
+            if 'ECONNREFUSED' in str(e):
+                return make_response('Unable to start the Playwright Browser session, is sockpuppetbrowser running? Network configuration is OK?', 401)
+            else:
+                # Other errors, bad URL syntax, bad reply etc
+                return make_response(str(e), 401)
+
         logger.debug("Starting connection with playwright - done")
         return {'browsersteps_session_id': browsersteps_session_id}
 
@@ -145,7 +172,6 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     def browsersteps_ui_update():
         import base64
         import playwright._impl._errors
-        global browsersteps_sessions
         from changedetectionio.blueprint.browser_steps import browser_steps
 
         remaining =0
@@ -159,38 +185,30 @@ def construct_blueprint(datastore: ChangeDetectionStore):
         if not browsersteps_sessions.get(browsersteps_session_id):
             return make_response('No session exists under that ID', 500)
 
-
+        is_last_step = False
         # Actions - step/apply/etc, do the thing and return state
         if request.method == 'POST':
             # @todo - should always be an existing session
             step_operation = request.form.get('operation')
             step_selector = request.form.get('selector')
             step_optional_value = request.form.get('optional_value')
-            step_n = int(request.form.get('step_n'))
             is_last_step = strtobool(request.form.get('is_last_step'))
 
-            # @todo try.. accept.. nice errors not popups..
             try:
-
-                browsersteps_sessions[browsersteps_session_id]['browserstepper'].call_action(action_name=step_operation,
-                                         selector=step_selector,
-                                         optional_value=step_optional_value)
+                # Run the async call_action method in the dedicated browser steps event loop
+                run_async_in_browser_loop(
+                    browsersteps_sessions[browsersteps_session_id]['browserstepper'].call_action(
+                        action_name=step_operation,
+                        selector=step_selector,
+                        optional_value=step_optional_value
+                    )
+                )
 
             except Exception as e:
                 logger.error(f"Exception when calling step operation {step_operation} {str(e)}")
                 # Try to find something of value to give back to the user
                 return make_response(str(e).splitlines()[0], 401)
 
-            # Get visual selector ready/update its data (also use the current filter info from the page?)
-            # When the last 'apply' button was pressed
-            # @todo this adds overhead because the xpath selection is happening twice
-            u = browsersteps_sessions[browsersteps_session_id]['browserstepper'].page.url
-            if is_last_step and u:
-                (screenshot, xpath_data) = browsersteps_sessions[browsersteps_session_id]['browserstepper'].request_visualselector_data()
-                watch = datastore.data['watching'].get(uuid)
-                if watch:
-                    watch.save_screenshot(screenshot=screenshot)
-                    watch.save_xpath_data(data=xpath_data)
 
 #        if not this_session.page:
 #            cleanup_playwright_session()
@@ -198,31 +216,36 @@ def construct_blueprint(datastore: ChangeDetectionStore):
 
         # Screenshots and other info only needed on requesting a step (POST)
         try:
-            state = browsersteps_sessions[browsersteps_session_id]['browserstepper'].get_current_state()
-        except playwright._impl._api_types.Error as e:
-            return make_response("Browser session ran out of time :( Please reload this page."+str(e), 401)
+            # Run the async get_current_state method in the dedicated browser steps event loop
+            (screenshot, xpath_data) = run_async_in_browser_loop(
+                browsersteps_sessions[browsersteps_session_id]['browserstepper'].get_current_state()
+            )
+                
+            if is_last_step:
+                watch = datastore.data['watching'].get(uuid)
+                u = browsersteps_sessions[browsersteps_session_id]['browserstepper'].page.url
+                if watch and u:
+                    watch.save_screenshot(screenshot=screenshot)
+                    watch.save_xpath_data(data=xpath_data)
 
-        # Use send_file() which is way faster than read/write loop on bytes
-        import json
-        from tempfile import mkstemp
-        from flask import send_file
-        tmp_fd, tmp_file = mkstemp(text=True, suffix=".json", prefix="changedetectionio-")
+        except Exception as e:
+            return make_response(f"Error fetching screenshot and element data - {str(e)}", 401)
 
-        output = json.dumps({'screenshot': "data:image/jpeg;base64,{}".format(
-            base64.b64encode(state[0]).decode('ascii')),
-            'xpath_data': state[1],
-            'session_age_start': browsersteps_sessions[browsersteps_session_id]['browserstepper'].age_start,
-            'browser_time_remaining': round(remaining)
-        })
+        # SEND THIS BACK TO THE BROWSER
+        output = {
+            "screenshot": f"data:image/jpeg;base64,{base64.b64encode(screenshot).decode('ascii')}",
+            "xpath_data": xpath_data,
+            "session_age_start": browsersteps_sessions[browsersteps_session_id]['browserstepper'].age_start,
+            "browser_time_remaining": round(remaining)
+        }
+        json_data = json.dumps(output)
 
-        with os.fdopen(tmp_fd, 'w') as f:
-            f.write(output)
+        # Generate an ETag (hash of the response body)
+        etag_hash = hashlib.md5(json_data.encode('utf-8')).hexdigest()
 
-        response = make_response(send_file(path_or_file=tmp_file,
-                                           mimetype='application/json; charset=UTF-8',
-                                           etag=True))
-        # No longer needed
-        os.unlink(tmp_file)
+        # Create the response with ETag
+        response = Response(json_data, mimetype="application/json; charset=UTF-8")
+        response.set_etag(etag_hash)
 
         return response
 

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -1,14 +1,15 @@
-#!/usr/bin/python3
-
 import os
 import time
 import re
 from random import randint
 from loguru import logger
 
+from changedetectionio.content_fetchers import SCREENSHOT_MAX_HEIGHT_DEFAULT
 from changedetectionio.content_fetchers.base import manage_user_agent
 from changedetectionio.safe_jinja import render as jinja_render
 
+
+
 # Two flags, tell the JS which of the "Selector" or "Value" field should be enabled in the front end
 # 0- off, 1- on
 browser_step_ui_config = {'Choose one': '0 0',
@@ -25,18 +26,22 @@ browser_step_ui_config = {'Choose one': '0 0',
                           'Click element if exists': '1 0',
                           'Click element': '1 0',
                           'Click element containing text': '0 1',
+                          'Click element containing text if exists': '0 1',
                           'Enter text in field': '1 1',
                           'Execute JS': '0 1',
 #                          'Extract text and use as filter': '1 0',
                           'Goto site': '0 0',
                           'Goto URL': '0 1',
+                          'Make all child elements visible': '1 0',
                           'Press Enter': '0 0',
                           'Select by label': '1 1',
+                          '<select> by option text': '1 1',
                           'Scroll down': '0 0',
                           'Uncheck checkbox': '1 0',
                           'Wait for seconds': '0 1',
                           'Wait for text': '0 1',
                           'Wait for text in element': '1 1',
+                          'Remove elements': '1 0',
                           #                          'Press Page Down': '0 0',
                           #                          'Press Page Up': '0 0',
                           # weird bug, come back to it later
@@ -50,12 +55,17 @@ browser_step_ui_config = {'Choose one': '0 0',
 class steppable_browser_interface():
     page = None
     start_url = None
+    action_timeout = 10 * 1000
 
     def __init__(self, start_url):
         self.start_url = start_url
 
     # Convert and perform "Click Button" for example
-    def call_action(self, action_name, selector=None, optional_value=None):
+    async def call_action(self, action_name, selector=None, optional_value=None):
+        if self.page is None:
+            logger.warning("Cannot call action on None page object")
+            return
+            
         now = time.time()
         call_action_name = re.sub('[^0-9a-zA-Z]+', '_', action_name.lower())
         if call_action_name == 'choose_one':
@@ -66,117 +76,227 @@ class steppable_browser_interface():
         if selector and selector.startswith('/') and not selector.startswith('//'):
             selector = "xpath=" + selector
 
+        # Check if action handler exists
+        if not hasattr(self, "action_" + call_action_name):
+            logger.warning(f"Action handler for '{call_action_name}' not found")
+            return
+            
         action_handler = getattr(self, "action_" + call_action_name)
 
         # Support for Jinja2 variables in the value and selector
-
         if selector and ('{%' in selector or '{{' in selector):
             selector = jinja_render(template_str=selector)
 
         if optional_value and ('{%' in optional_value or '{{' in optional_value):
             optional_value = jinja_render(template_str=optional_value)
 
-        action_handler(selector, optional_value)
-        self.page.wait_for_timeout(1.5 * 1000)
+        # Trigger click and cautiously handle potential navigation
+        # This means the page redirects/reloads/changes JS etc etc
+        if call_action_name.startswith('click_'):
+            try:
+                # Set up navigation expectation before the click (like sync version)
+                async with self.page.expect_event("framenavigated", timeout=3000) as navigation_info:
+                    await action_handler(selector, optional_value)
+                
+                # Check if navigation actually occurred
+                try:
+                    await navigation_info.value  # This waits for the navigation promise
+                    logger.debug(f"Navigation occurred on {call_action_name}.")
+                except Exception:
+                    logger.debug(f"No navigation occurred within timeout when calling {call_action_name}, that's OK, continuing.")
+                    
+            except Exception as e:
+                # If expect_event itself times out, that means no navigation occurred - that's OK
+                if "framenavigated" in str(e) and "exceeded" in str(e):
+                    logger.debug(f"No navigation occurred within timeout when calling {call_action_name}, that's OK, continuing.")
+                else:
+                    raise e
+        else:
+            # Some other action that probably a navigation is not expected
+            await action_handler(selector, optional_value)
+
+
+        # Safely wait for timeout
+        await self.page.wait_for_timeout(1.5 * 1000)
         logger.debug(f"Call action done in {time.time()-now:.2f}s")
 
-    def action_goto_url(self, selector=None, value=None):
-        # self.page.set_viewport_size({"width": 1280, "height": 5000})
+    async def action_goto_url(self, selector=None, value=None):
+        if not value:
+            logger.warning("No URL provided for goto_url action")
+            return None
+            
         now = time.time()
-        response = self.page.goto(value, timeout=0, wait_until='load')
-        # Should be the same as the puppeteer_fetch.js methods, means, load with no timeout set (skip timeout)
-        #and also wait for seconds ?
-        #await page.waitForTimeout(1000);
-        #await page.waitForTimeout(extra_wait_ms);
+        response = await self.page.goto(value, timeout=0, wait_until='load')
         logger.debug(f"Time to goto URL {time.time()-now:.2f}s")
         return response
 
     # Incase they request to go back to the start
-    def action_goto_site(self, selector=None, value=None):
-        return self.action_goto_url(value=self.start_url)
+    async def action_goto_site(self, selector=None, value=None):
+        return await self.action_goto_url(value=re.sub(r'^source:', '', self.start_url, flags=re.IGNORECASE))
 
-    def action_click_element_containing_text(self, selector=None, value=''):
-        if not len(value.strip()):
+    async def action_click_element_containing_text(self, selector=None, value=''):
+        logger.debug("Clicking element containing text")
+        if not value or not len(value.strip()):
             return
+            
         elem = self.page.get_by_text(value)
-        if elem.count():
-            elem.first.click(delay=randint(200, 500), timeout=3000)
+        if await elem.count():
+            await elem.first.click(delay=randint(200, 500), timeout=self.action_timeout)
 
-    def action_enter_text_in_field(self, selector, value):
-        if not len(selector.strip()):
+
+    async def action_click_element_containing_text_if_exists(self, selector=None, value=''):
+        logger.debug("Clicking element containing text if exists")
+        if not value or not len(value.strip()):
+            return
+            
+        elem = self.page.get_by_text(value)
+        count = await elem.count()
+        logger.debug(f"Clicking element containing text - {count} elements found")
+        if count:
+            await elem.first.click(delay=randint(200, 500), timeout=self.action_timeout)
+                
+
+    async def action_enter_text_in_field(self, selector, value):
+        if not selector or not len(selector.strip()):
             return
 
-        self.page.fill(selector, value, timeout=10 * 1000)
+        await self.page.fill(selector, value, timeout=self.action_timeout)
 
-    def action_execute_js(self, selector, value):
-        response = self.page.evaluate(value)
-        return response
+    async def action_execute_js(self, selector, value):
+        if not value:
+            return None
+            
+        return await self.page.evaluate(value)
 
-    def action_click_element(self, selector, value):
+    async def action_click_element(self, selector, value):
         logger.debug("Clicking element")
-        if not len(selector.strip()):
+        if not selector or not len(selector.strip()):
             return
 
-        self.page.click(selector=selector, timeout=30 * 1000, delay=randint(200, 500))
+        await self.page.click(selector=selector, timeout=self.action_timeout + 20 * 1000, delay=randint(200, 500))
 
-    def action_click_element_if_exists(self, selector, value):
+    async def action_click_element_if_exists(self, selector, value):
         import playwright._impl._errors as _api_types
         logger.debug("Clicking element if exists")
-        if not len(selector.strip()):
+        if not selector or not len(selector.strip()):
             return
+            
         try:
-            self.page.click(selector, timeout=10 * 1000, delay=randint(200, 500))
-        except _api_types.TimeoutError as e:
+            await self.page.click(selector, timeout=self.action_timeout, delay=randint(200, 500))
+        except _api_types.TimeoutError:
             return
-        except _api_types.Error as e:
+        except _api_types.Error:
             # Element was there, but page redrew and now its long long gone
             return
+                
 
-    def action_click_x_y(self, selector, value):
-        if not re.match(r'^\s?\d+\s?,\s?\d+\s?$', value):
-            raise Exception("'Click X,Y' step should be in the format of '100 , 90'")
+    async def action_click_x_y(self, selector, value):
+        if not value or not re.match(r'^\s?\d+\s?,\s?\d+\s?$', value):
+            logger.warning("'Click X,Y' step should be in the format of '100 , 90'")
+            return
 
-        x, y = value.strip().split(',')
-        x = int(float(x.strip()))
-        y = int(float(y.strip()))
-        self.page.mouse.click(x=x, y=y, delay=randint(200, 500))
+        try:
+            x, y = value.strip().split(',')
+            x = int(float(x.strip()))
+            y = int(float(y.strip()))
+            
+            await self.page.mouse.click(x=x, y=y, delay=randint(200, 500))
+                
+        except Exception as e:
+            logger.error(f"Error parsing x,y coordinates: {str(e)}")
 
-    def action_scroll_down(self, selector, value):
+    async def action__select_by_option_text(self, selector, value):
+        if not selector or not len(selector.strip()):
+            return
+
+        await self.page.select_option(selector, label=value, timeout=self.action_timeout)
+
+    async def action_scroll_down(self, selector, value):
         # Some sites this doesnt work on for some reason
-        self.page.mouse.wheel(0, 600)
-        self.page.wait_for_timeout(1000)
+        await self.page.mouse.wheel(0, 600)
+        await self.page.wait_for_timeout(1000)
 
-    def action_wait_for_seconds(self, selector, value):
-        self.page.wait_for_timeout(float(value.strip()) * 1000)
+    async def action_wait_for_seconds(self, selector, value):
+        try:
+            seconds = float(value.strip()) if value else 1.0
+            await self.page.wait_for_timeout(seconds * 1000)
+        except (ValueError, TypeError) as e:
+            logger.error(f"Invalid value for wait_for_seconds: {str(e)}")
 
-    def action_wait_for_text(self, selector, value):
+    async def action_wait_for_text(self, selector, value):
+        if not value:
+            return
+            
         import json
         v = json.dumps(value)
-        self.page.wait_for_function(f'document.querySelector("body").innerText.includes({v});', timeout=30000)
+        await self.page.wait_for_function(
+            f'document.querySelector("body").innerText.includes({v});',
+            timeout=30000
+        )
+            
 
-    def action_wait_for_text_in_element(self, selector, value):
+    async def action_wait_for_text_in_element(self, selector, value):
+        if not selector or not value:
+            return
+            
         import json
         s = json.dumps(selector)
         v = json.dumps(value)
-        self.page.wait_for_function(f'document.querySelector({s}).innerText.includes({v});', timeout=30000)
+        
+        await self.page.wait_for_function(
+            f'document.querySelector({s}).innerText.includes({v});',
+            timeout=30000
+        )
 
     # @todo - in the future make some popout interface to capture what needs to be set
     # https://playwright.dev/python/docs/api/class-keyboard
-    def action_press_enter(self, selector, value):
-        self.page.keyboard.press("Enter", delay=randint(200, 500))
+    async def action_press_enter(self, selector, value):
+        await self.page.keyboard.press("Enter", delay=randint(200, 500))
+            
 
-    def action_press_page_up(self, selector, value):
-        self.page.keyboard.press("PageUp", delay=randint(200, 500))
+    async def action_press_page_up(self, selector, value):
+        await self.page.keyboard.press("PageUp", delay=randint(200, 500))
 
-    def action_press_page_down(self, selector, value):
-        self.page.keyboard.press("PageDown", delay=randint(200, 500))
+    async def action_press_page_down(self, selector, value):
+        await self.page.keyboard.press("PageDown", delay=randint(200, 500))
 
-    def action_check_checkbox(self, selector, value):
-        self.page.locator(selector).check(timeout=1000)
+    async def action_check_checkbox(self, selector, value):
+        if not selector:
+            return
 
-    def action_uncheck_checkbox(self, selector, value):
-        self.page.locator(selector, timeout=1000).uncheck(timeout=1000)
+        await self.page.locator(selector).check(timeout=self.action_timeout)
 
+    async def action_uncheck_checkbox(self, selector, value):
+        if not selector:
+            return
+            
+        await self.page.locator(selector).uncheck(timeout=self.action_timeout)
+            
+
+    async def action_remove_elements(self, selector, value):
+        """Removes all elements matching the given selector from the DOM."""
+        if not selector:
+            return
+            
+        await self.page.locator(selector).evaluate_all("els => els.forEach(el => el.remove())")
+
+    async def action_make_all_child_elements_visible(self, selector, value):
+        """Recursively makes all child elements inside the given selector fully visible."""
+        if not selector:
+            return
+            
+        await self.page.locator(selector).locator("*").evaluate_all("""
+            els => els.forEach(el => {
+                el.style.display = 'block';   // Forces it to be displayed
+                el.style.visibility = 'visible';   // Ensures it's not hidden
+                el.style.opacity = '1';   // Fully opaque
+                el.style.position = 'relative';   // Avoids 'absolute' hiding
+                el.style.height = 'auto';   // Expands collapsed elements
+                el.style.width = 'auto';   // Ensures full visibility
+                el.removeAttribute('hidden');   // Removes hidden attribute
+                el.classList.remove('hidden', 'd-none');  // Removes common CSS hidden classes
+            })
+        """)
 
 # Responsible for maintaining a live 'context' with the chrome CDP
 # @todo - how long do contexts live for anyway?
@@ -188,7 +308,9 @@ class browsersteps_live_ui(steppable_browser_interface):
     # bump and kill this if idle after X sec
     age_start = 0
     headers = {}
-
+    # Track if resources are properly cleaned up
+    _is_cleaned_up = False
+    
     # use a special driver, maybe locally etc
     command_executor = os.getenv(
         "PLAYWRIGHT_BROWSERSTEPS_DRIVER_URL"
@@ -207,17 +329,23 @@ class browsersteps_live_ui(steppable_browser_interface):
         self.age_start = time.time()
         self.playwright_browser = playwright_browser
         self.start_url = start_url
-        if self.context is None:
-            self.connect(proxy=proxy)
+        self._is_cleaned_up = False
+        self.proxy = proxy
+        # Note: connect() is now async and must be called separately
+
+    def __del__(self):
+        # Ensure cleanup happens if object is garbage collected
+        # Note: cleanup is now async, so we can only mark as cleaned up here
+        self._is_cleaned_up = True
 
     # Connect and setup a new context
-    def connect(self, proxy=None):
+    async def connect(self, proxy=None):
         # Should only get called once - test that
         keep_open = 1000 * 60 * 5
         now = time.time()
 
         # @todo handle multiple contexts, bind a unique id from the browser on each req?
-        self.context = self.playwright_browser.new_context(
+        self.context = await self.playwright_browser.new_context(
             accept_downloads=False,  # Should never be needed
             bypass_csp=True,  # This is needed to enable JavaScript execution on GitHub and others
             extra_http_headers=self.headers,
@@ -228,72 +356,142 @@ class browsersteps_live_ui(steppable_browser_interface):
             user_agent=manage_user_agent(headers=self.headers),
         )
 
-
-        self.page = self.context.new_page()
+        self.page = await self.context.new_page()
 
         # self.page.set_default_navigation_timeout(keep_open)
         self.page.set_default_timeout(keep_open)
-        # @todo probably this doesnt work
-        self.page.on(
-            "close",
-            self.mark_as_closed,
-        )
+        # Set event handlers
+        self.page.on("close", self.mark_as_closed)
         # Listen for all console events and handle errors
         self.page.on("console", lambda msg: print(f"Browser steps console - {msg.type}: {msg.text} {msg.args}"))
 
         logger.debug(f"Time to browser setup {time.time()-now:.2f}s")
-        self.page.wait_for_timeout(1 * 1000)
+        await self.page.wait_for_timeout(1 * 1000)
 
     def mark_as_closed(self):
         logger.debug("Page closed, cleaning up..")
+        # Note: This is called from a sync context (event handler)
+        # so we'll just mark as cleaned up and let __del__ handle the rest
+        self._is_cleaned_up = True
+
+    async def cleanup(self):
+        """Properly clean up all resources to prevent memory leaks"""
+        if self._is_cleaned_up:
+            return
+            
+        logger.debug("Cleaning up browser steps resources")
+        
+        # Clean up page
+        if hasattr(self, 'page') and self.page is not None:
+            try:
+                # Force garbage collection before closing
+                await self.page.request_gc()
+            except Exception as e:
+                logger.debug(f"Error during page garbage collection: {str(e)}")
+                
+            try:
+                # Remove event listeners before closing
+                self.page.remove_listener("close", self.mark_as_closed)
+            except Exception as e:
+                logger.debug(f"Error removing event listeners: {str(e)}")
+                
+            try:
+                await self.page.close()
+            except Exception as e:
+                logger.debug(f"Error closing page: {str(e)}")
+            
+            self.page = None
+
+        # Clean up context
+        if hasattr(self, 'context') and self.context is not None:
+            try:
+                await self.context.close()
+            except Exception as e:
+                logger.debug(f"Error closing context: {str(e)}")
+            
+            self.context = None
+            
+        self._is_cleaned_up = True
+        logger.debug("Browser steps resources cleanup complete")
 
     @property
     def has_expired(self):
-        if not self.page:
+        if not self.page or self._is_cleaned_up:
             return True
+        
+        # Check if session has expired based on age
+        max_age_seconds = int(os.getenv("BROWSER_STEPS_MAX_AGE_SECONDS", 60 * 10))  # Default 10 minutes
+        if (time.time() - self.age_start) > max_age_seconds:
+            logger.debug(f"Browser steps session expired after {max_age_seconds} seconds")
+            return True
+            
+        return False
 
-
-    def get_current_state(self):
+    async def get_current_state(self):
         """Return the screenshot and interactive elements mapping, generally always called after action_()"""
         import importlib.resources
+        import json
+        # because we for now only run browser steps in playwright mode (not puppeteer mode)
+        from changedetectionio.content_fetchers.playwright import capture_full_page_async
+
+        # Safety check - don't proceed if resources are cleaned up
+        if self._is_cleaned_up or self.page is None:
+            logger.warning("Attempted to get current state after cleanup")
+            return (None, None)
+
         xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text()
 
         now = time.time()
-        self.page.wait_for_timeout(1 * 1000)
+        await self.page.wait_for_timeout(1 * 1000)
 
-        # The actual screenshot
-        screenshot = self.page.screenshot(type='jpeg', full_page=True, quality=40)
+        screenshot = None
+        xpath_data = None
+        
+        try:
+            # Get screenshot first
+            screenshot = await capture_full_page_async(page=self.page)
+            if not screenshot:
+                logger.error("No screenshot was retrieved :((")
 
-        self.page.evaluate("var include_filters=''")
-        # Go find the interactive elements
-        # @todo in the future, something smarter that can scan for elements with .click/focus etc event handlers?
-        elements = 'a,button,input,select,textarea,i,th,td,p,li,h1,h2,h3,h4,div,span'
-        xpath_element_js = xpath_element_js.replace('%ELEMENTS%', elements)
-        xpath_data = self.page.evaluate("async () => {" + xpath_element_js + "}")
-        # So the JS will find the smallest one first
-        xpath_data['size_pos'] = sorted(xpath_data['size_pos'], key=lambda k: k['width'] * k['height'], reverse=True)
-        logger.debug(f"Time to complete get_current_state of browser {time.time()-now:.2f}s")
-        # except
-        # playwright._impl._api_types.Error: Browser closed.
-        # @todo show some countdown timer?
+            logger.debug(f"Time to get screenshot from browser {time.time() - now:.2f}s")
+
+            # Then get interactive elements
+            now = time.time()
+            await self.page.evaluate("var include_filters=''")
+            await self.page.request_gc()
+
+            scan_elements = 'a,button,input,select,textarea,i,th,td,p,li,h1,h2,h3,h4,div,span'
+
+            MAX_TOTAL_HEIGHT = int(os.getenv("SCREENSHOT_MAX_HEIGHT", SCREENSHOT_MAX_HEIGHT_DEFAULT))
+            xpath_data = json.loads(await self.page.evaluate(xpath_element_js, {
+                "visualselector_xpath_selectors": scan_elements,
+                "max_height": MAX_TOTAL_HEIGHT
+            }))
+            await self.page.request_gc()
+
+            # Sort elements by size
+            xpath_data['size_pos'] = sorted(xpath_data['size_pos'], key=lambda k: k['width'] * k['height'], reverse=True)
+            logger.debug(f"Time to scrape xPath element data in browser {time.time()-now:.2f}s")
+            
+        except Exception as e:
+            logger.error(f"Error getting current state: {str(e)}")
+            # If the page has navigated (common with logins) then the context is destroyed on navigation, continue
+            # I'm not sure that this is required anymore because we have the "expect navigation wrapper" at the top
+            if "Execution context was destroyed" in str(e):
+                logger.debug("Execution context was destroyed, most likely because of navigation, continuing...")
+            pass
+
+            # Attempt recovery - force garbage collection
+            try:
+                await self.page.request_gc()
+            except:
+                pass
+        
+        # Request garbage collection one final time
+        try:
+            await self.page.request_gc()
+        except:
+            pass
+            
         return (screenshot, xpath_data)
 
-    def request_visualselector_data(self):
-        """
-        Does the same that the playwright operation in content_fetcher does
-        This is used to just bump the VisualSelector data so it' ready to go if they click on the tab
-        @todo refactor and remove duplicate code, add include_filters
-        :param xpath_data:
-        :param screenshot:
-        :param current_include_filters:
-        :return:
-        """
-        import importlib.resources
-        self.page.evaluate("var include_filters=''")
-        xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text()
-        from changedetectionio.content_fetchers import visualselector_xpath_selectors
-        xpath_element_js = xpath_element_js.replace('%ELEMENTS%', visualselector_xpath_selectors)
-        xpath_data = self.page.evaluate("async () => {" + xpath_element_js + "}")
-        screenshot = self.page.screenshot(type='jpeg', full_page=True, quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
-
-        return (screenshot, xpath_data)

changedetectionio/blueprint/browser_steps/nonContext.py

@@ -1,17 +0,0 @@
-from playwright.sync_api import PlaywrightContextManager
-
-# So playwright wants to run as a context manager, but we do something horrible and hacky
-# we are holding the session open for as long as possible, then shutting it down, and opening a new one
-# So it means we don't get to use PlaywrightContextManager' __enter__ __exit__
-# To work around this, make goodbye() act the same as the __exit__()
-#
-# But actually I think this is because the context is opened correctly with __enter__() but we timeout the connection
-# then theres some lock condition where we cant destroy it without it hanging
-
-class c_PlaywrightContextManager(PlaywrightContextManager):
-
-    def goodbye(self) -> None:
-        self.__exit__()
-
-def c_sync_playwright() -> PlaywrightContextManager:
-    return c_PlaywrightContextManager()

changedetectionio/blueprint/check_proxies/__init__.py

@@ -1,4 +1,7 @@
+import importlib
 from concurrent.futures import ThreadPoolExecutor
+
+from changedetectionio.processors.text_json_diff.processor import FilterNotFoundInResponse
 from changedetectionio.store import ChangeDetectionStore
 
 from functools import wraps
@@ -30,7 +33,6 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     def long_task(uuid, preferred_proxy):
         import time
         from changedetectionio.content_fetchers import exceptions as content_fetcher_exceptions
-        from changedetectionio.processors import text_json_diff
         from changedetectionio.safe_jinja import render as jinja_render
 
         status = {'status': '', 'length': 0, 'text': ''}
@@ -38,8 +40,12 @@ def construct_blueprint(datastore: ChangeDetectionStore):
         contents = ''
         now = time.time()
         try:
-            update_handler = text_json_diff.perform_site_check(datastore=datastore, watch_uuid=uuid)
-            update_handler.call_browser()
+            processor_module = importlib.import_module("changedetectionio.processors.text_json_diff.processor")
+            update_handler = processor_module.perform_site_check(datastore=datastore,
+                                                                 watch_uuid=uuid
+                                                                 )
+
+            update_handler.call_browser(preferred_proxy_id=preferred_proxy)
         # title, size is len contents not len xfer
         except content_fetcher_exceptions.Non200ErrorCodeReceived as e:
             if e.status_code == 404:
@@ -48,7 +54,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
                 status.update({'status': 'ERROR', 'length': len(contents), 'text': f"{e.status_code} - Access denied"})
             else:
                 status.update({'status': 'ERROR', 'length': len(contents), 'text': f"Status code: {e.status_code}"})
-        except text_json_diff.FilterNotFoundInResponse:
+        except FilterNotFoundInResponse:
             status.update({'status': 'OK', 'length': len(contents), 'text': f"OK but CSS/xPath filter not found (page changed layout?)"})
         except content_fetcher_exceptions.EmptyReply as e:
             if e.status_code == 403 or e.status_code == 401:

changedetectionio/blueprint/imports/__init__.py

@@ -0,0 +1,75 @@
+from flask import Blueprint, request, redirect, url_for, flash, render_template
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+from changedetectionio import worker_handler
+from changedetectionio.blueprint.imports.importer import (
+    import_url_list, 
+    import_distill_io_json, 
+    import_xlsx_wachete, 
+    import_xlsx_custom
+)
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMetaData):
+    import_blueprint = Blueprint('imports', __name__, template_folder="templates")
+    
+    @import_blueprint.route("/import", methods=['GET', 'POST'])
+    @login_optionally_required
+    def import_page():
+        remaining_urls = []
+        from changedetectionio import forms
+
+        if request.method == 'POST':
+            # URL List import
+            if request.values.get('urls') and len(request.values.get('urls').strip()):
+                # Import and push into the queue for immediate update check
+                importer_handler = import_url_list()
+                importer_handler.run(data=request.values.get('urls'), flash=flash, datastore=datastore, processor=request.values.get('processor', 'text_json_diff'))
+                for uuid in importer_handler.new_uuids:
+                    worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+
+                if len(importer_handler.remaining_data) == 0:
+                    return redirect(url_for('watchlist.index'))
+                else:
+                    remaining_urls = importer_handler.remaining_data
+
+            # Distill.io import
+            if request.values.get('distill-io') and len(request.values.get('distill-io').strip()):
+                # Import and push into the queue for immediate update check
+                d_importer = import_distill_io_json()
+                d_importer.run(data=request.values.get('distill-io'), flash=flash, datastore=datastore)
+                for uuid in d_importer.new_uuids:
+                    worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+
+            # XLSX importer
+            if request.files and request.files.get('xlsx_file'):
+                file = request.files['xlsx_file']
+
+                if request.values.get('file_mapping') == 'wachete':
+                    w_importer = import_xlsx_wachete()
+                    w_importer.run(data=file, flash=flash, datastore=datastore)
+                else:
+                    w_importer = import_xlsx_custom()
+                    # Building mapping of col # to col # type
+                    map = {}
+                    for i in range(10):
+                        c = request.values.get(f"custom_xlsx[col_{i}]")
+                        v = request.values.get(f"custom_xlsx[col_type_{i}]")
+                        if c and v:
+                            map[int(c)] = v
+
+                    w_importer.import_profile = map
+                    w_importer.run(data=file, flash=flash, datastore=datastore)
+
+                for uuid in w_importer.new_uuids:
+                    worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+
+        # Could be some remaining, or we could be on GET
+        form = forms.importForm(formdata=request.form if request.method == 'POST' else None)
+        output = render_template("import.html",
+                                form=form,
+                                import_url_list_remaining="\n".join(remaining_urls),
+                                original_distill_json=''
+                                )
+        return output
+
+    return import_blueprint

changedetectionio/blueprint/imports/importer.py

@@ -1,6 +1,5 @@
-from abc import ABC, abstractmethod
+from abc import abstractmethod
 import time
-import validators
 from wtforms import ValidationError
 from loguru import logger
 
@@ -241,7 +240,7 @@ class import_xlsx_custom(Importer):
             return
 
         # @todo cehck atleast 2 rows, same in other method
-        from .forms import validate_url
+        from changedetectionio.forms import validate_url
         row_i = 1
 
         try:
@@ -300,4 +299,4 @@ class import_xlsx_custom(Importer):
             row_i += 1
 
         flash(
-            "{} imported from custom .xlsx in {:.2f}s".format(len(self.new_uuids), time.time() - now))
+            "{} imported from custom .xlsx in {:.2f}s".format(len(self.new_uuids), time.time() - now))

changedetectionio/blueprint/imports/templates/import.html

@@ -13,29 +13,27 @@
     </div>
 
     <div class="box-wrap inner">
-        <form class="pure-form" action="{{url_for('import_page')}}" method="POST" enctype="multipart/form-data">
+        <form class="pure-form" action="{{url_for('imports.import_page')}}" method="POST" enctype="multipart/form-data">
             <input type="hidden" name="csrf_token" value="{{ csrf_token() }}">
             <div class="tab-pane-inner" id="url-list">
-                    <legend>
+                <div class="pure-control-group">
                         Enter one URL per line, and optionally add tags for each URL after a space, delineated by comma
                         (,):
                         <br>
-                        <code>https://example.com tag1, tag2, last tag</code>
-                        <br>
+                        <p><strong>Example:  </strong><code>https://example.com tag1, tag2, last tag</code></p>
                         URLs which do not pass validation will stay in the textarea.
-                    </legend>
+                </div>
                 {{ render_field(form.processor, class="processor") }}
-
+                
+                <div class="pure-control-group">
                     <textarea name="urls" class="pure-input-1-2" placeholder="https://"
                               style="width: 100%;
                                 font-family:monospace;
                                 white-space: pre;
                                 overflow-wrap: normal;
                                 overflow-x: scroll;" rows="25">{{ import_url_list_remaining }}</textarea>
-
-<div id="quick-watch-processor-type">
-
-                    </div>
+                 </div>
+                 <div id="quick-watch-processor-type"></div>
 
             </div>
 
@@ -43,7 +41,7 @@
 
 
 
-                    <legend>
+                    <div class="pure-control-group">
                         Copy and Paste your Distill.io watch 'export' file, this should be a JSON file.<br>
                         This is <i>experimental</i>, supported fields are <code>name</code>, <code>uri</code>, <code>tags</code>, <code>config:selections</code>, the rest (including <code>schedule</code>) are ignored.
                         <br>
@@ -51,7 +49,7 @@
                         How to export? <a href="https://distill.io/docs/web-monitor/how-export-and-import-monitors/">https://distill.io/docs/web-monitor/how-export-and-import-monitors/</a><br>
                         Be sure to set your default fetcher to Chrome if required.<br>
                         </p>
-                    </legend>
+                    </div>
 
 
                     <textarea name="distill-io" class="pure-input-1-2" style="width: 100%;
@@ -122,4 +120,4 @@
     </div>
 </div>
 
-{% endblock %}
+{% endblock %}

changedetectionio/blueprint/price_data_follower/__init__.py

@@ -4,6 +4,7 @@ from flask import Blueprint, flash, redirect, url_for
 from flask_login import login_required
 from changedetectionio.store import ChangeDetectionStore
 from changedetectionio import queuedWatchMetaData
+from changedetectionio import worker_handler
 from queue import PriorityQueue
 
 PRICE_DATA_TRACK_ACCEPT = 'accepted'
@@ -17,14 +18,16 @@ def construct_blueprint(datastore: ChangeDetectionStore, update_q: PriorityQueue
     @price_data_follower_blueprint.route("/<string:uuid>/accept", methods=['GET'])
     def accept(uuid):
         datastore.data['watching'][uuid]['track_ldjson_price_data'] = PRICE_DATA_TRACK_ACCEPT
-        update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
-        return redirect(url_for("index"))
+        datastore.data['watching'][uuid]['processor'] = 'restock_diff'
+        datastore.data['watching'][uuid].clear_watch()
+        worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+        return redirect(url_for("watchlist.index"))
 
     @login_required
     @price_data_follower_blueprint.route("/<string:uuid>/reject", methods=['GET'])
     def reject(uuid):
         datastore.data['watching'][uuid]['track_ldjson_price_data'] = PRICE_DATA_TRACK_REJECT
-        return redirect(url_for("index"))
+        return redirect(url_for("watchlist.index"))
 
 
     return price_data_follower_blueprint

changedetectionio/blueprint/rss/__init__.py

@@ -0,0 +1 @@
+RSS_FORMAT_TYPES = [('plaintext', 'Plain text'), ('html', 'HTML Color')]

changedetectionio/blueprint/rss/blueprint.py

@@ -0,0 +1,147 @@
+
+from changedetectionio.safe_jinja import render as jinja_render
+from changedetectionio.store import ChangeDetectionStore
+from feedgen.feed import FeedGenerator
+from flask import Blueprint, make_response, request, url_for, redirect
+from loguru import logger
+import datetime
+import pytz
+import re
+import time
+
+
+BAD_CHARS_REGEX=r'[\x00-\x08\x0B\x0C\x0E-\x1F]'
+
+# Anything that is not text/UTF-8 should be stripped before it breaks feedgen (such as binary data etc)
+def scan_invalid_chars_in_rss(content):
+    for match in re.finditer(BAD_CHARS_REGEX, content):
+        i = match.start()
+        bad_char = content[i]
+        hex_value = f"0x{ord(bad_char):02x}"
+        # Grab context
+        start = max(0, i - 20)
+        end = min(len(content), i + 21)
+        context = content[start:end].replace('\n', '\\n').replace('\r', '\\r')
+        logger.warning(f"Invalid char {hex_value} at pos {i}: ...{context}...")
+        # First match is enough
+        return True
+
+    return False
+
+
+def clean_entry_content(content):
+    cleaned = re.sub(BAD_CHARS_REGEX, '', content)
+    return cleaned
+
+def construct_blueprint(datastore: ChangeDetectionStore):
+    rss_blueprint = Blueprint('rss', __name__)
+
+    # Some RSS reader situations ended up with rss/ (forward slash after RSS) due
+    # to some earlier blueprint rerouting work, it should goto feed.
+    @rss_blueprint.route("/", methods=['GET'])
+    def extraslash():
+        return redirect(url_for('rss.feed'))
+
+    # Import the login decorator if needed
+    # from changedetectionio.auth_decorator import login_optionally_required
+    @rss_blueprint.route("", methods=['GET'])
+    def feed():
+        now = time.time()
+        # Always requires token set
+        app_rss_token = datastore.data['settings']['application'].get('rss_access_token')
+        rss_url_token = request.args.get('token')
+        if rss_url_token != app_rss_token:
+            return "Access denied, bad token", 403
+
+        from changedetectionio import diff
+        limit_tag = request.args.get('tag', '').lower().strip()
+        # Be sure limit_tag is a uuid
+        for uuid, tag in datastore.data['settings']['application'].get('tags', {}).items():
+            if limit_tag == tag.get('title', '').lower().strip():
+                limit_tag = uuid
+
+        # Sort by last_changed and add the uuid which is usually the key..
+        sorted_watches = []
+
+        # @todo needs a .itemsWithTag() or something - then we can use that in Jinaj2 and throw this away
+        for uuid, watch in datastore.data['watching'].items():
+            # @todo tag notification_muted skip also (improve Watch model)
+            if datastore.data['settings']['application'].get('rss_hide_muted_watches') and watch.get('notification_muted'):
+                continue
+            if limit_tag and not limit_tag in watch['tags']:
+                continue
+            watch['uuid'] = uuid
+            sorted_watches.append(watch)
+
+        sorted_watches.sort(key=lambda x: x.last_changed, reverse=False)
+
+        fg = FeedGenerator()
+        fg.title('changedetection.io')
+        fg.description('Feed description')
+        fg.link(href='https://changedetection.io')
+
+        html_colour_enable = False
+        if datastore.data['settings']['application'].get('rss_content_format') == 'html':
+            html_colour_enable = True
+
+        for watch in sorted_watches:
+
+            dates = list(watch.history.keys())
+            # Re #521 - Don't bother processing this one if theres less than 2 snapshots, means we never had a change detected.
+            if len(dates) < 2:
+                continue
+
+            if not watch.viewed:
+                # Re #239 - GUID needs to be individual for each event
+                # @todo In the future make this a configurable link back (see work on BASE_URL https://github.com/dgtlmoon/changedetection.io/pull/228)
+                guid = "{}/{}".format(watch['uuid'], watch.last_changed)
+                fe = fg.add_entry()
+
+                # Include a link to the diff page, they will have to login here to see if password protection is enabled.
+                # Description is the page you watch, link takes you to the diff JS UI page
+                # Dict val base_url will get overriden with the env var if it is set.
+                ext_base_url = datastore.data['settings']['application'].get('active_base_url')
+                # @todo fix
+
+                # Because we are called via whatever web server, flask should figure out the right path (
+                diff_link = {'href': url_for('ui.ui_views.diff_history_page', uuid=watch['uuid'], _external=True)}
+
+                fe.link(link=diff_link)
+
+                # @todo watch should be a getter - watch.get('title') (internally if URL else..)
+
+                watch_title = watch.get('title') if watch.get('title') else watch.get('url')
+                fe.title(title=watch_title)
+                try:
+
+                    html_diff = diff.render_diff(previous_version_file_contents=watch.get_history_snapshot(dates[-2]),
+                                                 newest_version_file_contents=watch.get_history_snapshot(dates[-1]),
+                                                 include_equal=False,
+                                                 line_feed_sep="<br>",
+                                                 html_colour=html_colour_enable
+                                                 )
+                except FileNotFoundError as e:
+                    html_diff = f"History snapshot file for watch {watch.get('uuid')}@{watch.last_changed} - '{watch.get('title')} not found."
+
+                # @todo Make this configurable and also consider html-colored markup
+                # @todo User could decide if <link> goes to the diff page, or to the watch link
+                rss_template = "<html><body>\n<h4><a href=\"{{watch_url}}\">{{watch_title}}</a></h4>\n<p>{{html_diff}}</p>\n</body></html>\n"
+
+                content = jinja_render(template_str=rss_template, watch_title=watch_title, html_diff=html_diff, watch_url=watch.link)
+
+                # Out of range chars could also break feedgen
+                if scan_invalid_chars_in_rss(content):
+                    content = clean_entry_content(content)
+
+                fe.content(content=content, type='CDATA')
+                fe.guid(guid, permalink=False)
+                dt = datetime.datetime.fromtimestamp(int(watch.newest_history_key))
+                dt = dt.replace(tzinfo=pytz.UTC)
+                fe.pubDate(dt)
+
+        response = make_response(fg.rss_str())
+        response.headers.set('Content-Type', 'application/rss+xml;charset=utf-8')
+        logger.trace(f"RSS generated in {time.time() - now:.3f}s")
+        return response
+
+    return rss_blueprint

changedetectionio/blueprint/settings/__init__.py

@@ -0,0 +1,145 @@
+import os
+from copy import deepcopy
+from datetime import datetime
+from zoneinfo import ZoneInfo, available_timezones
+import secrets
+import flask_login
+from flask import Blueprint, render_template, request, redirect, url_for, flash
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+
+
+def construct_blueprint(datastore: ChangeDetectionStore):
+    settings_blueprint = Blueprint('settings', __name__, template_folder="templates")
+
+    @settings_blueprint.route("", methods=['GET', "POST"])
+    @login_optionally_required
+    def settings_page():
+        from changedetectionio import forms
+
+        default = deepcopy(datastore.data['settings'])
+        if datastore.proxy_list is not None:
+            available_proxies = list(datastore.proxy_list.keys())
+            # When enabled
+            system_proxy = datastore.data['settings']['requests']['proxy']
+            # In the case it doesnt exist anymore
+            if not system_proxy in available_proxies:
+                system_proxy = None
+
+            default['requests']['proxy'] = system_proxy if system_proxy is not None else available_proxies[0]
+            # Used by the form handler to keep or remove the proxy settings
+            default['proxy_list'] = available_proxies[0]
+
+        # Don't use form.data on POST so that it doesnt overrid the checkbox status from the POST status
+        form = forms.globalSettingsForm(formdata=request.form if request.method == 'POST' else None,
+                                        data=default,
+                                        extra_notification_tokens=datastore.get_unique_notification_tokens_available()
+                                        )
+
+        # Remove the last option 'System default'
+        form.application.form.notification_format.choices.pop()
+
+        if datastore.proxy_list is None:
+            # @todo - Couldn't get setattr() etc dynamic addition working, so remove it instead
+            del form.requests.form.proxy
+        else:
+            form.requests.form.proxy.choices = []
+            for p in datastore.proxy_list:
+                form.requests.form.proxy.choices.append(tuple((p, datastore.proxy_list[p]['label'])))
+
+        if request.method == 'POST':
+            # Password unset is a GET, but we can lock the session to a salted env password to always need the password
+            if form.application.form.data.get('removepassword_button', False):
+                # SALTED_PASS means the password is "locked" to what we set in the Env var
+                if not os.getenv("SALTED_PASS", False):
+                    datastore.remove_password()
+                    flash("Password protection removed.", 'notice')
+                    flask_login.logout_user()
+                    return redirect(url_for('settings.settings_page'))
+
+            if form.validate():
+                # Don't set password to False when a password is set - should be only removed with the `removepassword` button
+                app_update = dict(deepcopy(form.data['application']))
+
+                # Never update password with '' or False (Added by wtforms when not in submission)
+                if 'password' in app_update and not app_update['password']:
+                    del (app_update['password'])
+
+                datastore.data['settings']['application'].update(app_update)
+                
+                # Handle dynamic worker count adjustment
+                old_worker_count = datastore.data['settings']['requests'].get('workers', 1)
+                new_worker_count = form.data['requests'].get('workers', 1)
+                
+                datastore.data['settings']['requests'].update(form.data['requests'])
+                
+                # Adjust worker count if it changed
+                if new_worker_count != old_worker_count:
+                    from changedetectionio import worker_handler
+                    from changedetectionio.flask_app import update_q, notification_q, app, datastore as ds
+                    
+                    result = worker_handler.adjust_async_worker_count(
+                        new_count=new_worker_count,
+                        update_q=update_q,
+                        notification_q=notification_q,
+                        app=app,
+                        datastore=ds
+                    )
+                    
+                    if result['status'] == 'success':
+                        flash(f"Worker count adjusted: {result['message']}", 'notice')
+                    elif result['status'] == 'not_supported':
+                        flash("Dynamic worker adjustment not supported for sync workers", 'warning')
+                    elif result['status'] == 'error':
+                        flash(f"Error adjusting workers: {result['message']}", 'error')
+
+                if not os.getenv("SALTED_PASS", False) and len(form.application.form.password.encrypted_password):
+                    datastore.data['settings']['application']['password'] = form.application.form.password.encrypted_password
+                    datastore.needs_write_urgent = True
+                    flash("Password protection enabled.", 'notice')
+                    flask_login.logout_user()
+                    return redirect(url_for('watchlist.index'))
+
+                datastore.needs_write_urgent = True
+                flash("Settings updated.")
+
+            else:
+                flash("An error occurred, please see below.", "error")
+
+        # Convert to ISO 8601 format, all date/time relative events stored as UTC time
+        utc_time = datetime.now(ZoneInfo("UTC")).isoformat()
+
+        output = render_template("settings.html",
+                                api_key=datastore.data['settings']['application'].get('api_access_token'),
+                                available_timezones=sorted(available_timezones()),
+                                emailprefix=os.getenv('NOTIFICATION_MAIL_BUTTON_PREFIX', False),
+                                extra_notification_token_placeholder_info=datastore.get_unique_notification_token_placeholders_available(),
+                                form=form,
+                                hide_remove_pass=os.getenv("SALTED_PASS", False),
+                                min_system_recheck_seconds=int(os.getenv('MINIMUM_SECONDS_RECHECK_TIME', 3)),
+                                settings_application=datastore.data['settings']['application'],
+                                timezone_default_config=datastore.data['settings']['application'].get('timezone'),
+                                utc_time=utc_time,
+                                )
+
+        return output
+
+    @settings_blueprint.route("/reset-api-key", methods=['GET'])
+    @login_optionally_required
+    def settings_reset_api_key():
+        secret = secrets.token_hex(16)
+        datastore.data['settings']['application']['api_access_token'] = secret
+        datastore.needs_write_urgent = True
+        flash("API Key was regenerated.")
+        return redirect(url_for('settings.settings_page')+'#api')
+        
+    @settings_blueprint.route("/notification-logs", methods=['GET'])
+    @login_optionally_required
+    def notification_logs():
+        from changedetectionio.flask_app import notification_debug_log
+        output = render_template("notification-log.html",
+                               logs=notification_debug_log if len(notification_debug_log) else ["Notification logs are empty - no notifications sent yet."])
+        return output
+
+    return settings_blueprint

changedetectionio/blueprint/settings/templates/settings.html

@@ -1,18 +1,20 @@
 {% extends 'base.html' %}
 
 {% block content %}
-{% from '_helpers.html' import render_field, render_checkbox_field, render_button %}
+{% from '_helpers.html' import render_field, render_checkbox_field, render_button, render_time_schedule_form %}
 {% from '_common_fields.html' import render_common_settings_form %}
 <script>
-    const notification_base_url="{{url_for('ajax_callback_send_notification_test', mode="global-settings")}}";
+    const notification_base_url="{{url_for('ui.ui_notification.ajax_callback_send_notification_test', mode="global-settings")}}";
 {% if emailprefix %}
     const email_notification_prefix=JSON.parse('{{emailprefix|tojson}}');
 {% endif %}
 </script>
 <script src="{{url_for('static_content', group='js', filename='tabs.js')}}" defer></script>
+<script src="{{url_for('static_content', group='js', filename='plugins.js')}}" defer></script>
 <script src="{{url_for('static_content', group='js', filename='notifications.js')}}" defer></script>
 <script src="{{url_for('static_content', group='js', filename='vis.js')}}" defer></script>
 <script src="{{url_for('static_content', group='js', filename='global-settings.js')}}" defer></script>
+<script src="{{url_for('static_content', group='js', filename='scheduler.js')}}" defer></script>
 <div class="edit-form">
     <div class="tabs collapsable">
         <ul>
@@ -20,18 +22,26 @@
             <li class="tab"><a href="#notifications">Notifications</a></li>
             <li class="tab"><a href="#fetching">Fetching</a></li>
             <li class="tab"><a href="#filters">Global Filters</a></li>
+            <li class="tab"><a href="#ui-options">UI Options</a></li>
             <li class="tab"><a href="#api">API</a></li>
+            <li class="tab"><a href="#timedate">Time &amp Date</a></li>
             <li class="tab"><a href="#proxies">CAPTCHA &amp; Proxies</a></li>
         </ul>
     </div>
     <div class="box-wrap inner">
-        <form class="pure-form pure-form-stacked settings" action="{{url_for('settings_page')}}" method="POST">
+        <form class="pure-form pure-form-stacked settings" action="{{url_for('settings.settings_page')}}" method="POST">
             <input type="hidden" name="csrf_token" value="{{ csrf_token() }}" >
             <div class="tab-pane-inner" id="general">
                 <fieldset>
                     <div class="pure-control-group">
                         {{ render_field(form.requests.form.time_between_check, class="time-check-widget") }}
                         <span class="pure-form-message-inline">Default recheck time for all watches, current system minimum is <i>{{min_system_recheck_seconds}}</i> seconds (<a href="https://github.com/dgtlmoon/changedetection.io/wiki/Misc-system-settings#enviroment-variables">more info</a>).</span>
+                            <div id="time-between-check-schedule">
+                                <!-- Start Time and End Time -->
+                                <div id="limit-between-time">
+                                    {{ render_time_schedule_form(form.requests, available_timezones, timezone_default_config) }}
+                                </div>
+                        </div>
                     </div>
                     <div class="pure-control-group">
                         {{ render_field(form.requests.form.jitter_seconds, class="jitter_seconds") }}
@@ -69,14 +79,17 @@
                         {{ render_field(form.application.form.pager_size) }}
                         <span class="pure-form-message-inline">Number of items per page in the watch overview list, 0 to disable.</span>
                     </div>
-
+                    <div class="pure-control-group">
+                        {{ render_field(form.application.form.rss_content_format) }}
+                        <span class="pure-form-message-inline">Love RSS? Does your reader support HTML? Set it here</span>
+                    </div>
                     <div class="pure-control-group">
                         {{ render_checkbox_field(form.application.form.extract_title_as_title) }}
                         <span class="pure-form-message-inline">Note: This will automatically apply to all existing watches.</span>
                     </div>
                     <div class="pure-control-group">
                         {{ render_checkbox_field(form.application.form.empty_pages_are_a_change) }}
-                        <span class="pure-form-message-inline">When a page contains HTML, but no renderable text appears (empty page), is this considered a change?</span>
+                        <span class="pure-form-message-inline">When a request returns no content, or the HTML does not contain any text, is this considered a change?</span>
                     </div>
                 {% if form.requests.proxy %}
                     <div class="pure-control-group inline-radio">
@@ -92,7 +105,7 @@
             <div class="tab-pane-inner" id="notifications">
                 <fieldset>
                     <div class="field-group">
-                        {{ render_common_settings_form(form.application.form, emailprefix, settings_application) }}
+                        {{ render_common_settings_form(form.application.form, emailprefix, settings_application, extra_notification_token_placeholder_info) }}
                     </div>
                 </fieldset>
                 <div class="pure-control-group" id="notification-base-url">
@@ -122,6 +135,12 @@
                         {{ render_field(form.application.form.webdriver_delay) }}
                     </div>
                 </fieldset>
+                <div class="pure-control-group">
+                    {{ render_field(form.requests.form.workers) }}
+                    {% set worker_info = get_worker_status_info() %}
+                    <span class="pure-form-message-inline">Number of concurrent workers to process watches. More workers = faster processing but higher memory usage.<br>
+                    Currently running: <strong>{{ worker_info.count }}</strong> operational {{ worker_info.type }} workers{% if worker_info.active_workers > 0 %} ({{ worker_info.active_workers }} actively processing){% endif %}.</span>
+                </div>
                 <div class="pure-control-group inline-radio">
                     {{ render_field(form.requests.form.default_ua) }}
                     <span class="pure-form-message-inline">
@@ -155,11 +174,13 @@
                       {{ render_field(form.application.form.global_subtractive_selectors, rows=5, placeholder="header
 footer
 nav
-.stockticker") }}
+.stockticker
+//*[contains(text(), 'Advertisement')]") }}
                       <span class="pure-form-message-inline">
                         <ul>
-                          <li> Remove HTML element(s) by CSS selector before text conversion. </li>
-                          <li> Add multiple elements or CSS selectors per line to ignore multiple parts of the HTML. </li>
+                          <li> Remove HTML element(s) by CSS and XPath selectors before text conversion. </li>
+                          <li> Don't paste HTML here, use only CSS and XPath selectors </li>
+                          <li> Add multiple elements, CSS or XPath selectors per line to ignore multiple parts of the HTML. </li>
                         </ul>
                       </span>
                     </fieldset>
@@ -170,11 +191,11 @@ nav
                     <span class="pure-form-message-inline">Note: This is applied globally in addition to the per-watch rules.</span><br>
                     <span class="pure-form-message-inline">
                         <ul>
+                            <li>Matching text will be <strong>ignored</strong> in the text snapshot (you can still see it but it wont trigger a change)</li>
                             <li>Note: This is applied globally in addition to the per-watch rules.</li>
                             <li>Each line processed separately, any line matching will be ignored (removed before creating the checksum)</li>
                             <li>Regular Expression support, wrap the entire line in forward slash <code>/regex/</code></li>
                             <li>Changing this will affect the comparison checksum which may trigger an alert</li>
-                            <li>Use the preview/show current tab to see ignores</li>
                         </ul>
                      </span>
                     </fieldset>
@@ -192,7 +213,7 @@ nav
                     </div>
                 </div>
                 <div class="pure-control-group">
-                    <a href="{{url_for('settings_reset_api_key')}}" class="pure-button button-small button-cancel">Regenerate API key</a>
+                    <a href="{{url_for('settings.settings_reset_api_key')}}" class="pure-button button-small button-cancel">Regenerate API key</a>
                 </div>
                 <div class="pure-control-group">
                     <h4>Chrome Extension</h4>
@@ -203,12 +224,39 @@ nav
                         <a id="chrome-extension-link"
                            title="Try our new Chrome Extension!"
                            href="https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop">
-                            <img src="{{ url_for('static_content', group='images', filename='Google-Chrome-icon.png') }}" alt="Chrome">
+                            <img alt="Chrome store icon" src="{{ url_for('static_content', group='images', filename='google-chrome-icon.png') }}" alt="Chrome">
                             Chrome Webstore
                         </a>
                     </p>
                 </div>
             </div>
+            <div class="tab-pane-inner" id="timedate">
+                <div class="pure-control-group">
+                    Ensure the settings below are correct, they are used to manage the time schedule for checking your web page watches.
+                </div>
+                <div class="pure-control-group">
+                    <p><strong>UTC Time &amp Date from Server:</strong> <span id="utc-time" >{{ utc_time }}</span></p>
+                    <p><strong>Local Time &amp Date in Browser:</strong> <span class="local-time" data-utc="{{ utc_time }}"></span></p>
+                    <p>
+                       {{ render_field(form.application.form.timezone) }}
+                        <datalist id="timezones" style="display: none;">
+                            {% for tz_name in available_timezones %}
+                                <option value="{{ tz_name }}">{{ tz_name }}</option>
+                            {% endfor %}
+                        </datalist>
+                    </p>
+                </div>
+            </div>
+            <div class="tab-pane-inner" id="ui-options">
+                <div class="pure-control-group">
+                    {{ render_checkbox_field(form.application.form.ui.form.open_diff_in_new_tab, class="open_diff_in_new_tab") }}
+                    <span class="pure-form-message-inline">Enable this setting to open the diff page in a new tab. If disabled, the diff page will open in the current tab.</span>
+                </div>
+                <div class="pure-control-group">
+                    {{ render_checkbox_field(form.application.form.ui.form.socket_io_enabled, class="socket_io_enabled") }}
+                    <span class="pure-form-message-inline">Realtime UI Updates Enabled - (Restart required if this is changed)</span>
+                </div>
+            </div>
             <div class="tab-pane-inner" id="proxies">
                 <div id="recommended-proxy">
                     <div>
@@ -252,9 +300,7 @@ nav
                         
                     </div>
                 </div>
-                <p>
-                    Your proxy provider may need to whitelist our IP of <code>204.15.192.195</code>
-                </p>
+
                <p><strong>Tip</strong>: "Residential" and "Mobile" proxy type can be more successfull than "Data Center" for blocked websites.
 
                 <div class="pure-control-group" id="extra-proxies-setting">
@@ -273,8 +319,8 @@ nav
             <div id="actions">
                 <div class="pure-control-group">
                     {{ render_button(form.save_button) }}
-                    <a href="{{url_for('index')}}" class="pure-button button-small button-cancel">Back</a>
-                    <a href="{{url_for('clear_all_history')}}" class="pure-button button-small button-cancel">Clear Snapshot History</a>
+                    <a href="{{url_for('watchlist.index')}}" class="pure-button button-small button-cancel">Back</a>
+                    <a href="{{url_for('ui.clear_all_history')}}" class="pure-button button-small button-error">Clear Snapshot History</a>
                 </div>
             </div>
         </form>

changedetectionio/blueprint/tags/__init__.py

@@ -1,4 +1,6 @@
-from flask import Blueprint, request, make_response, render_template, flash, url_for, redirect
+from flask import Blueprint, request, render_template, flash, url_for, redirect
+
+
 from changedetectionio.store import ChangeDetectionStore
 from changedetectionio.flask_app import login_optionally_required
 
@@ -11,6 +13,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     def tags_overview_page():
         from .form import SingleTag
         add_form = SingleTag(request.form)
+
         sorted_tags = sorted(datastore.data['settings']['application'].get('tags').items(), key=lambda x: x[1]['title'])
 
         from collections import Counter
@@ -96,22 +99,60 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     @tags_blueprint.route("/edit/<string:uuid>", methods=['GET'])
     @login_optionally_required
     def form_tag_edit(uuid):
-        from changedetectionio import forms
-
+        from changedetectionio.blueprint.tags.form import group_restock_settings_form
         if uuid == 'first':
             uuid = list(datastore.data['settings']['application']['tags'].keys()).pop()
 
         default = datastore.data['settings']['application']['tags'].get(uuid)
+        if not default:
+            flash("Tag not found", "error")
+            return redirect(url_for('watchlist.index'))
 
-        form = forms.watchForm(formdata=request.form if request.method == 'POST' else None,
-                               data=default,
-                               )
-        form.datastore=datastore # needed?
+        form = group_restock_settings_form(
+                                       formdata=request.form if request.method == 'POST' else None,
+                                       data=default,
+                                       extra_notification_tokens=datastore.get_unique_notification_tokens_available(),
+                                       default_system_settings = datastore.data['settings'],
+                                       )
+
+        template_args = {
+            'data': default,
+            'form': form,
+            'watch': default,
+            'extra_notification_token_placeholder_info': datastore.get_unique_notification_token_placeholders_available(),
+        }
+
+        included_content = {}
+        if form.extra_form_content():
+            # So that the extra panels can access _helpers.html etc, we set the environment to load from templates/
+            # And then render the code from the module
+            from jinja2 import Environment, FileSystemLoader
+            import importlib.resources
+            templates_dir = str(importlib.resources.files("changedetectionio").joinpath('templates'))
+            env = Environment(loader=FileSystemLoader(templates_dir))
+            template_str = """{% from '_helpers.html' import render_field, render_checkbox_field, render_button %}
+        <script>        
+            $(document).ready(function () {
+                toggleOpacity('#overrides_watch', '#restock-fieldset-price-group', true);
+            });
+        </script>            
+                <fieldset>
+                    <div class="pure-control-group">
+                        <fieldset class="pure-group">
+                        {{ render_checkbox_field(form.overrides_watch) }}
+                        <span class="pure-form-message-inline">Used for watches in "Restock & Price detection" mode</span>
+                        </fieldset>
+                </fieldset>
+                """
+            template_str += form.extra_form_content()
+            template = env.from_string(template_str)
+            included_content = template.render(**template_args)
 
         output = render_template("edit-tag.html",
-                                 data=default,
-                                 form=form,
                                  settings_application=datastore.data['settings']['application'],
+                                 extra_tab_content=form.extra_tab_content() if form.extra_tab_content() else None,
+                                 extra_form_content=included_content,
+                                 **template_args
                                  )
 
         return output
@@ -120,14 +161,15 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     @tags_blueprint.route("/edit/<string:uuid>", methods=['POST'])
     @login_optionally_required
     def form_tag_edit_submit(uuid):
-        from changedetectionio import forms
+        from changedetectionio.blueprint.tags.form import group_restock_settings_form
         if uuid == 'first':
             uuid = list(datastore.data['settings']['application']['tags'].keys()).pop()
 
         default = datastore.data['settings']['application']['tags'].get(uuid)
 
-        form = forms.watchForm(formdata=request.form if request.method == 'POST' else None,
+        form = group_restock_settings_form(formdata=request.form if request.method == 'POST' else None,
                                data=default,
+                               extra_notification_tokens=datastore.get_unique_notification_tokens_available()
                                )
         # @todo subclass form so validation works
         #if not form.validate():
@@ -136,6 +178,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
 #           return redirect(url_for('tags.form_tag_edit_submit', uuid=uuid))
 
         datastore.data['settings']['application']['tags'][uuid].update(form.data)
+        datastore.data['settings']['application']['tags'][uuid]['processor'] = 'restock_diff'
         datastore.needs_write_urgent = True
         flash("Updated")
 

changedetectionio/blueprint/tags/form.py

@@ -1,16 +1,15 @@
 from wtforms import (
-    BooleanField,
     Form,
-    IntegerField,
-    RadioField,
-    SelectField,
     StringField,
     SubmitField,
-    TextAreaField,
     validators,
 )
+from wtforms.fields.simple import BooleanField
 
+from changedetectionio.processors.restock_diff.forms import processor_settings_form as restock_settings_form
 
+class group_restock_settings_form(restock_settings_form):
+    overrides_watch = BooleanField('Activate for individual watches in this tag/group?', default=False)
 
 class SingleTag(Form):
 

changedetectionio/blueprint/tags/templates/edit-tag.html

@@ -3,7 +3,7 @@
 {% from '_helpers.html' import render_field, render_checkbox_field, render_button %}
 {% from '_common_fields.html' import render_common_settings_form %}
 <script>
-    const notification_base_url="{{url_for('ajax_callback_send_notification_test', mode="group-settings")}}";
+    const notification_base_url="{{url_for('ui.ui_notification.ajax_callback_send_notification_test', mode="group-settings")}}";
 </script>
 
 <script src="{{url_for('static_content', group='js', filename='tabs.js')}}" defer></script>
@@ -13,11 +13,11 @@
     /*const email_notification_prefix=JSON.parse('{{ emailprefix|tojson }}');*/
 /*{% endif %}*/
 
+{% set has_tag_filters_extra='' %}
 
 </script>
 
 <script src="{{url_for('static_content', group='js', filename='watch-settings.js')}}" defer></script>
-<!--<script src="{{url_for('static_content', group='js', filename='limit.js')}}" defer></script>-->
 <script src="{{url_for('static_content', group='js', filename='notifications.js')}}" defer></script>
 
 <div class="edit-form monospaced-textarea">
@@ -26,6 +26,9 @@
         <ul>
             <li class="tab" id=""><a href="#general">General</a></li>
             <li class="tab"><a href="#filters-and-triggers">Filters &amp; Triggers</a></li>
+            {% if extra_tab_content %}
+            <li class="tab"><a href="#extras_tab">{{ extra_tab_content }}</a></li>
+            {% endif %}
             <li class="tab"><a href="#notifications">Notifications</a></li>
         </ul>
     </div>
@@ -44,65 +47,26 @@
             </div>
 
             <div class="tab-pane-inner" id="filters-and-triggers">
-                    <div class="pure-control-group">
-                        {% set field = render_field(form.include_filters,
-                            rows=5,
-                            placeholder="#example
-xpath://body/div/span[contains(@class, 'example-class')]",
-                            class="m-d")
-                        %}
-                        {{ field }}
-                        {% if '/text()' in  field %}
-                          <span class="pure-form-message-inline"><strong>Note!: //text() function does not work where the &lt;element&gt; contains &lt;![CDATA[]]&gt;</strong></span><br>
-                        {% endif %}
-                        <span class="pure-form-message-inline">One rule per line, <i>any</i> rules that matches will be used.<br>
-
-                    <ul>
-                        <li>CSS - Limit text to this CSS rule, only text matching this CSS rule is included.</li>
-                        <li>JSON - Limit text to this JSON rule, using either <a href="https://pypi.org/project/jsonpath-ng/" target="new">JSONPath</a> or <a href="https://stedolan.github.io/jq/" target="new">jq</a> (if installed).
-                            <ul>
-                                <li>JSONPath: Prefix with <code>json:</code>, use <code>json:$</code> to force re-formatting if required,  <a href="https://jsonpath.com/" target="new">test your JSONPath here</a>.</li>
-                                {% if jq_support %}
-                                <li>jq: Prefix with <code>jq:</code> and <a href="https://jqplay.org/" target="new">test your jq here</a>. Using <a href="https://stedolan.github.io/jq/" target="new">jq</a> allows for complex filtering and processing of JSON data with built-in functions, regex, filtering, and more. See examples and documentation <a href="https://stedolan.github.io/jq/manual/" target="new">here</a>. Prefix <code>jqraw:</code> outputs the results as text instead of a JSON list.</li>
-                                {% else %}
-                                <li>jq support not installed</li>
-                                {% endif %}
-                            </ul>
-                        </li>
-                        <li>XPath - Limit text to this XPath rule, simply start with a forward-slash. To specify XPath to be used explicitly or the XPath rule starts with an XPath function: Prefix with <code>xpath:</code>
-                            <ul>
-                                <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath:count(//*[contains(@class, 'sametext')])</code>, <a
-                                href="http://xpather.com/" target="new">test your XPath here</a></li>
-                                <li>Example: Get all titles from an RSS feed <code>//title/text()</code></li>
-                                <li>To use XPath1.0: Prefix with <code>xpath1:</code></li>
-                            </ul>
-                            </li>
-                    </ul>
-                    Please be sure that you thoroughly understand how to write CSS, JSONPath, XPath{% if jq_support %}, or jq selector{%endif%} rules before filing an issue on GitHub! <a
-                                href="https://github.com/dgtlmoon/changedetection.io/wiki/CSS-Selector-help">here for more CSS selector help</a>.<br>
-                </span>
-                    </div>
-                <fieldset class="pure-control-group">
-                    {{ render_field(form.subtractive_selectors, rows=5, placeholder="header
-footer
-nav
-.stockticker") }}
-                    <span class="pure-form-message-inline">
-                        <ul>
-                          <li> Remove HTML element(s) by CSS selector before text conversion. </li>
-                          <li> Add multiple elements or CSS selectors per line to ignore multiple parts of the HTML. </li>
-                        </ul>
-                      </span>
-                </fieldset>
-
+                <p>These settings are <strong><i>added</i></strong> to any existing watch configurations.</p>
+                {% include "edit/include_subtract.html" %}
+                <div class="text-filtering border-fieldset">
+                    <h3>Text filtering</h3>
+                    {% include "edit/text-options.html" %}
+                </div>
             </div>
 
+        {# rendered sub Template #}
+        {% if extra_form_content %}
+            <div class="tab-pane-inner" id="extras_tab">
+            {{ extra_form_content|safe }}
+            </div>
+        {% endif %}
             <div class="tab-pane-inner" id="notifications">
                 <fieldset>
                     <div  class="pure-control-group inline-radio">
                       {{ render_checkbox_field(form.notification_muted) }}
                     </div>
-                    {% if is_html_webdriver %}
+                    {% if 1 %}
                     <div class="pure-control-group inline-radio">
                       {{ render_checkbox_field(form.notification_screenshot) }}
                         <span class="pure-form-message-inline">
@@ -114,12 +78,12 @@ nav
                         {% if has_default_notification_urls %}
                         <div class="inline-warning">
                             <img class="inline-warning-icon" src="{{url_for('static_content', group='images', filename='notice.svg')}}" alt="Look out!" title="Lookout!" >
-                            There are <a href="{{ url_for('settings_page')}}#notifications">system-wide notification URLs enabled</a>, this form will override notification settings for this watch only &dash; an empty Notification URL list here will still send notifications.
+                            There are <a href="{{ url_for('settings.settings_page')}}#notifications">system-wide notification URLs enabled</a>, this form will override notification settings for this watch only &dash; an empty Notification URL list here will still send notifications.
                         </div>
                         {% endif %}
                         <a href="#notifications" id="notification-setting-reset-to-default" class="pure-button button-xsmall" style="right: 20px; top: 20px; position: absolute; background-color: #5f42dd; border-radius: 4px; font-size: 70%; color: #fff">Use system defaults</a>
 
-                        {{ render_common_settings_form(form, emailprefix, settings_application) }}
+                        {{ render_common_settings_form(form, emailprefix, settings_application, extra_notification_token_placeholder_info) }}
                     </div>
                 </fieldset>
             </div>

changedetectionio/blueprint/tags/templates/groups-overview.html

@@ -10,7 +10,7 @@
             <legend>Add a new organisational tag</legend>
             <div id="watch-add-wrapper-zone">
                 <div>
-                    {{ render_simple_field(form.name, placeholder="watch label / tag") }}
+                    {{ render_simple_field(form.name, placeholder="Watch group / tag") }}
                 </div>
                 <div>
                     {{ render_simple_field(form.save_button, title="Save" ) }}
@@ -47,7 +47,7 @@
                     <a class="link-mute state-{{'on' if tag.notification_muted else 'off'}}" href="{{url_for('tags.mute', uuid=tag.uuid)}}"><img src="{{url_for('static_content', group='images', filename='bell-off.svg')}}" alt="Mute notifications" title="Mute notifications" class="icon icon-mute" ></a>
                 </td>
                 <td>{{ "{:,}".format(tag_count[uuid]) if uuid in tag_count else 0 }}</td>
-                <td class="title-col inline"> <a href="{{url_for('index', tag=uuid) }}">{{ tag.title }}</a></td>
+                <td class="title-col inline"> <a href="{{url_for('watchlist.index', tag=uuid) }}">{{ tag.title }}</a></td>
                 <td>
                     <a class="pure-button pure-button-primary" href="{{ url_for('tags.form_tag_edit', uuid=uuid) }}">Edit</a>&nbsp;
                     <a class="pure-button pure-button-primary" href="{{ url_for('tags.delete', uuid=uuid) }}" title="Deletes and removes tag">Delete</a>

changedetectionio/blueprint/ui/__init__.py

@@ -0,0 +1,325 @@
+import time
+from flask import Blueprint, request, redirect, url_for, flash, render_template, session
+from loguru import logger
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.blueprint.ui.edit import construct_blueprint as construct_edit_blueprint
+from changedetectionio.blueprint.ui.notification import construct_blueprint as construct_notification_blueprint
+from changedetectionio.blueprint.ui.views import construct_blueprint as construct_views_blueprint
+
+def _handle_operations(op, uuids, datastore, worker_handler, update_q, queuedWatchMetaData, watch_check_update, extra_data=None, emit_flash=True):
+    from flask import request, flash
+
+    if op == 'delete':
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.delete(uuid)
+        if emit_flash:
+            flash(f"{len(uuids)} watches deleted")
+
+    elif op == 'pause':
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid]['paused'] = True
+        if emit_flash:
+            flash(f"{len(uuids)} watches paused")
+
+    elif op == 'unpause':
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid.strip()]['paused'] = False
+        if emit_flash:
+            flash(f"{len(uuids)} watches unpaused")
+
+    elif (op == 'mark-viewed'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.set_last_viewed(uuid, int(time.time()))
+        if emit_flash:
+            flash(f"{len(uuids)} watches updated")
+
+    elif (op == 'mute'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid]['notification_muted'] = True
+        if emit_flash:
+            flash(f"{len(uuids)} watches muted")
+
+    elif (op == 'unmute'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid]['notification_muted'] = False
+        if emit_flash:
+            flash(f"{len(uuids)} watches un-muted")
+
+    elif (op == 'recheck'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                # Recheck and require a full reprocessing
+                worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+        if emit_flash:
+            flash(f"{len(uuids)} watches queued for rechecking")
+
+    elif (op == 'clear-errors'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid]["last_error"] = False
+        if emit_flash:
+            flash(f"{len(uuids)} watches errors cleared")
+
+    elif (op == 'clear-history'):
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.clear_watch_history(uuid)
+        if emit_flash:
+            flash(f"{len(uuids)} watches cleared/reset.")
+
+    elif (op == 'notification-default'):
+        from changedetectionio.notification import (
+            default_notification_format_for_watch
+        )
+        for uuid in uuids:
+            if datastore.data['watching'].get(uuid):
+                datastore.data['watching'][uuid]['notification_title'] = None
+                datastore.data['watching'][uuid]['notification_body'] = None
+                datastore.data['watching'][uuid]['notification_urls'] = []
+                datastore.data['watching'][uuid]['notification_format'] = default_notification_format_for_watch
+        if emit_flash:
+            flash(f"{len(uuids)} watches set to use default notification settings")
+
+    elif (op == 'assign-tag'):
+        op_extradata = extra_data
+        if op_extradata:
+            tag_uuid = datastore.add_tag(title=op_extradata)
+            if op_extradata and tag_uuid:
+                for uuid in uuids:
+                    if datastore.data['watching'].get(uuid):
+                        # Bug in old versions caused by bad edit page/tag handler
+                        if isinstance(datastore.data['watching'][uuid]['tags'], str):
+                            datastore.data['watching'][uuid]['tags'] = []
+
+                        datastore.data['watching'][uuid]['tags'].append(tag_uuid)
+        if emit_flash:
+            flash(f"{len(uuids)} watches were tagged")
+
+    if uuids:
+        for uuid in uuids:
+            watch_check_update.send(watch_uuid=uuid)
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q, worker_handler, queuedWatchMetaData, watch_check_update):
+    ui_blueprint = Blueprint('ui', __name__, template_folder="templates")
+    
+    # Register the edit blueprint
+    edit_blueprint = construct_edit_blueprint(datastore, update_q, queuedWatchMetaData)
+    ui_blueprint.register_blueprint(edit_blueprint)
+    
+    # Register the notification blueprint
+    notification_blueprint = construct_notification_blueprint(datastore)
+    ui_blueprint.register_blueprint(notification_blueprint)
+    
+    # Register the views blueprint
+    views_blueprint = construct_views_blueprint(datastore, update_q, queuedWatchMetaData, watch_check_update)
+    ui_blueprint.register_blueprint(views_blueprint)
+
+    # Import the login decorator
+    from changedetectionio.auth_decorator import login_optionally_required
+
+    @ui_blueprint.route("/clear_history/<string:uuid>", methods=['GET'])
+    @login_optionally_required
+    def clear_watch_history(uuid):
+        try:
+            datastore.clear_watch_history(uuid)
+        except KeyError:
+            flash('Watch not found', 'error')
+        else:
+            flash("Cleared snapshot history for watch {}".format(uuid))
+        return redirect(url_for('watchlist.index'))
+
+    @ui_blueprint.route("/clear_history", methods=['GET', 'POST'])
+    @login_optionally_required
+    def clear_all_history():
+        if request.method == 'POST':
+            confirmtext = request.form.get('confirmtext')
+
+            if confirmtext == 'clear':
+                for uuid in datastore.data['watching'].keys():
+                    datastore.clear_watch_history(uuid)
+                flash("Cleared snapshot history for all watches")
+            else:
+                flash('Incorrect confirmation text.', 'error')
+
+            return redirect(url_for('watchlist.index'))
+
+        output = render_template("clear_all_history.html")
+        return output
+
+    # Clear all statuses, so we do not see the 'unviewed' class
+    @ui_blueprint.route("/form/mark-all-viewed", methods=['GET'])
+    @login_optionally_required
+    def mark_all_viewed():
+        # Save the current newest history as the most recently viewed
+        with_errors = request.args.get('with_errors') == "1"
+        tag_limit = request.args.get('tag')
+        logger.debug(f"Limiting to tag {tag_limit}")
+        now = int(time.time())
+        for watch_uuid, watch in datastore.data['watching'].items():
+            if with_errors and not watch.get('last_error'):
+                continue
+
+            if tag_limit and ( not watch.get('tags') or tag_limit not in watch['tags'] ):
+                logger.debug(f"Skipping watch {watch_uuid}")
+                continue
+
+            datastore.set_last_viewed(watch_uuid, now)
+
+        return redirect(url_for('watchlist.index', tag=tag_limit))
+
+    @ui_blueprint.route("/delete", methods=['GET'])
+    @login_optionally_required
+    def form_delete():
+        uuid = request.args.get('uuid')
+
+        if uuid != 'all' and not uuid in datastore.data['watching'].keys():
+            flash('The watch by UUID {} does not exist.'.format(uuid), 'error')
+            return redirect(url_for('watchlist.index'))
+
+        # More for testing, possible to return the first/only
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+        datastore.delete(uuid)
+        flash('Deleted.')
+
+        return redirect(url_for('watchlist.index'))
+
+    @ui_blueprint.route("/clone", methods=['GET'])
+    @login_optionally_required
+    def form_clone():
+        uuid = request.args.get('uuid')
+        # More for testing, possible to return the first/only
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+
+        new_uuid = datastore.clone(uuid)
+
+        if not datastore.data['watching'].get(uuid).get('paused'):
+            worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=5, item={'uuid': new_uuid}))
+
+        flash('Cloned, you are editing the new watch.')
+
+        return redirect(url_for("ui.ui_edit.edit_page", uuid=new_uuid))
+
+    @ui_blueprint.route("/checknow", methods=['GET'])
+    @login_optionally_required
+    def form_watch_checknow():
+        # Forced recheck will skip the 'skip if content is the same' rule (, 'reprocess_existing_data': True})))
+        tag = request.args.get('tag')
+        uuid = request.args.get('uuid')
+        with_errors = request.args.get('with_errors') == "1"
+
+        i = 0
+
+        running_uuids = worker_handler.get_running_uuids()
+
+        if uuid:
+            if uuid not in running_uuids:
+                worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+                i += 1
+
+        else:
+            # Recheck all, including muted
+            # Get most overdue first
+            for k in sorted(datastore.data['watching'].items(), key=lambda item: item[1].get('last_checked', 0)):
+                watch_uuid = k[0]
+                watch = k[1]
+                if not watch['paused']:
+                    if watch_uuid not in running_uuids:
+                        if with_errors and not watch.get('last_error'):
+                            continue
+
+                        if tag != None and tag not in watch['tags']:
+                            continue
+
+                        worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': watch_uuid}))
+                        i += 1
+
+        if i == 1:
+            flash("Queued 1 watch for rechecking.")
+        if i > 1:
+            flash(f"Queued {i} watches for rechecking.")
+        if i == 0:
+            flash("No watches available to recheck.")
+
+        return redirect(url_for('watchlist.index'))
+
+    @ui_blueprint.route("/form/checkbox-operations", methods=['POST'])
+    @login_optionally_required
+    def form_watch_list_checkbox_operations():
+        op = request.form['op']
+        uuids = [u.strip() for u in request.form.getlist('uuids') if u]
+        extra_data = request.form.get('op_extradata', '').strip()
+        _handle_operations(
+            datastore=datastore,
+            extra_data=extra_data,
+            queuedWatchMetaData=queuedWatchMetaData,
+            uuids=uuids,
+            worker_handler=worker_handler,
+            update_q=update_q,
+            watch_check_update=watch_check_update,
+            op=op,
+        )
+
+        return redirect(url_for('watchlist.index'))
+
+
+    @ui_blueprint.route("/share-url/<string:uuid>", methods=['GET'])
+    @login_optionally_required
+    def form_share_put_watch(uuid):
+        """Given a watch UUID, upload the info and return a share-link
+           the share-link can be imported/added"""
+        import requests
+        import json
+        from copy import deepcopy
+
+        # more for testing
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+
+        # copy it to memory as trim off what we dont need (history)
+        watch = deepcopy(datastore.data['watching'].get(uuid))
+        # For older versions that are not a @property
+        if (watch.get('history')):
+            del (watch['history'])
+
+        # for safety/privacy
+        for k in list(watch.keys()):
+            if k.startswith('notification_'):
+                del watch[k]
+
+        for r in['uuid', 'last_checked', 'last_changed']:
+            if watch.get(r):
+                del (watch[r])
+
+        # Add the global stuff which may have an impact
+        watch['ignore_text'] += datastore.data['settings']['application']['global_ignore_text']
+        watch['subtractive_selectors'] += datastore.data['settings']['application']['global_subtractive_selectors']
+
+        watch_json = json.dumps(watch)
+
+        try:
+            r = requests.request(method="POST",
+                                 data={'watch': watch_json},
+                                 url="https://changedetection.io/share/share",
+                                 headers={'App-Guid': datastore.data['app_guid']})
+            res = r.json()
+
+            # Add to the flask session
+            session['share-link'] = f"https://changedetection.io/share/{res['share_key']}"
+
+
+        except Exception as e:
+            logger.error(f"Error sharing -{str(e)}")
+            flash(f"Could not share, something went wrong while communicating with the share server - {str(e)}", 'error')
+
+        return redirect(url_for('watchlist.index'))
+
+    return ui_blueprint

changedetectionio/blueprint/ui/edit.py

@@ -0,0 +1,339 @@
+import time
+from copy import deepcopy
+import os
+import importlib.resources
+from flask import Blueprint, request, redirect, url_for, flash, render_template, make_response, send_from_directory, abort
+from loguru import logger
+from jinja2 import Environment, FileSystemLoader
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+from changedetectionio.time_handler import is_within_schedule
+from changedetectionio import worker_handler
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMetaData):
+    edit_blueprint = Blueprint('ui_edit', __name__, template_folder="../ui/templates")
+    
+    def _watch_has_tag_options_set(watch):
+        """This should be fixed better so that Tag is some proper Model, a tag is just a Watch also"""
+        for tag_uuid, tag in datastore.data['settings']['application'].get('tags', {}).items():
+            if tag_uuid in watch.get('tags', []) and (tag.get('include_filters') or tag.get('subtractive_selectors')):
+                return True
+
+    @edit_blueprint.route("/edit/<string:uuid>", methods=['GET', 'POST'])
+    @login_optionally_required
+    # https://stackoverflow.com/questions/42984453/wtforms-populate-form-with-data-if-data-exists
+    # https://wtforms.readthedocs.io/en/3.0.x/forms/#wtforms.form.Form.populate_obj ?
+    def edit_page(uuid):
+        from changedetectionio import forms
+        from changedetectionio.blueprint.browser_steps.browser_steps import browser_step_ui_config
+        from changedetectionio import processors
+        import importlib
+
+        # More for testing, possible to return the first/only
+        if not datastore.data['watching'].keys():
+            flash("No watches to edit", "error")
+            return redirect(url_for('watchlist.index'))
+
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+
+        if not uuid in datastore.data['watching']:
+            flash("No watch with the UUID %s found." % (uuid), "error")
+            return redirect(url_for('watchlist.index'))
+
+        switch_processor = request.args.get('switch_processor')
+        if switch_processor:
+            for p in processors.available_processors():
+                if p[0] == switch_processor:
+                    datastore.data['watching'][uuid]['processor'] = switch_processor
+                    flash(f"Switched to mode - {p[1]}.")
+                    datastore.clear_watch_history(uuid)
+                    redirect(url_for('ui_edit.edit_page', uuid=uuid))
+
+        # be sure we update with a copy instead of accidently editing the live object by reference
+        default = deepcopy(datastore.data['watching'][uuid])
+
+        # Defaults for proxy choice
+        if datastore.proxy_list is not None:  # When enabled
+            # @todo
+            # Radio needs '' not None, or incase that the chosen one no longer exists
+            if default['proxy'] is None or not any(default['proxy'] in tup for tup in datastore.proxy_list):
+                default['proxy'] = ''
+        # proxy_override set to the json/text list of the items
+
+        # Does it use some custom form? does one exist?
+        processor_name = datastore.data['watching'][uuid].get('processor', '')
+        processor_classes = next((tpl for tpl in processors.find_processors() if tpl[1] == processor_name), None)
+        if not processor_classes:
+            flash(f"Cannot load the edit form for processor/plugin '{processor_classes[1]}', plugin missing?", 'error')
+            return redirect(url_for('watchlist.index'))
+
+        parent_module = processors.get_parent_module(processor_classes[0])
+
+        try:
+            # Get the parent of the "processor.py" go up one, get the form (kinda spaghetti but its reusing existing code)
+            forms_module = importlib.import_module(f"{parent_module.__name__}.forms")
+            # Access the 'processor_settings_form' class from the 'forms' module
+            form_class = getattr(forms_module, 'processor_settings_form')
+        except ModuleNotFoundError as e:
+            # .forms didnt exist
+            form_class = forms.processor_text_json_diff_form
+        except AttributeError as e:
+            # .forms exists but no useful form
+            form_class = forms.processor_text_json_diff_form
+
+        form = form_class(formdata=request.form if request.method == 'POST' else None,
+                          data=default,
+                          extra_notification_tokens=default.extra_notification_token_values(),
+                          default_system_settings=datastore.data['settings']
+                          )
+
+        # For the form widget tag UUID back to "string name" for the field
+        form.tags.datastore = datastore
+
+        # Used by some forms that need to dig deeper
+        form.datastore = datastore
+        form.watch = default
+
+        for p in datastore.extra_browsers:
+            form.fetch_backend.choices.append(p)
+
+        form.fetch_backend.choices.append(("system", 'System settings default'))
+
+        # form.browser_steps[0] can be assumed that we 'goto url' first
+
+        if datastore.proxy_list is None:
+            # @todo - Couldn't get setattr() etc dynamic addition working, so remove it instead
+            del form.proxy
+        else:
+            form.proxy.choices = [('', 'Default')]
+            for p in datastore.proxy_list:
+                form.proxy.choices.append(tuple((p, datastore.proxy_list[p]['label'])))
+
+
+        if request.method == 'POST' and form.validate():
+
+            # If they changed processor, it makes sense to reset it.
+            if datastore.data['watching'][uuid].get('processor') != form.data.get('processor'):
+                datastore.data['watching'][uuid].clear_watch()
+                flash("Reset watch history due to change of processor")
+
+            extra_update_obj = {
+                'consecutive_filter_failures': 0,
+                'last_error' : False
+            }
+
+            if request.args.get('unpause_on_save'):
+                extra_update_obj['paused'] = False
+
+            extra_update_obj['time_between_check'] = form.time_between_check.data
+
+             # Ignore text
+            form_ignore_text = form.ignore_text.data
+            datastore.data['watching'][uuid]['ignore_text'] = form_ignore_text
+
+            # Be sure proxy value is None
+            if datastore.proxy_list is not None and form.data['proxy'] == '':
+                extra_update_obj['proxy'] = None
+
+            # Unsetting all filter_text methods should make it go back to default
+            # This particularly affects tests running
+            if 'filter_text_added' in form.data and not form.data.get('filter_text_added') \
+                    and 'filter_text_replaced' in form.data and not form.data.get('filter_text_replaced') \
+                    and 'filter_text_removed' in form.data and not form.data.get('filter_text_removed'):
+                extra_update_obj['filter_text_added'] = True
+                extra_update_obj['filter_text_replaced'] = True
+                extra_update_obj['filter_text_removed'] = True
+
+            # Because wtforms doesn't support accessing other data in process_ , but we convert the CSV list of tags back to a list of UUIDs
+            tag_uuids = []
+            if form.data.get('tags'):
+                # Sometimes in testing this can be list, dont know why
+                if type(form.data.get('tags')) == list:
+                    extra_update_obj['tags'] = form.data.get('tags')
+                else:
+                    for t in form.data.get('tags').split(','):
+                        tag_uuids.append(datastore.add_tag(title=t))
+                    extra_update_obj['tags'] = tag_uuids
+
+            datastore.data['watching'][uuid].update(form.data)
+            datastore.data['watching'][uuid].update(extra_update_obj)
+
+            if not datastore.data['watching'][uuid].get('tags'):
+                # Force it to be a list, because form.data['tags'] will be string if nothing found
+                # And del(form.data['tags'] ) wont work either for some reason
+                datastore.data['watching'][uuid]['tags'] = []
+
+            # Recast it if need be to right data Watch handler
+            watch_class = processors.get_custom_watch_obj_for_processor(form.data.get('processor'))
+            datastore.data['watching'][uuid] = watch_class(datastore_path=datastore.datastore_path, default=datastore.data['watching'][uuid])
+            flash("Updated watch - unpaused!" if request.args.get('unpause_on_save') else "Updated watch.")
+
+            # Re #286 - We wait for syncing new data to disk in another thread every 60 seconds
+            # But in the case something is added we should save straight away
+            datastore.needs_write_urgent = True
+
+            # Do not queue on edit if its not within the time range
+
+            # @todo maybe it should never queue anyway on edit...
+            is_in_schedule = True
+            watch = datastore.data['watching'].get(uuid)
+
+            if watch.get('time_between_check_use_default'):
+                time_schedule_limit = datastore.data['settings']['requests'].get('time_schedule_limit', {})
+            else:
+                time_schedule_limit = watch.get('time_schedule_limit')
+
+            tz_name = time_schedule_limit.get('timezone')
+            if not tz_name:
+                tz_name = datastore.data['settings']['application'].get('timezone', 'UTC')
+
+            if time_schedule_limit and time_schedule_limit.get('enabled'):
+                try:
+                    is_in_schedule = is_within_schedule(time_schedule_limit=time_schedule_limit,
+                                                      default_tz=tz_name
+                                                      )
+                except Exception as e:
+                    logger.error(
+                        f"{uuid} - Recheck scheduler, error handling timezone, check skipped - TZ name '{tz_name}' - {str(e)}")
+                    return False
+
+            #############################
+            if not datastore.data['watching'][uuid].get('paused') and is_in_schedule:
+                # Queue the watch for immediate recheck, with a higher priority
+                worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid}))
+
+            # Diff page [edit] link should go back to diff page
+            if request.args.get("next") and request.args.get("next") == 'diff':
+                return redirect(url_for('ui.ui_views.diff_history_page', uuid=uuid))
+
+            return redirect(url_for('watchlist.index', tag=request.args.get("tag",'')))
+
+        else:
+            if request.method == 'POST' and not form.validate():
+                flash("An error occurred, please see below.", "error")
+
+            # JQ is difficult to install on windows and must be manually added (outside requirements.txt)
+            jq_support = True
+            try:
+                import jq
+            except ModuleNotFoundError:
+                jq_support = False
+
+            watch = datastore.data['watching'].get(uuid)
+
+            # if system or watch is configured to need a chrome type browser
+            system_uses_webdriver = datastore.data['settings']['application']['fetch_backend'] == 'html_webdriver'
+            watch_needs_selenium_or_playwright = False
+            if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
+                watch_needs_selenium_or_playwright = True
+
+
+            from zoneinfo import available_timezones
+
+            # Only works reliably with Playwright
+
+            # Import the global plugin system
+            from changedetectionio.pluggy_interface import collect_ui_edit_stats_extras
+            
+            template_args = {
+                'available_processors': processors.available_processors(),
+                'available_timezones': sorted(available_timezones()),
+                'browser_steps_config': browser_step_ui_config,
+                'emailprefix': os.getenv('NOTIFICATION_MAIL_BUTTON_PREFIX', False),
+                'extra_notification_token_placeholder_info': datastore.get_unique_notification_token_placeholders_available(),
+                'extra_processor_config': form.extra_tab_content(),
+                'extra_title': f" - Edit - {watch.label}",
+                'form': form,
+                'has_default_notification_urls': True if len(datastore.data['settings']['application']['notification_urls']) else False,
+                'has_extra_headers_file': len(datastore.get_all_headers_in_textfile_for_watch(uuid=uuid)) > 0,
+                'has_special_tag_options': _watch_has_tag_options_set(watch=watch),
+                'jq_support': jq_support,
+                'playwright_enabled': os.getenv('PLAYWRIGHT_DRIVER_URL', False),
+                'settings_application': datastore.data['settings']['application'],
+                'system_has_playwright_configured': os.getenv('PLAYWRIGHT_DRIVER_URL'),
+                'system_has_webdriver_configured': os.getenv('WEBDRIVER_URL'),
+                'ui_edit_stats_extras': collect_ui_edit_stats_extras(watch),
+                'visual_selector_data_ready': datastore.visualselector_data_is_ready(watch_uuid=uuid),
+                'timezone_default_config': datastore.data['settings']['application'].get('timezone'),
+                'using_global_webdriver_wait': not default['webdriver_delay'],
+                'uuid': uuid,
+                'watch': watch,
+                'watch_needs_selenium_or_playwright': watch_needs_selenium_or_playwright,
+            }
+
+            included_content = None
+            if form.extra_form_content():
+                # So that the extra panels can access _helpers.html etc, we set the environment to load from templates/
+                # And then render the code from the module
+                templates_dir = str(importlib.resources.files("changedetectionio").joinpath('templates'))
+                env = Environment(loader=FileSystemLoader(templates_dir))
+                template = env.from_string(form.extra_form_content())
+                included_content = template.render(**template_args)
+
+            output = render_template("edit.html",
+                                     extra_tab_content=form.extra_tab_content() if form.extra_tab_content() else None,
+                                     extra_form_content=included_content,
+                                     **template_args
+                                     )
+
+        return output
+
+    @edit_blueprint.route("/edit/<string:uuid>/get-html", methods=['GET'])
+    @login_optionally_required
+    def watch_get_latest_html(uuid):
+        from io import BytesIO
+        from flask import send_file
+        import brotli
+
+        watch = datastore.data['watching'].get(uuid)
+        if watch and watch.history.keys() and os.path.isdir(watch.watch_data_dir):
+            latest_filename = list(watch.history.keys())[-1]
+            html_fname = os.path.join(watch.watch_data_dir, f"{latest_filename}.html.br")
+            with open(html_fname, 'rb') as f:
+                if html_fname.endswith('.br'):
+                    # Read and decompress the Brotli file
+                    decompressed_data = brotli.decompress(f.read())
+                else:
+                    decompressed_data = f.read()
+
+            buffer = BytesIO(decompressed_data)
+
+            return send_file(buffer, as_attachment=True, download_name=f"{latest_filename}.html", mimetype='text/html')
+
+        # Return a 500 error
+        abort(500)
+
+    # Ajax callback
+    @edit_blueprint.route("/edit/<string:uuid>/preview-rendered", methods=['POST'])
+    @login_optionally_required
+    def watch_get_preview_rendered(uuid):
+        '''For when viewing the "preview" of the rendered text from inside of Edit'''
+        from flask import jsonify
+        from changedetectionio.processors.text_json_diff import prepare_filter_prevew
+        result = prepare_filter_prevew(watch_uuid=uuid, form_data=request.form, datastore=datastore)
+        return jsonify(result)
+
+    @edit_blueprint.route("/highlight_submit_ignore_url", methods=['POST'])
+    @login_optionally_required
+    def highlight_submit_ignore_url():
+        import re
+        mode = request.form.get('mode')
+        selection = request.form.get('selection')
+
+        uuid = request.args.get('uuid','')
+        if datastore.data["watching"].get(uuid):
+            if mode == 'exact':
+                for l in selection.splitlines():
+                    datastore.data["watching"][uuid]['ignore_text'].append(l.strip())
+            elif mode == 'digit-regex':
+                for l in selection.splitlines():
+                    # Replace any series of numbers with a regex
+                    s = re.escape(l.strip())
+                    s = re.sub(r'[0-9]+', r'\\d+', s)
+                    datastore.data["watching"][uuid]['ignore_text'].append('/' + s + '/')
+
+        return f"<a href={url_for('ui.ui_views.preview_page', uuid=uuid)}>Click to preview</a>"
+    
+    return edit_blueprint

changedetectionio/blueprint/ui/notification.py

@@ -0,0 +1,108 @@
+from flask import Blueprint, request, make_response
+import random
+from loguru import logger
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+
+def construct_blueprint(datastore: ChangeDetectionStore):
+    notification_blueprint = Blueprint('ui_notification', __name__, template_folder="../ui/templates")
+    
+    # AJAX endpoint for sending a test
+    @notification_blueprint.route("/notification/send-test/<string:watch_uuid>", methods=['POST'])
+    @notification_blueprint.route("/notification/send-test", methods=['POST'])
+    @notification_blueprint.route("/notification/send-test/", methods=['POST'])
+    @login_optionally_required
+    def ajax_callback_send_notification_test(watch_uuid=None):
+
+        # Watch_uuid could be unset in the case it`s used in tag editor, global settings
+        import apprise
+        from changedetectionio.notification.handler import process_notification
+        from changedetectionio.notification.apprise_plugin.assets import apprise_asset
+
+        from changedetectionio.notification.apprise_plugin.custom_handlers import apprise_http_custom_handler
+
+        apobj = apprise.Apprise(asset=apprise_asset)
+
+        is_global_settings_form = request.args.get('mode', '') == 'global-settings'
+        is_group_settings_form = request.args.get('mode', '') == 'group-settings'
+
+        # Use an existing random one on the global/main settings form
+        if not watch_uuid and (is_global_settings_form or is_group_settings_form) \
+                and datastore.data.get('watching'):
+            logger.debug(f"Send test notification - Choosing random Watch {watch_uuid}")
+            watch_uuid = random.choice(list(datastore.data['watching'].keys()))
+
+        if not watch_uuid:
+            return make_response("Error: You must have atleast one watch configured for 'test notification' to work", 400)
+
+        watch = datastore.data['watching'].get(watch_uuid)
+
+        notification_urls = None
+
+        if request.form.get('notification_urls'):
+            notification_urls = request.form['notification_urls'].strip().splitlines()
+
+        if not notification_urls:
+            logger.debug("Test notification - Trying by group/tag in the edit form if available")
+            # On an edit page, we should also fire off to the tags if they have notifications
+            if request.form.get('tags') and request.form['tags'].strip():
+                for k in request.form['tags'].split(','):
+                    tag = datastore.tag_exists_by_name(k.strip())
+                    notification_urls = tag.get('notifications_urls') if tag and tag.get('notifications_urls') else None
+
+        if not notification_urls and not is_global_settings_form and not is_group_settings_form:
+            # In the global settings, use only what is typed currently in the text box
+            logger.debug("Test notification - Trying by global system settings notifications")
+            if datastore.data['settings']['application'].get('notification_urls'):
+                notification_urls = datastore.data['settings']['application']['notification_urls']
+
+        if not notification_urls:
+            return 'Error: No Notification URLs set/found'
+
+        for n_url in notification_urls:
+            if len(n_url.strip()):
+                if not apobj.add(n_url):
+                    return f'Error:  {n_url} is not a valid AppRise URL.'
+
+        try:
+            # use the same as when it is triggered, but then override it with the form test values
+            n_object = {
+                'watch_url': request.form.get('window_url', "https://changedetection.io"),
+                'notification_urls': notification_urls
+            }
+
+            # Only use if present, if not set in n_object it should use the default system value
+            if 'notification_format' in request.form and request.form['notification_format'].strip():
+                n_object['notification_format'] = request.form.get('notification_format', '').strip()
+
+            if 'notification_title' in request.form and request.form['notification_title'].strip():
+                n_object['notification_title'] = request.form.get('notification_title', '').strip()
+            elif datastore.data['settings']['application'].get('notification_title'):
+                n_object['notification_title'] = datastore.data['settings']['application'].get('notification_title')
+            else:
+                n_object['notification_title'] = "Test title"
+
+            if 'notification_body' in request.form and request.form['notification_body'].strip():
+                n_object['notification_body'] = request.form.get('notification_body', '').strip()
+            elif datastore.data['settings']['application'].get('notification_body'):
+                n_object['notification_body'] = datastore.data['settings']['application'].get('notification_body')
+            else:
+                n_object['notification_body'] = "Test body"
+
+            n_object['as_async'] = False
+            n_object.update(watch.extra_notification_token_values())
+            sent_obj = process_notification(n_object, datastore)
+
+        except Exception as e:
+            e_str = str(e)
+            # Remove this text which is not important and floods the container
+            e_str = e_str.replace(
+                "DEBUG - <class 'apprise.decorators.base.CustomNotifyPlugin.instantiate_plugin.<locals>.CustomNotifyPluginWrapper'>",
+                '')
+
+            return make_response(e_str, 400)
+
+        return 'OK - Sent test notifications'
+
+    return notification_blueprint

changedetectionio/blueprint/ui/templates/clear_all_history.html

@@ -3,7 +3,7 @@
   <div class="box-wrap inner">
     <form
       class="pure-form pure-form-stacked"
-      action="{{url_for('clear_all_history')}}"
+      action="{{url_for('ui.clear_all_history')}}"
       method="POST"
     >
       <input type="hidden" name="csrf_token" value="{{ csrf_token() }}" >
@@ -37,7 +37,7 @@
         </div>
         <br />
         <div class="pure-control-group">
-          <a href="{{url_for('index')}}" class="pure-button button-cancel"
+          <a href="{{url_for('watchlist.index')}}" class="pure-button button-cancel"
             >Cancel</a
           >
         </div>

changedetectionio/blueprint/ui/views.py

@@ -0,0 +1,221 @@
+from flask import Blueprint, request, redirect, url_for, flash, render_template, make_response, send_from_directory, abort
+from flask_login import current_user
+import os
+import time
+from copy import deepcopy
+
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+from changedetectionio import html_tools
+from changedetectionio import worker_handler
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMetaData, watch_check_update):
+    views_blueprint = Blueprint('ui_views', __name__, template_folder="../ui/templates")
+    
+    @views_blueprint.route("/preview/<string:uuid>", methods=['GET'])
+    @login_optionally_required
+    def preview_page(uuid):
+        content = []
+        versions = []
+        timestamp = None
+
+        # More for testing, possible to return the first/only
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+
+        try:
+            watch = datastore.data['watching'][uuid]
+        except KeyError:
+            flash("No history found for the specified link, bad link?", "error")
+            return redirect(url_for('watchlist.index'))
+
+        system_uses_webdriver = datastore.data['settings']['application']['fetch_backend'] == 'html_webdriver'
+        extra_stylesheets = [url_for('static_content', group='styles', filename='diff.css')]
+
+        is_html_webdriver = False
+        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
+            is_html_webdriver = True
+        triggered_line_numbers = []
+        if datastore.data['watching'][uuid].history_n == 0 and (watch.get_error_text() or watch.get_error_snapshot()):
+            flash("Preview unavailable - No fetch/check completed or triggers not reached", "error")
+        else:
+            # So prepare the latest preview or not
+            preferred_version = request.args.get('version')
+            versions = list(watch.history.keys())
+            timestamp = versions[-1]
+            if preferred_version and preferred_version in versions:
+                timestamp = preferred_version
+
+            try:
+                versions = list(watch.history.keys())
+                content = watch.get_history_snapshot(timestamp)
+
+                triggered_line_numbers = html_tools.strip_ignore_text(content=content,
+                                                                      wordlist=watch['trigger_text'],
+                                                                      mode='line numbers'
+                                                                      )
+
+            except Exception as e:
+                content.append({'line': f"File doesnt exist or unable to read timestamp {timestamp}", 'classes': ''})
+
+        output = render_template("preview.html",
+                                 content=content,
+                                 current_version=timestamp,
+                                 history_n=watch.history_n,
+                                 extra_stylesheets=extra_stylesheets,
+                                 extra_title=f" - Diff - {watch.label} @ {timestamp}",
+                                 triggered_line_numbers=triggered_line_numbers,
+                                 current_diff_url=watch['url'],
+                                 screenshot=watch.get_screenshot(),
+                                 watch=watch,
+                                 uuid=uuid,
+                                 is_html_webdriver=is_html_webdriver,
+                                 last_error=watch['last_error'],
+                                 last_error_text=watch.get_error_text(),
+                                 last_error_screenshot=watch.get_error_snapshot(),
+                                 versions=versions
+                                )
+
+        return output
+
+    @views_blueprint.route("/diff/<string:uuid>", methods=['GET', 'POST'])
+    @login_optionally_required
+    def diff_history_page(uuid):
+        from changedetectionio import forms
+
+        # More for testing, possible to return the first/only
+        if uuid == 'first':
+            uuid = list(datastore.data['watching'].keys()).pop()
+
+        extra_stylesheets = [url_for('static_content', group='styles', filename='diff.css')]
+        try:
+            watch = datastore.data['watching'][uuid]
+        except KeyError:
+            flash("No history found for the specified link, bad link?", "error")
+            return redirect(url_for('watchlist.index'))
+
+        # For submission of requesting an extract
+        extract_form = forms.extractDataForm(request.form)
+        if request.method == 'POST':
+            if not extract_form.validate():
+                flash("An error occurred, please see below.", "error")
+
+            else:
+                extract_regex = request.form.get('extract_regex').strip()
+                output = watch.extract_regex_from_all_history(extract_regex)
+                if output:
+                    watch_dir = os.path.join(datastore.datastore_path, uuid)
+                    response = make_response(send_from_directory(directory=watch_dir, path=output, as_attachment=True))
+                    response.headers['Content-type'] = 'text/csv'
+                    response.headers['Cache-Control'] = 'no-cache, no-store, must-revalidate'
+                    response.headers['Pragma'] = 'no-cache'
+                    response.headers['Expires'] = 0
+                    return response
+
+                flash('Nothing matches that RegEx', 'error')
+                redirect(url_for('ui_views.diff_history_page', uuid=uuid)+'#extract')
+
+        history = watch.history
+        dates = list(history.keys())
+
+        if len(dates) < 2:
+            flash("Not enough saved change detection snapshots to produce a report.", "error")
+            return redirect(url_for('watchlist.index'))
+
+        # Save the current newest history as the most recently viewed
+        datastore.set_last_viewed(uuid, time.time())
+
+        # Read as binary and force decode as UTF-8
+        # Windows may fail decode in python if we just use 'r' mode (chardet decode exception)
+        from_version = request.args.get('from_version')
+        from_version_index = -2  # second newest
+        if from_version and from_version in dates:
+            from_version_index = dates.index(from_version)
+        else:
+            from_version = dates[from_version_index]
+
+        try:
+            from_version_file_contents = watch.get_history_snapshot(dates[from_version_index])
+        except Exception as e:
+            from_version_file_contents = f"Unable to read to-version at index {dates[from_version_index]}.\n"
+
+        to_version = request.args.get('to_version')
+        to_version_index = -1
+        if to_version and to_version in dates:
+            to_version_index = dates.index(to_version)
+        else:
+            to_version = dates[to_version_index]
+
+        try:
+            to_version_file_contents = watch.get_history_snapshot(dates[to_version_index])
+        except Exception as e:
+            to_version_file_contents = "Unable to read to-version at index{}.\n".format(dates[to_version_index])
+
+        screenshot_url = watch.get_screenshot()
+
+        system_uses_webdriver = datastore.data['settings']['application']['fetch_backend'] == 'html_webdriver'
+
+        is_html_webdriver = False
+        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
+            is_html_webdriver = True
+
+        password_enabled_and_share_is_off = False
+        if datastore.data['settings']['application'].get('password') or os.getenv("SALTED_PASS", False):
+            password_enabled_and_share_is_off = not datastore.data['settings']['application'].get('shared_diff_access')
+
+        output = render_template("diff.html",
+                                 current_diff_url=watch['url'],
+                                 from_version=str(from_version),
+                                 to_version=str(to_version),
+                                 extra_stylesheets=extra_stylesheets,
+                                 extra_title=f" - Diff - {watch.label}",
+                                 extract_form=extract_form,
+                                 is_html_webdriver=is_html_webdriver,
+                                 last_error=watch['last_error'],
+                                 last_error_screenshot=watch.get_error_snapshot(),
+                                 last_error_text=watch.get_error_text(),
+                                 left_sticky=True,
+                                 newest=to_version_file_contents,
+                                 newest_version_timestamp=dates[-1],
+                                 password_enabled_and_share_is_off=password_enabled_and_share_is_off,
+                                 from_version_file_contents=from_version_file_contents,
+                                 to_version_file_contents=to_version_file_contents,
+                                 screenshot=screenshot_url,
+                                 uuid=uuid,
+                                 versions=dates, # All except current/last
+                                 watch_a=watch
+                                 )
+
+        return output
+
+    @views_blueprint.route("/form/add/quickwatch", methods=['POST'])
+    @login_optionally_required
+    def form_quick_watch_add():
+        from changedetectionio import forms
+        form = forms.quickWatchForm(request.form)
+
+        if not form.validate():
+            for widget, l in form.errors.items():
+                flash(','.join(l), 'error')
+            return redirect(url_for('watchlist.index'))
+
+        url = request.form.get('url').strip()
+        if datastore.url_exists(url):
+            flash(f'Warning, URL {url} already exists', "notice")
+
+        add_paused = request.form.get('edit_and_watch_submit_button') != None
+        processor = request.form.get('processor', 'text_json_diff')
+        new_uuid = datastore.add_watch(url=url, tag=request.form.get('tags').strip(), extras={'paused': add_paused, 'processor': processor})
+
+        if new_uuid:
+            if add_paused:
+                flash('Watch added in Paused state, saving will unpause.')
+                return redirect(url_for('ui.ui_edit.edit_page', uuid=new_uuid, unpause_on_save=1, tag=request.args.get('tag')))
+            else:
+                # Straight into the queue.
+                worker_handler.queue_item_async_safe(update_q, queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid}))
+                flash("Watch added.")
+
+        return redirect(url_for('watchlist.index', tag=request.args.get('tag','')))
+
+    return views_blueprint

changedetectionio/blueprint/watchlist/__init__.py

@@ -0,0 +1,112 @@
+import os
+import time
+
+from flask import Blueprint, request, make_response, render_template, redirect, url_for, flash, session
+from flask_login import current_user
+from flask_paginate import Pagination, get_page_parameter
+
+from changedetectionio import forms
+from changedetectionio.store import ChangeDetectionStore
+from changedetectionio.auth_decorator import login_optionally_required
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMetaData):
+    watchlist_blueprint = Blueprint('watchlist', __name__, template_folder="templates")
+    
+    @watchlist_blueprint.route("/", methods=['GET'])
+    @login_optionally_required
+    def index():
+        active_tag_req = request.args.get('tag', '').lower().strip()
+        active_tag_uuid = active_tag = None
+
+        # Be sure limit_tag is a uuid
+        if active_tag_req:
+            for uuid, tag in datastore.data['settings']['application'].get('tags', {}).items():
+                if active_tag_req == tag.get('title', '').lower().strip() or active_tag_req == uuid:
+                    active_tag = tag
+                    active_tag_uuid = uuid
+                    break
+
+        # Redirect for the old rss path which used the /?rss=true
+        if request.args.get('rss'):
+            return redirect(url_for('rss.feed', tag=active_tag_uuid))
+
+        op = request.args.get('op')
+        if op:
+            uuid = request.args.get('uuid')
+            if op == 'pause':
+                datastore.data['watching'][uuid].toggle_pause()
+            elif op == 'mute':
+                datastore.data['watching'][uuid].toggle_mute()
+
+            datastore.needs_write = True
+            return redirect(url_for('watchlist.index', tag = active_tag_uuid))
+
+        # Sort by last_changed and add the uuid which is usually the key..
+        sorted_watches = []
+        with_errors = request.args.get('with_errors') == "1"
+        errored_count = 0
+        search_q = request.args.get('q').strip().lower() if request.args.get('q') else False
+        for uuid, watch in datastore.data['watching'].items():
+            if with_errors and not watch.get('last_error'):
+                continue
+
+            if active_tag_uuid and not active_tag_uuid in watch['tags']:
+                    continue
+            if watch.get('last_error'):
+                errored_count += 1
+
+            if search_q:
+                if (watch.get('title') and search_q in watch.get('title').lower()) or search_q in watch.get('url', '').lower():
+                    sorted_watches.append(watch)
+                elif watch.get('last_error') and search_q in watch.get('last_error').lower():
+                    sorted_watches.append(watch)
+            else:
+                sorted_watches.append(watch)
+
+        form = forms.quickWatchForm(request.form)
+        page = request.args.get(get_page_parameter(), type=int, default=1)
+        total_count = len(sorted_watches)
+
+        pagination = Pagination(page=page,
+                                total=total_count,
+                                per_page=datastore.data['settings']['application'].get('pager_size', 50), css_framework="semantic")
+
+        sorted_tags = sorted(datastore.data['settings']['application'].get('tags').items(), key=lambda x: x[1]['title'])
+
+        output = render_template(
+            "watch-overview.html",
+            active_tag=active_tag,
+            active_tag_uuid=active_tag_uuid,
+            app_rss_token=datastore.data['settings']['application'].get('rss_access_token'),
+            datastore=datastore,
+            errored_count=errored_count,
+            form=form,
+            guid=datastore.data['app_guid'],
+            has_proxies=datastore.proxy_list,
+            has_unviewed=datastore.has_unviewed,
+            hosted_sticky=os.getenv("SALTED_PASS", False) == False,
+            now_time_server=round(time.time()),
+            pagination=pagination,
+            queued_uuids=[q_uuid.item['uuid'] for q_uuid in update_q.queue],
+            search_q=request.args.get('q', '').strip(),
+            sort_attribute=request.args.get('sort') if request.args.get('sort') else request.cookies.get('sort'),
+            sort_order=request.args.get('order') if request.args.get('order') else request.cookies.get('order'),
+            system_default_fetcher=datastore.data['settings']['application'].get('fetch_backend'),
+            tags=sorted_tags,
+            watches=sorted_watches
+        )
+
+        if session.get('share-link'):
+            del (session['share-link'])
+
+        resp = make_response(output)
+
+        # The template can run on cookie or url query info
+        if request.args.get('sort'):
+            resp.set_cookie('sort', request.args.get('sort'))
+        if request.args.get('order'):
+            resp.set_cookie('order', request.args.get('order'))
+
+        return resp
+        
+    return watchlist_blueprint

changedetectionio/blueprint/watchlist/templates/watch-overview.html

@@ -0,0 +1,234 @@
+{%- extends 'base.html' -%}
+{%- block content -%}
+{%- from '_helpers.html' import render_simple_field, render_field, render_nolabel_field, sort_by_title -%}
+<script src="{{url_for('static_content', group='js', filename='jquery-3.6.0.min.js')}}"></script>
+<script src="{{url_for('static_content', group='js', filename='watch-overview.js')}}" defer></script>
+<script>let nowtimeserver={{ now_time_server }};</script>
+<script>
+// Initialize Feather icons after the page loads
+document.addEventListener('DOMContentLoaded', function() {
+    feather.replace();
+});
+</script>
+<style>
+.checking-now .last-checked {
+    background-image: linear-gradient(to bottom, transparent 0%, rgba(0,0,0,0.05) 40%, rgba(0,0,0,0.1) 100%);
+    background-size: 0 100%;
+    background-repeat: no-repeat;
+    transition: background-size 0.9s ease
+}
+</style>
+<div class="box" id="form-quick-watch-add">
+
+    <form class="pure-form" action="{{ url_for('ui.ui_views.form_quick_watch_add', tag=active_tag_uuid) }}" method="POST" id="new-watch-form">
+        <input type="hidden" name="csrf_token" value="{{ csrf_token() }}" >
+        <fieldset>
+            <legend>Add a new web page change detection watch</legend>
+            <div id="watch-add-wrapper-zone">
+                    {{ render_nolabel_field(form.url, placeholder="https://...", required=true) }}
+                    {{ render_nolabel_field(form.watch_submit_button, title="Watch this URL!" ) }}
+                    {{ render_nolabel_field(form.edit_and_watch_submit_button, title="Edit first then Watch") }}
+            </div>
+            <div id="watch-group-tag">
+               {{ render_field(form.tags, value=active_tag.title if active_tag_uuid else '', placeholder="Watch group / tag", class="transparent-field") }}
+            </div>
+            <div id="quick-watch-processor-type">
+                {{ render_simple_field(form.processor) }}
+            </div>
+
+        </fieldset>
+        <span style="color:#eee; font-size: 80%;"><img alt="Create a shareable link" style="height: 1em;display:inline-block;" src="{{url_for('static_content', group='images', filename='spread-white.svg')}}" > Tip: You can also add 'shared' watches. <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Sharing-a-Watch">More info</a></span>
+    </form>
+</div>
+<div class="box">
+    <form class="pure-form" action="{{ url_for('ui.form_watch_list_checkbox_operations') }}" method="POST" id="watch-list-form">
+    <input type="hidden" name="csrf_token" value="{{ csrf_token() }}" >
+    <input type="hidden" id="op_extradata" name="op_extradata" value="" >
+    <div id="checkbox-operations">
+        <button class="pure-button button-secondary button-xsmall"  name="op" value="pause"><i data-feather="pause" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Pause</button>
+        <button class="pure-button button-secondary button-xsmall"  name="op" value="unpause"><i data-feather="play" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>UnPause</button>
+        <button class="pure-button button-secondary button-xsmall"  name="op" value="mute"><i data-feather="volume-x" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Mute</button>
+        <button class="pure-button button-secondary button-xsmall"  name="op" value="unmute"><i data-feather="volume-2" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>UnMute</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="recheck"><i data-feather="refresh-cw" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Recheck</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="assign-tag" id="checkbox-assign-tag"><i data-feather="tag" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Tag</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="mark-viewed"><i data-feather="eye" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Mark viewed</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="notification-default"><i data-feather="bell" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Use default notification</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="clear-errors"><i data-feather="x-circle" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Clear errors</button>
+        <button class="pure-button button-secondary button-xsmall" style="background: #dd4242;" name="op" value="clear-history"><i data-feather="trash-2" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Clear/reset history</button>
+        <button class="pure-button button-secondary button-xsmall" style="background: #dd4242;" name="op" value="delete"><i data-feather="trash" style="width: 14px; height: 14px; stroke: white; margin-right: 4px;"></i>Delete</button>
+    </div>
+    {%- if watches|length >= pagination.per_page -%}
+        {{ pagination.info }}
+    {%- endif -%}
+    {%- if search_q -%}<div id="search-result-info">Searching "<strong><i>{{search_q}}</i></strong>"</div>{%- endif -%}
+    <div>
+        <a href="{{url_for('watchlist.index')}}" class="pure-button button-tag {{'active' if not active_tag_uuid }}">All</a>
+
+    <!-- tag list -->
+    {%- for uuid, tag in tags -%}
+        {%- if tag != "" -%}
+            <a href="{{url_for('watchlist.index', tag=uuid) }}" class="pure-button button-tag {{'active' if active_tag_uuid == uuid }}">{{ tag.title }}</a>
+        {%- endif -%}
+    {%- endfor -%}
+    </div>
+
+    {%- set sort_order = sort_order or 'asc' -%}
+    {%- set sort_attribute = sort_attribute or 'last_changed'  -%}
+    {%- set pagination_page = request.args.get('page', 0) -%}
+    {%- set cols_required = 6 -%}
+    {%- set any_has_restock_price_processor = datastore.any_watches_have_processor_by_name("restock_diff") -%}
+    {%- if any_has_restock_price_processor -%}
+        {%- set cols_required = cols_required + 1 -%}
+    {%- endif -%}
+
+    <div id="watch-table-wrapper">
+
+        <table class="pure-table pure-table-striped watch-table">
+            <thead>
+            <tr>
+                {%- set link_order = "desc" if sort_order  == 'asc' else "asc" -%}
+                {%- set arrow_span = "" -%}
+                <th><input style="vertical-align: middle" type="checkbox" id="check-all" > <a class="{{ 'active '+link_order if sort_attribute == 'date_created' else 'inactive' }}"  href="{{url_for('watchlist.index', sort='date_created', order=link_order, tag=active_tag_uuid)}}"># <span class='arrow {{link_order}}'></span></a></th>
+                <th class="empty-cell"></th>
+                <th><a class="{{ 'active '+link_order if sort_attribute == 'label' else 'inactive' }}" href="{{url_for('watchlist.index', sort='label', order=link_order, tag=active_tag_uuid)}}">Website <span class='arrow {{link_order}}'></span></a></th>
+             {%- if any_has_restock_price_processor -%}
+                <th>Restock &amp; Price</th>
+             {%- endif -%}
+                <th><a class="{{ 'active '+link_order if sort_attribute == 'last_checked' else 'inactive' }}" href="{{url_for('watchlist.index', sort='last_checked', order=link_order, tag=active_tag_uuid)}}"><span class="hide-on-mobile">Last</span> Checked <span class='arrow {{link_order}}'></span></a></th>
+                <th><a class="{{ 'active '+link_order if sort_attribute == 'last_changed' else 'inactive' }}" href="{{url_for('watchlist.index', sort='last_changed', order=link_order, tag=active_tag_uuid)}}"><span class="hide-on-mobile">Last</span> Changed <span class='arrow {{link_order}}'></span></a></th>
+                <th class="empty-cell"></th>
+            </tr>
+            </thead>
+            <tbody>
+            {%- if not watches|length -%}
+            <tr>
+                <td colspan="{{ cols_required }}" style="text-wrap: wrap;">No website watches configured, please add a URL in the box above, or <a href="{{ url_for('imports.import_page')}}" >import a list</a>.</td>
+            </tr>
+            {%- endif -%}
+            {%- for watch in (watches|sort(attribute=sort_attribute, reverse=sort_order == 'asc'))|pagination_slice(skip=pagination.skip) -%}
+                {%- set checking_now = is_checking_now(watch) -%}
+                {%- set history_n = watch.history_n -%}
+                {#  Mirror in changedetectionio/static/js/realtime.js for the frontend #}
+                {%- set row_classes = [
+                    loop.cycle('pure-table-odd', 'pure-table-even'),
+                    'processor-' ~ watch['processor'],
+                    'has-error' if watch.compile_error_texts()|length > 2 else '',
+                    'paused' if watch.paused is defined and watch.paused != False else '',
+                    'unviewed' if watch.has_unviewed else '',
+                    'has-restock-info' if watch.has_restock_info else 'no-restock-info',
+                    'in-stock' if watch.has_restock_info and watch['restock']['in_stock'] else '',
+                    'not-in-stock' if watch.has_restock_info and not watch['restock']['in_stock'] else '',
+                    'queued' if watch.uuid in queued_uuids else '',
+                    'checking-now' if checking_now else '',
+                    'notification_muted' if watch.notification_muted else '',
+                    'single-history' if history_n == 1 else '',
+                    'multiple-history' if history_n >= 2 else ''
+                ] -%}
+            <tr id="{{ watch.uuid }}" data-watch-uuid="{{ watch.uuid }}" class="{{ row_classes | reject('equalto', '') | join(' ') }}">
+                <td class="inline checkbox-uuid" ><input name="uuids"  type="checkbox" value="{{ watch.uuid}} " > <span>{{ loop.index+pagination.skip }}</span></td>
+                <td class="inline watch-controls">
+                    <a class="ajax-op state-off pause-toggle" data-op="pause" href="{{url_for('watchlist.index', op='pause', uuid=watch.uuid, tag=active_tag_uuid)}}"><img src="{{url_for('static_content', group='images', filename='pause.svg')}}" alt="Pause checks" title="Pause checks" class="icon icon-pause" ></a>
+                    <a class="ajax-op state-on pause-toggle"  data-op="pause" style="display: none" href="{{url_for('watchlist.index', op='pause', uuid=watch.uuid, tag=active_tag_uuid)}}"><img src="{{url_for('static_content', group='images', filename='play.svg')}}" alt="UnPause checks" title="UnPause checks" class="icon icon-unpause" ></a>
+                    <a class="ajax-op state-off mute-toggle" data-op="mute" href="{{url_for('watchlist.index', op='mute', uuid=watch.uuid, tag=active_tag_uuid)}}"><img src="{{url_for('static_content', group='images', filename='bell-off.svg')}}" alt="Mute notification" title="Mute notification" class="icon icon-mute" ></a>
+                    <a class="ajax-op state-on mute-toggle" data-op="mute"  style="display: none" href="{{url_for('watchlist.index', op='mute', uuid=watch.uuid, tag=active_tag_uuid)}}"><img src="{{url_for('static_content', group='images', filename='bell-off.svg')}}" alt="UnMute notification" title="UnMute notification" class="icon icon-mute" ></a>
+                </td>
+                <td class="title-col inline">{{watch.title if watch.title is not none and watch.title|length > 0 else watch.url}}
+                    <a class="external" target="_blank" rel="noopener" href="{{ watch.link.replace('source:','') }}">&nbsp;</a>
+                    <a class="link-spread" href="{{url_for('ui.form_share_put_watch', uuid=watch.uuid)}}"><img src="{{url_for('static_content', group='images', filename='spread.svg')}}" class="status-icon icon icon-spread" title="Create a link to share watch config with others" ></a>
+
+                    {%- if watch.get_fetch_backend == "html_webdriver"
+                         or ( watch.get_fetch_backend == "system" and system_default_fetcher == 'html_webdriver'  )
+                         or "extra_browser_" in watch.get_fetch_backend
+                    -%}
+                    <img class="status-icon" src="{{url_for('static_content', group='images', filename='google-chrome-icon.png')}}" alt="Using a Chrome browser" title="Using a Chrome browser" >
+                    {%- endif -%}
+
+                    {%- if watch.is_pdf  -%}<img class="status-icon" src="{{url_for('static_content', group='images', filename='pdf-icon.svg')}}" alt="Converting PDF to text" >{%- endif -%}
+                    {%- if watch.has_browser_steps -%}<img class="status-icon status-browsersteps" src="{{url_for('static_content', group='images', filename='steps.svg')}}" alt="Browser Steps is enabled" >{%- endif -%}
+
+                    <div class="error-text" style="display:none;">{{ watch.compile_error_texts(has_proxies=datastore.proxy_list)|safe }}</div>
+
+                    {%- if watch['processor'] == 'text_json_diff'  -%}
+                        {%- if watch['has_ldjson_price_data'] and not watch['track_ldjson_price_data']  -%}
+                        <div class="ldjson-price-track-offer">Switch to Restock & Price watch mode? <a href="{{url_for('price_data_follower.accept', uuid=watch.uuid)}}" class="pure-button button-xsmall">Yes</a> <a href="{{url_for('price_data_follower.reject', uuid=watch.uuid)}}" class="">No</a></div>
+                        {%- endif -%}
+                    {%- endif -%}
+                    {%- if watch['processor'] == 'restock_diff' -%}
+                        <span class="tracking-ldjson-price-data" title="Automatically following embedded price information"><img src="{{url_for('static_content', group='images', filename='price-tag-icon.svg')}}"  class="status-icon price-follow-tag-icon" > Price</span>
+                    {%- endif -%}
+                    {%- for watch_tag_uuid, watch_tag in datastore.get_all_tags_for_watch(watch['uuid']).items() -%}
+                      <span class="watch-tag-list">{{ watch_tag.title }}</span>
+                    {%- endfor -%}
+                </td>
+{%- if any_has_restock_price_processor -%}
+                <td class="restock-and-price">
+                    {%- if watch['processor'] == 'restock_diff'  -%}
+                        {%- if watch.has_restock_info -%}
+                            <span class="restock-label {{'in-stock' if watch['restock']['in_stock'] else 'not-in-stock' }}" title="Detecting restock and price">
+                                <!-- maybe some object watch['processor'][restock_diff] or.. -->
+                                 {%- if watch['restock']['in_stock']-%}  In stock {%- else-%}  Not in stock {%- endif -%}
+                            </span>
+                        {%- endif -%}
+
+                        {%- if watch.get('restock') and watch['restock']['price'] != None -%}
+                            {%- if watch['restock']['price'] != None -%}
+                                <span class="restock-label price" title="Price">
+                                {{ watch['restock']['price']|format_number_locale }} {{ watch['restock']['currency'] }}
+                                </span>
+                            {%- endif -%}
+                        {%- elif not watch.has_restock_info -%}
+                            <span class="restock-label error">No information</span>
+                        {%- endif -%}
+                    {%- endif -%}
+                </td>
+{%- endif -%}
+            {#last_checked becomes fetch-start-time#}
+                <td class="last-checked" data-timestamp="{{ watch.last_checked }}" data-fetchduration={{ watch.fetch_time }} data-eta_complete="{{ watch.last_checked+watch.fetch_time }}" >
+                    <div class="spinner-wrapper" style="display:none;" >
+                        <span class="spinner"></span><span>&nbsp;Checking now</span>
+                    </div>
+                    <span class="innertext">{{watch|format_last_checked_time|safe}}</span>
+                </td>
+                <td class="last-changed" data-timestamp="{{ watch.last_changed }}">{%- if watch.history_n >=2 and watch.last_changed >0 -%}
+                    {{watch.last_changed|format_timestamp_timeago}}
+                    {%- else -%}
+                    Not yet
+                    {%- endif -%}
+                </td>
+                <td>
+                    {%- set target_attr = ' target="' ~ watch.uuid ~ '"' if datastore.data['settings']['application']['ui'].get('open_diff_in_new_tab') else '' -%}
+                    <a href="" class="already-in-queue-button recheck pure-button pure-button-primary" style="display: none;" disabled="disabled">Queued</a>
+                    <a href="{{ url_for('ui.form_watch_checknow', uuid=watch.uuid, tag=request.args.get('tag')) }}" data-op='recheck' class="ajax-op recheck pure-button pure-button-primary">Recheck</a>
+                    <a href="{{ url_for('ui.ui_edit.edit_page', uuid=watch.uuid, tag=active_tag_uuid)}}#general" class="pure-button pure-button-primary">Edit</a>
+                    <a href="{{ url_for('ui.ui_views.diff_history_page', uuid=watch.uuid)}}" {{target_attr}} class="pure-button pure-button-primary history-link" style="display: none;">History</a>
+                    <a href="{{ url_for('ui.ui_views.preview_page', uuid=watch.uuid)}}" {{target_attr}} class="pure-button pure-button-primary preview-link" style="display: none;">Preview</a>
+                </td>
+            </tr>
+            {%- endfor -%}
+            </tbody>
+        </table>
+        <ul id="post-list-buttons">
+            <li id="post-list-with-errors" class="{%- if errored_count -%}has-error{%- endif -%}" style="display: none;" >
+                <a href="{{url_for('watchlist.index', with_errors=1, tag=request.args.get('tag')) }}" class="pure-button button-tag button-error">With errors ({{ errored_count }})</a>
+            </li>
+            <li id="post-list-mark-views" class="{%- if has_unviewed -%}has-unviewed{%- endif -%}" style="display: none;" >
+                <a href="{{url_for('ui.mark_all_viewed',with_errors=request.args.get('with_errors',0)) }}" class="pure-button button-tag " id="mark-all-viewed">Mark all viewed</a>
+            </li>
+        {%-  if active_tag_uuid -%}
+            <li id="post-list-mark-views-tag">
+                <a href="{{url_for('ui.mark_all_viewed', tag=active_tag_uuid) }}" class="pure-button button-tag " id="mark-all-viewed">Mark all viewed in '{{active_tag.title}}'</a>
+            </li>
+        {%-  endif -%}
+            <li>
+               <a href="{{ url_for('ui.form_watch_checknow', tag=active_tag_uuid, with_errors=request.args.get('with_errors',0)) }}" class="pure-button button-tag" id="recheck-all">Recheck
+                all {% if active_tag_uuid %}  in '{{active_tag.title}}'{%endif%}</a>
+            </li>
+            <li>
+                <a href="{{ url_for('rss.feed', tag=active_tag_uuid, token=app_rss_token)}}"><img alt="RSS Feed" id="feed-icon" src="{{url_for('static_content', group='images', filename='generic_feed-icon.svg')}}" height="15"></a>
+            </li>
+        </ul>
+        {{ pagination.links }}
+    </div>
+    </form>
+</div>
+{%- endblock -%}

changedetectionio/conditions/__init__.py

@@ -0,0 +1,170 @@
+from flask import Blueprint
+
+from json_logic.builtins import BUILTINS
+
+from .exceptions import EmptyConditionRuleRowNotUsable
+from .pluggy_interface import plugin_manager  # Import the pluggy plugin manager
+from . import default_plugin
+from loguru import logger
+# List of all supported JSON Logic operators
+operator_choices = [
+    (None, "Choose one - Operator"),
+    (">", "Greater Than"),
+    ("<", "Less Than"),
+    (">=", "Greater Than or Equal To"),
+    ("<=", "Less Than or Equal To"),
+    ("==", "Equals"),
+    ("!=", "Not Equals"),
+    ("in", "Contains"),
+    ("!in", "Does Not Contain"),
+]
+
+# Fields available in the rules
+field_choices = [
+    (None, "Choose one - Field"),
+]
+
+# The data we will feed the JSON Rules to see if it passes the test/conditions or not
+EXECUTE_DATA = {}
+
+
+# Define the extended operations dictionary
+CUSTOM_OPERATIONS = {
+    **BUILTINS,  # Include all standard operators
+}
+
+def filter_complete_rules(ruleset):
+    rules = [
+        rule for rule in ruleset
+        if all(value not in ("", False, "None", None) for value in [rule["operator"], rule["field"], rule["value"]])
+    ]
+    return rules
+
+def convert_to_jsonlogic(logic_operator: str, rule_dict: list):
+    """
+    Convert a structured rule dict into a JSON Logic rule.
+
+    :param rule_dict: Dictionary containing conditions.
+    :return: JSON Logic rule as a dictionary.
+    """
+
+
+    json_logic_conditions = []
+
+    for condition in rule_dict:
+        operator = condition["operator"]
+        field = condition["field"]
+        value = condition["value"]
+
+        if not operator or operator == 'None' or not value or not field:
+            raise EmptyConditionRuleRowNotUsable()
+
+        # Convert value to int/float if possible
+        try:
+            if isinstance(value, str) and "." in value and str != "None":
+                value = float(value)
+            else:
+                value = int(value)
+        except (ValueError, TypeError):
+            pass  # Keep as a string if conversion fails
+
+        # Handle different JSON Logic operators properly
+        if operator == "in":
+            json_logic_conditions.append({"in": [value, {"var": field}]})  # value first
+        elif operator in ("!", "!!", "-"):
+            json_logic_conditions.append({operator: [{"var": field}]})  # Unary operators
+        elif operator in ("min", "max", "cat"):
+            json_logic_conditions.append({operator: value})  # Multi-argument operators
+        else:
+            json_logic_conditions.append({operator: [{"var": field}, value]})  # Standard binary operators
+
+    return {logic_operator: json_logic_conditions} if len(json_logic_conditions) > 1 else json_logic_conditions[0]
+
+
+def execute_ruleset_against_all_plugins(current_watch_uuid: str, application_datastruct, ephemeral_data={} ):
+    """
+    Build our data and options by calling our plugins then pass it to jsonlogic and see if the conditions pass
+
+    :param ruleset: JSON Logic rule dictionary.
+    :param extracted_data: Dictionary containing the facts.   <-- maybe the app struct+uuid
+    :return: Dictionary of plugin results.
+    """
+    from json_logic import jsonLogic
+
+    EXECUTE_DATA = {}
+    result = True
+    
+    watch = application_datastruct['watching'].get(current_watch_uuid)
+
+    if watch and watch.get("conditions"):
+        logic_operator = "and" if watch.get("conditions_match_logic", "ALL") == "ALL" else "or"
+        complete_rules = filter_complete_rules(watch['conditions'])
+        if complete_rules:
+            # Give all plugins a chance to update the data dict again (that we will test the conditions against)
+            for plugin in plugin_manager.get_plugins():
+                try:
+                    import concurrent.futures
+                    import time
+                    
+                    with concurrent.futures.ThreadPoolExecutor() as executor:
+                        future = executor.submit(
+                            plugin.add_data,
+                            current_watch_uuid=current_watch_uuid,
+                            application_datastruct=application_datastruct,
+                            ephemeral_data=ephemeral_data
+                        )
+                        logger.debug(f"Trying plugin {plugin}....")
+
+                        # Set a timeout of 10 seconds
+                        try:
+                            new_execute_data = future.result(timeout=10)
+                            if new_execute_data and isinstance(new_execute_data, dict):
+                                EXECUTE_DATA.update(new_execute_data)
+
+                        except concurrent.futures.TimeoutError:
+                            # The plugin took too long, abort processing for this watch
+                            raise Exception(f"Plugin {plugin.__class__.__name__} took more than 10 seconds to run.")
+                except Exception as e:
+                    # Log the error but continue with the next plugin
+                    import logging
+                    logging.error(f"Error executing plugin {plugin.__class__.__name__}: {str(e)}")
+                    continue
+
+            # Create the ruleset
+            ruleset = convert_to_jsonlogic(logic_operator=logic_operator, rule_dict=complete_rules)
+            
+            # Pass the custom operations dictionary to jsonLogic
+            if not jsonLogic(logic=ruleset, data=EXECUTE_DATA, operations=CUSTOM_OPERATIONS):
+                result = False
+
+    return {'executed_data': EXECUTE_DATA, 'result': result}
+
+# Load plugins dynamically
+for plugin in plugin_manager.get_plugins():
+    new_ops = plugin.register_operators()
+    if isinstance(new_ops, dict):
+        CUSTOM_OPERATIONS.update(new_ops)
+
+    new_operator_choices = plugin.register_operator_choices()
+    if isinstance(new_operator_choices, list):
+        operator_choices.extend(new_operator_choices)
+
+    new_field_choices = plugin.register_field_choices()
+    if isinstance(new_field_choices, list):
+        field_choices.extend(new_field_choices)
+
+def collect_ui_edit_stats_extras(watch):
+    """Collect and combine HTML content from all plugins that implement ui_edit_stats_extras"""
+    extras_content = []
+    
+    for plugin in plugin_manager.get_plugins():
+        try:
+            content = plugin.ui_edit_stats_extras(watch=watch)
+            if content:
+                extras_content.append(content)
+        except Exception as e:
+            # Skip plugins that don't implement the hook or have errors
+            pass
+            
+    return "\n".join(extras_content) if extras_content else ""
+

changedetectionio/conditions/blueprint.py

@@ -0,0 +1,81 @@
+# Flask Blueprint Definition
+import json
+
+from flask import Blueprint
+
+from changedetectionio.conditions import execute_ruleset_against_all_plugins
+
+
+def construct_blueprint(datastore):
+    from changedetectionio.flask_app import login_optionally_required
+
+    conditions_blueprint = Blueprint('conditions', __name__, template_folder="templates")
+
+    @conditions_blueprint.route("/<string:watch_uuid>/verify-condition-single-rule", methods=['POST'])
+    @login_optionally_required
+    def verify_condition_single_rule(watch_uuid):
+        """Verify a single condition rule against the current snapshot"""
+        from changedetectionio.processors.text_json_diff import prepare_filter_prevew
+        from flask import request, jsonify
+        from copy import deepcopy
+
+        ephemeral_data = {}
+
+        # Get the watch data
+        watch = datastore.data['watching'].get(watch_uuid)
+        if not watch:
+            return jsonify({'status': 'error', 'message': 'Watch not found'}), 404
+
+        # First use prepare_filter_prevew to process the form data
+        # This will return text_after_filter which is after all current form settings are applied
+        # Create ephemeral data with the text from the current snapshot
+
+        try:
+            # Call prepare_filter_prevew to get a processed version of the content with current form settings
+            # We'll ignore the returned response and just use the datastore which is modified by the function
+
+            # this should apply all filters etc so then we can run the CONDITIONS against the final output text
+            result = prepare_filter_prevew(datastore=datastore,
+                                           form_data=request.form,
+                                           watch_uuid=watch_uuid)
+
+            ephemeral_data['text'] = result.get('after_filter', '')
+            # Create a temporary watch data structure with this single rule
+            tmp_watch_data = deepcopy(datastore.data['watching'].get(watch_uuid))
+
+            # Override the conditions in the temporary watch
+            rule_json = request.args.get("rule")
+            rule = json.loads(rule_json) if rule_json else None
+
+            # Should be key/value of field, operator, value
+            tmp_watch_data['conditions'] = [rule]
+            tmp_watch_data['conditions_match_logic'] = "ALL"  # Single rule, so use ALL
+
+            # Create a temporary application data structure for the rule check
+            temp_app_data = {
+                'watching': {
+                    watch_uuid: tmp_watch_data
+                }
+            }
+
+            # Execute the rule against the current snapshot with form data
+            result = execute_ruleset_against_all_plugins(
+                current_watch_uuid=watch_uuid,
+                application_datastruct=temp_app_data,
+                ephemeral_data=ephemeral_data
+            )
+
+            return jsonify({
+                'status': 'success',
+                'result': result.get('result'),
+                'data': result.get('executed_data'),
+                'message': 'Condition passes' if result else 'Condition does not pass'
+            })
+
+        except Exception as e:
+            return jsonify({
+                'status': 'error',
+                'message': f'Error verifying condition: {str(e)}'
+            }), 500
+
+    return conditions_blueprint

changedetectionio/conditions/default_plugin.py

@@ -0,0 +1,78 @@
+import re
+
+import pluggy
+from price_parser import Price
+from loguru import logger
+
+hookimpl = pluggy.HookimplMarker("changedetectionio_conditions")
+
+
+@hookimpl
+def register_operators():
+    def starts_with(_, text, prefix):
+        return text.lower().strip().startswith(str(prefix).strip().lower())
+
+    def ends_with(_, text, suffix):
+        return text.lower().strip().endswith(str(suffix).strip().lower())
+
+    def length_min(_, text, strlen):
+        return len(text) >= int(strlen)
+
+    def length_max(_, text, strlen):
+        return len(text) <= int(strlen)
+
+    # ✅ Custom function for case-insensitive regex matching
+    def contains_regex(_, text, pattern):
+        """Returns True if `text` contains `pattern` (case-insensitive regex match)."""
+        return bool(re.search(pattern, str(text), re.IGNORECASE))
+
+    # ✅ Custom function for NOT matching case-insensitive regex
+    def not_contains_regex(_, text, pattern):
+        """Returns True if `text` does NOT contain `pattern` (case-insensitive regex match)."""
+        return not bool(re.search(pattern, str(text), re.IGNORECASE))
+
+    return {
+        "!contains_regex": not_contains_regex,
+        "contains_regex": contains_regex,
+        "ends_with": ends_with,
+        "length_max": length_max,
+        "length_min": length_min,
+        "starts_with": starts_with,
+    }
+
+@hookimpl
+def register_operator_choices():
+    return [
+        ("starts_with", "Text Starts With"),
+        ("ends_with", "Text Ends With"),
+        ("length_min", "Length minimum"),
+        ("length_max", "Length maximum"),
+        ("contains_regex", "Text Matches Regex"),
+        ("!contains_regex", "Text Does NOT Match Regex"),
+    ]
+
+@hookimpl
+def register_field_choices():
+    return [
+        ("extracted_number", "Extracted number after 'Filters & Triggers'"),
+#        ("meta_description", "Meta Description"),
+#        ("meta_keywords", "Meta Keywords"),
+        ("page_filtered_text", "Page text after 'Filters & Triggers'"),
+        #("page_title", "Page <title>"), # actual page title <title>
+    ]
+
+@hookimpl
+def add_data(current_watch_uuid, application_datastruct, ephemeral_data):
+
+    res = {}
+    if 'text' in ephemeral_data:
+        res['page_filtered_text'] = ephemeral_data['text']
+
+        # Better to not wrap this in try/except so that the UI can see any errors
+        price = Price.fromstring(ephemeral_data.get('text'))
+        if price and price.amount != None:
+            # This is slightly misleading, it's extracting a PRICE not a Number..
+            res['extracted_number'] = float(price.amount)
+            logger.debug(f"Extracted number result: '{price}' - returning float({res['extracted_number']})")
+
+    return res

changedetectionio/conditions/exceptions.py

@@ -0,0 +1,6 @@
+class EmptyConditionRuleRowNotUsable(Exception):
+    def __init__(self):
+        super().__init__("One of the 'conditions' rulesets is incomplete, cannot run.")
+
+    def __str__(self):
+        return self.args[0]

changedetectionio/conditions/form.py

@@ -0,0 +1,44 @@
+# Condition Rule Form (for each rule row)
+from wtforms import Form, SelectField, StringField, validators
+from wtforms import validators
+
+class ConditionFormRow(Form):
+
+    # ✅ Ensure Plugins Are Loaded BEFORE Importing Choices
+    from changedetectionio.conditions import plugin_manager
+    from changedetectionio.conditions import operator_choices, field_choices
+    field = SelectField(
+        "Field",
+        choices=field_choices,
+        validators=[validators.Optional()]
+    )
+
+    operator = SelectField(
+        "Operator",
+        choices=operator_choices,
+        validators=[validators.Optional()]
+    )
+
+    value = StringField("Value", validators=[validators.Optional()], render_kw={"placeholder": "A value"})
+
+    def validate(self, extra_validators=None):
+        # First, run the default validators
+        if not super().validate(extra_validators):
+            return False
+
+        # Custom validation logic
+        # If any of the operator/field/value is set, then they must be all set
+        if any(value not in ("", False, "None", None) for value in [self.operator.data, self.field.data, self.value.data]):
+            if not self.operator.data or self.operator.data == 'None':
+                self.operator.errors.append("Operator is required.")
+                return False
+
+            if not self.field.data or self.field.data == 'None':
+                self.field.errors.append("Field is required.")
+                return False
+
+            if not self.value.data:
+                self.value.errors.append("Value is required.")
+                return False
+
+        return True  # Only return True if all conditions pass

changedetectionio/conditions/pluggy_interface.py

@@ -0,0 +1,74 @@
+import pluggy
+import os
+import importlib
+import sys
+from . import default_plugin
+
+# ✅ Ensure that the namespace in HookspecMarker matches PluginManager
+PLUGIN_NAMESPACE = "changedetectionio_conditions"
+
+hookspec = pluggy.HookspecMarker(PLUGIN_NAMESPACE)
+hookimpl = pluggy.HookimplMarker(PLUGIN_NAMESPACE)
+
+
+class ConditionsSpec:
+    """Hook specifications for extending JSON Logic conditions."""
+
+    @hookspec
+    def register_operators():
+        """Return a dictionary of new JSON Logic operators."""
+        pass
+
+    @hookspec
+    def register_operator_choices():
+        """Return a list of new operator choices."""
+        pass
+
+    @hookspec
+    def register_field_choices():
+        """Return a list of new field choices."""
+        pass
+
+    @hookspec
+    def add_data(current_watch_uuid, application_datastruct, ephemeral_data):
+        """Add to the datadict"""
+        pass
+        
+    @hookspec
+    def ui_edit_stats_extras(watch):
+        """Return HTML content to add to the stats tab in the edit view"""
+        pass
+
+# ✅ Set up Pluggy Plugin Manager
+plugin_manager = pluggy.PluginManager(PLUGIN_NAMESPACE)
+
+# ✅ Register hookspecs (Ensures they are detected)
+plugin_manager.add_hookspecs(ConditionsSpec)
+
+# ✅ Register built-in plugins manually
+plugin_manager.register(default_plugin, "default_plugin")
+
+# ✅ Load plugins from the plugins directory
+def load_plugins_from_directory():
+    plugins_dir = os.path.join(os.path.dirname(__file__), 'plugins')
+    if not os.path.exists(plugins_dir):
+        return
+        
+    # Get all Python files (excluding __init__.py)
+    for filename in os.listdir(plugins_dir):
+        if filename.endswith(".py") and filename != "__init__.py":
+            module_name = filename[:-3]  # Remove .py extension
+            module_path = f"changedetectionio.conditions.plugins.{module_name}"
+            
+            try:
+                module = importlib.import_module(module_path)
+                # Register the plugin with pluggy
+                plugin_manager.register(module, module_name)
+            except (ImportError, AttributeError) as e:
+                print(f"Error loading plugin {module_name}: {e}")
+
+# Load plugins from the plugins directory
+load_plugins_from_directory()
+
+# ✅ Discover installed plugins from external packages (if any)
+plugin_manager.load_setuptools_entrypoints(PLUGIN_NAMESPACE)

changedetectionio/conditions/plugins/__init__.py

@@ -0,0 +1 @@
+# Import plugins package to make them discoverable

changedetectionio/conditions/plugins/levenshtein_plugin.py

@@ -0,0 +1,107 @@
+import pluggy
+from loguru import logger
+
+# Support both plugin systems
+conditions_hookimpl = pluggy.HookimplMarker("changedetectionio_conditions")
+global_hookimpl = pluggy.HookimplMarker("changedetectionio")
+
+def levenshtein_ratio_recent_history(watch, incoming_text=None):
+    try:
+        from Levenshtein import ratio, distance
+        k = list(watch.history.keys())
+        a = None
+        b = None
+
+        # When called from ui_edit_stats_extras, we don't have incoming_text
+        if incoming_text is None:
+            a = watch.get_history_snapshot(timestamp=k[-1])  # Latest snapshot
+            b = watch.get_history_snapshot(timestamp=k[-2])  # Previous snapshot
+
+        # Needs atleast one snapshot
+        elif len(k) >= 1: # Should be atleast one snapshot to compare against
+            a = watch.get_history_snapshot(timestamp=k[-1]) # Latest saved snapshot
+            b = incoming_text if incoming_text else k[-2]
+
+        if a and b:
+            distance_value = distance(a, b)
+            ratio_value = ratio(a, b)
+            return {
+                'distance': distance_value,
+                'ratio': ratio_value,
+                'percent_similar': round(ratio_value * 100, 2)
+            }
+    except Exception as e:
+        logger.warning(f"Unable to calc similarity: {str(e)}")
+
+    return ''
+
+@conditions_hookimpl
+def register_operators():
+    pass
+
+@conditions_hookimpl
+def register_operator_choices():
+    pass
+
+
+@conditions_hookimpl
+def register_field_choices():
+    return [
+        ("levenshtein_ratio", "Levenshtein - Text similarity ratio"),
+        ("levenshtein_distance", "Levenshtein - Text change distance"),
+    ]
+
+@conditions_hookimpl
+def add_data(current_watch_uuid, application_datastruct, ephemeral_data):
+    res = {}
+    watch = application_datastruct['watching'].get(current_watch_uuid)
+    # ephemeral_data['text'] will be the current text after filters, they may have edited filters but not saved them yet etc
+
+    if watch and 'text' in ephemeral_data:
+        lev_data = levenshtein_ratio_recent_history(watch, ephemeral_data.get('text',''))
+        if isinstance(lev_data, dict):
+            res['levenshtein_ratio'] = lev_data.get('ratio', 0)
+            res['levenshtein_similarity'] = lev_data.get('percent_similar', 0)
+            res['levenshtein_distance'] = lev_data.get('distance', 0)
+
+    return res
+
+@global_hookimpl
+def ui_edit_stats_extras(watch):
+    """Add Levenshtein stats to the UI using the global plugin system"""
+    """Generate the HTML for Levenshtein stats - shared by both plugin systems"""
+    if len(watch.history.keys()) < 2:
+        return "<p>Not enough history to calculate Levenshtein metrics</p>"
+    
+    try:
+        lev_data = levenshtein_ratio_recent_history(watch)
+        if not lev_data or not isinstance(lev_data, dict):
+            return "<p>Unable to calculate Levenshtein metrics</p>"
+            
+        html = f"""
+        <div class="levenshtein-stats">
+            <h4>Levenshtein Text Similarity Details</h4>
+            <table class="pure-table">
+                <tbody>
+                    <tr>
+                        <td>Raw distance (edits needed)</td>
+                        <td>{lev_data['distance']}</td>
+                    </tr>
+                    <tr>
+                        <td>Similarity ratio</td>
+                        <td>{lev_data['ratio']:.4f}</td>
+                    </tr>
+                    <tr>
+                        <td>Percent similar</td>
+                        <td>{lev_data['percent_similar']}%</td>
+                    </tr>
+                </tbody>
+            </table>
+            <p style="font-size: 80%;">Levenshtein metrics compare the last two snapshots, measuring how many character edits are needed to transform one into the other.</p>
+        </div>
+        """
+        return html
+    except Exception as e:
+        logger.error(f"Error generating Levenshtein UI extras: {str(e)}")
+        return "<p>Error calculating Levenshtein metrics</p>"
+        

changedetectionio/conditions/plugins/wordcount_plugin.py

@@ -0,0 +1,82 @@
+import pluggy
+from loguru import logger
+
+# Support both plugin systems
+conditions_hookimpl = pluggy.HookimplMarker("changedetectionio_conditions")
+global_hookimpl = pluggy.HookimplMarker("changedetectionio")
+
+def count_words_in_history(watch, incoming_text=None):
+    """Count words in snapshot text"""
+    try:
+        if incoming_text is not None:
+            # When called from add_data with incoming text
+            return len(incoming_text.split())
+        elif watch.history.keys():
+            # When called from UI extras to count latest snapshot
+            latest_key = list(watch.history.keys())[-1]
+            latest_content = watch.get_history_snapshot(latest_key)
+            return len(latest_content.split())
+        return 0
+    except Exception as e:
+        logger.error(f"Error counting words: {str(e)}")
+        return 0
+
+# Implement condition plugin hooks
+@conditions_hookimpl
+def register_operators():
+    # No custom operators needed
+    return {}
+
+@conditions_hookimpl
+def register_operator_choices():
+    # No custom operator choices needed
+    return []
+
+@conditions_hookimpl
+def register_field_choices():
+    # Add a field that will be available in conditions
+    return [
+        ("word_count", "Word count of content"),
+    ]
+
+@conditions_hookimpl
+def add_data(current_watch_uuid, application_datastruct, ephemeral_data):
+    """Add word count data for conditions"""
+    result = {}
+    watch = application_datastruct['watching'].get(current_watch_uuid)
+    
+    if watch and 'text' in ephemeral_data:
+        word_count = count_words_in_history(watch, ephemeral_data['text'])
+        result['word_count'] = word_count
+    
+    return result
+
+def _generate_stats_html(watch):
+    """Generate the HTML content for the stats tab"""
+    word_count = count_words_in_history(watch)
+    
+    html = f"""
+    <div class="word-count-stats">
+        <h4>Content Analysis</h4>
+        <table class="pure-table">
+            <tbody>
+                <tr>
+                    <td>Word count (latest snapshot)</td>
+                    <td>{word_count}</td>
+                </tr>
+            </tbody>
+        </table>
+        <p style="font-size: 80%;">Word count is a simple measure of content length, calculated by splitting text on whitespace.</p>
+    </div>
+    """
+    return html
+
+@conditions_hookimpl
+def ui_edit_stats_extras(watch):
+    """Add word count stats to the UI through conditions plugin system"""
+    return _generate_stats_html(watch)
+
+@global_hookimpl
+def ui_edit_stats_extras(watch):
+    """Add word count stats to the UI using the global plugin system"""
+    return _generate_stats_html(watch)

changedetectionio/content_fetchers/__init__.py

@@ -4,12 +4,32 @@ from loguru import logger
 from changedetectionio.content_fetchers.exceptions import BrowserStepsStepException
 import os
 
-visualselector_xpath_selectors = 'div,span,form,table,tbody,tr,td,a,p,ul,li,h1,h2,h3,h4,header,footer,section,article,aside,details,main,nav,section,summary'
+# Visual Selector scraper - 'Button' is there because some sites have <button>OUT OF STOCK</button>.
+visualselector_xpath_selectors = 'div,span,form,table,tbody,tr,td,a,p,ul,li,h1,h2,h3,h4,header,footer,section,article,aside,details,main,nav,section,summary,button'
+
+SCREENSHOT_MAX_HEIGHT_DEFAULT = 20000
+SCREENSHOT_DEFAULT_QUALITY = 40
+
+# Maximum total height for the final image (When in stitch mode).
+# We limit this to 16000px due to the huge amount of RAM that was being used
+# Example: 16000 × 1400 × 3 = 67,200,000 bytes ≈ 64.1 MB (not including buffers in PIL etc)
+SCREENSHOT_MAX_TOTAL_HEIGHT = int(os.getenv("SCREENSHOT_MAX_HEIGHT", SCREENSHOT_MAX_HEIGHT_DEFAULT))
+
+# The size at which we will switch to stitching method, when below this (and
+# MAX_TOTAL_HEIGHT which can be set by a user) we will use the default
+# screenshot method.
+SCREENSHOT_SIZE_STITCH_THRESHOLD = 8000
 
 # available_fetchers() will scan this implementation looking for anything starting with html_
 # this information is used in the form selections
 from changedetectionio.content_fetchers.requests import fetcher as html_requests
 
+
+import importlib.resources
+XPATH_ELEMENT_JS = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text(encoding='utf-8')
+INSTOCK_DATA_JS = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('stock-not-in-stock.js').read_text(encoding='utf-8')
+
+
 def available_fetchers():
     # See the if statement at the bottom of this file for how we switch between playwright and webdriver
     import inspect

changedetectionio/content_fetchers/base.py

@@ -63,17 +63,12 @@ class Fetcher():
     # Time ONTOP of the system defined env minimum time
     render_extract_delay = 0
 
-    def __init__(self):
-        import importlib.resources
-        self.xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text()
-        self.instock_data_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('stock-not-in-stock.js').read_text()
-
     @abstractmethod
     def get_error(self):
         return self.error
 
     @abstractmethod
-    def run(self,
+    async def run(self,
             url,
             timeout,
             request_headers,
@@ -81,12 +76,13 @@ class Fetcher():
             request_method,
             ignore_status_codes=False,
             current_include_filters=None,
-            is_binary=False):
+            is_binary=False,
+            empty_pages_are_a_change=False):
         # Should set self.error, self.status_code and self.content
         pass
 
     @abstractmethod
-    def quit(self):
+    def quit(self, watch=None):
         return
 
     @abstractmethod
@@ -95,6 +91,9 @@ class Fetcher():
 
     @abstractmethod
     def screenshot_step(self, step_n):
+        if self.browser_steps_screenshot_path and not os.path.isdir(self.browser_steps_screenshot_path):
+            logger.debug(f"> Creating data dir {self.browser_steps_screenshot_path}")
+            os.mkdir(self.browser_steps_screenshot_path)
         return None
 
     @abstractmethod
@@ -123,7 +122,7 @@ class Fetcher():
 
         return None
 
-    def iterate_browser_steps(self, start_url=None):
+    async def iterate_browser_steps(self, start_url=None):
         from changedetectionio.blueprint.browser_steps.browser_steps import steppable_browser_interface
         from playwright._impl._errors import TimeoutError, Error
         from changedetectionio.safe_jinja import render as jinja_render
@@ -137,8 +136,9 @@ class Fetcher():
             for step in valid_steps:
                 step_n += 1
                 logger.debug(f">> Iterating check - browser Step n {step_n} - {step['operation']}...")
-                self.screenshot_step("before-" + str(step_n))
-                self.save_step_html("before-" + str(step_n))
+                await self.screenshot_step("before-" + str(step_n))
+                await self.save_step_html("before-" + str(step_n))
+
                 try:
                     optional_value = step['optional_value']
                     selector = step['selector']
@@ -148,11 +148,11 @@ class Fetcher():
                     if '{%' in step['selector'] or '{{' in step['selector']:
                         selector = jinja_render(template_str=step['selector'])
 
-                    getattr(interface, "call_action")(action_name=step['operation'],
+                    await getattr(interface, "call_action")(action_name=step['operation'],
                                                       selector=selector,
                                                       optional_value=optional_value)
-                    self.screenshot_step(step_n)
-                    self.save_step_html(step_n)
+                    await self.screenshot_step(step_n)
+                    await self.save_step_html(step_n)
                 except (Error, TimeoutError) as e:
                     logger.debug(str(e))
                     # Stop processing here
@@ -168,5 +168,8 @@ class Fetcher():
                 if os.path.isfile(f):
                     os.unlink(f)
 
-    def save_step_html(self, param):
+    def save_step_html(self, step_n):
+        if self.browser_steps_screenshot_path and not os.path.isdir(self.browser_steps_screenshot_path):
+            logger.debug(f"> Creating data dir {self.browser_steps_screenshot_path}")
+            os.mkdir(self.browser_steps_screenshot_path)
         pass

changedetectionio/content_fetchers/exceptions/__init__.py

@@ -1,6 +1,5 @@
 from loguru import logger
 
-
 class Non200ErrorCodeReceived(Exception):
     def __init__(self, status_code, url, screenshot=None, xpath_data=None, page_html=None):
         # Set this so we can use it in other parts of the app
@@ -81,7 +80,7 @@ class ScreenshotUnavailable(Exception):
         self.status_code = status_code
         self.url = url
         if page_html:
-            from html_tools import html_to_text
+            from changedetectionio.html_tools import html_to_text
             self.page_text = html_to_text(page_html)
         return
 

changedetectionio/content_fetchers/playwright.py

@@ -4,9 +4,75 @@ from urllib.parse import urlparse
 
 from loguru import logger
 
+from changedetectionio.content_fetchers import SCREENSHOT_MAX_HEIGHT_DEFAULT, visualselector_xpath_selectors, \
+    SCREENSHOT_SIZE_STITCH_THRESHOLD, SCREENSHOT_MAX_TOTAL_HEIGHT, XPATH_ELEMENT_JS, INSTOCK_DATA_JS
 from changedetectionio.content_fetchers.base import Fetcher, manage_user_agent
 from changedetectionio.content_fetchers.exceptions import PageUnloadable, Non200ErrorCodeReceived, EmptyReply, ScreenshotUnavailable
 
+async def capture_full_page_async(page):
+    import os
+    import time
+    from multiprocessing import Process, Pipe
+
+    start = time.time()
+
+    page_height = await page.evaluate("document.documentElement.scrollHeight")
+    page_width = await page.evaluate("document.documentElement.scrollWidth")
+    original_viewport = page.viewport_size
+
+    logger.debug(f"Playwright viewport size {page.viewport_size} page height {page_height} page width {page_width}")
+
+    # Use an approach similar to puppeteer: set a larger viewport and take screenshots in chunks
+    step_size = SCREENSHOT_SIZE_STITCH_THRESHOLD # Size that won't cause GPU to overflow
+    screenshot_chunks = []
+    y = 0
+
+    if page_height > page.viewport_size['height']:
+        if page_height < step_size:
+            step_size = page_height # Incase page is bigger than default viewport but smaller than proposed step size
+        logger.debug(f"Setting bigger viewport to step through large page width W{page.viewport_size['width']}xH{step_size} because page_height > viewport_size")
+        # Set viewport to a larger size to capture more content at once
+        await page.set_viewport_size({'width': page.viewport_size['width'], 'height': step_size})
+
+    # Capture screenshots in chunks up to the max total height
+    while y < min(page_height, SCREENSHOT_MAX_TOTAL_HEIGHT):
+        await page.request_gc()
+        await page.evaluate(f"window.scrollTo(0, {y})")
+        await page.request_gc()
+        screenshot_chunks.append(await page.screenshot(
+            type="jpeg",
+            full_page=False,
+            quality=int(os.getenv("SCREENSHOT_QUALITY", 72))
+        ))
+        y += step_size
+        await page.request_gc()
+
+    # Restore original viewport size
+    await page.set_viewport_size({'width': original_viewport['width'], 'height': original_viewport['height']})
+
+    # If we have multiple chunks, stitch them together
+    if len(screenshot_chunks) > 1:
+        from changedetectionio.content_fetchers.screenshot_handler import stitch_images_worker
+        logger.debug(f"Screenshot stitching {len(screenshot_chunks)} chunks together")
+        parent_conn, child_conn = Pipe()
+        p = Process(target=stitch_images_worker, args=(child_conn, screenshot_chunks, page_height, SCREENSHOT_MAX_TOTAL_HEIGHT))
+        p.start()
+        screenshot = parent_conn.recv_bytes()
+        p.join()
+        logger.debug(
+            f"Screenshot (chunked/stitched) - Page height: {page_height} Capture height: {SCREENSHOT_MAX_TOTAL_HEIGHT} - Stitched together in {time.time() - start:.2f}s")
+        # Explicit cleanup
+        del screenshot_chunks
+        del p
+        del parent_conn, child_conn
+        screenshot_chunks = None
+        return screenshot
+
+    logger.debug(
+        f"Screenshot Page height: {page_height} Capture height: {SCREENSHOT_MAX_TOTAL_HEIGHT} - Stitched together in {time.time() - start:.2f}s")
+
+    return screenshot_chunks[0]
+
 class fetcher(Fetcher):
     fetcher_description = "Playwright {}/Javascript".format(
         os.getenv("PLAYWRIGHT_BROWSER_TYPE", 'chromium').capitalize()
@@ -57,8 +123,10 @@ class fetcher(Fetcher):
                 self.proxy['username'] = parsed.username
                 self.proxy['password'] = parsed.password
 
-    def screenshot_step(self, step_n=''):
-        screenshot = self.page.screenshot(type='jpeg', full_page=True, quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
+    async def screenshot_step(self, step_n=''):
+        super().screenshot_step(step_n=step_n)
+        screenshot = await capture_full_page_async(page=self.page)
+
 
         if self.browser_steps_screenshot_path is not None:
             destination = os.path.join(self.browser_steps_screenshot_path, 'step_{}.jpeg'.format(step_n))
@@ -66,14 +134,15 @@ class fetcher(Fetcher):
             with open(destination, 'wb') as f:
                 f.write(screenshot)
 
-    def save_step_html(self, step_n):
-        content = self.page.content()
+    async def save_step_html(self, step_n):
+        super().save_step_html(step_n=step_n)
+        content = await self.page.content()
         destination = os.path.join(self.browser_steps_screenshot_path, 'step_{}.html'.format(step_n))
         logger.debug(f"Saving step HTML to {destination}")
         with open(destination, 'w') as f:
             f.write(content)
 
-    def run(self,
+    async def run(self,
             url,
             timeout,
             request_headers,
@@ -81,28 +150,29 @@ class fetcher(Fetcher):
             request_method,
             ignore_status_codes=False,
             current_include_filters=None,
-            is_binary=False):
+            is_binary=False,
+            empty_pages_are_a_change=False):
 
-        from playwright.sync_api import sync_playwright
+        from playwright.async_api import async_playwright
         import playwright._impl._errors
-        from changedetectionio.content_fetchers import visualselector_xpath_selectors
+        import time
         self.delete_browser_steps_screenshots()
         response = None
 
-        with sync_playwright() as p:
+        async with async_playwright() as p:
             browser_type = getattr(p, self.browser_type)
 
             # Seemed to cause a connection Exception even tho I can see it connect
             # self.browser = browser_type.connect(self.command_executor, timeout=timeout*1000)
             # 60,000 connection timeout only
-            browser = browser_type.connect_over_cdp(self.browser_connection_url, timeout=60000)
+            browser = await browser_type.connect_over_cdp(self.browser_connection_url, timeout=60000)
 
             # SOCKS5 with authentication is not supported (yet)
             # https://github.com/microsoft/playwright/issues/10567
 
             # Set user agent to prevent Cloudflare from blocking the browser
             # Use the default one configured in the App.py model that's passed from fetch_site_status.py
-            context = browser.new_context(
+            context = await browser.new_context(
                 accept_downloads=False,  # Should never be needed
                 bypass_csp=True,  # This is needed to enable JavaScript execution on GitHub and others
                 extra_http_headers=request_headers,
@@ -112,41 +182,47 @@ class fetcher(Fetcher):
                 user_agent=manage_user_agent(headers=request_headers),
             )
 
-            self.page = context.new_page()
+            self.page = await context.new_page()
 
             # Listen for all console events and handle errors
-            self.page.on("console", lambda msg: print(f"Playwright console: Watch URL: {url} {msg.type}: {msg.text} {msg.args}"))
+            self.page.on("console", lambda msg: logger.debug(f"Playwright console: Watch URL: {url} {msg.type}: {msg.text} {msg.args}"))
 
             # Re-use as much code from browser steps as possible so its the same
             from changedetectionio.blueprint.browser_steps.browser_steps import steppable_browser_interface
             browsersteps_interface = steppable_browser_interface(start_url=url)
             browsersteps_interface.page = self.page
 
-            response = browsersteps_interface.action_goto_url(value=url)
-            self.headers = response.all_headers()
+            response = await browsersteps_interface.action_goto_url(value=url)
 
             if response is None:
-                context.close()
-                browser.close()
-                logger.debug("Content Fetcher > Response object was none")
+                await context.close()
+                await browser.close()
+                logger.debug("Content Fetcher > Response object from the browser communication was none")
                 raise EmptyReply(url=url, status_code=None)
 
+            # In async_playwright, all_headers() returns a coroutine
+            try:
+                self.headers = await response.all_headers()
+            except TypeError:
+                # Fallback for sync version
+                self.headers = response.all_headers()
+
             try:
                 if self.webdriver_js_execute_code is not None and len(self.webdriver_js_execute_code):
-                    browsersteps_interface.action_execute_js(value=self.webdriver_js_execute_code, selector=None)
+                    await browsersteps_interface.action_execute_js(value=self.webdriver_js_execute_code, selector=None)
             except playwright._impl._errors.TimeoutError as e:
-                context.close()
-                browser.close()
+                await context.close()
+                await browser.close()
                 # This can be ok, we will try to grab what we could retrieve
                 pass
             except Exception as e:
                 logger.debug(f"Content Fetcher > Other exception when executing custom JS code {str(e)}")
-                context.close()
-                browser.close()
+                await context.close()
+                await browser.close()
                 raise PageUnloadable(url=url, status_code=None, message=str(e))
 
             extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
-            self.page.wait_for_timeout(extra_wait * 1000)
+            await self.page.wait_for_timeout(extra_wait * 1000)
 
             try:
                 self.status_code = response.status
@@ -154,39 +230,50 @@ class fetcher(Fetcher):
                 # https://github.com/dgtlmoon/changedetection.io/discussions/2122#discussioncomment-8241962
                 logger.critical(f"Response from the browser/Playwright did not have a status_code! Response follows.")
                 logger.critical(response)
-                context.close()
-                browser.close()
+                await context.close()
+                await browser.close()
                 raise PageUnloadable(url=url, status_code=None, message=str(e))
 
             if self.status_code != 200 and not ignore_status_codes:
-                screenshot = self.page.screenshot(type='jpeg', full_page=True,
-                                                  quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
-
+                screenshot = await capture_full_page_async(self.page)
                 raise Non200ErrorCodeReceived(url=url, status_code=self.status_code, screenshot=screenshot)
 
-            if len(self.page.content().strip()) == 0:
-                context.close()
-                browser.close()
-                logger.debug("Content Fetcher > Content was empty")
+            if not empty_pages_are_a_change and len((await self.page.content()).strip()) == 0:
+                logger.debug("Content Fetcher > Content was empty, empty_pages_are_a_change = False")
+                await context.close()
+                await browser.close()
                 raise EmptyReply(url=url, status_code=response.status)
 
             # Run Browser Steps here
             if self.browser_steps_get_valid_steps():
-                self.iterate_browser_steps(start_url=url)
+                await self.iterate_browser_steps(start_url=url)
 
-            self.page.wait_for_timeout(extra_wait * 1000)
+            await self.page.wait_for_timeout(extra_wait * 1000)
 
+            now = time.time()
             # So we can find an element on the page where its selector was entered manually (maybe not xPath etc)
             if current_include_filters is not None:
-                self.page.evaluate("var include_filters={}".format(json.dumps(current_include_filters)))
+                await self.page.evaluate("var include_filters={}".format(json.dumps(current_include_filters)))
             else:
-                self.page.evaluate("var include_filters=''")
+                await self.page.evaluate("var include_filters=''")
+            await self.page.request_gc()
 
-            self.xpath_data = self.page.evaluate(
-                "async () => {" + self.xpath_element_js.replace('%ELEMENTS%', visualselector_xpath_selectors) + "}")
-            self.instock_data = self.page.evaluate("async () => {" + self.instock_data_js + "}")
+            # request_gc before and after evaluate to free up memory
+            # @todo browsersteps etc
+            MAX_TOTAL_HEIGHT = int(os.getenv("SCREENSHOT_MAX_HEIGHT", SCREENSHOT_MAX_HEIGHT_DEFAULT))
+            self.xpath_data = await self.page.evaluate(XPATH_ELEMENT_JS, {
+                "visualselector_xpath_selectors": visualselector_xpath_selectors,
+                "max_height": MAX_TOTAL_HEIGHT
+            })
+            await self.page.request_gc()
+
+            self.instock_data = await self.page.evaluate(INSTOCK_DATA_JS)
+            await self.page.request_gc()
+
+            self.content = await self.page.content()
+            await self.page.request_gc()
+            logger.debug(f"Scrape xPath element data in browser done in {time.time() - now:.2f}s")
 
-            self.content = self.page.content()
             # Bug 3 in Playwright screenshot handling
             # Some bug where it gives the wrong screenshot size, but making a request with the clip set first seems to solve it
             # JPEG is better here because the screenshots can be very very large
@@ -196,13 +283,41 @@ class fetcher(Fetcher):
             # acceptable screenshot quality here
             try:
                 # The actual screenshot - this always base64 and needs decoding! horrible! huge CPU usage
-                self.screenshot = self.page.screenshot(type='jpeg',
-                                                       full_page=True,
-                                                       quality=int(os.getenv("SCREENSHOT_QUALITY", 72)),
-                                                       )
+                self.screenshot = await capture_full_page_async(page=self.page)
+
             except Exception as e:
                 # It's likely the screenshot was too long/big and something crashed
                 raise ScreenshotUnavailable(url=url, status_code=self.status_code)
             finally:
-                context.close()
-                browser.close()
+                # Request garbage collection one more time before closing
+                try:
+                    await self.page.request_gc()
+                except:
+                    pass
+                
+                # Clean up resources properly
+                try:
+                    await self.page.request_gc()
+                except:
+                    pass
+
+                try:
+                    await self.page.close()
+                except:
+                    pass
+                self.page = None
+
+                try:
+                    await context.close()
+                except:
+                    pass
+                context = None
+
+                try:
+                    await browser.close()
+                except:
+                    pass
+                browser = None
+
+
+

changedetectionio/content_fetchers/puppeteer.py

@@ -6,8 +6,85 @@ from urllib.parse import urlparse
 
 from loguru import logger
 
+from changedetectionio.content_fetchers import SCREENSHOT_MAX_HEIGHT_DEFAULT, visualselector_xpath_selectors, \
+    SCREENSHOT_SIZE_STITCH_THRESHOLD, SCREENSHOT_DEFAULT_QUALITY, XPATH_ELEMENT_JS, INSTOCK_DATA_JS, \
+    SCREENSHOT_MAX_TOTAL_HEIGHT
 from changedetectionio.content_fetchers.base import Fetcher, manage_user_agent
-from changedetectionio.content_fetchers.exceptions import PageUnloadable, Non200ErrorCodeReceived, EmptyReply, BrowserFetchTimedOut, BrowserConnectError
+from changedetectionio.content_fetchers.exceptions import PageUnloadable, Non200ErrorCodeReceived, EmptyReply, BrowserFetchTimedOut, \
+    BrowserConnectError
+
+
+# Bug 3 in Playwright screenshot handling
+# Some bug where it gives the wrong screenshot size, but making a request with the clip set first seems to solve it
+
+# Screenshots also travel via the ws:// (websocket) meaning that the binary data is base64 encoded
+# which will significantly increase the IO size between the server and client, it's recommended to use the lowest
+# acceptable screenshot quality here
+async def capture_full_page(page):
+    import os
+    import time
+    from multiprocessing import Process, Pipe
+
+    start = time.time()
+
+    page_height = await page.evaluate("document.documentElement.scrollHeight")
+    page_width = await page.evaluate("document.documentElement.scrollWidth")
+    original_viewport = page.viewport
+
+    logger.debug(f"Puppeteer viewport size {page.viewport} page height {page_height} page width {page_width}")
+
+    # Bug 3 in Playwright screenshot handling
+    # Some bug where it gives the wrong screenshot size, but making a request with the clip set first seems to solve it
+    # JPEG is better here because the screenshots can be very very large
+
+    # Screenshots also travel via the ws:// (websocket) meaning that the binary data is base64 encoded
+    # which will significantly increase the IO size between the server and client, it's recommended to use the lowest
+    # acceptable screenshot quality here
+
+
+    step_size = SCREENSHOT_SIZE_STITCH_THRESHOLD # Something that will not cause the GPU to overflow when taking the screenshot
+    screenshot_chunks = []
+    y = 0
+    if page_height > page.viewport['height']:
+        if page_height < step_size:
+            step_size = page_height # Incase page is bigger than default viewport but smaller than proposed step size
+        await page.setViewport({'width': page.viewport['width'], 'height': step_size})
+
+    while y < min(page_height, SCREENSHOT_MAX_TOTAL_HEIGHT):
+        # better than scrollTo incase they override it in the page
+        await page.evaluate(
+            """(y) => {
+                document.documentElement.scrollTop = y;
+                document.body.scrollTop = y;
+            }""",
+            y
+        )
+
+        screenshot_chunks.append(await page.screenshot(type_='jpeg',
+                                                       fullPage=False,
+                                                       quality=int(os.getenv("SCREENSHOT_QUALITY", 72))))
+        y += step_size
+
+    await page.setViewport({'width': original_viewport['width'], 'height': original_viewport['height']})
+
+    if len(screenshot_chunks) > 1:
+        from changedetectionio.content_fetchers.screenshot_handler import stitch_images_worker
+        logger.debug(f"Screenshot stitching {len(screenshot_chunks)} chunks together")
+        parent_conn, child_conn = Pipe()
+        p = Process(target=stitch_images_worker, args=(child_conn, screenshot_chunks, page_height, SCREENSHOT_MAX_TOTAL_HEIGHT))
+        p.start()
+        screenshot = parent_conn.recv_bytes()
+        p.join()
+        logger.debug(
+            f"Screenshot (chunked/stitched) - Page height: {page_height} Capture height: {SCREENSHOT_MAX_TOTAL_HEIGHT} - Stitched together in {time.time() - start:.2f}s")
+
+        screenshot_chunks = None
+        return screenshot
+
+    logger.debug(
+        f"Screenshot Page height: {page_height} Capture height: {SCREENSHOT_MAX_TOTAL_HEIGHT} - Stitched together in {time.time() - start:.2f}s")
+    return screenshot_chunks[0]
+
 
 class fetcher(Fetcher):
     fetcher_description = "Puppeteer/direct {}/Javascript".format(
@@ -75,12 +152,16 @@ class fetcher(Fetcher):
                          request_method,
                          ignore_status_codes,
                          current_include_filters,
-                         is_binary
+                         is_binary,
+                         empty_pages_are_a_change
                          ):
-
-        from changedetectionio.content_fetchers import visualselector_xpath_selectors
+        import re
         self.delete_browser_steps_screenshots()
-        extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
+
+        n = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
+        extra_wait = min(n, 15)
+
+        logger.debug(f"Extra wait set to {extra_wait}s, requested was {n}s.")
 
         from pyppeteer import Pyppeteer
         pyppeteer_instance = Pyppeteer()
@@ -96,13 +177,24 @@ class fetcher(Fetcher):
         except websockets.exceptions.InvalidURI:
             raise BrowserConnectError(msg=f"Error connecting to the browser, check your browser connection address (should be ws:// or wss://")
         except Exception as e:
-            raise BrowserConnectError(msg=f"Error connecting to the browser {str(e)}")
+            raise BrowserConnectError(msg=f"Error connecting to the browser - Exception '{str(e)}'")
 
         # Better is to launch chrome with the URL as arg
         # non-headless - newPage() will launch an extra tab/window, .browser should already contain 1 page/tab
         # headless - ask a new page
         self.page = (pages := await browser.pages) and len(pages) or await browser.newPage()
 
+        if '--window-size' in self.browser_connection_url:
+            # Be sure the viewport is always the window-size, this is often not the same thing
+            match = re.search(r'--window-size=(\d+),(\d+)', self.browser_connection_url)
+            if match:
+                logger.debug(f"Setting viewport to same as --window-size in browser connection URL {int(match.group(1))},{int(match.group(2))}")
+                await self.page.setViewport({
+                    "width": int(match.group(1)),
+                    "height": int(match.group(2))
+                })
+                logger.debug(f"Puppeteer viewport size {self.page.viewport}")
+
         try:
             from pyppeteerstealth import inject_evasions_into_page
         except ImportError:
@@ -147,14 +239,35 @@ class fetcher(Fetcher):
         #            browsersteps_interface = steppable_browser_interface()
         #            browsersteps_interface.page = self.page
 
-        response = await self.page.goto(url, waitUntil="load")
+        async def handle_frame_navigation(event):
+            logger.debug(f"Frame navigated: {event}")
+            w = extra_wait - 2 if extra_wait > 4 else 2
+            logger.debug(f"Waiting {w} seconds before calling Page.stopLoading...")
+            await asyncio.sleep(w)
+            logger.debug("Issuing stopLoading command...")
+            await self.page._client.send('Page.stopLoading')
+            logger.debug("stopLoading command sent!")
 
+        self.page._client.on('Page.frameStartedNavigating', lambda event: asyncio.create_task(handle_frame_navigation(event)))
+        self.page._client.on('Page.frameStartedLoading', lambda event: asyncio.create_task(handle_frame_navigation(event)))
+        self.page._client.on('Page.frameStoppedLoading', lambda event: logger.debug(f"Frame stopped loading: {event}"))
 
-        if response is None:
-            await self.page.close()
-            await browser.close()
-            logger.warning("Content Fetcher > Response object was none")
-            raise EmptyReply(url=url, status_code=None)
+        response = None
+        attempt=0
+        while not response:
+            logger.debug(f"Attempting page fetch {url} attempt {attempt}")
+            response = await self.page.goto(url)
+            await asyncio.sleep(1 + extra_wait)
+            if response:
+                break
+            if not response:
+                logger.warning("Page did not fetch! trying again!")
+            if response is None and attempt>=2:
+                await self.page.close()
+                await browser.close()
+                logger.warning(f"Content Fetcher > Response object was none (as in, the response from the browser was empty, not just the content) exiting attmpt {attempt}")
+                raise EmptyReply(url=url, status_code=None)
+            attempt+=1
 
         self.headers = response.headers
 
@@ -180,16 +293,16 @@ class fetcher(Fetcher):
             raise PageUnloadable(url=url, status_code=None, message=str(e))
 
         if self.status_code != 200 and not ignore_status_codes:
-            screenshot = await self.page.screenshot(type_='jpeg',
-                                                    fullPage=True,
-                                                    quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
+            screenshot = await capture_full_page(page=self.page)
 
             raise Non200ErrorCodeReceived(url=url, status_code=self.status_code, screenshot=screenshot)
+
         content = await self.page.content
-        if len(content.strip()) == 0:
+
+        if not empty_pages_are_a_change and len(content.strip()) == 0:
+            logger.error("Content Fetcher > Content was empty (empty_pages_are_a_change is False), closing browsers")
             await self.page.close()
             await browser.close()
-            logger.error("Content Fetcher > Content was empty")
             raise EmptyReply(url=url, status_code=response.status)
 
         # Run Browser Steps here
@@ -197,64 +310,48 @@ class fetcher(Fetcher):
         #            if self.browser_steps_get_valid_steps():
         #                self.iterate_browser_steps()
 
-        await asyncio.sleep(1 + extra_wait)
 
         # So we can find an element on the page where its selector was entered manually (maybe not xPath etc)
         # Setup the xPath/VisualSelector scraper
-        if current_include_filters is not None:
+        if current_include_filters:
             js = json.dumps(current_include_filters)
             await self.page.evaluate(f"var include_filters={js}")
         else:
             await self.page.evaluate(f"var include_filters=''")
 
-        self.xpath_data = await self.page.evaluate(
-            "async () => {" + self.xpath_element_js.replace('%ELEMENTS%', visualselector_xpath_selectors) + "}")
-        self.instock_data = await self.page.evaluate("async () => {" + self.instock_data_js + "}")
+        MAX_TOTAL_HEIGHT = int(os.getenv("SCREENSHOT_MAX_HEIGHT", SCREENSHOT_MAX_HEIGHT_DEFAULT))
+        self.xpath_data = await self.page.evaluate(XPATH_ELEMENT_JS, {
+            "visualselector_xpath_selectors": visualselector_xpath_selectors,
+            "max_height": MAX_TOTAL_HEIGHT
+        })
+        if not self.xpath_data:
+            raise Exception(f"Content Fetcher > xPath scraper failed. Please report this URL so we can fix it :)")
+
+        self.instock_data = await self.page.evaluate(INSTOCK_DATA_JS)
 
         self.content = await self.page.content
-        # Bug 3 in Playwright screenshot handling
-        # Some bug where it gives the wrong screenshot size, but making a request with the clip set first seems to solve it
-        # JPEG is better here because the screenshots can be very very large
 
-        # Screenshots also travel via the ws:// (websocket) meaning that the binary data is base64 encoded
-        # which will significantly increase the IO size between the server and client, it's recommended to use the lowest
-        # acceptable screenshot quality here
-        try:
-            self.screenshot = await self.page.screenshot(type_='jpeg',
-                                                         fullPage=True,
-                                                         quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
-        except Exception as e:
-            logger.error("Error fetching screenshot")
-            # // May fail on very large pages with 'WARNING: tile memory limits exceeded, some content may not draw'
-            # // @ todo after text extract, we can place some overlay text with red background to say 'croppped'
-            logger.error('ERROR: content-fetcher page was maybe too large for a screenshot, reverting to viewport only screenshot')
-            try:
-                self.screenshot = await self.page.screenshot(type_='jpeg',
-                                                             fullPage=False,
-                                                             quality=int(os.getenv("SCREENSHOT_QUALITY", 72)))
-            except Exception as e:
-                logger.error('ERROR: Failed to get viewport-only reduced screenshot :(')
-                pass
-        finally:
-            # It's good to log here in the case that the browser crashes on shutting down but we still get the data we need
-            logger.success(f"Fetching '{url}' complete, closing page")
-            await self.page.close()
-            logger.success(f"Fetching '{url}' complete, closing browser")
-            await browser.close()
+        self.screenshot = await capture_full_page(page=self.page)
+
+        # It's good to log here in the case that the browser crashes on shutting down but we still get the data we need
+        logger.success(f"Fetching '{url}' complete, closing page")
+        await self.page.close()
+        logger.success(f"Fetching '{url}' complete, closing browser")
+        await browser.close()
         logger.success(f"Fetching '{url}' complete, exiting puppeteer fetch.")
 
     async def main(self, **kwargs):
         await self.fetch_page(**kwargs)
 
-    def run(self, url, timeout, request_headers, request_body, request_method, ignore_status_codes=False,
-            current_include_filters=None, is_binary=False):
+    async def run(self, url, timeout, request_headers, request_body, request_method, ignore_status_codes=False,
+            current_include_filters=None, is_binary=False, empty_pages_are_a_change=False):
 
         #@todo make update_worker async which could run any of these content_fetchers within memory and time constraints
-        max_time = os.getenv('PUPPETEER_MAX_PROCESSING_TIMEOUT_SECONDS', 180)
+        max_time = int(os.getenv('PUPPETEER_MAX_PROCESSING_TIMEOUT_SECONDS', 180))
 
-        # This will work in 3.10 but not >= 3.11 because 3.11 wants tasks only
+        # Now we run this properly in async context since we're called from async worker
         try:
-            asyncio.run(asyncio.wait_for(self.main(
+            await asyncio.wait_for(self.main(
                 url=url,
                 timeout=timeout,
                 request_headers=request_headers,
@@ -262,8 +359,9 @@ class fetcher(Fetcher):
                 request_method=request_method,
                 ignore_status_codes=ignore_status_codes,
                 current_include_filters=current_include_filters,
-                is_binary=is_binary
-            ), timeout=max_time))
+                is_binary=is_binary,
+                empty_pages_are_a_change=empty_pages_are_a_change
+            ), timeout=max_time)
         except asyncio.TimeoutError:
             raise(BrowserFetchTimedOut(msg=f"Browser connected but was unable to process the page in {max_time} seconds."))
 

changedetectionio/content_fetchers/requests.py

@@ -1,9 +1,8 @@
+from loguru import logger
 import hashlib
 import os
-
-import chardet
-import requests
-
+import asyncio
+from changedetectionio import strtobool
 from changedetectionio.content_fetchers.exceptions import BrowserStepsInUnsupportedFetcher, EmptyReply, Non200ErrorCodeReceived
 from changedetectionio.content_fetchers.base import Fetcher
 
@@ -17,7 +16,7 @@ class fetcher(Fetcher):
         self.proxy_override = proxy_override
         # browser_connection_url is none because its always 'launched locally'
 
-    def run(self,
+    def _run_sync(self,
             url,
             timeout,
             request_headers,
@@ -25,7 +24,13 @@ class fetcher(Fetcher):
             request_method,
             ignore_status_codes=False,
             current_include_filters=None,
-            is_binary=False):
+            is_binary=False,
+            empty_pages_are_a_change=False):
+        """Synchronous version of run - the original requests implementation"""
+
+        import chardet
+        import requests
+        from requests.exceptions import ProxyError, ConnectionError, RequestException
 
         if self.browser_steps_get_valid_steps():
             raise BrowserStepsInUnsupportedFetcher(url=url)
@@ -33,7 +38,6 @@ class fetcher(Fetcher):
         proxies = {}
 
         # Allows override the proxy on a per-request basis
-
         # https://requests.readthedocs.io/en/latest/user/advanced/#socks
         # Should also work with `socks5://user:pass@host:port` type syntax.
 
@@ -45,13 +49,24 @@ class fetcher(Fetcher):
             if self.system_https_proxy:
                 proxies['https'] = self.system_https_proxy
 
-        r = requests.request(method=request_method,
-                             data=request_body,
-                             url=url,
-                             headers=request_headers,
-                             timeout=timeout,
-                             proxies=proxies,
-                             verify=False)
+        session = requests.Session()
+
+        if strtobool(os.getenv('ALLOW_FILE_URI', 'false')) and url.startswith('file://'):
+            from requests_file import FileAdapter
+            session.mount('file://', FileAdapter())
+        try:
+            r = session.request(method=request_method,
+                                data=request_body.encode('utf-8') if type(request_body) is str else request_body,
+                                url=url,
+                                headers=request_headers,
+                                timeout=timeout,
+                                proxies=proxies,
+                                verify=False)
+        except Exception as e:
+            msg = str(e)
+            if proxies and 'SOCKSHTTPSConnectionPool' in msg:
+                msg = f"Proxy connection failed? {msg}"
+            raise Exception(msg) from e
 
         # If the response did not tell us what encoding format to expect, Then use chardet to override what `requests` thinks.
         # For example - some sites don't tell us it's utf-8, but return utf-8 content
@@ -67,7 +82,11 @@ class fetcher(Fetcher):
         self.headers = r.headers
 
         if not r.content or not len(r.content):
-            raise EmptyReply(url=url, status_code=r.status_code)
+            logger.debug(f"Requests returned empty content for '{url}'")
+            if not empty_pages_are_a_change:
+                raise EmptyReply(url=url, status_code=r.status_code)
+            else:
+                logger.debug(f"URL {url} gave zero byte content reply with Status Code {r.status_code}, but empty_pages_are_a_change = True")
 
         # @todo test this
         # @todo maybe you really want to test zero-byte return pages?
@@ -82,5 +101,48 @@ class fetcher(Fetcher):
         else:
             self.content = r.text
 
-
         self.raw_content = r.content
+
+    async def run(self,
+            url,
+            timeout,
+            request_headers,
+            request_body,
+            request_method,
+            ignore_status_codes=False,
+            current_include_filters=None,
+            is_binary=False,
+            empty_pages_are_a_change=False):
+        """Async wrapper that runs the synchronous requests code in a thread pool"""
+        
+        loop = asyncio.get_event_loop()
+        
+        # Run the synchronous _run_sync in a thread pool to avoid blocking the event loop
+        await loop.run_in_executor(
+            None,  # Use default ThreadPoolExecutor
+            lambda: self._run_sync(
+                url=url,
+                timeout=timeout,
+                request_headers=request_headers,
+                request_body=request_body,
+                request_method=request_method,
+                ignore_status_codes=ignore_status_codes,
+                current_include_filters=current_include_filters,
+                is_binary=is_binary,
+                empty_pages_are_a_change=empty_pages_are_a_change
+            )
+        )
+
+    def quit(self, watch=None):
+
+        # In case they switched to `requests` fetcher from something else
+        # Then the screenshot could be old, in any case, it's not used here.
+        # REMOVE_REQUESTS_OLD_SCREENSHOTS - Mainly used for testing
+        if strtobool(os.getenv("REMOVE_REQUESTS_OLD_SCREENSHOTS", 'true')):
+            screenshot = watch.get_screenshot()
+            if screenshot:
+                try:
+                    os.unlink(screenshot)
+                except Exception as e:
+                    logger.warning(f"Failed to unlink screenshot: {screenshot} - {e}")
+

changedetectionio/content_fetchers/res/puppeteer_fetch.js

@@ -1,190 +0,0 @@
-module.exports = async ({page, context}) => {
-
-    var {
-        url,
-        execute_js,
-        user_agent,
-        extra_wait_ms,
-        req_headers,
-        include_filters,
-        xpath_element_js,
-        screenshot_quality,
-        proxy_username,
-        proxy_password,
-        disk_cache_dir,
-        no_cache_list,
-        block_url_list,
-    } = context;
-
-    await page.setBypassCSP(true)
-    await page.setExtraHTTPHeaders(req_headers);
-
-    if (user_agent) {
-        await page.setUserAgent(user_agent);
-    }
-    // https://ourcodeworld.com/articles/read/1106/how-to-solve-puppeteer-timeouterror-navigation-timeout-of-30000-ms-exceeded
-
-    await page.setDefaultNavigationTimeout(0);
-
-    if (proxy_username) {
-        // Setting Proxy-Authentication header is deprecated, and doing so can trigger header change errors from Puppeteer
-        // https://github.com/puppeteer/puppeteer/issues/676 ?
-        // https://help.brightdata.com/hc/en-us/articles/12632549957649-Proxy-Manager-How-to-Guides#h_01HAKWR4Q0AFS8RZTNYWRDFJC2
-        // https://cri.dev/posts/2020-03-30-How-to-solve-Puppeteer-Chrome-Error-ERR_INVALID_ARGUMENT/
-        await page.authenticate({
-            username: proxy_username,
-            password: proxy_password
-        });
-    }
-
-    await page.setViewport({
-        width: 1024,
-        height: 768,
-        deviceScaleFactor: 1,
-    });
-
-    await page.setRequestInterception(true);
-    if (disk_cache_dir) {
-        console.log(">>>>>>>>>>>>>>> LOCAL DISK CACHE ENABLED <<<<<<<<<<<<<<<<<<<<<");
-    }
-    const fs = require('fs');
-    const crypto = require('crypto');
-
-    function file_is_expired(file_path) {
-        if (!fs.existsSync(file_path)) {
-            return true;
-        }
-        var stats = fs.statSync(file_path);
-        const now_date = new Date();
-        const expire_seconds = 300;
-        if ((now_date / 1000) - (stats.mtime.getTime() / 1000) > expire_seconds) {
-            console.log("CACHE EXPIRED: " + file_path);
-            return true;
-        }
-        return false;
-
-    }
-
-    page.on('request', async (request) => {
-        // General blocking of requests that waste traffic
-        if (block_url_list.some(substring => request.url().toLowerCase().includes(substring))) return request.abort();
-
-        if (disk_cache_dir) {
-            const url = request.url();
-            const key = crypto.createHash('md5').update(url).digest("hex");
-            const dir_path = disk_cache_dir + key.slice(0, 1) + '/' + key.slice(1, 2) + '/' + key.slice(2, 3) + '/';
-
-            // https://stackoverflow.com/questions/4482686/check-synchronously-if-file-directory-exists-in-node-js
-
-            if (fs.existsSync(dir_path + key)) {
-                console.log("* CACHE HIT , using - " + dir_path + key + " - " + url);
-                const cached_data = fs.readFileSync(dir_path + key);
-                // @todo headers can come from dir_path+key+".meta" json file
-                request.respond({
-                    status: 200,
-                    //contentType: 'text/html', //@todo
-                    body: cached_data
-                });
-                return;
-            }
-        }
-        request.continue();
-    });
-
-
-    if (disk_cache_dir) {
-        page.on('response', async (response) => {
-            const url = response.url();
-            // Basic filtering for sane responses
-            if (response.request().method() != 'GET' || response.request().resourceType() == 'xhr' || response.request().resourceType() == 'document' || response.status() != 200) {
-                console.log("Skipping (not useful) - Status:" + response.status() + " Method:" + response.request().method() + " ResourceType:" + response.request().resourceType() + " " + url);
-                return;
-            }
-            if (no_cache_list.some(substring => url.toLowerCase().includes(substring))) {
-                console.log("Skipping (no_cache_list) - " + url);
-                return;
-            }
-            if (url.toLowerCase().includes('data:')) {
-                console.log("Skipping (embedded-data) - " + url);
-                return;
-            }
-            response.buffer().then(buffer => {
-                if (buffer.length > 100) {
-                    console.log("Cache - Saving " + response.request().method() + " - " + url + " - " + response.request().resourceType());
-
-                    const key = crypto.createHash('md5').update(url).digest("hex");
-                    const dir_path = disk_cache_dir + key.slice(0, 1) + '/' + key.slice(1, 2) + '/' + key.slice(2, 3) + '/';
-
-                    if (!fs.existsSync(dir_path)) {
-                        fs.mkdirSync(dir_path, {recursive: true})
-                    }
-
-                    if (fs.existsSync(dir_path + key)) {
-                        if (file_is_expired(dir_path + key)) {
-                            fs.writeFileSync(dir_path + key, buffer);
-                        }
-                    } else {
-                        fs.writeFileSync(dir_path + key, buffer);
-                    }
-                }
-            });
-        });
-    }
-
-    const r = await page.goto(url, {
-        waitUntil: 'load'
-    });
-
-    await page.waitForTimeout(1000);
-    await page.waitForTimeout(extra_wait_ms);
-
-    if (execute_js) {
-        await page.evaluate(execute_js);
-        await page.waitForTimeout(200);
-    }
-
-    var xpath_data;
-    var instock_data;
-    try {
-        // Not sure the best way here, in the future this should be a new package added to npm then run in evaluatedCode
-        // (Once the old playwright is removed)
-        xpath_data = await page.evaluate((include_filters) => {%xpath_scrape_code%}, include_filters);
-        instock_data = await page.evaluate(() => {%instock_scrape_code%});
-    } catch (e) {
-        console.log(e);
-    }
-
-    // Protocol error (Page.captureScreenshot): Cannot take screenshot with 0 width can come from a proxy auth failure
-    // Wrap it here (for now)
-
-    var b64s = false;
-    try {
-        b64s = await page.screenshot({encoding: "base64", fullPage: true, quality: screenshot_quality, type: 'jpeg'});
-    } catch (e) {
-        console.log(e);
-    }
-
-    // May fail on very large pages with 'WARNING: tile memory limits exceeded, some content may not draw'
-    if (!b64s) {
-        // @todo after text extract, we can place some overlay text with red background to say 'croppped'
-        console.error('ERROR: content-fetcher page was maybe too large for a screenshot, reverting to viewport only screenshot');
-        try {
-            b64s = await page.screenshot({encoding: "base64", quality: screenshot_quality, type: 'jpeg'});
-        } catch (e) {
-            console.log(e);
-        }
-    }
-
-    var html = await page.content();
-    return {
-        data: {
-            'content': html,
-            'headers': r.headers(),
-            'instock_data': instock_data,
-            'screenshot': b64s,
-            'status_code': r.status(),
-            'xpath_data': xpath_data
-        },
-        type: 'application/json',
-    };
-};

changedetectionio/content_fetchers/res/stock-not-in-stock.js

@@ -1,204 +1,236 @@
-// Restock Detector
-// (c) Leigh Morresi dgtlmoon@gmail.com
-//
-// Assumes the product is in stock to begin with, unless the following appears above the fold ;
-// - outOfStockTexts appears above the fold (out of stock)
-// - negateOutOfStockRegex (really is in stock)
+async () => {
 
-function isItemInStock() {
-    // @todo Pass these in so the same list can be used in non-JS fetchers
-    const outOfStockTexts = [
-        ' أخبرني عندما يتوفر',
-        '0 in stock',
-        'actuellement indisponible',
-        'agotado',
-        'article épuisé',
-        'artikel zurzeit vergriffen',
-        'as soon as stock is available',
-        'ausverkauft', // sold out
-        'available for back order',
-        'awaiting stock',
-        'back in stock soon',
-        'back-order or out of stock',
-        'backordered',
-        'benachrichtigt mich', // notify me
-        'brak na stanie',
-        'brak w magazynie',
-        'coming soon',
-        'currently have any tickets for this',
-        'currently unavailable',
-        'dieser artikel ist bald wieder verfügbar',
-        'dostępne wkrótce',
-        'en rupture de stock',
-        'isn\'t in stock right now',
-        'isnt in stock right now',
-        'isn’t in stock right now',
-        'item is no longer available',
-        'let me know when it\'s available',
-        'mail me when available',
-        'message if back in stock',
-        'nachricht bei',
-        'nicht auf lager',
-        'nicht lagernd',
-        'nicht lieferbar',
-        'nicht verfügbar',
-        'nicht vorrätig',
-        'nicht zur verfügung',
-        'nie znaleziono produktów',
-        'niet beschikbaar',
-        'niet leverbaar',
-        'niet op voorraad',
-        'no disponible temporalmente',
-        'no longer in stock',
-        'no tickets available',
-        'not available',
-        'not currently available',
-        'not in stock',
-        'notify me when available',
-        'notify me',
-        'notify when available',
-        'não estamos a aceitar encomendas',
-        'out of stock',
-        'out-of-stock',
-        'prodotto esaurito',
-        'produkt niedostępny',
-        'sold out',
-        'sold-out',
-        'temporarily out of stock',
-        'temporarily unavailable',
-        'there were no search results for',
-        'this item is currently unavailable',
-        'tickets unavailable',
-        'tijdelijk uitverkocht',
-        'unavailable nearby',
-        'unavailable tickets',
-        'vergriffen',
-        'vorbestellen',
-        'vorbestellung ist bald möglich',
-        'we couldn\'t find any products that match',
-        'we do not currently have an estimate of when this product will be back in stock.',
-        'we don\'t know when or if this item will be back in stock.',
-        'we were not able to find a match',
-        'when this arrives in stock',
-        'zur zeit nicht an lager',
-        '品切れ',
-        '已售',
-        '已售完',
-        '품절'
-    ];
+    function isItemInStock() {
+        // @todo Pass these in so the same list can be used in non-JS fetchers
+        const outOfStockTexts = [
+            ' أخبرني عندما يتوفر',
+            '0 in stock',
+            'actuellement indisponible',
+            'agotado',
+            'article épuisé',
+            'artikel zurzeit vergriffen',
+            'as soon as stock is available',
+            'aucune offre n\'est disponible',
+            'ausverkauft', // sold out
+            'available for back order',
+            'awaiting stock',
+            'back in stock soon',
+            'back-order or out of stock',
+            'backordered',
+            'benachrichtigt mich', // notify me
+            'binnenkort leverbaar', // coming soon
+            'brak na stanie',
+            'brak w magazynie',
+            'coming soon',
+            'currently have any tickets for this',
+            'currently unavailable',
+            'dieser artikel ist bald wieder verfügbar',
+            'dostępne wkrótce',
+            'en rupture',
+            'esgotado',
+            'in kürze lieferbar',
+            'indisponible',
+            'indisponível',
+            'isn\'t in stock right now',
+            'isnt in stock right now',
+            'isn’t in stock right now',
+            'item is no longer available',
+            'let me know when it\'s available',
+            'mail me when available',
+            'message if back in stock',
+            'mevcut değil',
+            'nachricht bei',
+            'nicht auf lager',
+            'nicht lagernd',
+            'nicht lieferbar',
+            'nicht verfügbar',
+            'nicht vorrätig',
+            'nicht zur verfügung',
+            'nie znaleziono produktów',
+            'niet beschikbaar',
+            'niet leverbaar',
+            'niet op voorraad',
+            'no disponible',
+            'no featured offers available',
+            'no longer available',
+            'no longer in stock',
+            'no tickets available',
+            'non disponibile',
+            'non disponible',
+            'not available',
+            'not currently available',
+            'not in stock',
+            'notify me when available',
+            'notify me',
+            'notify when available',
+            'não disponível',
+            'não estamos a aceitar encomendas',
+            'out of stock',
+            'out-of-stock',
+            'plus disponible',
+            'prodotto esaurito',
+            'produkt niedostępny',
+            'rupture',
+            'sold out',
+            'sold-out',
+            'stok habis',
+            'stok kosong',
+            'stok varian ini habis',
+            'stokta yok',
+            'temporarily out of stock',
+            'temporarily unavailable',
+            'there were no search results for',
+            'this item is currently unavailable',
+            'tickets unavailable',
+            'tidak dijual',
+            'tidak tersedia',
+            'tijdelijk uitverkocht',
+            'tiket tidak tersedia',
+            'to subscribe to back in stock',
+            'tükendi',
+            'unavailable nearby',
+            'unavailable tickets',
+            'vergriffen',
+            'vorbestellen',
+            'vorbestellung ist bald möglich',
+            'we couldn\'t find any products that match',
+            'we do not currently have an estimate of when this product will be back in stock.',
+            'we don\'t currently have any',
+            'we don\'t know when or if this item will be back in stock.',
+            'we were not able to find a match',
+            'when this arrives in stock',
+            'when this item is available to order',
+            'zur zeit nicht an lager',
+            'épuisé',
+            '品切れ',
+            '已售',
+            '已售完',
+            '품절'
+        ];
 
 
-    const vh = Math.max(document.documentElement.clientHeight || 0, window.innerHeight || 0);
+        const vh = Math.max(document.documentElement.clientHeight || 0, window.innerHeight || 0);
 
-    function getElementBaseText(element) {
-        // .textContent can include text from children which may give the wrong results
-        // scan only immediate TEXT_NODEs, which will be a child of the element
-        var text = "";
-        for (var i = 0; i < element.childNodes.length; ++i)
-            if (element.childNodes[i].nodeType === Node.TEXT_NODE)
-                text += element.childNodes[i].textContent;
-        return text.toLowerCase().trim();
-    }
+        function getElementBaseText(element) {
+            // .textContent can include text from children which may give the wrong results
+            // scan only immediate TEXT_NODEs, which will be a child of the element
+            var text = "";
+            for (var i = 0; i < element.childNodes.length; ++i)
+                if (element.childNodes[i].nodeType === Node.TEXT_NODE)
+                    text += element.childNodes[i].textContent;
+            return text.toLowerCase().trim();
+        }
 
-    const negateOutOfStockRegex = new RegExp('^([0-9] in stock|add to cart|in stock)', 'ig');
-
-    // The out-of-stock or in-stock-text is generally always above-the-fold
-    // and often below-the-fold is a list of related products that may or may not contain trigger text
-    // so it's good to filter to just the 'above the fold' elements
-    // and it should be atleast 100px from the top to ignore items in the toolbar, sometimes menu items like "Coming soon" exist
+        const negateOutOfStockRegex = new RegExp('^([0-9] in stock|add to cart|in stock|arrives approximately)', 'ig');
+        // The out-of-stock or in-stock-text is generally always above-the-fold
+        // and often below-the-fold is a list of related products that may or may not contain trigger text
+        // so it's good to filter to just the 'above the fold' elements
+        // and it should be atleast 100px from the top to ignore items in the toolbar, sometimes menu items like "Coming soon" exist
 
+        function elementIsInEyeBallRange(element) {
+            // outside the 'fold' or some weird text in the heading area
+            // .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
+            // Note: theres also an automated test that places the 'out of stock' text fairly low down
+            // Skip text that could be in the header area
+            if (element.getBoundingClientRect().bottom + window.scrollY <= 300 ) {
+                return false;
+            }
+            // Skip text that could be much further down (like a list of "you may like" products that have 'sold out' in there
+            if (element.getBoundingClientRect().bottom + window.scrollY >= 1300 ) {
+                return false;
+            }
+            return true;
+        }
 
 // @todo - if it's SVG or IMG, go into image diff mode
-// %ELEMENTS% replaced at injection time because different interfaces use it with different settings
 
-    console.log("Scanning %ELEMENTS%");
+        function collectVisibleElements(parent, visibleElements) {
+            if (!parent) return; // Base case: if parent is null or undefined, return
 
-    function collectVisibleElements(parent, visibleElements) {
-        if (!parent) return; // Base case: if parent is null or undefined, return
+            // Add the parent itself to the visible elements array if it's of the specified types
+            visibleElements.push(parent);
 
-        // Add the parent itself to the visible elements array if it's of the specified types
-        visibleElements.push(parent);
-
-        // Iterate over the parent's children
-        const children = parent.children;
-        for (let i = 0; i < children.length; i++) {
-            const child = children[i];
-            if (
-                child.nodeType === Node.ELEMENT_NODE &&
-                window.getComputedStyle(child).display !== 'none' &&
-                window.getComputedStyle(child).visibility !== 'hidden' &&
-                child.offsetWidth >= 0 &&
-                child.offsetHeight >= 0 &&
-                window.getComputedStyle(child).contentVisibility !== 'hidden'
-            ) {
-                // If the child is an element and is visible, recursively collect visible elements
-                collectVisibleElements(child, visibleElements);
-            }
-        }
-    }
-
-    const elementsToScan = [];
-    collectVisibleElements(document.body, elementsToScan);
-
-    var elementText = "";
-
-    // REGEXS THAT REALLY MEAN IT'S IN STOCK
-    for (let i = elementsToScan.length - 1; i >= 0; i--) {
-        const element = elementsToScan[i];
-
-        // outside the 'fold' or some weird text in the heading area
-        // .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
-        if (element.getBoundingClientRect().top + window.scrollY >= vh || element.getBoundingClientRect().top + window.scrollY <= 100) {
-            continue
-        }
-
-        elementText = "";
-        if (element.tagName.toLowerCase() === "input") {
-            elementText = element.value.toLowerCase().trim();
-        } else {
-            elementText = getElementBaseText(element);
-        }
-
-        if (elementText.length) {
-            // try which ones could mean its in stock
-            if (negateOutOfStockRegex.test(elementText) && !elementText.includes('(0 products)')) {
-                console.log(`Negating/overriding 'Out of Stock' back to "Possibly in stock" found "${elementText}"`)
-                return 'Possibly in stock';
-            }
-        }
-    }
-
-    // OTHER STUFF THAT COULD BE THAT IT'S OUT OF STOCK
-    for (let i = elementsToScan.length - 1; i >= 0; i--) {
-        const element = elementsToScan[i];
-        // outside the 'fold' or some weird text in the heading area
-        // .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
-        if (element.getBoundingClientRect().top + window.scrollY >= vh + 150 || element.getBoundingClientRect().top + window.scrollY <= 100) {
-            continue
-        }
-        elementText = "";
-        if (element.tagName.toLowerCase() === "input") {
-            elementText = element.value.toLowerCase().trim();
-        } else {
-            elementText = getElementBaseText(element);
-        }
-
-        if (elementText.length) {
-            // and these mean its out of stock
-            for (const outOfStockText of outOfStockTexts) {
-                if (elementText.includes(outOfStockText)) {
-                    console.log(`Selected 'Out of Stock' - found text "${outOfStockText}" - "${elementText}"`)
-                    return outOfStockText; // item is out of stock
+            // Iterate over the parent's children
+            const children = parent.children;
+            for (let i = 0; i < children.length; i++) {
+                const child = children[i];
+                if (
+                    child.nodeType === Node.ELEMENT_NODE &&
+                    window.getComputedStyle(child).display !== 'none' &&
+                    window.getComputedStyle(child).visibility !== 'hidden' &&
+                    child.offsetWidth >= 0 &&
+                    child.offsetHeight >= 0 &&
+                    window.getComputedStyle(child).contentVisibility !== 'hidden'
+                ) {
+                    // If the child is an element and is visible, recursively collect visible elements
+                    collectVisibleElements(child, visibleElements);
                 }
             }
         }
+
+        const elementsToScan = [];
+        collectVisibleElements(document.body, elementsToScan);
+
+        var elementText = "";
+
+        // REGEXS THAT REALLY MEAN IT'S IN STOCK
+        for (let i = elementsToScan.length - 1; i >= 0; i--) {
+            const element = elementsToScan[i];
+
+            if (!elementIsInEyeBallRange(element)) {
+                continue
+            }
+
+            elementText = "";
+            try {
+                if (element.tagName.toLowerCase() === "input") {
+                    elementText = element.value.toLowerCase().trim();
+                } else {
+                    elementText = getElementBaseText(element);
+                }
+            } catch (e) {
+                console.warn('stock-not-in-stock.js scraper - handling element for gettext failed', e);
+            }
+            if (elementText.length) {
+                // try which ones could mean its in stock
+                if (negateOutOfStockRegex.test(elementText) && !elementText.includes('(0 products)')) {
+                    console.log(`Negating/overriding 'Out of Stock' back to "Possibly in stock" found "${elementText}"`)
+                    element.style.border = "2px solid green"; // highlight the element that was detected as in stock
+                    return 'Possibly in stock';
+                }
+            }
+        }
+
+        // OTHER STUFF THAT COULD BE THAT IT'S OUT OF STOCK
+        for (let i = elementsToScan.length - 1; i >= 0; i--) {
+            const element = elementsToScan[i];
+
+            if (!elementIsInEyeBallRange(element)) {
+                continue
+            }
+            elementText = "";
+            if (element.tagName.toLowerCase() === "input") {
+                elementText = element.value.toLowerCase().trim();
+            } else {
+                elementText = getElementBaseText(element);
+            }
+
+            if (elementText.length) {
+                // and these mean its out of stock
+                for (const outOfStockText of outOfStockTexts) {
+                    if (elementText.includes(outOfStockText)) {
+                        console.log(`Selected 'Out of Stock' - found text "${outOfStockText}" - "${elementText}" - offset top ${element.getBoundingClientRect().top}, page height is ${vh}`)
+                        element.style.border = "2px solid red"; // highlight the element that was detected as out of stock
+                        return outOfStockText; // item is out of stock
+                    }
+                }
+            }
+        }
+
+        console.log(`Returning 'Possibly in stock' - cant' find any useful matching text`)
+        return 'Possibly in stock'; // possibly in stock, cant decide otherwise.
     }
 
-    console.log(`Returning 'Possibly in stock' - cant' find any useful matching text`)
-    return 'Possibly in stock'; // possibly in stock, cant decide otherwise.
-}
-
 // returns the element text that makes it think it's out of stock
-return isItemInStock().trim()
-
-
+    return isItemInStock().trim()
+}

changedetectionio/content_fetchers/res/xpath_element_scraper.js

@@ -1,266 +1,284 @@
-// Copyright (C) 2021 Leigh Morresi (dgtlmoon@gmail.com)
-// All rights reserved.
+async (options) => {
 
-// @file Scrape the page looking for elements of concern (%ELEMENTS%)
-// http://matatk.agrip.org.uk/tests/position-and-width/
-// https://stackoverflow.com/questions/26813480/when-is-element-getboundingclientrect-guaranteed-to-be-updated-accurate
-//
-// Some pages like https://www.londonstockexchange.com/stock/NCCL/ncondezi-energy-limited/analysis
-// will automatically force a scroll somewhere, so include the position offset
-// Lets hope the position doesnt change while we iterate the bbox's, but this is better than nothing
-var scroll_y = 0;
-try {
-    scroll_y = +document.documentElement.scrollTop || document.body.scrollTop
-} catch (e) {
-    console.log(e);
-}
+    let visualselector_xpath_selectors = options.visualselector_xpath_selectors
+    let max_height = options.max_height
 
-
-// Include the getXpath script directly, easier than fetching
-function getxpath(e) {
-    var n = e;
-    if (n && n.id) return '//*[@id="' + n.id + '"]';
-    for (var o = []; n && Node.ELEMENT_NODE === n.nodeType;) {
-        for (var i = 0, r = !1, d = n.previousSibling; d;) d.nodeType !== Node.DOCUMENT_TYPE_NODE && d.nodeName === n.nodeName && i++, d = d.previousSibling;
-        for (d = n.nextSibling; d;) {
-            if (d.nodeName === n.nodeName) {
-                r = !0;
-                break
-            }
-            d = d.nextSibling
-        }
-        o.push((n.prefix ? n.prefix + ":" : "") + n.localName + (i || r ? "[" + (i + 1) + "]" : "")), n = n.parentNode
-    }
-    return o.length ? "/" + o.reverse().join("/") : ""
-}
-
-const findUpTag = (el) => {
-    let r = el
-    chained_css = [];
-    depth = 0;
-
-    //  Strategy 1: If it's an input, with name, and there's only one, prefer that
-    if (el.name !== undefined && el.name.length) {
-        var proposed = el.tagName + "[name=" + el.name + "]";
-        var proposed_element = window.document.querySelectorAll(proposed);
-        if (proposed_element.length) {
-            if (proposed_element.length === 1) {
-                return proposed;
-            } else {
-                // Some sites change ID but name= stays the same, we can hit it if we know the index
-                // Find all the elements that match and work out the input[n]
-                var n = Array.from(proposed_element).indexOf(el);
-                // Return a Playwright selector for nthinput[name=zipcode]
-                return proposed + " >> nth=" + n;
-            }
-        }
-    }
-
-    // Strategy 2: Keep going up until we hit an ID tag, imagine it's like  #list-widget div h4
-    while (r.parentNode) {
-        if (depth === 5) {
-            break;
-        }
-        if ('' !== r.id) {
-            chained_css.unshift("#" + CSS.escape(r.id));
-            final_selector = chained_css.join(' > ');
-            // Be sure theres only one, some sites have multiples of the same ID tag :-(
-            if (window.document.querySelectorAll(final_selector).length === 1) {
-                return final_selector;
-            }
-            return null;
-        } else {
-            chained_css.unshift(r.tagName.toLowerCase());
-        }
-        r = r.parentNode;
-        depth += 1;
-    }
-    return null;
-}
-
-
-// @todo - if it's SVG or IMG, go into image diff mode
-// %ELEMENTS% replaced at injection time because different interfaces use it with different settings
-
-var size_pos = [];
-// after page fetch, inject this JS
-// build a map of all elements and their positions (maybe that only include text?)
-var bbox;
-console.log("Scanning %ELEMENTS%");
-
-function collectVisibleElements(parent, visibleElements) {
-    if (!parent) return; // Base case: if parent is null or undefined, return
-
-
-    // Add the parent itself to the visible elements array if it's of the specified types
-    const tagName = parent.tagName.toLowerCase();
-    if ("%ELEMENTS%".split(',').includes(tagName)) {
-        visibleElements.push(parent);
-    }
-
-    // Iterate over the parent's children
-    const children = parent.children;
-    for (let i = 0; i < children.length; i++) {
-        const child = children[i];
-        if (
-            child.nodeType === Node.ELEMENT_NODE &&
-            window.getComputedStyle(child).display !== 'none' &&
-            window.getComputedStyle(child).visibility !== 'hidden' &&
-            child.offsetWidth >= 0 &&
-            child.offsetHeight >= 0 &&
-            window.getComputedStyle(child).contentVisibility !== 'hidden'
-        ) {
-            // If the child is an element and is visible, recursively collect visible elements
-            collectVisibleElements(child, visibleElements);
-        }
-    }
-}
-
-// Create an array to hold the visible elements
-const visibleElementsArray = [];
-
-// Call collectVisibleElements with the starting parent element
-collectVisibleElements(document.body, visibleElementsArray);
-
-
-visibleElementsArray.forEach(function (element) {
-
-    bbox = element.getBoundingClientRect();
-
-    // Skip really small ones, and where width or height ==0
-    if (bbox['width'] * bbox['height'] < 10) {
-        return
-    }
-
-    // Don't include elements that are offset from canvas
-    if (bbox['top'] + scroll_y < 0 || bbox['left'] < 0) {
-        return
-    }
-
-    // @todo the getXpath kind of sucks, it doesnt know when there is for example just one ID sometimes
-    // it should not traverse when we know we can anchor off just an ID one level up etc..
-    // maybe, get current class or id, keep traversing up looking for only class or id until there is just one match
-
-    // 1st primitive - if it has class, try joining it all and select, if theres only one.. well thats us.
-    xpath_result = false;
+    var scroll_y = 0;
     try {
-        var d = findUpTag(element);
-        if (d) {
-            xpath_result = d;
-        }
+        scroll_y = +document.documentElement.scrollTop || document.body.scrollTop
     } catch (e) {
         console.log(e);
     }
-    // You could swap it and default to getXpath and then try the smarter one
-    // default back to the less intelligent one
-    if (!xpath_result) {
-        try {
-            // I've seen on FB and eBay that this doesnt work
-            // ReferenceError: getXPath is not defined at eval (eval at evaluate (:152:29), <anonymous>:67:20) at UtilityScript.evaluate (<anonymous>:159:18) at UtilityScript.<anonymous> (<anonymous>:1:44)
-            xpath_result = getxpath(element);
-        } catch (e) {
-            console.log(e);
-            return
+
+// Include the getXpath script directly, easier than fetching
+    function getxpath(e) {
+        var n = e;
+        if (n && n.id) return '//*[@id="' + n.id + '"]';
+        for (var o = []; n && Node.ELEMENT_NODE === n.nodeType;) {
+            for (var i = 0, r = !1, d = n.previousSibling; d;) d.nodeType !== Node.DOCUMENT_TYPE_NODE && d.nodeName === n.nodeName && i++, d = d.previousSibling;
+            for (d = n.nextSibling; d;) {
+                if (d.nodeName === n.nodeName) {
+                    r = !0;
+                    break
+                }
+                d = d.nextSibling
+            }
+            o.push((n.prefix ? n.prefix + ":" : "") + n.localName + (i || r ? "[" + (i + 1) + "]" : "")), n = n.parentNode
         }
+        return o.length ? "/" + o.reverse().join("/") : ""
+    }
+
+    const findUpTag = (el) => {
+        let r = el
+        chained_css = [];
+        depth = 0;
+
+        //  Strategy 1: If it's an input, with name, and there's only one, prefer that
+        if (el.name !== undefined && el.name.length) {
+            var proposed = el.tagName + "[name=\"" + CSS.escape(el.name) + "\"]";
+            var proposed_element = window.document.querySelectorAll(proposed);
+            if (proposed_element.length) {
+                if (proposed_element.length === 1) {
+                    return proposed;
+                } else {
+                    // Some sites change ID but name= stays the same, we can hit it if we know the index
+                    // Find all the elements that match and work out the input[n]
+                    var n = Array.from(proposed_element).indexOf(el);
+                    // Return a Playwright selector for nthinput[name=zipcode]
+                    return proposed + " >> nth=" + n;
+                }
+            }
+        }
+
+        // Strategy 2: Keep going up until we hit an ID tag, imagine it's like  #list-widget div h4
+        while (r.parentNode) {
+            if (depth === 5) {
+                break;
+            }
+            if ('' !== r.id) {
+                chained_css.unshift("#" + CSS.escape(r.id));
+                final_selector = chained_css.join(' > ');
+                // Be sure theres only one, some sites have multiples of the same ID tag :-(
+                if (window.document.querySelectorAll(final_selector).length === 1) {
+                    return final_selector;
+                }
+                return null;
+            } else {
+                chained_css.unshift(r.tagName.toLowerCase());
+            }
+            r = r.parentNode;
+            depth += 1;
+        }
+        return null;
     }
 
 
-    size_pos.push({
-        xpath: xpath_result,
-        width: Math.round(bbox['width']),
-        height: Math.round(bbox['height']),
-        left: Math.floor(bbox['left']),
-        top: Math.floor(bbox['top']) + scroll_y,
-        tagName: (element.tagName) ? element.tagName.toLowerCase() : '',
-        tagtype: (element.tagName.toLowerCase() === 'input' && element.type) ? element.type.toLowerCase() : '',
-        isClickable: window.getComputedStyle(element).cursor == "pointer"
-    });
+// @todo - if it's SVG or IMG, go into image diff mode
 
-});
+    var size_pos = [];
+// after page fetch, inject this JS
+// build a map of all elements and their positions (maybe that only include text?)
+    var bbox;
+    console.log(`Scanning for "${visualselector_xpath_selectors}"`);
+
+    function collectVisibleElements(parent, visibleElements) {
+        if (!parent) return; // Base case: if parent is null or undefined, return
+
+
+        // Add the parent itself to the visible elements array if it's of the specified types
+        const tagName = parent.tagName.toLowerCase();
+        if (visualselector_xpath_selectors.split(',').includes(tagName)) {
+            visibleElements.push(parent);
+        }
+
+        // Iterate over the parent's children
+        const children = parent.children;
+        for (let i = 0; i < children.length; i++) {
+            const child = children[i];
+            const computedStyle = window.getComputedStyle(child);
+
+            if (
+                child.nodeType === Node.ELEMENT_NODE &&
+                computedStyle.display !== 'none' &&
+                computedStyle.visibility !== 'hidden' &&
+                child.offsetWidth >= 0 &&
+                child.offsetHeight >= 0 &&
+                computedStyle.contentVisibility !== 'hidden'
+            ) {
+                // If the child is an element and is visible, recursively collect visible elements
+                collectVisibleElements(child, visibleElements);
+            }
+        }
+    }
+
+// Create an array to hold the visible elements
+    const visibleElementsArray = [];
+
+// Call collectVisibleElements with the starting parent element
+    collectVisibleElements(document.body, visibleElementsArray);
+
+
+    visibleElementsArray.forEach(function (element) {
+
+        bbox = element.getBoundingClientRect();
+
+        // Skip really small ones, and where width or height ==0
+        if (bbox['width'] * bbox['height'] < 10) {
+            return
+        }
+
+        // Don't include elements that are offset from canvas
+        if (bbox['top'] + scroll_y < 0 || bbox['left'] < 0) {
+            return
+        }
+
+        // @todo the getXpath kind of sucks, it doesnt know when there is for example just one ID sometimes
+        // it should not traverse when we know we can anchor off just an ID one level up etc..
+        // maybe, get current class or id, keep traversing up looking for only class or id until there is just one match
+
+        // 1st primitive - if it has class, try joining it all and select, if theres only one.. well thats us.
+        xpath_result = false;
+        try {
+            var d = findUpTag(element);
+            if (d) {
+                xpath_result = d;
+            }
+        } catch (e) {
+            console.log(e);
+        }
+        // You could swap it and default to getXpath and then try the smarter one
+        // default back to the less intelligent one
+        if (!xpath_result) {
+            try {
+                // I've seen on FB and eBay that this doesnt work
+                // ReferenceError: getXPath is not defined at eval (eval at evaluate (:152:29), <anonymous>:67:20) at UtilityScript.evaluate (<anonymous>:159:18) at UtilityScript.<anonymous> (<anonymous>:1:44)
+                xpath_result = getxpath(element);
+            } catch (e) {
+                console.log(e);
+                return
+            }
+        }
+
+        let label = "not-interesting" // A placeholder, the actual labels for training are done by hand for now
+
+        let text = element.textContent.trim().slice(0, 30).trim();
+        while (/\n{2,}|\t{2,}/.test(text)) {
+            text = text.replace(/\n{2,}/g, '\n').replace(/\t{2,}/g, '\t')
+        }
+
+        // Try to identify any possible currency amounts "Sale: 4000" or "Sale now 3000 Kc", can help with the training.
+        const hasDigitCurrency = (/\d/.test(text.slice(0, 6)) || /\d/.test(text.slice(-6))) && /([€£$¥₩₹]|USD|AUD|EUR|Kč|kr|SEK|,–)/.test(text);
+        const computedStyle = window.getComputedStyle(element);
+
+        if (Math.floor(bbox['top']) + scroll_y > max_height) {
+            return
+        }
+
+        size_pos.push({
+            xpath: xpath_result,
+            width: Math.round(bbox['width']),
+            height: Math.round(bbox['height']),
+            left: Math.floor(bbox['left']),
+            top: Math.floor(bbox['top']) + scroll_y,
+            // tagName used by Browser Steps
+            tagName: (element.tagName) ? element.tagName.toLowerCase() : '',
+            // tagtype used by Browser Steps
+            tagtype: (element.tagName.toLowerCase() === 'input' && element.type) ? element.type.toLowerCase() : '',
+            isClickable: computedStyle.cursor === "pointer",
+            // Used by the keras trainer
+            fontSize: computedStyle.getPropertyValue('font-size'),
+            fontWeight: computedStyle.getPropertyValue('font-weight'),
+            hasDigitCurrency: hasDigitCurrency,
+            label: label,
+        });
+
+    });
 
 
 // Inject the current one set in the include_filters, which may be a CSS rule
 // used for displaying the current one in VisualSelector, where its not one we generated.
-if (include_filters.length) {
-    let results;
-    // Foreach filter, go and find it on the page and add it to the results so we can visualise it again
-    for (const f of include_filters) {
-        bbox = false;
-        q = false;
+    if (include_filters.length) {
+        let results;
+        // Foreach filter, go and find it on the page and add it to the results so we can visualise it again
+        for (const f of include_filters) {
+            bbox = false;
 
-        if (!f.length) {
-            console.log("xpath_element_scraper: Empty filter, skipping");
-            continue;
-        }
-
-        try {
-            // is it xpath?
-            if (f.startsWith('/') || f.startsWith('xpath')) {
-                var qry_f = f.replace(/xpath(:|\d:)/, '')
-                console.log("[xpath] Scanning for included filter " + qry_f)
-                let xpathResult = document.evaluate(qry_f, document, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null);
-                results = [];
-                for (let i = 0; i < xpathResult.snapshotLength; i++) {
-                    results.push(xpathResult.snapshotItem(i));
-                }
-            } else {
-                console.log("[css] Scanning for included filter " + f)
-                console.log("[css] Scanning for included filter " + f);
-                results = document.querySelectorAll(f);
+            if (!f.length) {
+                console.log("xpath_element_scraper: Empty filter, skipping");
+                continue;
             }
-        } catch (e) {
-            // Maybe catch DOMException and alert?
-            console.log("xpath_element_scraper: Exception selecting element from filter " + f);
-            console.log(e);
-        }
 
-        if (results.length) {
-
-            // Iterate over the results
-            results.forEach(node => {
-                // Try to resolve //something/text() back to its /something so we can atleast get the bounding box
-                try {
-                    if (typeof node.nodeName == 'string' && node.nodeName === '#text') {
-                        node = node.parentElement
+            try {
+                // is it xpath?
+                if (f.startsWith('/') || f.startsWith('xpath')) {
+                    var qry_f = f.replace(/xpath(:|\d:)/, '')
+                    console.log("[xpath] Scanning for included filter " + qry_f)
+                    let xpathResult = document.evaluate(qry_f, document, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null);
+                    results = [];
+                    for (let i = 0; i < xpathResult.snapshotLength; i++) {
+                        results.push(xpathResult.snapshotItem(i));
                     }
-                } catch (e) {
-                    console.log(e)
-                    console.log("xpath_element_scraper: #text resolver")
-                }
-
-                // #1231 - IN the case XPath attribute filter is applied, we will have to traverse up and find the element.
-                if (typeof node.getBoundingClientRect == 'function') {
-                    bbox = node.getBoundingClientRect();
-                    console.log("xpath_element_scraper: Got filter element, scroll from top was " + scroll_y)
                 } else {
+                    console.log("[css] Scanning for included filter " + f)
+                    console.log("[css] Scanning for included filter " + f);
+                    results = document.querySelectorAll(f);
+                }
+            } catch (e) {
+                // Maybe catch DOMException and alert?
+                console.log("xpath_element_scraper: Exception selecting element from filter " + f);
+                console.log(e);
+            }
+
+            if (results != null && results.length) {
+
+                // Iterate over the results
+                results.forEach(node => {
+                    // Try to resolve //something/text() back to its /something so we can atleast get the bounding box
                     try {
-                        // Try and see we can find its ownerElement
-                        bbox = node.ownerElement.getBoundingClientRect();
-                        console.log("xpath_element_scraper: Got filter by ownerElement element, scroll from top was " + scroll_y)
+                        if (typeof node.nodeName == 'string' && node.nodeName === '#text') {
+                            node = node.parentElement
+                        }
                     } catch (e) {
                         console.log(e)
-                        console.log("xpath_element_scraper: error looking up q.ownerElement")
+                        console.log("xpath_element_scraper: #text resolver")
                     }
-                }
 
-                if (bbox && bbox['width'] > 0 && bbox['height'] > 0) {
-                    size_pos.push({
-                        xpath: f,
-                        width: parseInt(bbox['width']),
-                        height: parseInt(bbox['height']),
-                        left: parseInt(bbox['left']),
-                        top: parseInt(bbox['top']) + scroll_y,
-                        highlight_as_custom_filter: true
-                    });
-                }
-            });
+                    // #1231 - IN the case XPath attribute filter is applied, we will have to traverse up and find the element.
+                    if (typeof node.getBoundingClientRect == 'function') {
+                        bbox = node.getBoundingClientRect();
+                        console.log("xpath_element_scraper: Got filter element, scroll from top was " + scroll_y)
+                    } else {
+                        try {
+                            // Try and see we can find its ownerElement
+                            bbox = node.ownerElement.getBoundingClientRect();
+                            console.log("xpath_element_scraper: Got filter by ownerElement element, scroll from top was " + scroll_y)
+                        } catch (e) {
+                            console.log(e)
+                            console.log("xpath_element_scraper: error looking up node.ownerElement")
+                        }
+                    }
+
+                    if (bbox && bbox['width'] > 0 && bbox['height'] > 0) {
+                        size_pos.push({
+                            xpath: f,
+                            width: parseInt(bbox['width']),
+                            height: parseInt(bbox['height']),
+                            left: parseInt(bbox['left']),
+                            top: parseInt(bbox['top']) + scroll_y,
+                            highlight_as_custom_filter: true
+                        });
+                    }
+                });
+            }
         }
     }
-}
 
 // Sort the elements so we find the smallest one first, in other words, we find the smallest one matching in that area
 // so that we dont select the wrapping element by mistake and be unable to select what we want
-size_pos.sort((a, b) => (a.width * a.height > b.width * b.height) ? 1 : -1)
+    size_pos.sort((a, b) => (a.width * a.height > b.width * b.height) ? 1 : -1)
+
+// browser_width required for proper scaling in the frontend
+    // Return as a string to save playwright for juggling thousands of objects
+    return JSON.stringify({'size_pos': size_pos, 'browser_width': window.innerWidth});
+}
 
-// Window.width required for proper scaling in the frontend
-return {'size_pos': size_pos, 'browser_width': window.innerWidth};

changedetectionio/content_fetchers/screenshot_handler.py

@@ -0,0 +1,73 @@
+# Pages with a vertical height longer than this will use the 'stitch together' method.
+
+# - Many GPUs have a max texture size of 16384x16384px (or lower on older devices).
+# - If a page is taller than ~8000–10000px, it risks exceeding GPU memory limits.
+# - This is especially important on headless Chromium, where Playwright may fail to allocate a massive full-page buffer.
+
+from loguru import logger
+
+from changedetectionio.content_fetchers import SCREENSHOT_MAX_HEIGHT_DEFAULT, SCREENSHOT_DEFAULT_QUALITY
+
+
+def stitch_images_worker(pipe_conn, chunks_bytes, original_page_height, capture_height):
+    import os
+    import io
+    from PIL import Image, ImageDraw, ImageFont
+
+    try:
+
+        # Load images from byte chunks
+        images = [Image.open(io.BytesIO(b)) for b in chunks_bytes]
+        total_height = sum(im.height for im in images)
+        max_width = max(im.width for im in images)
+
+        # Create stitched image
+        stitched = Image.new('RGB', (max_width, total_height))
+        y_offset = 0
+        for im in images:
+            stitched.paste(im, (0, y_offset))
+            y_offset += im.height
+
+        # Draw caption on top (overlaid, not extending canvas)
+        draw = ImageDraw.Draw(stitched)
+
+        if original_page_height > capture_height:
+            caption_text = f"WARNING: Screenshot was {original_page_height}px but trimmed to {capture_height}px because it was too long"
+            padding = 10
+            font_size = 35
+            font_color = (255, 0, 0)
+            background_color = (255, 255, 255)
+
+
+            # Try to load a proper font
+            try:
+                font = ImageFont.truetype("arial.ttf", font_size)
+            except IOError:
+                font = ImageFont.load_default()
+
+            bbox = draw.textbbox((0, 0), caption_text, font=font)
+            text_width = bbox[2] - bbox[0]
+            text_height = bbox[3] - bbox[1]
+
+            # Draw white rectangle background behind text
+            rect_top = 0
+            rect_bottom = text_height + 2 * padding
+            draw.rectangle([(0, rect_top), (max_width, rect_bottom)], fill=background_color)
+
+            # Draw text centered horizontally, 10px padding from top of the rectangle
+            text_x = (max_width - text_width) // 2
+            text_y = padding
+            draw.text((text_x, text_y), caption_text, font=font, fill=font_color)
+
+        # Encode and send image
+        output = io.BytesIO()
+        stitched.save(output, format="JPEG", quality=int(os.getenv("SCREENSHOT_QUALITY", SCREENSHOT_DEFAULT_QUALITY)))
+        pipe_conn.send_bytes(output.getvalue())
+
+        stitched.close()
+    except Exception as e:
+        pipe_conn.send(f"error:{e}")
+    finally:
+        pipe_conn.close()
+
+

changedetectionio/content_fetchers/webdriver_selenium.py

@@ -10,16 +10,13 @@ class fetcher(Fetcher):
     else:
         fetcher_description = "WebDriver Chrome/Javascript"
 
-    # Configs for Proxy setup
-    # In the ENV vars, is prefixed with "webdriver_", so it is for example "webdriver_sslProxy"
-    selenium_proxy_settings_mappings = ['proxyType', 'ftpProxy', 'httpProxy', 'noProxy',
-                                        'proxyAutoconfigUrl', 'sslProxy', 'autodetect',
-                                        'socksProxy', 'socksVersion', 'socksUsername', 'socksPassword']
     proxy = None
+    proxy_url = None
 
     def __init__(self, proxy_override=None, custom_browser_connection_url=None):
         super().__init__()
-        from selenium.webdriver.common.proxy import Proxy as SeleniumProxy
+        from urllib.parse import urlparse
+        from selenium.webdriver.common.proxy import Proxy
 
         # .strip('"') is going to save someone a lot of time when they accidently wrap the env value
         if not custom_browser_connection_url:
@@ -28,27 +25,29 @@ class fetcher(Fetcher):
             self.browser_connection_is_custom = True
             self.browser_connection_url = custom_browser_connection_url
 
-        # If any proxy settings are enabled, then we should setup the proxy object
-        proxy_args = {}
-        for k in self.selenium_proxy_settings_mappings:
-            v = os.getenv('webdriver_' + k, False)
-            if v:
-                proxy_args[k] = v.strip('"')
 
-        # Map back standard HTTP_ and HTTPS_PROXY to webDriver httpProxy/sslProxy
-        if not proxy_args.get('webdriver_httpProxy') and self.system_http_proxy:
-            proxy_args['httpProxy'] = self.system_http_proxy
-        if not proxy_args.get('webdriver_sslProxy') and self.system_https_proxy:
-            proxy_args['httpsProxy'] = self.system_https_proxy
+        ##### PROXY SETUP #####
 
-        # Allows override the proxy on a per-request basis
-        if proxy_override is not None:
-            proxy_args['httpProxy'] = proxy_override
+        proxy_sources = [
+            self.system_http_proxy,
+            self.system_https_proxy,
+            os.getenv('webdriver_proxySocks'),
+            os.getenv('webdriver_socksProxy'),
+            os.getenv('webdriver_proxyHttp'),
+            os.getenv('webdriver_httpProxy'),
+            os.getenv('webdriver_proxyHttps'),
+            os.getenv('webdriver_httpsProxy'),
+            os.getenv('webdriver_sslProxy'),
+            proxy_override, # last one should override
+        ]
+        # The built in selenium proxy handling is super unreliable!!! so we just grab which ever proxy setting we can find and throw it in --proxy-server=
+        for k in filter(None, proxy_sources):
+            if not k:
+                continue
+            self.proxy_url = k.strip()
 
-        if proxy_args:
-            self.proxy = SeleniumProxy(raw=proxy_args)
 
-    def run(self,
+    async def run(self,
             url,
             timeout,
             request_headers,
@@ -56,64 +55,89 @@ class fetcher(Fetcher):
             request_method,
             ignore_status_codes=False,
             current_include_filters=None,
-            is_binary=False):
+            is_binary=False,
+            empty_pages_are_a_change=False):
 
-        from selenium import webdriver
-        from selenium.webdriver.chrome.options import Options as ChromeOptions
-        from selenium.common.exceptions import WebDriverException
-        # request_body, request_method unused for now, until some magic in the future happens.
+        import asyncio
+        
+        # Wrap the entire selenium operation in a thread executor
+        def _run_sync():
+            from selenium.webdriver.chrome.options import Options as ChromeOptions
+            # request_body, request_method unused for now, until some magic in the future happens.
 
-        options = ChromeOptions()
-        if self.proxy:
-            options.proxy = self.proxy
+            options = ChromeOptions()
 
-        self.driver = webdriver.Remote(
-            command_executor=self.browser_connection_url,
-            options=options)
+            # Load Chrome options from env
+            CHROME_OPTIONS = [
+                line.strip()
+                for line in os.getenv("CHROME_OPTIONS", "").strip().splitlines()
+                if line.strip()
+            ]
 
-        try:
-            self.driver.get(url)
-        except WebDriverException as e:
-            # Be sure we close the session window
-            self.quit()
-            raise
+            for opt in CHROME_OPTIONS:
+                options.add_argument(opt)
 
-        self.driver.set_window_size(1280, 1024)
-        self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+            # 1. proxy_config /Proxy(proxy_config) selenium object is REALLY unreliable
+            # 2. selenium-wire cant be used because the websocket version conflicts with pypeteer-ng
+            # 3. selenium only allows ONE runner at a time by default!
+            # 4. driver must use quit() or it will continue to block/hold the selenium process!!
 
-        if self.webdriver_js_execute_code is not None:
-            self.driver.execute_script(self.webdriver_js_execute_code)
-            # Selenium doesn't automatically wait for actions as good as Playwright, so wait again
-            self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+            if self.proxy_url:
+                options.add_argument(f'--proxy-server={self.proxy_url}')
 
-        # @todo - how to check this? is it possible?
-        self.status_code = 200
-        # @todo somehow we should try to get this working for WebDriver
-        # raise EmptyReply(url=url, status_code=r.status_code)
-
-        # @todo - dom wait loaded?
-        time.sleep(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay)
-        self.content = self.driver.page_source
-        self.headers = {}
-
-        self.screenshot = self.driver.get_screenshot_as_png()
-
-    # Does the connection to the webdriver work? run a test connection.
-    def is_ready(self):
-        from selenium import webdriver
-        from selenium.webdriver.chrome.options import Options as ChromeOptions
-
-        self.driver = webdriver.Remote(
-            command_executor=self.command_executor,
-            options=ChromeOptions())
-
-        # driver.quit() seems to cause better exceptions
-        self.quit()
-        return True
-
-    def quit(self):
-        if self.driver:
+            from selenium.webdriver.remote.remote_connection import RemoteConnection
+            from selenium.webdriver.remote.webdriver import WebDriver as RemoteWebDriver
+            driver = None
             try:
-                self.driver.quit()
+                # Create the RemoteConnection and set timeout (e.g., 30 seconds)
+                remote_connection = RemoteConnection(
+                    self.browser_connection_url,
+                )
+                remote_connection.set_timeout(30)  # seconds
+
+                # Now create the driver with the RemoteConnection
+                driver = RemoteWebDriver(
+                    command_executor=remote_connection,
+                    options=options
+                )
+
+                driver.set_page_load_timeout(int(os.getenv("WEBDRIVER_PAGELOAD_TIMEOUT", 45)))
             except Exception as e:
-                logger.debug(f"Content Fetcher > Exception in chrome shutdown/quit {str(e)}")
+                if driver:
+                    driver.quit()
+                raise e
+
+            try:
+                driver.get(url)
+
+                if not "--window-size" in os.getenv("CHROME_OPTIONS", ""):
+                    driver.set_window_size(1280, 1024)
+
+                driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+
+                if self.webdriver_js_execute_code is not None:
+                    driver.execute_script(self.webdriver_js_execute_code)
+                    # Selenium doesn't automatically wait for actions as good as Playwright, so wait again
+                    driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+
+                # @todo - how to check this? is it possible?
+                self.status_code = 200
+                # @todo somehow we should try to get this working for WebDriver
+                # raise EmptyReply(url=url, status_code=r.status_code)
+
+                # @todo - dom wait loaded?
+                import time
+                time.sleep(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay)
+                self.content = driver.page_source
+                self.headers = {}
+                self.screenshot = driver.get_screenshot_as_png()
+            except Exception as e:
+                driver.quit()
+                raise e
+
+            driver.quit()
+
+        # Run the selenium operations in a thread pool to avoid blocking the event loop
+        loop = asyncio.get_event_loop()
+        await loop.run_in_executor(None, _run_sync)
+

changedetectionio/custom_queue.py

@@ -0,0 +1,535 @@
+import queue
+import asyncio
+from blinker import signal
+from loguru import logger
+
+
+class NotificationQueue(queue.Queue):
+    """
+    Extended Queue that sends a 'notification_event' signal when notifications are added.
+    
+    This class extends the standard Queue and adds a signal emission after a notification
+    is put into the queue. The signal includes the watch UUID if available.
+    """
+    
+    def __init__(self, maxsize=0):
+        super().__init__(maxsize)
+        try:
+            self.notification_event_signal = signal('notification_event')
+        except Exception as e:
+            logger.critical(f"Exception creating notification_event signal: {e}")
+
+    def put(self, item, block=True, timeout=None):
+        # Call the parent's put method first
+        super().put(item, block, timeout)
+        
+        # After putting the notification in the queue, emit signal with watch UUID
+        try:
+            if self.notification_event_signal and isinstance(item, dict):
+                watch_uuid = item.get('uuid')
+                if watch_uuid:
+                    # Send the notification_event signal with the watch UUID
+                    self.notification_event_signal.send(watch_uuid=watch_uuid)
+                    logger.trace(f"NotificationQueue: Emitted notification_event signal for watch UUID {watch_uuid}")
+                else:
+                    # Send signal without UUID for system notifications
+                    self.notification_event_signal.send()
+                    logger.trace("NotificationQueue: Emitted notification_event signal for system notification")
+        except Exception as e:
+            logger.error(f"Exception emitting notification_event signal: {e}")
+
+class SignalPriorityQueue(queue.PriorityQueue):
+    """
+    Extended PriorityQueue that sends a signal when items with a UUID are added.
+    
+    This class extends the standard PriorityQueue and adds a signal emission
+    after an item is put into the queue. If the item contains a UUID, the signal
+    is sent with that UUID as a parameter.
+    """
+    
+    def __init__(self, maxsize=0):
+        super().__init__(maxsize)
+        try:
+            self.queue_length_signal = signal('queue_length')
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+
+    def put(self, item, block=True, timeout=None):
+        # Call the parent's put method first
+        super().put(item, block, timeout)
+        
+        # After putting the item in the queue, check if it has a UUID and emit signal
+        if hasattr(item, 'item') and isinstance(item.item, dict) and 'uuid' in item.item:
+            uuid = item.item['uuid']
+            # Get the signal and send it if it exists
+            watch_check_update = signal('watch_check_update')
+            if watch_check_update:
+                # Send the watch_uuid parameter
+                watch_check_update.send(watch_uuid=uuid)
+        
+        # Send queue_length signal with current queue size
+        try:
+
+            if self.queue_length_signal:
+                self.queue_length_signal.send(length=self.qsize())
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+
+    def get(self, block=True, timeout=None):
+        # Call the parent's get method first
+        item = super().get(block, timeout)
+        
+        # Send queue_length signal with current queue size
+        try:
+            if self.queue_length_signal:
+                self.queue_length_signal.send(length=self.qsize())
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+        return item
+    
+    def get_uuid_position(self, target_uuid):
+        """
+        Find the position of a watch UUID in the priority queue.
+        Optimized for large queues - O(n) complexity instead of O(n log n).
+        
+        Args:
+            target_uuid: The UUID to search for
+            
+        Returns:
+            dict: Contains position info or None if not found
+                - position: 0-based position in queue (0 = next to be processed)
+                - total_items: total number of items in queue
+                - priority: the priority value of the found item
+        """
+        with self.mutex:
+            queue_list = list(self.queue)
+            total_items = len(queue_list)
+            
+            if total_items == 0:
+                return {
+                    'position': None,
+                    'total_items': 0,
+                    'priority': None,
+                    'found': False
+                }
+            
+            # Find the target item and its priority first - O(n)
+            target_item = None
+            target_priority = None
+            
+            for item in queue_list:
+                if (hasattr(item, 'item') and 
+                    isinstance(item.item, dict) and 
+                    item.item.get('uuid') == target_uuid):
+                    target_item = item
+                    target_priority = item.priority
+                    break
+            
+            if target_item is None:
+                return {
+                    'position': None,
+                    'total_items': total_items,
+                    'priority': None,
+                    'found': False
+                }
+            
+            # Count how many items have higher priority (lower numbers) - O(n)
+            position = 0
+            for item in queue_list:
+                # Items with lower priority numbers are processed first
+                if item.priority < target_priority:
+                    position += 1
+                elif item.priority == target_priority and item != target_item:
+                    # For same priority, count items that come before this one
+                    # (Note: this is approximate since heap order isn't guaranteed for equal priorities)
+                    position += 1
+            
+            return {
+                'position': position,
+                'total_items': total_items,
+                'priority': target_priority,
+                'found': True
+            }
+    
+    def get_all_queued_uuids(self, limit=None, offset=0):
+        """
+        Get UUIDs currently in the queue with their positions.
+        For large queues, use limit/offset for pagination.
+        
+        Args:
+            limit: Maximum number of items to return (None = all)
+            offset: Number of items to skip (for pagination)
+        
+        Returns:
+            dict: Contains items and metadata
+                - items: List of dicts with uuid, position, and priority
+                - total_items: Total number of items in queue
+                - returned_items: Number of items returned
+                - has_more: Whether there are more items after this page
+        """
+        with self.mutex:
+            queue_list = list(self.queue)
+            total_items = len(queue_list)
+            
+            if total_items == 0:
+                return {
+                    'items': [],
+                    'total_items': 0,
+                    'returned_items': 0,
+                    'has_more': False
+                }
+            
+            # For very large queues, warn about performance
+            if total_items > 1000 and limit is None:
+                logger.warning(f"Getting all {total_items} queued items without limit - this may be slow")
+            
+            # Sort only if we need exact positions (expensive for large queues)
+            if limit is not None and limit <= 100:
+                # For small requests, we can afford to sort
+                queue_items = sorted(queue_list)
+                end_idx = min(offset + limit, len(queue_items)) if limit else len(queue_items)
+                items_to_process = queue_items[offset:end_idx]
+                
+                result = []
+                for position, item in enumerate(items_to_process, start=offset):
+                    if (hasattr(item, 'item') and 
+                        isinstance(item.item, dict) and 
+                        'uuid' in item.item):
+                        
+                        result.append({
+                            'uuid': item.item['uuid'],
+                            'position': position,
+                            'priority': item.priority
+                        })
+                
+                return {
+                    'items': result,
+                    'total_items': total_items,
+                    'returned_items': len(result),
+                    'has_more': (offset + len(result)) < total_items
+                }
+            else:
+                # For large requests, return items with approximate positions
+                # This is much faster O(n) instead of O(n log n)
+                result = []
+                processed = 0
+                skipped = 0
+                
+                for item in queue_list:
+                    if (hasattr(item, 'item') and 
+                        isinstance(item.item, dict) and 
+                        'uuid' in item.item):
+                        
+                        if skipped < offset:
+                            skipped += 1
+                            continue
+                        
+                        if limit and processed >= limit:
+                            break
+                        
+                        # Approximate position based on priority comparison
+                        approx_position = sum(1 for other in queue_list if other.priority < item.priority)
+                        
+                        result.append({
+                            'uuid': item.item['uuid'],
+                            'position': approx_position,  # Approximate
+                            'priority': item.priority
+                        })
+                        processed += 1
+                
+                return {
+                    'items': result,
+                    'total_items': total_items,
+                    'returned_items': len(result),
+                    'has_more': (offset + len(result)) < total_items,
+                    'note': 'Positions are approximate for performance with large queues'
+                }
+    
+    def get_queue_summary(self):
+        """
+        Get a quick summary of queue state without expensive operations.
+        O(n) complexity - fast even for large queues.
+        
+        Returns:
+            dict: Queue summary statistics
+        """
+        with self.mutex:
+            queue_list = list(self.queue)
+            total_items = len(queue_list)
+            
+            if total_items == 0:
+                return {
+                    'total_items': 0,
+                    'priority_breakdown': {},
+                    'immediate_items': 0,
+                    'clone_items': 0,
+                    'scheduled_items': 0
+                }
+            
+            # Count items by priority type - O(n)
+            immediate_items = 0  # priority 1
+            clone_items = 0      # priority 5  
+            scheduled_items = 0  # priority > 100 (timestamps)
+            priority_counts = {}
+            
+            for item in queue_list:
+                priority = item.priority
+                priority_counts[priority] = priority_counts.get(priority, 0) + 1
+                
+                if priority == 1:
+                    immediate_items += 1
+                elif priority == 5:
+                    clone_items += 1
+                elif priority > 100:
+                    scheduled_items += 1
+            
+            return {
+                'total_items': total_items,
+                'priority_breakdown': priority_counts,
+                'immediate_items': immediate_items,
+                'clone_items': clone_items,
+                'scheduled_items': scheduled_items,
+                'min_priority': min(priority_counts.keys()) if priority_counts else None,
+                'max_priority': max(priority_counts.keys()) if priority_counts else None
+            }
+
+
+class AsyncSignalPriorityQueue(asyncio.PriorityQueue):
+    """
+    Async version of SignalPriorityQueue that sends signals when items are added/removed.
+    
+    This class extends asyncio.PriorityQueue and maintains the same signal behavior
+    as the synchronous version for real-time UI updates.
+    """
+    
+    def __init__(self, maxsize=0):
+        super().__init__(maxsize)
+        try:
+            self.queue_length_signal = signal('queue_length')
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+
+    async def put(self, item):
+        # Call the parent's put method first
+        await super().put(item)
+        
+        # After putting the item in the queue, check if it has a UUID and emit signal
+        if hasattr(item, 'item') and isinstance(item.item, dict) and 'uuid' in item.item:
+            uuid = item.item['uuid']
+            # Get the signal and send it if it exists
+            watch_check_update = signal('watch_check_update')
+            if watch_check_update:
+                # Send the watch_uuid parameter
+                watch_check_update.send(watch_uuid=uuid)
+        
+        # Send queue_length signal with current queue size
+        try:
+            if self.queue_length_signal:
+                self.queue_length_signal.send(length=self.qsize())
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+
+    async def get(self):
+        # Call the parent's get method first
+        item = await super().get()
+        
+        # Send queue_length signal with current queue size
+        try:
+            if self.queue_length_signal:
+                self.queue_length_signal.send(length=self.qsize())
+        except Exception as e:
+            logger.critical(f"Exception: {e}")
+        return item
+    
+    @property
+    def queue(self):
+        """
+        Provide compatibility with sync PriorityQueue.queue access
+        Returns the internal queue for template access
+        """
+        return self._queue if hasattr(self, '_queue') else []
+    
+    def get_uuid_position(self, target_uuid):
+        """
+        Find the position of a watch UUID in the async priority queue.
+        Optimized for large queues - O(n) complexity instead of O(n log n).
+        
+        Args:
+            target_uuid: The UUID to search for
+            
+        Returns:
+            dict: Contains position info or None if not found
+                - position: 0-based position in queue (0 = next to be processed)
+                - total_items: total number of items in queue
+                - priority: the priority value of the found item
+        """
+        queue_list = list(self._queue)
+        total_items = len(queue_list)
+        
+        if total_items == 0:
+            return {
+                'position': None,
+                'total_items': 0,
+                'priority': None,
+                'found': False
+            }
+        
+        # Find the target item and its priority first - O(n)
+        target_item = None
+        target_priority = None
+        
+        for item in queue_list:
+            if (hasattr(item, 'item') and 
+                isinstance(item.item, dict) and 
+                item.item.get('uuid') == target_uuid):
+                target_item = item
+                target_priority = item.priority
+                break
+        
+        if target_item is None:
+            return {
+                'position': None,
+                'total_items': total_items,
+                'priority': None,
+                'found': False
+            }
+        
+        # Count how many items have higher priority (lower numbers) - O(n)
+        position = 0
+        for item in queue_list:
+            if item.priority < target_priority:
+                position += 1
+            elif item.priority == target_priority and item != target_item:
+                position += 1
+        
+        return {
+            'position': position,
+            'total_items': total_items,
+            'priority': target_priority,
+            'found': True
+        }
+    
+    def get_all_queued_uuids(self, limit=None, offset=0):
+        """
+        Get UUIDs currently in the async queue with their positions.
+        For large queues, use limit/offset for pagination.
+        
+        Args:
+            limit: Maximum number of items to return (None = all)
+            offset: Number of items to skip (for pagination)
+        
+        Returns:
+            dict: Contains items and metadata (same structure as sync version)
+        """
+        queue_list = list(self._queue)
+        total_items = len(queue_list)
+        
+        if total_items == 0:
+            return {
+                'items': [],
+                'total_items': 0,
+                'returned_items': 0,
+                'has_more': False
+            }
+        
+        # Same logic as sync version but without mutex
+        if limit is not None and limit <= 100:
+            queue_items = sorted(queue_list)
+            end_idx = min(offset + limit, len(queue_items)) if limit else len(queue_items)
+            items_to_process = queue_items[offset:end_idx]
+            
+            result = []
+            for position, item in enumerate(items_to_process, start=offset):
+                if (hasattr(item, 'item') and 
+                    isinstance(item.item, dict) and 
+                    'uuid' in item.item):
+                    
+                    result.append({
+                        'uuid': item.item['uuid'],
+                        'position': position,
+                        'priority': item.priority
+                    })
+            
+            return {
+                'items': result,
+                'total_items': total_items,
+                'returned_items': len(result),
+                'has_more': (offset + len(result)) < total_items
+            }
+        else:
+            # Fast approximate positions for large queues
+            result = []
+            processed = 0
+            skipped = 0
+            
+            for item in queue_list:
+                if (hasattr(item, 'item') and 
+                    isinstance(item.item, dict) and 
+                    'uuid' in item.item):
+                    
+                    if skipped < offset:
+                        skipped += 1
+                        continue
+                    
+                    if limit and processed >= limit:
+                        break
+                    
+                    approx_position = sum(1 for other in queue_list if other.priority < item.priority)
+                    
+                    result.append({
+                        'uuid': item.item['uuid'],
+                        'position': approx_position,
+                        'priority': item.priority
+                    })
+                    processed += 1
+            
+            return {
+                'items': result,
+                'total_items': total_items,
+                'returned_items': len(result),
+                'has_more': (offset + len(result)) < total_items,
+                'note': 'Positions are approximate for performance with large queues'
+            }
+    
+    def get_queue_summary(self):
+        """
+        Get a quick summary of async queue state.
+        O(n) complexity - fast even for large queues.
+        """
+        queue_list = list(self._queue)
+        total_items = len(queue_list)
+        
+        if total_items == 0:
+            return {
+                'total_items': 0,
+                'priority_breakdown': {},
+                'immediate_items': 0,
+                'clone_items': 0,
+                'scheduled_items': 0
+            }
+        
+        immediate_items = 0
+        clone_items = 0
+        scheduled_items = 0
+        priority_counts = {}
+        
+        for item in queue_list:
+            priority = item.priority
+            priority_counts[priority] = priority_counts.get(priority, 0) + 1
+            
+            if priority == 1:
+                immediate_items += 1
+            elif priority == 5:
+                clone_items += 1
+            elif priority > 100:
+                scheduled_items += 1
+        
+        return {
+            'total_items': total_items,
+            'priority_breakdown': priority_counts,
+            'immediate_items': immediate_items,
+            'clone_items': clone_items,
+            'scheduled_items': scheduled_items,
+            'min_priority': min(priority_counts.keys()) if priority_counts else None,
+            'max_priority': max(priority_counts.keys()) if priority_counts else None
+        }

changedetectionio/diff.py

@@ -1,6 +1,9 @@
 import difflib
 from typing import List, Iterator, Union
 
+REMOVED_STYLE = "background-color: #fadad7; color: #b30000;"
+ADDED_STYLE = "background-color: #eaf2c2; color: #406619;"
+
 def same_slicer(lst: List[str], start: int, end: int) -> List[str]:
     """Return a slice of the list, or a single element if start == end."""
     return lst[start:end] if start != end else [lst[start]]
@@ -12,11 +15,12 @@ def customSequenceMatcher(
     include_removed: bool = True,
     include_added: bool = True,
     include_replaced: bool = True,
-    include_change_type_prefix: bool = True
+    include_change_type_prefix: bool = True,
+    html_colour: bool = False
 ) -> Iterator[List[str]]:
     """
     Compare two sequences and yield differences based on specified parameters.
-    
+
     Args:
         before (List[str]): Original sequence
         after (List[str]): Modified sequence
@@ -25,26 +29,35 @@ def customSequenceMatcher(
         include_added (bool): Include added parts
         include_replaced (bool): Include replaced parts
         include_change_type_prefix (bool): Add prefixes to indicate change types
-    
+        html_colour (bool): Use HTML background colors for differences
+
     Yields:
         List[str]: Differences between sequences
     """
     cruncher = difflib.SequenceMatcher(isjunk=lambda x: x in " \t", a=before, b=after)
-    
+
+
+
     for tag, alo, ahi, blo, bhi in cruncher.get_opcodes():
         if include_equal and tag == 'equal':
             yield before[alo:ahi]
         elif include_removed and tag == 'delete':
-            prefix = "(removed) " if include_change_type_prefix else ''
-            yield [f"{prefix}{line}" for line in same_slicer(before, alo, ahi)]
+            if html_colour:
+                yield [f'<span style="{REMOVED_STYLE}">{line}</span>' for line in same_slicer(before, alo, ahi)]
+            else:
+                yield [f"(removed) {line}" for line in same_slicer(before, alo, ahi)] if include_change_type_prefix else same_slicer(before, alo, ahi)
         elif include_replaced and tag == 'replace':
-            prefix_changed = "(changed) " if include_change_type_prefix else ''
-            prefix_into = "(into) " if include_change_type_prefix else ''
-            yield [f"{prefix_changed}{line}" for line in same_slicer(before, alo, ahi)] + \
-                  [f"{prefix_into}{line}" for line in same_slicer(after, blo, bhi)]
+            if html_colour:
+                yield [f'<span style="{REMOVED_STYLE}">{line}</span>' for line in same_slicer(before, alo, ahi)] + \
+                      [f'<span style="{ADDED_STYLE}">{line}</span>' for line in same_slicer(after, blo, bhi)]
+            else:
+                yield [f"(changed) {line}" for line in same_slicer(before, alo, ahi)] + \
+                      [f"(into) {line}" for line in same_slicer(after, blo, bhi)] if include_change_type_prefix else same_slicer(before, alo, ahi) + same_slicer(after, blo, bhi)
         elif include_added and tag == 'insert':
-            prefix = "(added) " if include_change_type_prefix else ''
-            yield [f"{prefix}{line}" for line in same_slicer(after, blo, bhi)]
+            if html_colour:
+                yield [f'<span style="{ADDED_STYLE}">{line}</span>' for line in same_slicer(after, blo, bhi)]
+            else:
+                yield [f"(added) {line}" for line in same_slicer(after, blo, bhi)] if include_change_type_prefix else same_slicer(after, blo, bhi)
 
 def render_diff(
     previous_version_file_contents: str,
@@ -55,11 +68,12 @@ def render_diff(
     include_replaced: bool = True,
     line_feed_sep: str = "\n",
     include_change_type_prefix: bool = True,
-    patch_format: bool = False
+    patch_format: bool = False,
+    html_colour: bool = False
 ) -> str:
     """
     Render the difference between two file contents.
-    
+
     Args:
         previous_version_file_contents (str): Original file contents
         newest_version_file_contents (str): Modified file contents
@@ -70,7 +84,8 @@ def render_diff(
         line_feed_sep (str): Separator for lines in output
         include_change_type_prefix (bool): Add prefixes to indicate change types
         patch_format (bool): Use patch format for output
-    
+        html_colour (bool): Use HTML background colors for differences
+
     Returns:
         str: Rendered difference
     """
@@ -88,10 +103,11 @@ def render_diff(
         include_removed=include_removed,
         include_added=include_added,
         include_replaced=include_replaced,
-        include_change_type_prefix=include_change_type_prefix
+        include_change_type_prefix=include_change_type_prefix,
+        html_colour=html_colour
     )
 
     def flatten(lst: List[Union[str, List[str]]]) -> str:
         return line_feed_sep.join(flatten(x) if isinstance(x, list) else x for x in lst)
 
-    return flatten(rendered_diff)
+    return flatten(rendered_diff)

changedetectionio/forms.py

@@ -1,10 +1,16 @@
 import os
 import re
+from loguru import logger
+from wtforms.widgets.core import TimeInput
+
+from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES
+from changedetectionio.conditions.form import ConditionFormRow
 from changedetectionio.strtobool import strtobool
 
 from wtforms import (
     BooleanField,
     Form,
+    Field,
     IntegerField,
     RadioField,
     SelectField,
@@ -123,6 +129,87 @@ class StringTagUUID(StringField):
 
         return 'error'
 
+class TimeDurationForm(Form):
+    hours = SelectField(choices=[(f"{i}", f"{i}") for i in range(0, 25)], default="24",  validators=[validators.Optional()])
+    minutes = SelectField(choices=[(f"{i}", f"{i}") for i in range(0, 60)], default="00", validators=[validators.Optional()])
+
+class TimeStringField(Field):
+    """
+    A WTForms field for time inputs (HH:MM) that stores the value as a string.
+    """
+    widget = TimeInput()  # Use the built-in time input widget
+
+    def _value(self):
+        """
+        Returns the value for rendering in the form.
+        """
+        return self.data if self.data is not None else ""
+
+    def process_formdata(self, valuelist):
+        """
+        Processes the raw input from the form and stores it as a string.
+        """
+        if valuelist:
+            time_str = valuelist[0]
+            # Simple validation for HH:MM format
+            if not time_str or len(time_str.split(":")) != 2:
+                raise ValidationError("Invalid time format. Use HH:MM.")
+            self.data = time_str
+
+
+class validateTimeZoneName(object):
+    """
+       Flask wtform validators wont work with basic auth
+    """
+
+    def __init__(self, message=None):
+        self.message = message
+
+    def __call__(self, form, field):
+        from zoneinfo import available_timezones
+        python_timezones = available_timezones()
+        if field.data and field.data not in python_timezones:
+            raise ValidationError("Not a valid timezone name")
+
+class ScheduleLimitDaySubForm(Form):
+    enabled = BooleanField("not set", default=True)
+    start_time = TimeStringField("Start At", default="00:00", validators=[validators.Optional()])
+    duration = FormField(TimeDurationForm, label="Run duration")
+
+class ScheduleLimitForm(Form):
+    enabled = BooleanField("Use time scheduler", default=False)
+    # Because the label for=""" doesnt line up/work with the actual checkbox
+    monday = FormField(ScheduleLimitDaySubForm, label="")
+    tuesday = FormField(ScheduleLimitDaySubForm, label="")
+    wednesday = FormField(ScheduleLimitDaySubForm, label="")
+    thursday = FormField(ScheduleLimitDaySubForm, label="")
+    friday = FormField(ScheduleLimitDaySubForm, label="")
+    saturday = FormField(ScheduleLimitDaySubForm, label="")
+    sunday = FormField(ScheduleLimitDaySubForm, label="")
+
+    timezone = StringField("Optional timezone to run in",
+                                  render_kw={"list": "timezones"},
+                                  validators=[validateTimeZoneName()]
+                                  )
+    def __init__(
+        self,
+        formdata=None,
+        obj=None,
+        prefix="",
+        data=None,
+        meta=None,
+        **kwargs,
+    ):
+        super().__init__(formdata, obj, prefix, data, meta, **kwargs)
+        self.monday.form.enabled.label.text="Monday"
+        self.tuesday.form.enabled.label.text = "Tuesday"
+        self.wednesday.form.enabled.label.text = "Wednesday"
+        self.thursday.form.enabled.label.text = "Thursday"
+        self.friday.form.enabled.label.text = "Friday"
+        self.saturday.form.enabled.label.text = "Saturday"
+        self.sunday.form.enabled.label.text = "Sunday"
+
+
 class TimeBetweenCheckForm(Form):
     weeks = IntegerField('Weeks', validators=[validators.Optional(), validators.NumberRange(min=0, message="Should contain zero or more seconds")])
     days = IntegerField('Days', validators=[validators.Optional(), validators.NumberRange(min=0, message="Should contain zero or more seconds")])
@@ -137,27 +224,37 @@ class StringDictKeyValue(StringField):
 
     def _value(self):
         if self.data:
-            output = u''
-            for k in self.data.keys():
-                output += "{}: {}\r\n".format(k, self.data[k])
-
+            output = ''
+            for k, v in self.data.items():
+                output += f"{k}: {v}\r\n"
             return output
         else:
-            return u''
+            return ''
 
-    # incoming
+    # incoming data processing + validation
     def process_formdata(self, valuelist):
+        self.data = {}
+        errors = []
         if valuelist:
-            self.data = {}
-            # Remove empty strings
-            cleaned = list(filter(None, valuelist[0].split("\n")))
-            for s in cleaned:
-                parts = s.strip().split(':', 1)
-                if len(parts) == 2:
-                    self.data.update({parts[0].strip(): parts[1].strip()})
+            # Remove empty strings (blank lines)
+            cleaned = [line.strip() for line in valuelist[0].split("\n") if line.strip()]
+            for idx, s in enumerate(cleaned, start=1):
+                if ':' not in s:
+                    errors.append(f"Line {idx} is missing a ':' separator.")
+                    continue
+                parts = s.split(':', 1)
+                key = parts[0].strip()
+                value = parts[1].strip()
 
-        else:
-            self.data = {}
+                if not key:
+                    errors.append(f"Line {idx} has an empty key.")
+                if not value:
+                    errors.append(f"Line {idx} has an empty value.")
+
+                self.data[key] = value
+
+        if errors:
+            raise ValidationError("Invalid input:\n" + "\n".join(errors))
 
 class ValidateContentFetcherIsReady(object):
     """
@@ -219,20 +316,24 @@ class ValidateAppRiseServers(object):
 
     def __call__(self, form, field):
         import apprise
-        apobj = apprise.Apprise()
+        from .notification.apprise_plugin.assets import apprise_asset
+        from .notification.apprise_plugin.custom_handlers import apprise_http_custom_handler  # noqa: F401
+
+        apobj = apprise.Apprise(asset=apprise_asset)
 
         for server_url in field.data:
-            if not apobj.add(server_url):
-                message = field.gettext('\'%s\' is not a valid AppRise URL.' % (server_url))
+            url = server_url.strip()
+            if url.startswith("#"):
+                continue
+
+            if not apobj.add(url):
+                message = field.gettext('\'%s\' is not a valid AppRise URL.' % (url))
                 raise ValidationError(message)
 
 class ValidateJinja2Template(object):
     """
     Validates that a {token} is from a valid set
     """
-    def __init__(self, message=None):
-        self.message = message
-
     def __call__(self, form, field):
         from changedetectionio import notification
 
@@ -247,6 +348,10 @@ class ValidateJinja2Template(object):
         try:
             jinja2_env = ImmutableSandboxedEnvironment(loader=BaseLoader)
             jinja2_env.globals.update(notification.valid_tokens)
+            # Extra validation tokens provided on the form_class(... extra_tokens={}) setup
+            if hasattr(field, 'extra_notification_tokens'):
+                jinja2_env.globals.update(field.extra_notification_tokens)
+
             jinja2_env.from_string(joined_data).render()
         except TemplateSyntaxError as e:
             raise ValidationError(f"This is not a valid Jinja2 template: {e}") from e
@@ -275,6 +380,7 @@ class validateURL(object):
         # This should raise a ValidationError() or not
         validate_url(field.data)
 
+
 def validate_url(test_url):
     # If hosts that only contain alphanumerics are allowed ("localhost" for example)
     try:
@@ -417,17 +523,28 @@ class quickWatchForm(Form):
     edit_and_watch_submit_button = SubmitField('Edit > Watch', render_kw={"class": "pure-button pure-button-primary"})
 
 
+
 # Common to a single watch and the global settings
 class commonSettingsForm(Form):
+    from . import processors
 
-    notification_urls = StringListField('Notification URL List', validators=[validators.Optional(), ValidateAppRiseServers(), ValidateJinja2Template()])
-    notification_title = StringField('Notification Title', default='ChangeDetection.io Notification - {{ watch_url }}', validators=[validators.Optional(), ValidateJinja2Template()])
+    def __init__(self, formdata=None, obj=None, prefix="", data=None, meta=None, **kwargs):
+        super().__init__(formdata, obj, prefix, data, meta, **kwargs)
+        self.notification_body.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
+        self.notification_title.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
+        self.notification_urls.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
+
+    extract_title_as_title = BooleanField('Extract <title> from document and use as watch title', default=False)
+    fetch_backend = RadioField(u'Fetch Method', choices=content_fetchers.available_fetchers(), validators=[ValidateContentFetcherIsReady()])
     notification_body = TextAreaField('Notification Body', default='{{ watch_url }} had a change.', validators=[validators.Optional(), ValidateJinja2Template()])
     notification_format = SelectField('Notification format', choices=valid_notification_formats.keys())
-    fetch_backend = RadioField(u'Fetch Method', choices=content_fetchers.available_fetchers(), validators=[ValidateContentFetcherIsReady()])
-    extract_title_as_title = BooleanField('Extract <title> from document and use as watch title', default=False)
-    webdriver_delay = IntegerField('Wait seconds before extracting text', validators=[validators.Optional(), validators.NumberRange(min=1,
-                                                                                                                                    message="Should contain one or more seconds")])
+    notification_title = StringField('Notification Title', default='ChangeDetection.io Notification - {{ watch_url }}', validators=[validators.Optional(), ValidateJinja2Template()])
+    notification_urls = StringListField('Notification URL List', validators=[validators.Optional(), ValidateAppRiseServers(), ValidateJinja2Template()])
+    processor = RadioField( label=u"Processor - What do you want to achieve?", choices=processors.available_processors(), default="text_json_diff")
+    timezone = StringField("Timezone for watch schedule", render_kw={"list": "timezones"}, validators=[validateTimeZoneName()])
+    webdriver_delay = IntegerField('Wait seconds before extracting text', validators=[validators.Optional(), validators.NumberRange(min=1, message="Should contain one or more seconds")])
+
+
 class importForm(Form):
     from . import processors
     processor = RadioField(u'Processor', choices=processors.available_processors(), default="text_json_diff")
@@ -435,7 +552,6 @@ class importForm(Form):
     xlsx_file = FileField('Upload .xlsx file', validators=[FileAllowed(['xlsx'], 'Must be .xlsx file!')])
     file_mapping = SelectField('File mapping', [validators.DataRequired()], choices={('wachete', 'Wachete mapping'), ('custom','Custom mapping')})
 
-
 class SingleBrowserStep(Form):
 
     operation = SelectField('Operation', [validators.Optional()], choices=browser_step_ui_config.keys())
@@ -447,44 +563,46 @@ class SingleBrowserStep(Form):
 #    remove_button = SubmitField('-', render_kw={"type": "button", "class": "pure-button pure-button-primary", 'title': 'Remove'})
 #    add_button = SubmitField('+', render_kw={"type": "button", "class": "pure-button pure-button-primary", 'title': 'Add new step after'})
 
-class watchForm(commonSettingsForm):
+class processor_text_json_diff_form(commonSettingsForm):
 
     url = fields.URLField('URL', validators=[validateURL()])
     tags = StringTagUUID('Group tag', [validators.Optional()], default='')
 
     time_between_check = FormField(TimeBetweenCheckForm)
+
+    time_schedule_limit = FormField(ScheduleLimitForm)
+
     time_between_check_use_default = BooleanField('Use global settings for time between check', default=False)
 
     include_filters = StringListField('CSS/JSONPath/JQ/XPath Filters', [ValidateCSSJSONXPATHInput()], default='')
 
-    subtractive_selectors = StringListField('Remove elements', [ValidateCSSJSONXPATHInput(allow_xpath=False, allow_json=False)])
+    subtractive_selectors = StringListField('Remove elements', [ValidateCSSJSONXPATHInput(allow_json=False)])
 
     extract_text = StringListField('Extract text', [ValidateListRegex()])
 
     title = StringField('Title', default='')
 
-    ignore_text = StringListField('Ignore text', [ValidateListRegex()])
+    ignore_text = StringListField('Ignore lines containing', [ValidateListRegex()])
     headers = StringDictKeyValue('Request headers')
     body = TextAreaField('Request body', [validators.Optional()])
     method = SelectField('Request method', choices=valid_method, default=default_method)
     ignore_status_codes = BooleanField('Ignore status codes (process non-2xx status codes as normal)', default=False)
-    check_unique_lines = BooleanField('Only trigger when unique lines appear', default=False)
+    check_unique_lines = BooleanField('Only trigger when unique lines appear in all history', default=False)
+    remove_duplicate_lines = BooleanField('Remove duplicate lines of text', default=False)
     sort_text_alphabetically =  BooleanField('Sort text alphabetically', default=False)
+    trim_text_whitespace = BooleanField('Trim whitespace before and after text', default=False)
 
     filter_text_added = BooleanField('Added lines', default=True)
     filter_text_replaced = BooleanField('Replaced/changed lines', default=True)
     filter_text_removed = BooleanField('Removed lines', default=True)
 
-    # @todo this class could be moved to its own text_json_diff_watchForm and this goes to restock_diff_Watchform perhaps
-    in_stock_only = BooleanField('Only trigger when product goes BACK to in-stock', default=True)
-
-    trigger_text = StringListField('Trigger/wait for text', [validators.Optional(), ValidateListRegex()])
+    trigger_text = StringListField('Keyword triggers - Trigger/wait for text', [validators.Optional(), ValidateListRegex()])
     if os.getenv("PLAYWRIGHT_DRIVER_URL"):
         browser_steps = FieldList(FormField(SingleBrowserStep), min_entries=10)
     text_should_not_be_present = StringListField('Block change-detection while text matches', [validators.Optional(), ValidateListRegex()])
     webdriver_js_execute_code = TextAreaField('Execute JavaScript before change detection', render_kw={"rows": "5"}, validators=[validators.Optional()])
 
-    save_button = SubmitField('Save', render_kw={"class": "pure-button pure-button-primary"})
+    save_button = SubmitField('Save', render_kw={"class": "pure-button button-small pure-button-primary"})
 
     proxy = RadioField('Proxy')
     filter_failure_notification_send = BooleanField(
@@ -493,10 +611,21 @@ class watchForm(commonSettingsForm):
     notification_muted = BooleanField('Notifications Muted / Off', default=False)
     notification_screenshot = BooleanField('Attach screenshot to notification (where possible)', default=False)
 
+    conditions_match_logic = RadioField(u'Match', choices=[('ALL', 'Match all of the following'),('ANY', 'Match any of the following')], default='ALL')
+    conditions = FieldList(FormField(ConditionFormRow), min_entries=1)  # Add rule logic here
+
+
+    def extra_tab_content(self):
+        return None
+
+    def extra_form_content(self):
+        return None
+
     def validate(self, **kwargs):
         if not super().validate():
             return False
 
+        from changedetectionio.safe_jinja import render as jinja_render
         result = True
 
         # Fail form validation when a body is set for a GET
@@ -506,13 +635,64 @@ class watchForm(commonSettingsForm):
 
         # Attempt to validate jinja2 templates in the URL
         try:
-            from changedetectionio.safe_jinja import render as jinja_render
             jinja_render(template_str=self.url.data)
-        except Exception as e:
-            self.url.errors.append('Invalid template syntax')
+        except ModuleNotFoundError as e:
+            # incase jinja2_time or others is missing
+            logger.error(e)
+            self.url.errors.append(f'Invalid template syntax configuration: {e}')
             result = False
+        except Exception as e:
+            logger.error(e)
+            self.url.errors.append(f'Invalid template syntax: {e}')
+            result = False
+
+        # Attempt to validate jinja2 templates in the body
+        if self.body.data and self.body.data.strip():
+            try:
+                jinja_render(template_str=self.body.data)
+            except ModuleNotFoundError as e:
+                # incase jinja2_time or others is missing
+                logger.error(e)
+                self.body.errors.append(f'Invalid template syntax configuration: {e}')
+                result = False
+            except Exception as e:
+                logger.error(e)
+                self.body.errors.append(f'Invalid template syntax: {e}')
+                result = False
+
+        # Attempt to validate jinja2 templates in the headers
+        if len(self.headers.data) > 0:
+            try:
+                for header, value in self.headers.data.items():
+                    jinja_render(template_str=value)
+            except ModuleNotFoundError as e:
+                # incase jinja2_time or others is missing
+                logger.error(e)
+                self.headers.errors.append(f'Invalid template syntax configuration: {e}')
+                result = False
+            except Exception as e:
+                logger.error(e)
+                self.headers.errors.append(f'Invalid template syntax in "{header}" header: {e}')
+                result = False
+
         return result
 
+    def __init__(
+            self,
+            formdata=None,
+            obj=None,
+            prefix="",
+            data=None,
+            meta=None,
+            **kwargs,
+    ):
+        super().__init__(formdata, obj, prefix, data, meta, **kwargs)
+        if kwargs and kwargs.get('default_system_settings'):
+            default_tz = kwargs.get('default_system_settings').get('application', {}).get('timezone')
+            if default_tz:
+                self.time_schedule_limit.form.timezone.render_kw['placeholder'] = default_tz
+
+
 
 class SingleExtraProxy(Form):
 
@@ -534,10 +714,17 @@ class DefaultUAInputForm(Form):
 # datastore.data['settings']['requests']..
 class globalSettingsRequestForm(Form):
     time_between_check = FormField(TimeBetweenCheckForm)
+    time_schedule_limit = FormField(ScheduleLimitForm)
     proxy = RadioField('Proxy')
     jitter_seconds = IntegerField('Random jitter seconds ± check',
                                   render_kw={"style": "width: 5em;"},
                                   validators=[validators.NumberRange(min=0, message="Should contain zero or more seconds")])
+    
+    workers = IntegerField('Number of fetch workers',
+                          render_kw={"style": "width: 5em;"},
+                          validators=[validators.NumberRange(min=1, max=50,
+                                                             message="Should be between 1 and 50")])
+    
     extra_proxies = FieldList(FormField(SingleExtraProxy), min_entries=5)
     extra_browsers = FieldList(FormField(SingleExtraBrowser), min_entries=5)
 
@@ -550,6 +737,9 @@ class globalSettingsRequestForm(Form):
                     self.extra_proxies.errors.append('Both a name, and a Proxy URL is required.')
                     return False
 
+class globalSettingsApplicationUIForm(Form):
+    open_diff_in_new_tab = BooleanField("Open 'History' page in a new tab", default=True, validators=[validators.Optional()])
+    socket_io_enabled = BooleanField('Realtime UI Updates Enabled', default=True, validators=[validators.Optional()])
 
 # datastore.data['settings']['application']..
 class globalSettingsApplicationForm(commonSettingsForm):
@@ -562,13 +752,16 @@ class globalSettingsApplicationForm(commonSettingsForm):
     empty_pages_are_a_change =  BooleanField('Treat empty pages as a change?', default=False)
     fetch_backend = RadioField('Fetch Method', default="html_requests", choices=content_fetchers.available_fetchers(), validators=[ValidateContentFetcherIsReady()])
     global_ignore_text = StringListField('Ignore Text', [ValidateListRegex()])
-    global_subtractive_selectors = StringListField('Remove elements', [ValidateCSSJSONXPATHInput(allow_xpath=False, allow_json=False)])
+    global_subtractive_selectors = StringListField('Remove elements', [ValidateCSSJSONXPATHInput(allow_json=False)])
     ignore_whitespace = BooleanField('Ignore whitespace')
     password = SaltyPasswordField()
     pager_size = IntegerField('Pager size',
                               render_kw={"style": "width: 5em;"},
                               validators=[validators.NumberRange(min=0,
                                                                  message="Should be atleast zero (disabled)")])
+
+    rss_content_format = SelectField('RSS Content format', choices=RSS_FORMAT_TYPES)
+
     removepassword_button = SubmitField('Remove password', render_kw={"class": "pure-button pure-button-primary"})
     render_anchor_tag_content = BooleanField('Render anchor tag content', default=False)
     shared_diff_access = BooleanField('Allow access to view diff page when password is enabled', default=False, validators=[validators.Optional()])
@@ -578,16 +771,22 @@ class globalSettingsApplicationForm(commonSettingsForm):
                                                                   render_kw={"style": "width: 5em;"},
                                                                   validators=[validators.NumberRange(min=0,
                                                                                                      message="Should contain zero or more attempts")])
+    ui = FormField(globalSettingsApplicationUIForm)
 
 
 class globalSettingsForm(Form):
     # Define these as FormFields/"sub forms", this way it matches the JSON storage
     # datastore.data['settings']['application']..
     # datastore.data['settings']['requests']..
+    def __init__(self, formdata=None, obj=None, prefix="", data=None, meta=None, **kwargs):
+        super().__init__(formdata, obj, prefix, data, meta, **kwargs)
+        self.application.notification_body.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
+        self.application.notification_title.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
+        self.application.notification_urls.extra_notification_tokens = kwargs.get('extra_notification_tokens', {})
 
     requests = FormField(globalSettingsRequestForm)
     application = FormField(globalSettingsApplicationForm)
-    save_button = SubmitField('Save', render_kw={"class": "pure-button pure-button-primary"})
+    save_button = SubmitField('Save', render_kw={"class": "pure-button button-small pure-button-primary"})
 
 
 class extractDataForm(Form):

changedetectionio/gc_cleanup.py

@@ -0,0 +1,162 @@
+#!/usr/bin/env python3
+
+import ctypes
+import gc
+import re
+import psutil
+import sys
+import threading
+import importlib
+from loguru import logger
+
+def memory_cleanup(app=None):
+    """
+    Perform comprehensive memory cleanup operations and log memory usage
+    at each step with nicely formatted numbers.
+    
+    Args:
+        app: Optional Flask app instance for clearing Flask-specific caches
+        
+    Returns:
+        str: Status message
+    """
+    # Get current process
+    process = psutil.Process()
+    
+    # Log initial memory usage with nicely formatted numbers
+    current_memory = process.memory_info().rss / 1024 / 1024
+    logger.debug(f"Memory cleanup started - Current memory usage: {current_memory:,.2f} MB")
+
+    # 1. Standard garbage collection - force full collection on all generations
+    gc.collect(0)  # Collect youngest generation
+    gc.collect(1)  # Collect middle generation
+    gc.collect(2)  # Collect oldest generation
+
+    # Run full collection again to ensure maximum cleanup
+    gc.collect()
+    current_memory = process.memory_info().rss / 1024 / 1024
+    logger.debug(f"After full gc.collect() - Memory usage: {current_memory:,.2f} MB")
+    
+
+    # 3. Call libc's malloc_trim to release memory back to the OS
+    libc = ctypes.CDLL("libc.so.6")
+    libc.malloc_trim(0)
+    current_memory = process.memory_info().rss / 1024 / 1024
+    logger.debug(f"After malloc_trim(0) - Memory usage: {current_memory:,.2f} MB")
+    
+    # 4. Clear Python's regex cache
+    re.purge()
+    current_memory = process.memory_info().rss / 1024 / 1024
+    logger.debug(f"After re.purge() - Memory usage: {current_memory:,.2f} MB")
+
+    # 5. Reset thread-local storage
+    # Create a new thread local object to encourage cleanup of old ones
+    threading.local()
+    current_memory = process.memory_info().rss / 1024 / 1024
+    logger.debug(f"After threading.local() - Memory usage: {current_memory:,.2f} MB")
+
+    # 6. Clear sys.intern cache if Python version supports it
+    try:
+        sys.intern.clear()
+        current_memory = process.memory_info().rss / 1024 / 1024
+        logger.debug(f"After sys.intern.clear() - Memory usage: {current_memory:,.2f} MB")
+    except (AttributeError, TypeError):
+        logger.debug("sys.intern.clear() not supported in this Python version")
+    
+    # 7. Clear XML/lxml caches if available
+    try:
+        # Check if lxml.etree is in use
+        lxml_etree = sys.modules.get('lxml.etree')
+        if lxml_etree:
+            # Clear module-level caches
+            if hasattr(lxml_etree, 'clear_error_log'):
+                lxml_etree.clear_error_log()
+            
+            # Check for _ErrorLog and _RotatingErrorLog objects and clear them
+            for obj in gc.get_objects():
+                if hasattr(obj, '__class__') and hasattr(obj.__class__, '__name__'):
+                    class_name = obj.__class__.__name__
+                    if class_name in ('_ErrorLog', '_RotatingErrorLog', '_DomainErrorLog') and hasattr(obj, 'clear'):
+                        try:
+                            obj.clear()
+                        except (AttributeError, TypeError):
+                            pass
+                    
+                    # Clear Element objects which can hold references to documents
+                    elif class_name in ('_Element', 'ElementBase') and hasattr(obj, 'clear'):
+                        try:
+                            obj.clear()
+                        except (AttributeError, TypeError):
+                            pass
+            
+            current_memory = process.memory_info().rss / 1024 / 1024
+            logger.debug(f"After lxml.etree cleanup - Memory usage: {current_memory:,.2f} MB")
+
+        # Check if lxml.html is in use
+        lxml_html = sys.modules.get('lxml.html')
+        if lxml_html:
+            # Clear HTML-specific element types
+            for obj in gc.get_objects():
+                if hasattr(obj, '__class__') and hasattr(obj.__class__, '__name__'):
+                    class_name = obj.__class__.__name__
+                    if class_name in ('HtmlElement', 'FormElement', 'InputElement',
+                                    'SelectElement', 'TextareaElement', 'CheckboxGroup',
+                                    'RadioGroup', 'MultipleSelectOptions', 'FieldsDict') and hasattr(obj, 'clear'):
+                        try:
+                            obj.clear()
+                        except (AttributeError, TypeError):
+                            pass
+
+            current_memory = process.memory_info().rss / 1024 / 1024
+            logger.debug(f"After lxml.html cleanup - Memory usage: {current_memory:,.2f} MB")
+    except (ImportError, AttributeError):
+        logger.debug("lxml cleanup not applicable")
+    
+    # 8. Clear JSON parser caches if applicable
+    try:
+        # Check if json module is being used and try to clear its cache
+        json_module = sys.modules.get('json')
+        if json_module and hasattr(json_module, '_default_encoder'):
+            json_module._default_encoder.markers.clear()
+            current_memory = process.memory_info().rss / 1024 / 1024
+            logger.debug(f"After JSON parser cleanup - Memory usage: {current_memory:,.2f} MB")
+    except (AttributeError, KeyError):
+        logger.debug("JSON cleanup not applicable")
+    
+    # 9. Force Python's memory allocator to release unused memory
+    try:
+        if hasattr(sys, 'pypy_version_info'):
+            # PyPy has different memory management
+            gc.collect()
+        else:
+            # CPython - try to release unused memory
+            ctypes.pythonapi.PyGC_Collect()
+            current_memory = process.memory_info().rss / 1024 / 1024
+            logger.debug(f"After PyGC_Collect - Memory usage: {current_memory:,.2f} MB")
+    except (AttributeError, TypeError):
+        logger.debug("PyGC_Collect not supported")
+    
+    # 10. Clear Flask-specific caches if applicable
+    if app:
+        try:
+            # Clear Flask caches if they exist
+            for key in list(app.config.get('_cache', {}).keys()):
+                app.config['_cache'].pop(key, None)
+            
+            # Clear Jinja2 template cache if available
+            if hasattr(app, 'jinja_env') and hasattr(app.jinja_env, 'cache'):
+                app.jinja_env.cache.clear()
+            
+            current_memory = process.memory_info().rss / 1024 / 1024
+            logger.debug(f"After Flask cache clear - Memory usage: {current_memory:,.2f} MB")
+        except (AttributeError, KeyError):
+            logger.debug("No Flask cache to clear")
+    
+    # Final garbage collection pass
+    gc.collect()
+    libc.malloc_trim(0)
+    
+    # Log final memory usage
+    final_memory = process.memory_info().rss / 1024 / 1024
+    logger.info(f"Memory cleanup completed - Final memory usage: {final_memory:,.2f} MB")
+    return "cleaned"

changedetectionio/html_tools.py

@@ -1,18 +1,14 @@
-
-from bs4 import BeautifulSoup
-from inscriptis import get_text
-from jsonpath_ng.ext import parse
+from loguru import logger
+from lxml import etree
 from typing import List
-from inscriptis.model.config import ParserConfig
-from xml.sax.saxutils import escape as xml_escape
 import json
 import re
 
-
 # HTML added to be sure each result matching a filter (.example) gets converted to a new line by Inscriptis
 TEXT_FILTER_LIST_LINE_SUFFIX = "<br>"
-
+TRANSLATE_WHITESPACE_TABLE = str.maketrans('', '', '\r\n\t ')
 PERL_STYLE_REGEX = r'^/(.*?)/([a-z]*)?$'
+
 # 'price' , 'lowPrice', 'highPrice' are usually under here
 # All of those may or may not appear on different websites - I didnt find a way todo case-insensitive searching here
 LD_JSON_PRODUCT_OFFER_SELECTORS = ["json:$..offers", "json:$..Offers"]
@@ -39,6 +35,7 @@ def perl_style_slash_enclosed_regex_to_options(regex):
 
 # Given a CSS Rule, and a blob of HTML, return the blob of HTML that matches
 def include_filters(include_filters, html_content, append_pretty_line_formatting=False):
+    from bs4 import BeautifulSoup
     soup = BeautifulSoup(html_content, "html.parser")
     html_block = ""
     r = soup.select(include_filters, separator="")
@@ -56,16 +53,67 @@ def include_filters(include_filters, html_content, append_pretty_line_formatting
     return html_block
 
 def subtractive_css_selector(css_selector, html_content):
+    from bs4 import BeautifulSoup
     soup = BeautifulSoup(html_content, "html.parser")
-    for item in soup.select(css_selector):
+
+    # So that the elements dont shift their index, build a list of elements here which will be pointers to their place in the DOM
+    elements_to_remove = soup.select(css_selector)
+
+    # Then, remove them in a separate loop
+    for item in elements_to_remove:
         item.decompose()
+
     return str(soup)
 
+def subtractive_xpath_selector(selectors: List[str], html_content: str) -> str:
+    # Parse the HTML content using lxml
+    html_tree = etree.HTML(html_content)
+
+    # First, collect all elements to remove
+    elements_to_remove = []
+
+    # Iterate over the list of XPath selectors
+    for selector in selectors:
+        # Collect elements for each selector
+        elements_to_remove.extend(html_tree.xpath(selector))
+
+    # Then, remove them in a separate loop
+    for element in elements_to_remove:
+        if element.getparent() is not None:  # Ensure the element has a parent before removing
+            element.getparent().remove(element)
+
+    # Convert the modified HTML tree back to a string
+    modified_html = etree.tostring(html_tree, method="html").decode("utf-8")
+    return modified_html
+
 
 def element_removal(selectors: List[str], html_content):
-    """Joins individual filters into one css filter."""
-    selector = ",".join(selectors)
-    return subtractive_css_selector(selector, html_content)
+    """Removes elements that match a list of CSS or XPath selectors."""
+    modified_html = html_content
+    css_selectors = []
+    xpath_selectors = []
+
+    for selector in selectors:
+        if selector.startswith(('xpath:', 'xpath1:', '//')):
+            # Handle XPath selectors separately
+            xpath_selector = selector.removeprefix('xpath:').removeprefix('xpath1:')
+            xpath_selectors.append(xpath_selector)
+        else:
+            # Collect CSS selectors as one "hit", see comment in subtractive_css_selector
+            css_selectors.append(selector.strip().strip(","))
+
+    if xpath_selectors:
+        modified_html = subtractive_xpath_selector(xpath_selectors, modified_html)
+
+    if css_selectors:
+        # Remove duplicates, then combine all CSS selectors into one string, separated by commas
+        # This stops the elements index shifting
+        unique_selectors = list(set(css_selectors))  # Ensure uniqueness
+        combined_css_selector = " , ".join(unique_selectors)
+        modified_html = subtractive_css_selector(combined_css_selector, modified_html)
+
+
+    return modified_html
 
 def elementpath_tostring(obj):
     """
@@ -181,6 +229,7 @@ def xpath1_filter(xpath_filter, html_content, append_pretty_line_formatting=Fals
 
 # Extract/find element
 def extract_element(find='title', html_content=''):
+    from bs4 import BeautifulSoup
 
     #Re #106, be sure to handle when its not found
     element_text = None
@@ -194,6 +243,8 @@ def extract_element(find='title', html_content=''):
 
 #
 def _parse_json(json_data, json_filter):
+    from jsonpath_ng.ext import parse
+
     if json_filter.startswith("json:"):
         jsonpath_expression = parse(json_filter.replace('json:', ''))
         match = jsonpath_expression.find(json_data)
@@ -242,22 +293,26 @@ def _get_stripped_text_from_json_match(match):
 # json_filter - ie json:$..price
 # ensure_is_ldjson_info_type - str "product", optional, "@type == product" (I dont know how to do that as a json selector)
 def extract_json_as_string(content, json_filter, ensure_is_ldjson_info_type=None):
-    stripped_text_from_html = False
+    from bs4 import BeautifulSoup
 
+    stripped_text_from_html = False
+# https://github.com/dgtlmoon/changedetection.io/pull/2041#issuecomment-1848397161w
     # Try to parse/filter out the JSON, if we get some parser error, then maybe it's embedded within HTML tags
     try:
-        stripped_text_from_html = _parse_json(json.loads(content), json_filter)
-    except json.JSONDecodeError:
+        # .lstrip("\ufeff") strings ByteOrderMark from UTF8 and still lets the UTF work
+        stripped_text_from_html = _parse_json(json.loads(content.lstrip("\ufeff") ), json_filter)
+    except json.JSONDecodeError as e:
+        logger.warning(str(e))
 
         # Foreach <script json></script> blob.. just return the first that matches json_filter
         # As a last resort, try to parse the whole <body>
         soup = BeautifulSoup(content, 'html.parser')
 
         if ensure_is_ldjson_info_type:
-            bs_result = soup.findAll('script', {"type": "application/ld+json"})
+            bs_result = soup.find_all('script', {"type": "application/ld+json"})
         else:
-            bs_result = soup.findAll('script')
-        bs_result += soup.findAll('body')
+            bs_result = soup.find_all('script')
+        bs_result += soup.find_all('body')
 
         bs_jsons = []
         for result in bs_result:
@@ -282,17 +337,19 @@ def extract_json_as_string(content, json_filter, ensure_is_ldjson_info_type=None
                 if isinstance(json_data, dict):
                     # If it has LD JSON 'key' @type, and @type is 'product', and something was found for the search
                     # (Some sites have multiple of the same ld+json @type='product', but some have the review part, some have the 'price' part)
-                    # @type could also be a list (Product, SubType)
+                    # @type could also be a list although non-standard ("@type": ["Product", "SubType"],)
                     # LD_JSON auto-extract also requires some content PLUS the ldjson to be present
                     # 1833 - could be either str or dict, should not be anything else
-                    if json_data.get('@type') and stripped_text_from_html:
-                        try:
-                            if json_data.get('@type') == str or json_data.get('@type') == dict:
-                                types = [json_data.get('@type')] if isinstance(json_data.get('@type'), str) else json_data.get('@type')
-                                if ensure_is_ldjson_info_type.lower() in [x.lower().strip() for x in types]:
-                                    break
-                        except:
-                            continue
+
+                    t = json_data.get('@type')
+                    if t and stripped_text_from_html:
+
+                        if isinstance(t, str) and t.lower() == ensure_is_ldjson_info_type.lower():
+                            break
+                        # The non-standard part, some have a list
+                        elif isinstance(t, list):
+                            if ensure_is_ldjson_info_type.lower() in [x.lower().strip() for x in t]:
+                                break
 
             elif stripped_text_from_html:
                 break
@@ -307,49 +364,70 @@ def extract_json_as_string(content, json_filter, ensure_is_ldjson_info_type=None
 #          - "line numbers" return a list of line numbers that match (int list)
 #
 # wordlist - list of regex's (str) or words (str)
+# Preserves all linefeeds and other whitespacing, its not the job of this to remove that
 def strip_ignore_text(content, wordlist, mode="content"):
-    i = 0
-    output = []
     ignore_text = []
     ignore_regex = []
-    ignored_line_numbers = []
+    ignore_regex_multiline = []
+    ignored_lines = []
 
     for k in wordlist:
         # Is it a regex?
         res = re.search(PERL_STYLE_REGEX, k, re.IGNORECASE)
         if res:
-            ignore_regex.append(re.compile(perl_style_slash_enclosed_regex_to_options(k)))
+            res = re.compile(perl_style_slash_enclosed_regex_to_options(k))
+            if res.flags & re.DOTALL or res.flags & re.MULTILINE:
+                ignore_regex_multiline.append(res)
+            else:
+                ignore_regex.append(res)
         else:
             ignore_text.append(k.strip())
 
-    for line in content.splitlines():
-        i += 1
+    for r in ignore_regex_multiline:
+        for match in r.finditer(content):
+            content_lines = content[:match.end()].splitlines(keepends=True)
+            match_lines = content[match.start():match.end()].splitlines(keepends=True)
+
+            end_line = len(content_lines)
+            start_line = end_line - len(match_lines)
+
+            if end_line - start_line <= 1:
+                # Match is empty or in the middle of the line
+                ignored_lines.append(start_line)
+            else:
+                for i in range(start_line, end_line):
+                    ignored_lines.append(i)
+
+    line_index = 0
+    lines = content.splitlines(keepends=True)
+    for line in lines:
         # Always ignore blank lines in this mode. (when this function gets called)
         got_match = False
-        if len(line.strip()):
-            for l in ignore_text:
-                if l.lower() in line.lower():
+        for l in ignore_text:
+            if l.lower() in line.lower():
+                got_match = True
+
+        if not got_match:
+            for r in ignore_regex:
+                if r.search(line):
                     got_match = True
 
-            if not got_match:
-                for r in ignore_regex:
-                    if r.search(line):
-                        got_match = True
+        if got_match:
+            ignored_lines.append(line_index)
 
-            if not got_match:
-                # Not ignored
-                output.append(line.encode('utf8'))
-            else:
-                ignored_line_numbers.append(i)
+        line_index += 1
 
+    ignored_lines = set([i for i in ignored_lines if i >= 0 and i < len(lines)])
 
     # Used for finding out what to highlight
     if mode == "line numbers":
-        return ignored_line_numbers
+        return [i + 1 for i in ignored_lines]
 
-    return "\n".encode('utf8').join(output)
+    output_lines = set(range(len(lines))) - ignored_lines
+    return ''.join([lines[i] for i in output_lines])
 
 def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False) -> str:
+    from xml.sax.saxutils import escape as xml_escape
     pattern = '<!\[CDATA\[(\s*(?:.(?<!\]\]>)\s*)*)\]\]>'
     def repl(m):
         text = m.group(1)
@@ -357,60 +435,50 @@ def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False
 
     return re.sub(pattern, repl, html_content)
 
-def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False) -> str:
-    """Converts html string to a string with just the text. If ignoring
-    rendering anchor tag content is enable, anchor tag content are also
-    included in the text
 
-    :param html_content: string with html content
-    :param render_anchor_tag_content: boolean flag indicating whether to extract
-    hyperlinks (the anchor tag content) together with text. This refers to the
-    'href' inside 'a' tags.
-    Anchor tag content is rendered in the following manner:
-    '[ text ](anchor tag content)'
-    :return: extracted text from the HTML
-    """
-    #  if anchor tag content flag is set to True define a config for
-    #  extracting this content
+# NOTE!! ANYTHING LIBXML, HTML5LIB ETC WILL CAUSE SOME SMALL MEMORY LEAK IN THE LOCAL "LIB" IMPLEMENTATION OUTSIDE PYTHON
+
+
+def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False, timeout=10) -> str:
+    from inscriptis import get_text
+    from inscriptis.model.config import ParserConfig
+
     if render_anchor_tag_content:
         parser_config = ParserConfig(
             annotation_rules={"a": ["hyperlink"]},
             display_links=True
         )
-    # otherwise set config to None/default
     else:
         parser_config = None
 
-    # RSS Mode - Inscriptis will treat `title` as something else.
-    # Make it as a regular block display element (//item/title)
-    # This is a bit of a hack - the real way it to use XSLT to convert it to HTML #1874
     if is_rss:
         html_content = re.sub(r'<title([\s>])', r'<h1\1', html_content)
         html_content = re.sub(r'</title>', r'</h1>', html_content)
 
     text_content = get_text(html_content, config=parser_config)
-
     return text_content
 
-
 # Does LD+JSON exist with a @type=='product' and a .price set anywhere?
 def has_ldjson_product_info(content):
-    pricing_data = ''
-
     try:
-        if not 'application/ld+json' in content:
-            return False
-
-        for filter in LD_JSON_PRODUCT_OFFER_SELECTORS:
-            pricing_data += extract_json_as_string(content=content,
-                                                  json_filter=filter,
-                                                  ensure_is_ldjson_info_type="product")
+        # Better than .lower() which can use a lot of ram
+        if (re.search(r'application/ld\+json', content, re.IGNORECASE) and
+            re.search(r'"price"', content, re.IGNORECASE) and
+            re.search(r'"pricecurrency"', content, re.IGNORECASE)):
+            return True
 
+#       On some pages this is really terribly expensive when they dont really need it
+#       (For example you never want price monitoring, but this runs on every watch to suggest it)
+#        for filter in LD_JSON_PRODUCT_OFFER_SELECTORS:
+#            pricing_data += extract_json_as_string(content=content,
+#                                                  json_filter=filter,
+#                                                  ensure_is_ldjson_info_type="product")
     except Exception as e:
-        # Totally fine
+        # OK too
         return False
-    x=bool(pricing_data)
-    return x
+
+    return False
+
 
 
 def workarounds_for_obfuscations(content):

changedetectionio/model/App.py

@@ -1,4 +1,7 @@
 from os import getenv
+
+from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES
+
 from changedetectionio.notification import (
     default_notification_body,
     default_notification_format,
@@ -9,6 +12,8 @@ from changedetectionio.notification import (
 _FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT = 6
 DEFAULT_SETTINGS_HEADERS_USERAGENT='Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.66 Safari/537.36'
 
+
+
 class model(dict):
     base_config = {
             'note': "Hello! If you change this file manually, please be sure to restart your changedetection.io instance!",
@@ -48,11 +53,17 @@ class model(dict):
                     'password': False,
                     'render_anchor_tag_content': False,
                     'rss_access_token': None,
+                    'rss_content_format': RSS_FORMAT_TYPES[0][0],
                     'rss_hide_muted_watches': True,
                     'schema_version' : 0,
                     'shared_diff_access': False,
                     'webdriver_delay': None , # Extra delay in seconds before extracting text
-                    'tags': {} #@todo use Tag.model initialisers
+                    'tags': {}, #@todo use Tag.model initialisers
+                    'timezone': None, # Default IANA timezone name
+                    'ui': {
+                        'open_diff_in_new_tab': True,
+                        'socket_io_enabled': True
+                    },
                 }
             }
         }
@@ -68,7 +79,7 @@ def parse_headers_from_text_file(filepath):
         for l in f.readlines():
             l = l.strip()
             if not l.startswith('#') and ':' in l:
-                (k, v) = l.split(':')
+                (k, v) = l.split(':', 1)  # Split only on the first colon
                 headers[k.strip()] = v.strip()
 
     return headers

changedetectionio/model/Tag.py

@@ -1,19 +1,14 @@
-from .Watch import base_config
-import uuid
 
-class model(dict):
+from changedetectionio.model import watch_base
+
+
+class model(watch_base):
 
     def __init__(self, *arg, **kw):
+        super(model, self).__init__(*arg, **kw)
 
-        self.update(base_config)
-
-        self['uuid'] = str(uuid.uuid4())
+        self['overrides_watch'] = kw.get('default', {}).get('overrides_watch')
 
         if kw.get('default'):
             self.update(kw['default'])
             del kw['default']
-
-
-        # Goes at the end so we update the default object with the initialiser
-        super(model, self).__init__(*arg, **kw)
-

changedetectionio/model/Watch.py

@@ -1,13 +1,15 @@
+from blinker import signal
+
 from changedetectionio.strtobool import strtobool
 from changedetectionio.safe_jinja import render as jinja_render
-
+from . import watch_base
 import os
 import re
-import time
-import uuid
 from pathlib import Path
 from loguru import logger
 
+from ..html_tools import TRANSLATE_WHITESPACE_TABLE
+
 # Allowable protocols, protects against javascript: etc
 # file:// is further checked by ALLOW_FILE_URI
 SAFE_PROTOCOL_REGEX='^(http|https|ftp|file):'
@@ -15,69 +17,6 @@ SAFE_PROTOCOL_REGEX='^(http|https|ftp|file):'
 minimum_seconds_recheck_time = int(os.getenv('MINIMUM_SECONDS_RECHECK_TIME', 3))
 mtable = {'seconds': 1, 'minutes': 60, 'hours': 3600, 'days': 86400, 'weeks': 86400 * 7}
 
-from changedetectionio.notification import (
-    default_notification_format_for_watch
-)
-
-base_config = {
-    'body': None,
-    'browser_steps': [],
-    'browser_steps_last_error_step': None,
-    'check_unique_lines': False,  # On change-detected, compare against all history if its something new
-    'check_count': 0,
-    'date_created': None,
-    'consecutive_filter_failures': 0,  # Every time the CSS/xPath filter cannot be located, reset when all is fine.
-    'extract_text': [],  # Extract text by regex after filters
-    'extract_title_as_title': False,
-    'fetch_backend': 'system', # plaintext, playwright etc
-    'fetch_time': 0.0,
-    'processor': 'text_json_diff', # could be restock_diff or others from .processors
-    'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
-    'filter_text_added': True,
-    'filter_text_replaced': True,
-    'filter_text_removed': True,
-    'has_ldjson_price_data': None,
-    'track_ldjson_price_data': None,
-    'headers': {},  # Extra headers to send
-    'ignore_text': [],  # List of text to ignore when calculating the comparison checksum
-    'in_stock' : None,
-    'in_stock_only' : True, # Only trigger change on going to instock from out-of-stock
-    'include_filters': [],
-    'last_checked': 0,
-    'last_error': False,
-    'last_viewed': 0,  # history key value of the last viewed via the [diff] link
-    'method': 'GET',
-    'notification_alert_count': 0,
-    # Custom notification content
-    'notification_body': None,
-    'notification_format': default_notification_format_for_watch,
-    'notification_muted': False,
-    'notification_title': None,
-    'notification_screenshot': False,  # Include the latest screenshot if available and supported by the apprise URL
-    'notification_urls': [],  # List of URLs to add to the notification Queue (Usually AppRise)
-    'paused': False,
-    'previous_md5': False,
-    'previous_md5_before_filters': False,  # Used for skipping changedetection entirely
-    'proxy': None,  # Preferred proxy connection
-    'remote_server_reply': None, # From 'server' reply header
-    'sort_text_alphabetically': False,
-    'subtractive_selectors': [],
-    'tag': '', # Old system of text name for a tag, to be removed
-    'tags': [], # list of UUIDs to App.Tags
-    'text_should_not_be_present': [],  # Text that should not present
-    # Re #110, so then if this is set to None, we know to use the default value instead
-    # Requires setting to None on submit if it's the same as the default
-    # Should be all None by default, so we use the system default in this case.
-    'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
-    'time_between_check_use_default': True,
-    'title': None,
-    'trigger_text': [],  # List of text or regex to wait for until a change is detected
-    'url': '',
-    'uuid': str(uuid.uuid4()),
-    'webdriver_delay': None,
-    'webdriver_js_execute_code': None,  # Run before change-detection
-}
-
 
 def is_safe_url(test_url):
     # See https://github.com/dgtlmoon/changedetection.io/issues/1358
@@ -94,30 +33,28 @@ def is_safe_url(test_url):
 
     return True
 
-class model(dict):
+
+class model(watch_base):
     __newest_history_key = None
     __history_n = 0
     jitter_seconds = 0
 
     def __init__(self, *arg, **kw):
-
-        self.update(base_config)
-        self.__datastore_path = kw['datastore_path']
-
-        self['uuid'] = str(uuid.uuid4())
-
-        del kw['datastore_path']
-
+        self.__datastore_path = kw.get('datastore_path')
+        if kw.get('datastore_path'):
+            del kw['datastore_path']
+            
+        super(model, self).__init__(*arg, **kw)
         if kw.get('default'):
             self.update(kw['default'])
             del kw['default']
 
+        if self.get('default'):
+            del self['default']
+
         # Be sure the cached timestamp is ready
         bump = self.history
 
-        # Goes at the end so we update the default object with the initialiser
-        super(model, self).__init__(*arg, **kw)
-
     @property
     def viewed(self):
         # Don't return viewed when last_viewed is 0 and newest_key is 0
@@ -126,6 +63,10 @@ class model(dict):
 
         return False
 
+    @property
+    def has_unviewed(self):
+        return int(self.newest_history_key) > int(self['last_viewed']) and self.__history_n >= 2
+
     def ensure_data_dir_exists(self):
         if not os.path.isdir(self.watch_data_dir):
             logger.debug(f"> Creating data dir {self.watch_data_dir}")
@@ -149,14 +90,49 @@ class model(dict):
                     flash, Markup, url_for
                 )
                 message = Markup('<a href="{}#general">The URL {} is invalid and cannot be used, click to edit</a>'.format(
-                    url_for('edit_page', uuid=self.get('uuid')), self.get('url', '')))
+                    url_for('ui.ui_edit.edit_page', uuid=self.get('uuid')), self.get('url', '')))
                 flash(message, 'error')
                 return ''
 
         if ready_url.startswith('source:'):
             ready_url=ready_url.replace('source:', '')
+
+        # Also double check it after any Jinja2 formatting just incase
+        if not is_safe_url(ready_url):
+            return 'DISABLED'
         return ready_url
 
+    def clear_watch(self):
+        import pathlib
+
+        # JSON Data, Screenshots, Textfiles (history index and snapshots), HTML in the future etc
+        for item in pathlib.Path(str(self.watch_data_dir)).rglob("*.*"):
+            os.unlink(item)
+
+        # Force the attr to recalculate
+        bump = self.history
+
+        # Do this last because it will trigger a recheck due to last_checked being zero
+        self.update({
+            'browser_steps_last_error_step': None,
+            'check_count': 0,
+            'fetch_time': 0.0,
+            'has_ldjson_price_data': None,
+            'last_checked': 0,
+            'last_error': False,
+            'last_notification_error': False,
+            'last_viewed': 0,
+            'previous_md5': False,
+            'previous_md5_before_filters': False,
+            'remote_server_reply': None,
+            'track_ldjson_price_data': None
+        })
+        watch_check_update = signal('watch_check_update')
+        if watch_check_update:
+            watch_check_update.send(watch_uuid=self.get('uuid'))
+
+        return
+
     @property
     def is_source_type_url(self):
         return self.get('url', '').startswith('source:')
@@ -213,6 +189,10 @@ class model(dict):
         """
         tmp_history = {}
 
+        # In the case we are only using the watch for processing without history
+        if not self.watch_data_dir:
+            return []
+
         # Read the history file as a dict
         fname = os.path.join(self.watch_data_dir, "history.txt")
         if os.path.isfile(fname):
@@ -258,6 +238,13 @@ class model(dict):
 
         return has_browser_steps
 
+    @property
+    def has_restock_info(self):
+        if self.get('restock') and self['restock'].get('in_stock') != None:
+                return True
+
+        return False
+
     # Returns the newest key, but if theres only 1 record, then it's counted as not being new, so return 0.
     @property
     def newest_history_key(self):
@@ -271,37 +258,32 @@ class model(dict):
         bump = self.history
         return self.__newest_history_key
 
-    # Given an arbitrary timestamp, find the closest next key
-    # For example, last_viewed = 1000 so it should return the next 1001 timestamp
-    #
-    # used for the [diff] button so it can preset a smarter from_version
+    # Given an arbitrary timestamp, find the best history key for the [diff] button so it can preset a smarter from_version
     @property
-    def get_next_snapshot_key_to_last_viewed(self):
+    def get_from_version_based_on_last_viewed(self):
 
         """Unfortunately for now timestamp is stored as string key"""
         keys = list(self.history.keys())
         if not keys:
             return None
+        if len(keys) == 1:
+            return keys[0]
 
         last_viewed = int(self.get('last_viewed'))
-        prev_k = keys[0]
         sorted_keys = sorted(keys, key=lambda x: int(x))
         sorted_keys.reverse()
 
-        # When the 'last viewed' timestamp is greater than the newest snapshot, return second last
-        if last_viewed > int(sorted_keys[0]):
+        # When the 'last viewed' timestamp is greater than or equal the newest snapshot, return second newest
+        if last_viewed >= int(sorted_keys[0]):
             return sorted_keys[1]
+        
+        # When the 'last viewed' timestamp is between snapshots, return the older snapshot
+        for newer, older in list(zip(sorted_keys[0:], sorted_keys[1:])):
+            if last_viewed < int(newer) and last_viewed >= int(older):
+                return older
 
-        for k in sorted_keys:
-            if int(k) < last_viewed:
-                if prev_k == sorted_keys[0]:
-                    # Return the second last one so we dont recommend the same version compares itself
-                    return sorted_keys[1]
-
-                return prev_k
-            prev_k = k
-
-        return keys[0]
+        # When the 'last viewed' timestamp is less than the oldest snapshot, return oldest
+        return sorted_keys[-1]
 
     def get_history_snapshot(self, timestamp):
         import brotli
@@ -325,11 +307,11 @@ class model(dict):
         with open(filepath, 'r', encoding='utf-8', errors='ignore') as f:
             return f.read()
 
-    # Save some text file to the appropriate path and bump the history
+   # Save some text file to the appropriate path and bump the history
     # result_obj from fetch_site_status.run()
     def save_history_text(self, contents, timestamp, snapshot_id):
         import brotli
-
+        import tempfile
         logger.trace(f"{self.get('uuid')} - Updating history.txt with timestamp {timestamp}")
 
         self.ensure_data_dir_exists()
@@ -337,33 +319,43 @@ class model(dict):
         threshold = int(os.getenv('SNAPSHOT_BROTLI_COMPRESSION_THRESHOLD', 1024))
         skip_brotli = strtobool(os.getenv('DISABLE_BROTLI_TEXT_SNAPSHOT', 'False'))
 
+        # Decide on snapshot filename and destination path
         if not skip_brotli and len(contents) > threshold:
             snapshot_fname = f"{snapshot_id}.txt.br"
-            dest = os.path.join(self.watch_data_dir, snapshot_fname)
-            if not os.path.exists(dest):
-                with open(dest, 'wb') as f:
-                    f.write(brotli.compress(contents, mode=brotli.MODE_TEXT))
+            encoded_data = brotli.compress(contents.encode('utf-8'), mode=brotli.MODE_TEXT)
         else:
             snapshot_fname = f"{snapshot_id}.txt"
-            dest = os.path.join(self.watch_data_dir, snapshot_fname)
-            if not os.path.exists(dest):
-                with open(dest, 'wb') as f:
-                    f.write(contents)
+            encoded_data = contents.encode('utf-8')
 
-        # Append to index
-        # @todo check last char was \n
+        dest = os.path.join(self.watch_data_dir, snapshot_fname)
+
+        # Write snapshot file atomically if it doesn't exist
+        if not os.path.exists(dest):
+            with tempfile.NamedTemporaryFile('wb', delete=False, dir=self.watch_data_dir) as tmp:
+                tmp.write(encoded_data)
+                tmp.flush()
+                os.fsync(tmp.fileno())
+                tmp_path = tmp.name
+            os.rename(tmp_path, dest)
+
+        # Append to history.txt atomically
         index_fname = os.path.join(self.watch_data_dir, "history.txt")
-        with open(index_fname, 'a') as f:
-            f.write("{},{}\n".format(timestamp, snapshot_fname))
-            f.close()
+        index_line = f"{timestamp},{snapshot_fname}\n"
 
+        # Lets try force flush here since it's usually a very small file
+        # If this still fails in the future then try reading all to memory first, re-writing etc
+        with open(index_fname, 'a', encoding='utf-8') as f:
+            f.write(index_line)
+            f.flush()
+            os.fsync(f.fileno())
+
+        # Update internal state
         self.__newest_history_key = timestamp
         self.__history_n += 1
 
         # @todo bump static cache of the last timestamp so we dont need to examine the file to set a proper ''viewed'' status
         return snapshot_fname
 
-    @property
     @property
     def has_empty_checktime(self):
         # using all() + dictionary comprehension
@@ -380,14 +372,32 @@ class model(dict):
         return seconds
 
     # Iterate over all history texts and see if something new exists
-    def lines_contain_something_unique_compared_to_history(self, lines: list):
-        local_lines = set([l.decode('utf-8').strip().lower() for l in lines])
+    # Always applying .strip() to start/end but optionally replace any other whitespace
+    def lines_contain_something_unique_compared_to_history(self, lines: list, ignore_whitespace=False):
+        local_lines = set([])
+        if lines:
+            if ignore_whitespace:
+                if isinstance(lines[0], str): # Can be either str or bytes depending on what was on the disk
+                    local_lines = set([l.translate(TRANSLATE_WHITESPACE_TABLE).lower() for l in lines])
+                else:
+                    local_lines = set([l.decode('utf-8').translate(TRANSLATE_WHITESPACE_TABLE).lower() for l in lines])
+            else:
+                if isinstance(lines[0], str): # Can be either str or bytes depending on what was on the disk
+                    local_lines = set([l.strip().lower() for l in lines])
+                else:
+                    local_lines = set([l.decode('utf-8').strip().lower() for l in lines])
+
 
         # Compare each lines (set) against each history text file (set) looking for something new..
         existing_history = set({})
         for k, v in self.history.items():
             content = self.get_history_snapshot(k)
-            alist = set([line.strip().lower() for line in content.splitlines()])
+
+            if ignore_whitespace:
+                alist = set([line.translate(TRANSLATE_WHITESPACE_TABLE).lower() for line in content.splitlines()])
+            else:
+                alist = set([line.strip().lower() for line in content.splitlines()])
+
             existing_history = existing_history.union(alist)
 
         # Check that everything in local_lines(new stuff) already exists in existing_history - it should
@@ -431,8 +441,8 @@ class model(dict):
     @property
     def watch_data_dir(self):
         # The base dir of the watch data
-        return os.path.join(self.__datastore_path, self['uuid'])
-    
+        return os.path.join(self.__datastore_path, self['uuid']) if self.__datastore_path else None
+
     def get_error_text(self):
         """Return the text saved from a previous request that resulted in a non-200 error"""
         fname = os.path.join(self.watch_data_dir, "last-error.txt")
@@ -467,6 +477,17 @@ class model(dict):
     def toggle_mute(self):
         self['notification_muted'] ^= True
 
+    def extra_notification_token_values(self):
+        # Used for providing extra tokens
+        # return {'widget': 555}
+        return {}
+
+    def extra_notification_token_placeholder_info(self):
+        # Used for providing extra tokens
+        # return [('widget', "Get widget amounts")]
+        return []
+
+
     def extract_regex_from_all_history(self, regex):
         import csv
         import re
@@ -528,21 +549,25 @@ class model(dict):
     def save_error_text(self, contents):
         self.ensure_data_dir_exists()
         target_path = os.path.join(self.watch_data_dir, "last-error.txt")
-        with open(target_path, 'w') as f:
+        with open(target_path, 'w', encoding='utf-8') as f:
             f.write(contents)
 
     def save_xpath_data(self, data, as_error=False):
         import json
+        import zlib
 
         if as_error:
-            target_path = os.path.join(self.watch_data_dir, "elements-error.json")
+            target_path = os.path.join(str(self.watch_data_dir), "elements-error.deflate")
         else:
-            target_path = os.path.join(self.watch_data_dir, "elements.json")
+            target_path = os.path.join(str(self.watch_data_dir), "elements.deflate")
 
         self.ensure_data_dir_exists()
 
-        with open(target_path, 'w') as f:
-            f.write(json.dumps(data))
+        with open(target_path, 'wb') as f:
+            if not isinstance(data, str):
+                f.write(zlib.compress(json.dumps(data).encode()))
+            else:
+                f.write(zlib.compress(data.encode()))
             f.close()
 
     # Save as PNG, PNG is larger but better for doing visual diff in the future
@@ -564,7 +589,7 @@ class model(dict):
         import brotli
         filepath = os.path.join(self.watch_data_dir, 'last-fetched.br')
 
-        if not os.path.isfile(filepath):
+        if not os.path.isfile(filepath) or os.path.getsize(filepath) == 0:
             # If a previous attempt doesnt yet exist, just snarf the previous snapshot instead
             dates = list(self.history.keys())
             if len(dates):
@@ -634,3 +659,44 @@ class model(dict):
             if step_n:
                 available.append(step_n.group(1))
         return available
+
+    def compile_error_texts(self, has_proxies=None):
+        """Compile error texts for this watch.
+        Accepts has_proxies parameter to ensure it works even outside app context"""
+        from flask import url_for
+        from markupsafe import Markup
+
+        output = []  # Initialize as list since we're using append
+        last_error = self.get('last_error','')
+
+        try:
+            url_for('settings.settings_page')
+        except Exception as e:
+            has_app_context = False
+        else:
+            has_app_context = True
+
+        # has app+request context, we can use url_for()
+        if has_app_context:
+            if last_error:
+                if '403' in last_error:
+                    if has_proxies:
+                        output.append(str(Markup(f"{last_error} - <a href=\"{url_for('settings.settings_page', uuid=self.get('uuid'))}\">Try other proxies/location</a>&nbsp;'")))
+                    else:
+                        output.append(str(Markup(f"{last_error} - <a href=\"{url_for('settings.settings_page', uuid=self.get('uuid'))}\">Try adding external proxies/locations</a>&nbsp;'")))
+                else:
+                    output.append(str(Markup(last_error)))
+
+            if self.get('last_notification_error'):
+                output.append(str(Markup(f"<div class=\"notification-error\"><a href=\"{url_for('settings.notification_logs')}\">{ self.get('last_notification_error') }</a></div>")))
+
+        else:
+            # Lo_Fi version
+            if last_error:
+                output.append(str(Markup(last_error)))
+            if self.get('last_notification_error'):
+                output.append(str(Markup(self.get('last_notification_error'))))
+
+        res = "\n".join(output)
+        return res
+

changedetectionio/model/__init__.py

@@ -0,0 +1,136 @@
+import os
+import uuid
+
+from changedetectionio import strtobool
+default_notification_format_for_watch = 'System default'
+
+class watch_base(dict):
+
+    def __init__(self, *arg, **kw):
+        self.update({
+            # Custom notification content
+            # Re #110, so then if this is set to None, we know to use the default value instead
+            # Requires setting to None on submit if it's the same as the default
+            # Should be all None by default, so we use the system default in this case.
+            'body': None,
+            'browser_steps': [],
+            'browser_steps_last_error_step': None,
+            'check_count': 0,
+            'check_unique_lines': False,  # On change-detected, compare against all history if its something new
+            'consecutive_filter_failures': 0,  # Every time the CSS/xPath filter cannot be located, reset when all is fine.
+            'content-type': None,
+            'date_created': None,
+            'extract_text': [],  # Extract text by regex after filters
+            'extract_title_as_title': False,
+            'fetch_backend': 'system',  # plaintext, playwright etc
+            'fetch_time': 0.0,
+            'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
+            'filter_text_added': True,
+            'filter_text_removed': True,
+            'filter_text_replaced': True,
+            'follow_price_changes': True,
+            'has_ldjson_price_data': None,
+            'headers': {},  # Extra headers to send
+            'ignore_text': [],  # List of text to ignore when calculating the comparison checksum
+            'in_stock_only': True,  # Only trigger change on going to instock from out-of-stock
+            'include_filters': [],
+            'last_checked': 0,
+            'last_error': False,
+            'last_notification_error': None,
+            'last_viewed': 0,  # history key value of the last viewed via the [diff] link
+            'method': 'GET',
+            'notification_alert_count': 0,
+            'notification_body': None,
+            'notification_format': default_notification_format_for_watch,
+            'notification_muted': False,
+            'notification_screenshot': False,  # Include the latest screenshot if available and supported by the apprise URL
+            'notification_title': None,
+            'notification_urls': [],  # List of URLs to add to the notification Queue (Usually AppRise)
+            'paused': False,
+            'previous_md5': False,
+            'previous_md5_before_filters': False,  # Used for skipping changedetection entirely
+            'processor': 'text_json_diff',  # could be restock_diff or others from .processors
+            'price_change_threshold_percent': None,
+            'proxy': None,  # Preferred proxy connection
+            'remote_server_reply': None,  # From 'server' reply header
+            'sort_text_alphabetically': False,
+            'subtractive_selectors': [],
+            'tag': '',  # Old system of text name for a tag, to be removed
+            'tags': [],  # list of UUIDs to App.Tags
+            'text_should_not_be_present': [],  # Text that should not present
+            'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
+            'time_between_check_use_default': True,
+            "time_schedule_limit": {
+                "enabled": False,
+                "monday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "tuesday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "wednesday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "thursday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "friday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "saturday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+                "sunday": {
+                    "enabled": True,
+                    "start_time": "00:00",
+                    "duration": {
+                        "hours": "24",
+                        "minutes": "00"
+                    }
+                },
+            },
+            'title': None,
+            'track_ldjson_price_data': None,
+            'trim_text_whitespace': False,
+            'remove_duplicate_lines': False,
+            'trigger_text': [],  # List of text or regex to wait for until a change is detected
+            'url': '',
+            'uuid': str(uuid.uuid4()),
+            'webdriver_delay': None,
+            'webdriver_js_execute_code': None,  # Run before change-detection
+        })
+
+        super(watch_base, self).__init__(*arg, **kw)
+
+        if self.get('default'):
+            del self['default']

changedetectionio/notification/__init__.py

@@ -0,0 +1,35 @@
+from changedetectionio.model import default_notification_format_for_watch
+
+ult_notification_format_for_watch = 'System default'
+default_notification_format = 'HTML Color'
+default_notification_body = '{{watch_url}} had a change.\n---\n{{diff}}\n---\n'
+default_notification_title = 'ChangeDetection.io Notification - {{watch_url}}'
+
+# The values (markdown etc) are from apprise NotifyFormat,
+# But to avoid importing the whole heavy module just use the same strings here.
+valid_notification_formats = {
+    'Text': 'text',
+    'Markdown': 'markdown',
+    'HTML': 'html',
+    'HTML Color': 'htmlcolor',
+    # Used only for editing a watch (not for global)
+    default_notification_format_for_watch: default_notification_format_for_watch
+}
+
+
+valid_tokens = {
+    'base_url': '',
+    'current_snapshot': '',
+    'diff': '',
+    'diff_added': '',
+    'diff_full': '',
+    'diff_patch': '',
+    'diff_removed': '',
+    'diff_url': '',
+    'preview_url': '',
+    'triggered_text': '',
+    'watch_tag': '',
+    'watch_title': '',
+    'watch_url': '',
+    'watch_uuid': '',
+}

changedetectionio/notification/apprise_plugin/assets.py

@@ -0,0 +1,16 @@
+from apprise import AppriseAsset
+
+# Refer to:
+# https://github.com/caronc/apprise/wiki/Development_API#the-apprise-asset-object
+
+APPRISE_APP_ID = "changedetection.io"
+APPRISE_APP_DESC = "ChangeDetection.io best and simplest website monitoring and change detection"
+APPRISE_APP_URL = "https://changedetection.io"
+APPRISE_AVATAR_URL = "https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/changedetectionio/static/images/avatar-256x256.png"
+
+apprise_asset = AppriseAsset(
+    app_id=APPRISE_APP_ID,
+    app_desc=APPRISE_APP_DESC,
+    app_url=APPRISE_APP_URL,
+    image_url_logo=APPRISE_AVATAR_URL,
+)

changedetectionio/notification/apprise_plugin/custom_handlers.py

@@ -0,0 +1,112 @@
+import json
+import re
+from urllib.parse import unquote_plus
+
+import requests
+from apprise.decorators import notify
+from apprise.utils.parse import parse_url as apprise_parse_url
+from loguru import logger
+from requests.structures import CaseInsensitiveDict
+
+SUPPORTED_HTTP_METHODS = {"get", "post", "put", "delete", "patch", "head"}
+
+
+def notify_supported_methods(func):
+    for method in SUPPORTED_HTTP_METHODS:
+        func = notify(on=method)(func)
+        # Add support for https, for each supported http method
+        func = notify(on=f"{method}s")(func)
+    return func
+
+
+def _get_auth(parsed_url: dict) -> str | tuple[str, str]:
+    user: str | None = parsed_url.get("user")
+    password: str | None = parsed_url.get("password")
+
+    if user is not None and password is not None:
+        return (unquote_plus(user), unquote_plus(password))
+
+    if user is not None:
+        return unquote_plus(user)
+
+    return ""
+
+
+def _get_headers(parsed_url: dict, body: str) -> CaseInsensitiveDict:
+    headers = CaseInsensitiveDict(
+        {unquote_plus(k).title(): unquote_plus(v) for k, v in parsed_url["qsd+"].items()}
+    )
+
+    # If Content-Type is not specified, guess if the body is a valid JSON
+    if headers.get("Content-Type") is None:
+        try:
+            json.loads(body)
+            headers["Content-Type"] = "application/json; charset=utf-8"
+        except Exception:
+            pass
+
+    return headers
+
+
+def _get_params(parsed_url: dict) -> CaseInsensitiveDict:
+    # https://github.com/caronc/apprise/wiki/Notify_Custom_JSON#get-parameter-manipulation
+    # In Apprise, it relies on prefixing each request arg with "-", because it uses say &method=update as a flag for apprise
+    # but here we are making straight requests, so we need todo convert this against apprise's logic
+    params = CaseInsensitiveDict(
+        {
+            unquote_plus(k): unquote_plus(v)
+            for k, v in parsed_url["qsd"].items()
+            if k.strip("-") not in parsed_url["qsd-"]
+            and k.strip("+") not in parsed_url["qsd+"]
+        }
+    )
+
+    return params
+
+
+@notify_supported_methods
+def apprise_http_custom_handler(
+    body: str,
+    title: str,
+    notify_type: str,
+    meta: dict,
+    *args,
+    **kwargs,
+) -> bool:
+    url: str = meta.get("url")
+    schema: str = meta.get("schema")
+    method: str = re.sub(r"s$", "", schema).upper()
+
+    # Convert /foobar?+some-header=hello to proper header dictionary
+    parsed_url: dict[str, str | dict | None] | None = apprise_parse_url(url)
+    if parsed_url is None:
+        return False
+
+    auth = _get_auth(parsed_url=parsed_url)
+    headers = _get_headers(parsed_url=parsed_url, body=body)
+    params = _get_params(parsed_url=parsed_url)
+
+    url = re.sub(rf"^{schema}", "https" if schema.endswith("s") else "http", parsed_url.get("url"))
+
+    try:
+        response = requests.request(
+            method=method,
+            url=url,
+            auth=auth,
+            headers=headers,
+            params=params,
+            data=body.encode("utf-8") if isinstance(body, str) else body,
+        )
+
+        response.raise_for_status()
+
+        logger.info(f"Successfully sent custom notification to {url}")
+        return True
+
+    except requests.RequestException as e:
+        logger.error(f"Remote host error while sending custom notification to {url}: {e}")
+        return False
+
+    except Exception as e:
+        logger.error(f"Unexpected error occurred while sending custom notification to {url}: {e}")
+        return False

changedetectionio/notification/handler.py

@@ -1,121 +1,15 @@
-import apprise
+
 import time
-from apprise import NotifyFormat
-import json
+import apprise
 from loguru import logger
-
-valid_tokens = {
-    'base_url': '',
-    'current_snapshot': '',
-    'diff': '',
-    'diff_added': '',
-    'diff_full': '',
-    'diff_patch': '',
-    'diff_removed': '',
-    'diff_url': '',
-    'preview_url': '',
-    'triggered_text': '',
-    'watch_tag': '',
-    'watch_title': '',
-    'watch_url': '',
-    'watch_uuid': '',
-}
-
-default_notification_format_for_watch = 'System default'
-default_notification_format = 'Text'
-default_notification_body = '{{watch_url}} had a change.\n---\n{{diff}}\n---\n'
-default_notification_title = 'ChangeDetection.io Notification - {{watch_url}}'
-
-valid_notification_formats = {
-    'Text': NotifyFormat.TEXT,
-    'Markdown': NotifyFormat.MARKDOWN,
-    'HTML': NotifyFormat.HTML,
-    # Used only for editing a watch (not for global)
-    default_notification_format_for_watch: default_notification_format_for_watch
-}
-
-# include the decorator
-from apprise.decorators import notify
-
-@notify(on="delete")
-@notify(on="deletes")
-@notify(on="get")
-@notify(on="gets")
-@notify(on="post")
-@notify(on="posts")
-@notify(on="put")
-@notify(on="puts")
-def apprise_custom_api_call_wrapper(body, title, notify_type, *args, **kwargs):
-    import requests
-    from apprise.utils import parse_url as apprise_parse_url
-    from apprise import URLBase
-
-    url = kwargs['meta'].get('url')
-
-    if url.startswith('post'):
-        r = requests.post
-    elif url.startswith('get'):
-        r = requests.get
-    elif url.startswith('put'):
-        r = requests.put
-    elif url.startswith('delete'):
-        r = requests.delete
-
-    url = url.replace('post://', 'http://')
-    url = url.replace('posts://', 'https://')
-    url = url.replace('put://', 'http://')
-    url = url.replace('puts://', 'https://')
-    url = url.replace('get://', 'http://')
-    url = url.replace('gets://', 'https://')
-    url = url.replace('put://', 'http://')
-    url = url.replace('puts://', 'https://')
-    url = url.replace('delete://', 'http://')
-    url = url.replace('deletes://', 'https://')
-
-    headers = {}
-    params = {}
-    auth = None
-
-    # Convert /foobar?+some-header=hello to proper header dictionary
-    results = apprise_parse_url(url)
-    if results:
-        # Add our headers that the user can potentially over-ride if they wish
-        # to to our returned result set and tidy entries by unquoting them
-        headers = {URLBase.unquote(x): URLBase.unquote(y)
-                   for x, y in results['qsd+'].items()}
-
-        # https://github.com/caronc/apprise/wiki/Notify_Custom_JSON#get-parameter-manipulation
-        # In Apprise, it relies on prefixing each request arg with "-", because it uses say &method=update as a flag for apprise
-        # but here we are making straight requests, so we need todo convert this against apprise's logic
-        for k, v in results['qsd'].items():
-            if not k.strip('+-') in results['qsd+'].keys():
-                params[URLBase.unquote(k)] = URLBase.unquote(v)
-
-        # Determine Authentication
-        auth = ''
-        if results.get('user') and results.get('password'):
-            auth = (URLBase.unquote(results.get('user')), URLBase.unquote(results.get('user')))
-        elif results.get('user'):
-            auth = (URLBase.unquote(results.get('user')))
-
-    # Try to auto-guess if it's JSON
-    try:
-        json.loads(body)
-        headers['Content-Type'] = 'application/json; charset=utf-8'
-    except ValueError as e:
-        pass
-
-    r(results.get('url'),
-      auth=auth,
-      data=body,
-      headers=headers,
-      params=params
-      )
-
+from .apprise_plugin.assets import apprise_asset, APPRISE_AVATAR_URL
 
 def process_notification(n_object, datastore):
+    from changedetectionio.safe_jinja import render as jinja_render
+    from . import default_notification_format_for_watch, default_notification_format, valid_notification_formats
+    # be sure its registered
+    from .apprise_plugin.custom_handlers import apprise_http_custom_handler
 
-    from .safe_jinja import render as jinja_render
     now = time.time()
     if n_object.get('notification_timestamp'):
         logger.trace(f"Time since queued {now-n_object['notification_timestamp']:.3f}s")
@@ -132,15 +26,18 @@ def process_notification(n_object, datastore):
         # Initially text or whatever
         n_format = datastore.data['settings']['application'].get('notification_format', valid_notification_formats[default_notification_format])
 
-    logger.trace(f"Complete notification body including Jinja and placeholders calculated in  {time.time() - now:.3f}s")
+    logger.trace(f"Complete notification body including Jinja and placeholders calculated in  {time.time() - now:.2f}s")
 
     # https://github.com/caronc/apprise/wiki/Development_LogCapture
     # Anything higher than or equal to WARNING (which covers things like Connection errors)
     # raise it as an exception
 
     sent_objs = []
-    from .apprise_asset import asset
-    apobj = apprise.Apprise(debug=True, asset=asset)
+
+    if 'as_async' in n_object:
+        apprise_asset.async_mode = n_object.get('as_async')
+
+    apobj = apprise.Apprise(debug=True, asset=apprise_asset)
 
     if not n_object.get('notification_urls'):
         return None
@@ -150,14 +47,21 @@ def process_notification(n_object, datastore):
 
             # Get the notification body from datastore
             n_body = jinja_render(template_str=n_object.get('notification_body', ''), **notification_parameters)
+            if n_object.get('notification_format', '').startswith('HTML'):
+                n_body = n_body.replace("\n", '<br>')
+
             n_title = jinja_render(template_str=n_object.get('notification_title', ''), **notification_parameters)
 
             url = url.strip()
+            if url.startswith('#'):
+                logger.trace(f"Skipping commented out notification URL - {url}")
+                continue
+
             if not url:
                 logger.warning(f"Process Notification: skipping empty notification URL.")
                 continue
 
-            logger.info(">> Process Notification: AppRise notifying {}".format(url))
+            logger.info(f">> Process Notification: AppRise notifying {url}")
             url = jinja_render(template_str=url, **notification_parameters)
 
             # Re 323 - Limit discord length to their 2000 char limit total or it wont send.
@@ -174,7 +78,7 @@ def process_notification(n_object, datastore):
                     and not url.startswith('get') \
                     and not url.startswith('delete') \
                     and not url.startswith('put'):
-                url += k + 'avatar_url=https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/changedetectionio/static/images/avatar-256x256.png'
+                url += k + f"avatar_url={APPRISE_AVATAR_URL}"
 
             if url.startswith('tgram://'):
                 # Telegram only supports a limit subset of HTML, remove the '<br>' we place in.
@@ -223,13 +127,12 @@ def process_notification(n_object, datastore):
             attach=n_object.get('screenshot', None)
         )
 
-        # Give apprise time to register an error
-        time.sleep(3)
 
         # Returns empty string if nothing found, multi-line string otherwise
         log_value = logs.getvalue()
 
         if log_value and 'WARNING' in log_value or 'ERROR' in log_value:
+            logger.critical(log_value)
             raise Exception(log_value)
 
     # Return what was sent for better logging - after the for loop
@@ -240,6 +143,7 @@ def process_notification(n_object, datastore):
 # ( Where we prepare the tokens in the notification to be replaced with actual values )
 def create_notification_parameters(n_object, datastore):
     from copy import deepcopy
+    from . import valid_tokens
 
     # in the case we send a test notification from the main settings, there is no UUID.
     uuid = n_object['uuid'] if 'uuid' in n_object else ''
@@ -272,19 +176,18 @@ def create_notification_parameters(n_object, datastore):
     tokens.update(
         {
             'base_url': base_url,
-            'current_snapshot': n_object.get('current_snapshot', ''),
-            'diff': n_object.get('diff', ''),  # Null default in the case we use a test
-            'diff_added': n_object.get('diff_added', ''),  # Null default in the case we use a test
-            'diff_full': n_object.get('diff_full', ''),  # Null default in the case we use a test
-            'diff_patch': n_object.get('diff_patch', ''),  # Null default in the case we use a test
-            'diff_removed': n_object.get('diff_removed', ''),  # Null default in the case we use a test
             'diff_url': diff_url,
             'preview_url': preview_url,
-            'triggered_text': n_object.get('triggered_text', ''),
             'watch_tag': watch_tag if watch_tag is not None else '',
             'watch_title': watch_title if watch_title is not None else '',
             'watch_url': watch_url,
             'watch_uuid': uuid,
         })
 
+    # n_object will contain diff, diff_added etc etc
+    tokens.update(n_object)
+
+    if uuid:
+        tokens.update(datastore.data['watching'].get(uuid).extra_notification_token_values())
+
     return tokens

changedetectionio/notification_service.py

@@ -0,0 +1,246 @@
+#!/usr/bin/env python3
+
+"""
+Notification Service Module
+Extracted from update_worker.py to provide standalone notification functionality
+for both sync and async workers
+"""
+
+import time
+from loguru import logger
+
+
+class NotificationService:
+    """
+    Standalone notification service that handles all notification functionality
+    previously embedded in the update_worker class
+    """
+    
+    def __init__(self, datastore, notification_q):
+        self.datastore = datastore
+        self.notification_q = notification_q
+    
+    def queue_notification_for_watch(self, n_object, watch):
+        """
+        Queue a notification for a watch with full diff rendering and template variables
+        """
+        from changedetectionio import diff
+        from changedetectionio.notification import default_notification_format_for_watch
+
+        dates = []
+        trigger_text = ''
+
+        now = time.time()
+
+        if watch:
+            watch_history = watch.history
+            dates = list(watch_history.keys())
+            trigger_text = watch.get('trigger_text', [])
+
+        # Add text that was triggered
+        if len(dates):
+            snapshot_contents = watch.get_history_snapshot(dates[-1])
+        else:
+            snapshot_contents = "No snapshot/history available, the watch should fetch atleast once."
+
+        # If we ended up here with "System default"
+        if n_object.get('notification_format') == default_notification_format_for_watch:
+            n_object['notification_format'] = self.datastore.data['settings']['application'].get('notification_format')
+
+        html_colour_enable = False
+        # HTML needs linebreak, but MarkDown and Text can use a linefeed
+        if n_object.get('notification_format') == 'HTML':
+            line_feed_sep = "<br>"
+            # Snapshot will be plaintext on the disk, convert to some kind of HTML
+            snapshot_contents = snapshot_contents.replace('\n', line_feed_sep)
+        elif n_object.get('notification_format') == 'HTML Color':
+            line_feed_sep = "<br>"
+            # Snapshot will be plaintext on the disk, convert to some kind of HTML
+            snapshot_contents = snapshot_contents.replace('\n', line_feed_sep)
+            html_colour_enable = True
+        else:
+            line_feed_sep = "\n"
+
+        triggered_text = ''
+        if len(trigger_text):
+            from . import html_tools
+            triggered_text = html_tools.get_triggered_text(content=snapshot_contents, trigger_text=trigger_text)
+            if triggered_text:
+                triggered_text = line_feed_sep.join(triggered_text)
+
+        # Could be called as a 'test notification' with only 1 snapshot available
+        prev_snapshot = "Example text: example test\nExample text: change detection is cool\nExample text: some more examples\n"
+        current_snapshot = "Example text: example test\nExample text: change detection is fantastic\nExample text: even more examples\nExample text: a lot more examples"
+
+        if len(dates) > 1:
+            prev_snapshot = watch.get_history_snapshot(dates[-2])
+            current_snapshot = watch.get_history_snapshot(dates[-1])
+
+        n_object.update({
+            'current_snapshot': snapshot_contents,
+            'diff': diff.render_diff(prev_snapshot, current_snapshot, line_feed_sep=line_feed_sep, html_colour=html_colour_enable),
+            'diff_added': diff.render_diff(prev_snapshot, current_snapshot, include_removed=False, line_feed_sep=line_feed_sep),
+            'diff_full': diff.render_diff(prev_snapshot, current_snapshot, include_equal=True, line_feed_sep=line_feed_sep, html_colour=html_colour_enable),
+            'diff_patch': diff.render_diff(prev_snapshot, current_snapshot, line_feed_sep=line_feed_sep, patch_format=True),
+            'diff_removed': diff.render_diff(prev_snapshot, current_snapshot, include_added=False, line_feed_sep=line_feed_sep),
+            'notification_timestamp': now,
+            'screenshot': watch.get_screenshot() if watch and watch.get('notification_screenshot') else None,
+            'triggered_text': triggered_text,
+            'uuid': watch.get('uuid') if watch else None,
+            'watch_url': watch.get('url') if watch else None,
+        })
+
+        if watch:
+            n_object.update(watch.extra_notification_token_values())
+
+        logger.trace(f"Main rendered notification placeholders (diff_added etc) calculated in {time.time()-now:.3f}s")
+        logger.debug("Queued notification for sending")
+        self.notification_q.put(n_object)
+
+    def _check_cascading_vars(self, var_name, watch):
+        """
+        Check notification variables in cascading priority:
+        Individual watch settings > Tag settings > Global settings
+        """
+        from changedetectionio.notification import (
+            default_notification_format_for_watch,
+            default_notification_body,
+            default_notification_title
+        )
+
+        # Would be better if this was some kind of Object where Watch can reference the parent datastore etc
+        v = watch.get(var_name)
+        if v and not watch.get('notification_muted'):
+            if var_name == 'notification_format' and v == default_notification_format_for_watch:
+                return self.datastore.data['settings']['application'].get('notification_format')
+
+            return v
+
+        tags = self.datastore.get_all_tags_for_watch(uuid=watch.get('uuid'))
+        if tags:
+            for tag_uuid, tag in tags.items():
+                v = tag.get(var_name)
+                if v and not tag.get('notification_muted'):
+                    return v
+
+        if self.datastore.data['settings']['application'].get(var_name):
+            return self.datastore.data['settings']['application'].get(var_name)
+
+        # Otherwise could be defaults
+        if var_name == 'notification_format':
+            return default_notification_format_for_watch
+        if var_name == 'notification_body':
+            return default_notification_body
+        if var_name == 'notification_title':
+            return default_notification_title
+
+        return None
+
+    def send_content_changed_notification(self, watch_uuid):
+        """
+        Send notification when content changes are detected
+        """
+        n_object = {}
+        watch = self.datastore.data['watching'].get(watch_uuid)
+        if not watch:
+            return
+
+        watch_history = watch.history
+        dates = list(watch_history.keys())
+        # Theoretically it's possible that this could be just 1 long,
+        # - In the case that the timestamp key was not unique
+        if len(dates) == 1:
+            raise ValueError(
+                "History index had 2 or more, but only 1 date loaded, timestamps were not unique? maybe two of the same timestamps got written, needs more delay?"
+            )
+
+        # Should be a better parent getter in the model object
+
+        # Prefer - Individual watch settings > Tag settings >  Global settings (in that order)
+        n_object['notification_urls'] = self._check_cascading_vars('notification_urls', watch)
+        n_object['notification_title'] = self._check_cascading_vars('notification_title', watch)
+        n_object['notification_body'] = self._check_cascading_vars('notification_body', watch)
+        n_object['notification_format'] = self._check_cascading_vars('notification_format', watch)
+
+        # (Individual watch) Only prepare to notify if the rules above matched
+        queued = False
+        if n_object and n_object.get('notification_urls'):
+            queued = True
+
+            count = watch.get('notification_alert_count', 0) + 1
+            self.datastore.update_watch(uuid=watch_uuid, update_obj={'notification_alert_count': count})
+
+            self.queue_notification_for_watch(n_object=n_object, watch=watch)
+
+        return queued
+
+    def send_filter_failure_notification(self, watch_uuid):
+        """
+        Send notification when CSS/XPath filters fail consecutively
+        """
+        threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts')
+        watch = self.datastore.data['watching'].get(watch_uuid)
+        if not watch:
+            return
+
+        n_object = {'notification_title': 'Changedetection.io - Alert - CSS/xPath filter was not present in the page',
+                    'notification_body': "Your configured CSS/xPath filters of '{}' for {{{{watch_url}}}} did not appear on the page after {} attempts, did the page change layout?\n\nLink: {{{{base_url}}}}/edit/{{{{watch_uuid}}}}\n\nThanks - Your omniscient changedetection.io installation :)\n".format(
+                        ", ".join(watch['include_filters']),
+                        threshold),
+                    'notification_format': 'text'}
+
+        if len(watch['notification_urls']):
+            n_object['notification_urls'] = watch['notification_urls']
+
+        elif len(self.datastore.data['settings']['application']['notification_urls']):
+            n_object['notification_urls'] = self.datastore.data['settings']['application']['notification_urls']
+
+        # Only prepare to notify if the rules above matched
+        if 'notification_urls' in n_object:
+            n_object.update({
+                'watch_url': watch['url'],
+                'uuid': watch_uuid,
+                'screenshot': None
+            })
+            self.notification_q.put(n_object)
+            logger.debug(f"Sent filter not found notification for {watch_uuid}")
+        else:
+            logger.debug(f"NOT sending filter not found notification for {watch_uuid} - no notification URLs")
+
+    def send_step_failure_notification(self, watch_uuid, step_n):
+        """
+        Send notification when browser steps fail consecutively
+        """
+        watch = self.datastore.data['watching'].get(watch_uuid, False)
+        if not watch:
+            return
+        threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts')
+        n_object = {'notification_title': "Changedetection.io - Alert - Browser step at position {} could not be run".format(step_n+1),
+                    'notification_body': "Your configured browser step at position {} for {{{{watch_url}}}} "
+                                         "did not appear on the page after {} attempts, did the page change layout? "
+                                         "Does it need a delay added?\n\nLink: {{{{base_url}}}}/edit/{{{{watch_uuid}}}}\n\n"
+                                         "Thanks - Your omniscient changedetection.io installation :)\n".format(step_n+1, threshold),
+                    'notification_format': 'text'}
+
+        if len(watch['notification_urls']):
+            n_object['notification_urls'] = watch['notification_urls']
+
+        elif len(self.datastore.data['settings']['application']['notification_urls']):
+            n_object['notification_urls'] = self.datastore.data['settings']['application']['notification_urls']
+
+        # Only prepare to notify if the rules above matched
+        if 'notification_urls' in n_object:
+            n_object.update({
+                'watch_url': watch['url'],
+                'uuid': watch_uuid
+            })
+            self.notification_q.put(n_object)
+            logger.error(f"Sent step not found notification for {watch_uuid}")
+
+
+# Convenience functions for creating notification service instances
+def create_notification_service(datastore, notification_q):
+    """
+    Factory function to create a NotificationService instance
+    """
+    return NotificationService(datastore, notification_q)

changedetectionio/pluggy_interface.py

@@ -0,0 +1,82 @@
+import pluggy
+import os
+import importlib
+import sys
+
+# Global plugin namespace for changedetection.io
+PLUGIN_NAMESPACE = "changedetectionio"
+
+hookspec = pluggy.HookspecMarker(PLUGIN_NAMESPACE)
+hookimpl = pluggy.HookimplMarker(PLUGIN_NAMESPACE)
+
+
+class ChangeDetectionSpec:
+    """Hook specifications for extending changedetection.io functionality."""
+
+    @hookspec
+    def ui_edit_stats_extras(watch):
+        """Return HTML content to add to the stats tab in the edit view.
+        
+        Args:
+            watch: The watch object being edited
+            
+        Returns:
+            str: HTML content to be inserted in the stats tab
+        """
+        pass
+
+
+# Set up Plugin Manager
+plugin_manager = pluggy.PluginManager(PLUGIN_NAMESPACE)
+
+# Register hookspecs
+plugin_manager.add_hookspecs(ChangeDetectionSpec)
+
+# Load plugins from subdirectories
+def load_plugins_from_directories():
+    # Dictionary of directories to scan for plugins
+    plugin_dirs = {
+        'conditions': os.path.join(os.path.dirname(__file__), 'conditions', 'plugins'),
+        # Add more plugin directories here as needed
+    }
+    
+    # Note: Removed the direct import of example_word_count_plugin as it's now in the conditions/plugins directory
+    
+    for dir_name, dir_path in plugin_dirs.items():
+        if not os.path.exists(dir_path):
+            continue
+            
+        # Get all Python files (excluding __init__.py)
+        for filename in os.listdir(dir_path):
+            if filename.endswith(".py") and filename != "__init__.py":
+                module_name = filename[:-3]  # Remove .py extension
+                module_path = f"changedetectionio.{dir_name}.plugins.{module_name}"
+                
+                try:
+                    module = importlib.import_module(module_path)
+                    # Register the plugin with pluggy
+                    plugin_manager.register(module, module_name)
+                except (ImportError, AttributeError) as e:
+                    print(f"Error loading plugin {module_name}: {e}")
+
+# Load plugins
+load_plugins_from_directories()
+
+# Discover installed plugins from external packages (if any)
+plugin_manager.load_setuptools_entrypoints(PLUGIN_NAMESPACE)
+
+# Helper function to collect UI stats extras from all plugins
+def collect_ui_edit_stats_extras(watch):
+    """Collect and combine HTML content from all plugins that implement ui_edit_stats_extras"""
+    extras_content = []
+    
+    # Get all plugins that implement the ui_edit_stats_extras hook
+    results = plugin_manager.hook.ui_edit_stats_extras(watch=watch)
+    
+    # If we have results, add them to our content
+    if results:
+        for result in results:
+            if result:  # Skip empty results
+                extras_content.append(result)
+            
+    return "\n".join(extras_content) if extras_content else ""

changedetectionio/processors/README.md

@@ -8,4 +8,8 @@ The concept here is to be able to switch between different domain specific probl
 Some suggestions for the future
 
 - `graphical` 
-- `restock_and_price` - extract price AND stock text
+
+## Todo
+
+- Make each processor return a extra list of sub-processed (so you could configure a single processor in different ways)
+- move restock_diff to its own pip/github repo

changedetectionio/processors/__init__.py

@@ -1,10 +1,13 @@
 from abc import abstractmethod
+from changedetectionio.content_fetchers.base import Fetcher
 from changedetectionio.strtobool import strtobool
-from changedetectionio.model import Watch
 from copy import deepcopy
 from loguru import logger
 import hashlib
+import importlib
+import inspect
 import os
+import pkgutil
 import re
 
 class difference_detection_processor():
@@ -15,28 +18,33 @@ class difference_detection_processor():
     screenshot = None
     watch = None
     xpath_data = None
+    preferred_proxy = None
 
     def __init__(self, *args, datastore, watch_uuid, **kwargs):
         super().__init__(*args, **kwargs)
         self.datastore = datastore
         self.watch = deepcopy(self.datastore.data['watching'].get(watch_uuid))
+        # Generic fetcher that should be extended (requests, playwright etc)
+        self.fetcher = Fetcher()
+
+    async def call_browser(self, preferred_proxy_id=None):
 
-    def call_browser(self):
         from requests.structures import CaseInsensitiveDict
-        # Protect against file:// access
-        if re.search(r'^file://', self.watch.get('url', '').strip(), re.IGNORECASE):
+
+        url = self.watch.link
+
+        # Protect against file:, file:/, file:// access, check the real "link" without any meta "source:" etc prepended.
+        if re.search(r'^file:', url.strip(), re.IGNORECASE):
             if not strtobool(os.getenv('ALLOW_FILE_URI', 'false')):
                 raise Exception(
                     "file:// type access is denied for security reasons."
                 )
 
-        url = self.watch.link
-
         # Requests, playwright, other browser via wss:// etc, fetch_extra_something
         prefer_fetch_backend = self.watch.get('fetch_backend', 'system')
 
         # Proxy ID "key"
-        preferred_proxy_id = self.datastore.get_preferred_proxy_for_watch(uuid=self.watch.get('uuid'))
+        preferred_proxy_id = preferred_proxy_id if preferred_proxy_id else self.datastore.get_preferred_proxy_for_watch(uuid=self.watch.get('uuid'))
 
         # Pluggable content self.fetcher
         if not prefer_fetch_backend or prefer_fetch_backend == 'system':
@@ -81,7 +89,7 @@ class difference_detection_processor():
                 proxy_url = self.datastore.proxy_list.get(preferred_proxy_id).get('url')
                 logger.debug(f"Selected proxy key '{preferred_proxy_id}' as proxy URL '{proxy_url}' for {url}")
             else:
-                logger.debug(f"Skipping adding proxy data when custom Browser endpoint is specified. ")
+                logger.debug("Skipping adding proxy data when custom Browser endpoint is specified. ")
 
         # Now call the fetcher (playwright/requests/etc) with arguments that only a fetcher would need.
         # When browser_connection_url is None, it method should default to working out whats the best defaults (os env vars etc)
@@ -94,6 +102,7 @@ class difference_detection_processor():
             self.fetcher.browser_steps_screenshot_path = os.path.join(self.datastore.datastore_path, self.watch.get('uuid'))
 
         # Tweak the base config with the per-watch ones
+        from changedetectionio.safe_jinja import render as jinja_render
         request_headers = CaseInsensitiveDict()
 
         ua = self.datastore.data['settings']['requests'].get('default_ua')
@@ -110,9 +119,15 @@ class difference_detection_processor():
         if 'Accept-Encoding' in request_headers and "br" in request_headers['Accept-Encoding']:
             request_headers['Accept-Encoding'] = request_headers['Accept-Encoding'].replace(', br', '')
 
+        for header_name in request_headers:
+            request_headers.update({header_name: jinja_render(template_str=request_headers.get(header_name))})
+
         timeout = self.datastore.data['settings']['requests'].get('timeout')
 
         request_body = self.watch.get('body')
+        if request_body:
+            request_body = jinja_render(template_str=self.watch.get('body'))
+        
         request_method = self.watch.get('method')
         ignore_status_codes = self.watch.get('ignore_status_codes', False)
 
@@ -130,16 +145,27 @@ class difference_detection_processor():
         is_binary = self.watch.is_pdf
 
         # And here we go! call the right browser with browser-specific settings
-        self.fetcher.run(url, timeout, request_headers, request_body, request_method, ignore_status_codes, self.watch.get('include_filters'),
-                    is_binary=is_binary)
+        empty_pages_are_a_change = self.datastore.data['settings']['application'].get('empty_pages_are_a_change', False)
+
+        # All fetchers are now async
+        await self.fetcher.run(url=url,
+                               timeout=timeout,
+                               request_headers=request_headers,
+                               request_body=request_body,
+                               request_method=request_method,
+                               ignore_status_codes=ignore_status_codes,
+                               current_include_filters=self.watch.get('include_filters'),
+                               is_binary=is_binary,
+                               empty_pages_are_a_change=empty_pages_are_a_change
+                               )
 
         #@todo .quit here could go on close object, so we can run JS if change-detected
-        self.fetcher.quit()
+        self.fetcher.quit(watch=self.watch)
 
         # After init, call run_changedetection() which will do the actual change-detection
 
     @abstractmethod
-    def run_changedetection(self, watch: Watch, skip_when_checksum_same=True):
+    def run_changedetection(self, watch):
         update_obj = {'last_notification_error': False, 'last_error': False}
         some_data = 'xxxxx'
         update_obj["previous_md5"] = hashlib.md5(some_data.encode('utf-8')).hexdigest()
@@ -147,8 +173,83 @@ class difference_detection_processor():
         return changed_detected, update_obj, ''.encode('utf-8')
 
 
+def find_sub_packages(package_name):
+    """
+    Find all sub-packages within the given package.
+
+    :param package_name: The name of the base package to scan for sub-packages.
+    :return: A list of sub-package names.
+    """
+    package = importlib.import_module(package_name)
+    return [name for _, name, is_pkg in pkgutil.iter_modules(package.__path__) if is_pkg]
+
+
+def find_processors():
+    """
+    Find all subclasses of DifferenceDetectionProcessor in the specified package.
+
+    :param package_name: The name of the package to scan for processor modules.
+    :return: A list of (module, class) tuples.
+    """
+    package_name = "changedetectionio.processors"  # Name of the current package/module
+
+    processors = []
+    sub_packages = find_sub_packages(package_name)
+
+    for sub_package in sub_packages:
+        module_name = f"{package_name}.{sub_package}.processor"
+        try:
+            module = importlib.import_module(module_name)
+
+            # Iterate through all classes in the module
+            for name, obj in inspect.getmembers(module, inspect.isclass):
+                if issubclass(obj, difference_detection_processor) and obj is not difference_detection_processor:
+                    processors.append((module, sub_package))
+        except (ModuleNotFoundError, ImportError) as e:
+            logger.warning(f"Failed to import module {module_name}: {e} (find_processors())")
+
+    return processors
+
+
+def get_parent_module(module):
+    module_name = module.__name__
+    if '.' not in module_name:
+        return None  # Top-level module has no parent
+    parent_module_name = module_name.rsplit('.', 1)[0]
+    try:
+        return importlib.import_module(parent_module_name)
+    except Exception as e:
+        pass
+
+    return False
+
+
+
+def get_custom_watch_obj_for_processor(processor_name):
+    from changedetectionio.model import Watch
+    watch_class = Watch.model
+    processor_classes = find_processors()
+    custom_watch_obj = next((tpl for tpl in processor_classes if tpl[1] == processor_name), None)
+    if custom_watch_obj:
+        # Parent of .processor.py COULD have its own Watch implementation
+        parent_module = get_parent_module(custom_watch_obj[0])
+        if hasattr(parent_module, 'Watch'):
+            watch_class = parent_module.Watch
+
+    return watch_class
+
+
 def available_processors():
-    from . import restock_diff, text_json_diff
-    x=[('text_json_diff', text_json_diff.name), ('restock_diff', restock_diff.name)]
-    # @todo Make this smarter with introspection of sorts.
-    return x
+    """
+    Get a list of processors by name and description for the UI elements
+    :return: A list :)
+    """
+
+    processor_classes = find_processors()
+
+    available = []
+    for package, processor_class in processor_classes:
+        available.append((processor_class, package.name))
+
+    return available
+

changedetectionio/processors/exceptions.py

@@ -0,0 +1,10 @@
+class ProcessorException(Exception):
+    def __init__(self, message=None, status_code=None, url=None, screenshot=None, has_filters=False, html_content='', xpath_data=None):
+        self.message = message
+        self.status_code = status_code
+        self.url = url
+        self.screenshot = screenshot
+        self.has_filters = has_filters
+        self.html_content = html_content
+        self.xpath_data = xpath_data
+        return

changedetectionio/processors/restock_diff.py

@@ -1,62 +0,0 @@
-
-from . import difference_detection_processor
-from loguru import logger
-import hashlib
-import urllib3
-
-urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
-
-name = 'Re-stock detection for single product pages'
-description = 'Detects if the product goes back to in-stock'
-
-class UnableToExtractRestockData(Exception):
-    def __init__(self, status_code):
-        # Set this so we can use it in other parts of the app
-        self.status_code = status_code
-        return
-
-class perform_site_check(difference_detection_processor):
-    screenshot = None
-    xpath_data = None
-
-    def run_changedetection(self, watch, skip_when_checksum_same=True):
-
-        if not watch:
-            raise Exception("Watch no longer exists.")
-
-        # Unset any existing notification error
-        update_obj = {'last_notification_error': False, 'last_error': False}
-
-        self.screenshot = self.fetcher.screenshot
-        self.xpath_data = self.fetcher.xpath_data
-
-        # Track the content type
-        update_obj['content_type'] = self.fetcher.headers.get('Content-Type', '')
-        update_obj["last_check_status"] = self.fetcher.get_last_status_code()
-
-        # Main detection method
-        fetched_md5 = None
-        if self.fetcher.instock_data:
-            fetched_md5 = hashlib.md5(self.fetcher.instock_data.encode('utf-8')).hexdigest()
-            # 'Possibly in stock' comes from stock-not-in-stock.js when no string found above the fold.
-            update_obj["in_stock"] = True if self.fetcher.instock_data == 'Possibly in stock' else False
-            logger.debug(f"Watch UUID {watch.get('uuid')} restock check returned '{self.fetcher.instock_data}' from JS scraper.")
-        else:
-            raise UnableToExtractRestockData(status_code=self.fetcher.status_code)
-
-        # The main thing that all this at the moment comes down to :)
-        changed_detected = False
-        logger.debug(f"Watch UUID {watch.get('uuid')} restock check - Previous MD5: {watch.get('previous_md5')}, Fetched MD5 {fetched_md5}")
-
-        if watch.get('previous_md5') and watch.get('previous_md5') != fetched_md5:
-            # Yes if we only care about it going to instock, AND we are in stock
-            if watch.get('in_stock_only') and update_obj["in_stock"]:
-                changed_detected = True
-
-            if not watch.get('in_stock_only'):
-                # All cases
-                changed_detected = True
-
-        # Always record the new checksum
-        update_obj["previous_md5"] = fetched_md5
-        return changed_detected, update_obj, self.fetcher.instock_data.encode('utf-8').strip()

changedetectionio/processors/restock_diff/__init__.py

@@ -0,0 +1,84 @@
+
+from babel.numbers import parse_decimal
+from changedetectionio.model.Watch import model as BaseWatch
+from typing import Union
+import re
+
+class Restock(dict):
+
+    def parse_currency(self, raw_value: str) -> Union[float, None]:
+        # Clean and standardize the value (ie 1,400.00 should be 1400.00), even better would be store the whole thing as an integer.
+        standardized_value = raw_value
+
+        if ',' in standardized_value and '.' in standardized_value:
+            # Identify the correct decimal separator
+            if standardized_value.rfind('.') > standardized_value.rfind(','):
+                standardized_value = standardized_value.replace(',', '')
+            else:
+                standardized_value = standardized_value.replace('.', '').replace(',', '.')
+        else:
+            standardized_value = standardized_value.replace(',', '.')
+
+        # Remove any non-numeric characters except for the decimal point
+        standardized_value = re.sub(r'[^\d.-]', '', standardized_value)
+
+        if standardized_value:
+            # Convert to float
+            return float(parse_decimal(standardized_value, locale='en'))
+
+        return None
+
+    def __init__(self, *args, **kwargs):
+        # Define default values
+        default_values = {
+            'in_stock': None,
+            'price': None,
+            'currency': None,
+            'original_price': None
+        }
+
+        # Initialize the dictionary with default values
+        super().__init__(default_values)
+
+        # Update with any provided positional arguments (dictionaries)
+        if args:
+            if len(args) == 1 and isinstance(args[0], dict):
+                self.update(args[0])
+            else:
+                raise ValueError("Only one positional argument of type 'dict' is allowed")
+
+    def __setitem__(self, key, value):
+        # Custom logic to handle setting price and original_price
+        if key == 'price' or key == 'original_price':
+            if isinstance(value, str):
+                value = self.parse_currency(raw_value=value)
+
+        super().__setitem__(key, value)
+
+class Watch(BaseWatch):
+    def __init__(self, *arg, **kw):
+        super().__init__(*arg, **kw)
+        self['restock'] = Restock(kw['default']['restock']) if kw.get('default') and kw['default'].get('restock') else Restock()
+
+        self['restock_settings'] = kw['default']['restock_settings'] if kw.get('default',{}).get('restock_settings') else {
+            'follow_price_changes': True,
+            'in_stock_processing' : 'in_stock_only'
+        } #@todo update
+
+    def clear_watch(self):
+        super().clear_watch()
+        self.update({'restock': Restock()})
+
+    def extra_notification_token_values(self):
+        values = super().extra_notification_token_values()
+        values['restock'] = self.get('restock', {})
+        return values
+
+    def extra_notification_token_placeholder_info(self):
+        values = super().extra_notification_token_placeholder_info()
+
+        values.append(('restock.price', "Price detected"))
+        values.append(('restock.original_price', "Original price at first check"))
+
+        return values
+

changedetectionio/processors/restock_diff/forms.py

@@ -0,0 +1,81 @@
+from wtforms import (
+    BooleanField,
+    validators,
+    FloatField
+)
+from wtforms.fields.choices import RadioField
+from wtforms.fields.form import FormField
+from wtforms.form import Form
+
+from changedetectionio.forms import processor_text_json_diff_form
+
+
+class RestockSettingsForm(Form):
+    in_stock_processing = RadioField(label='Re-stock detection', choices=[
+        ('in_stock_only', "In Stock only (Out Of Stock -> In Stock only)"),
+        ('all_changes', "Any availability changes"),
+        ('off', "Off, don't follow availability/restock"),
+    ], default="in_stock_only")
+
+    price_change_min = FloatField('Below price to trigger notification', [validators.Optional()],
+                                  render_kw={"placeholder": "No limit", "size": "10"})
+    price_change_max = FloatField('Above price to trigger notification', [validators.Optional()],
+                                  render_kw={"placeholder": "No limit", "size": "10"})
+    price_change_threshold_percent = FloatField('Threshold in % for price changes since the original price', validators=[
+
+        validators.Optional(),
+        validators.NumberRange(min=0, max=100, message="Should be between 0 and 100"),
+    ], render_kw={"placeholder": "0%", "size": "5"})
+
+    follow_price_changes = BooleanField('Follow price changes', default=True)
+
+class processor_settings_form(processor_text_json_diff_form):
+    restock_settings = FormField(RestockSettingsForm)
+
+    def extra_tab_content(self):
+        return 'Restock & Price Detection'
+
+    def extra_form_content(self):
+        output = ""
+
+        if getattr(self, 'watch', None) and getattr(self, 'datastore'):
+            for tag_uuid in self.watch.get('tags'):
+                tag = self.datastore.data['settings']['application']['tags'].get(tag_uuid, {})
+                if tag.get('overrides_watch'):
+                    # @todo - Quick and dirty, cant access 'url_for' here because its out of scope somehow
+                    output = f"""<p><strong>Note! A Group tag overrides the restock and price detection here.</strong></p><style>#restock-fieldset-price-group {{ opacity: 0.6; }}</style>"""
+
+        output += """
+        {% from '_helpers.html' import render_field, render_checkbox_field, render_button %}
+        <script>        
+            $(document).ready(function () {
+                toggleOpacity('#restock_settings-follow_price_changes', '.price-change-minmax', true);
+            });
+        </script>
+
+        <fieldset id="restock-fieldset-price-group">
+            <div class="pure-control-group">
+                <fieldset class="pure-group inline-radio">
+                    {{ render_field(form.restock_settings.in_stock_processing) }}
+                </fieldset>
+                <fieldset class="pure-group">
+                    {{ render_checkbox_field(form.restock_settings.follow_price_changes) }}
+                    <span class="pure-form-message-inline">Changes in price should trigger a notification</span>
+                </fieldset>
+                <fieldset class="pure-group price-change-minmax">               
+                    {{ render_field(form.restock_settings.price_change_min, placeholder=watch.get('restock', {}).get('price')) }}
+                    <span class="pure-form-message-inline">Minimum amount, Trigger a change/notification when the price drops <i>below</i> this value.</span>
+                </fieldset>
+                <fieldset class="pure-group price-change-minmax">
+                    {{ render_field(form.restock_settings.price_change_max, placeholder=watch.get('restock', {}).get('price')) }}
+                    <span class="pure-form-message-inline">Maximum amount, Trigger a change/notification when the price rises <i>above</i> this value.</span>
+                </fieldset>
+                <fieldset class="pure-group price-change-minmax">
+                    {{ render_field(form.restock_settings.price_change_threshold_percent) }}
+                    <span class="pure-form-message-inline">Price must change more than this % to trigger a change since the first check.</span><br>
+                    <span class="pure-form-message-inline">For example, If the product is $1,000 USD originally, <strong>2%</strong> would mean it has to change more than $20 since the first check.</span><br>
+                </fieldset>                
+            </div>
+        </fieldset>
+        """
+        return output

changedetectionio/processors/restock_diff/processor.py

@@ -0,0 +1,314 @@
+from .. import difference_detection_processor
+from ..exceptions import ProcessorException
+from . import Restock
+from loguru import logger
+
+import urllib3
+import time
+
+urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+name = 'Re-stock & Price detection for pages with a SINGLE product'
+description = 'Detects if the product goes back to in-stock'
+
+class UnableToExtractRestockData(Exception):
+    def __init__(self, status_code):
+        # Set this so we can use it in other parts of the app
+        self.status_code = status_code
+        return
+
+class MoreThanOnePriceFound(Exception):
+    def __init__(self):
+        return
+
+def _search_prop_by_value(matches, value):
+    for properties in matches:
+        for prop in properties:
+            if value in prop[0]:
+                return prop[1]  # Yield the desired value and exit the function
+
+def _deduplicate_prices(data):
+    import re
+
+    '''
+    Some price data has multiple entries, OR it has a single entry with ['$159', '159', 159, "$ 159"] or just "159"
+    Get all the values, clean it and add it to a set then return the unique values
+    '''
+    unique_data = set()
+
+    # Return the complete 'datum' where its price was not seen before
+    for datum in data:
+
+        if isinstance(datum.value, list):
+            # Process each item in the list
+            normalized_value = set([float(re.sub(r'[^\d.]', '', str(item))) for item in datum.value if str(item).strip()])
+            unique_data.update(normalized_value)
+        else:
+            # Process single value
+            v = float(re.sub(r'[^\d.]', '', str(datum.value)))
+            unique_data.add(v)
+
+    return list(unique_data)
+
+
+# should return Restock()
+# add casting?
+def get_itemprop_availability(html_content) -> Restock:
+    """
+    Kind of funny/cool way to find price/availability in one many different possibilities.
+    Use 'extruct' to find any possible RDFa/microdata/json-ld data, make a JSON string from the output then search it.
+    """
+    from jsonpath_ng import parse
+
+    import re
+    now = time.time()
+    import extruct
+    logger.trace(f"Imported extruct module in {time.time() - now:.3f}s")
+
+    now = time.time()
+
+    # Extruct is very slow, I'm wondering if some ML is going to be faster (800ms on my i7), 'rdfa' seems to be the heaviest.
+    syntaxes = ['dublincore', 'json-ld', 'microdata', 'microformat', 'opengraph']
+    try:
+        data = extruct.extract(html_content, syntaxes=syntaxes)
+    except Exception as e:
+        logger.warning(f"Unable to extract data, document parsing with extruct failed with {type(e).__name__} - {str(e)}")
+        return Restock()
+
+    logger.trace(f"Extruct basic extract of all metadata done in {time.time() - now:.3f}s")
+
+    # First phase, dead simple scanning of anything that looks useful
+    value = Restock()
+    if data:
+        logger.debug("Using jsonpath to find price/availability/etc")
+        price_parse = parse('$..(price|Price)')
+        pricecurrency_parse = parse('$..(pricecurrency|currency|priceCurrency )')
+        availability_parse = parse('$..(availability|Availability)')
+
+        price_result = _deduplicate_prices(price_parse.find(data))
+        if price_result:
+            # Right now, we just support single product items, maybe we will store the whole actual metadata seperately in teh future and
+            # parse that for the UI?
+            if len(price_result) > 1 and len(price_result) > 1:
+                # See of all prices are different, in the case that one product has many embedded data types with the same price
+                # One might have $121.95 and another 121.95 etc
+                logger.warning(f"More than one price found {price_result}, throwing exception, cant use this plugin.")
+                raise MoreThanOnePriceFound()
+
+            value['price'] = price_result[0]
+
+        pricecurrency_result = pricecurrency_parse.find(data)
+        if pricecurrency_result:
+            value['currency'] = pricecurrency_result[0].value
+
+        availability_result = availability_parse.find(data)
+        if availability_result:
+            value['availability'] = availability_result[0].value
+
+        if value.get('availability'):
+            value['availability'] = re.sub(r'(?i)^(https|http)://schema.org/', '',
+                                           value.get('availability').strip(' "\'').lower()) if value.get('availability') else None
+
+        # Second, go dig OpenGraph which is something that jsonpath_ng cant do because of the tuples and double-dots (:)
+        if not value.get('price') or value.get('availability'):
+            logger.debug("Alternatively digging through OpenGraph properties for restock/price info..")
+            jsonpath_expr = parse('$..properties')
+
+            for match in jsonpath_expr.find(data):
+                if not value.get('price'):
+                    value['price'] = _search_prop_by_value([match.value], "price:amount")
+                if not value.get('availability'):
+                    value['availability'] = _search_prop_by_value([match.value], "product:availability")
+                if not value.get('currency'):
+                    value['currency'] = _search_prop_by_value([match.value], "price:currency")
+    logger.trace(f"Processed with Extruct in {time.time()-now:.3f}s")
+
+    return value
+
+
+def is_between(number, lower=None, upper=None):
+    """
+    Check if a number is between two values.
+
+    Parameters:
+    number (float): The number to check.
+    lower (float or None): The lower bound (inclusive). If None, no lower bound.
+    upper (float or None): The upper bound (inclusive). If None, no upper bound.
+
+    Returns:
+    bool: True if the number is between the lower and upper bounds, False otherwise.
+    """
+    return (lower is None or lower <= number) and (upper is None or number <= upper)
+
+
+class perform_site_check(difference_detection_processor):
+    screenshot = None
+    xpath_data = None
+
+    def run_changedetection(self, watch):
+        import hashlib
+
+        if not watch:
+            raise Exception("Watch no longer exists.")
+
+        # Unset any existing notification error
+        update_obj = {'last_notification_error': False, 'last_error': False, 'restock':  Restock()}
+
+        self.screenshot = self.fetcher.screenshot
+        self.xpath_data = self.fetcher.xpath_data
+
+        # Track the content type
+        update_obj['content_type'] = self.fetcher.headers.get('Content-Type', '')
+        update_obj["last_check_status"] = self.fetcher.get_last_status_code()
+
+        # Only try to process restock information (like scraping for keywords) if the page was actually rendered correctly.
+        # Otherwise it will assume "in stock" because nothing suggesting the opposite was found
+        from ...html_tools import html_to_text
+        text = html_to_text(self.fetcher.content)
+        logger.debug(f"Length of text after conversion: {len(text)}")
+        if not len(text):
+            from ...content_fetchers.exceptions import ReplyWithContentButNoText
+            raise ReplyWithContentButNoText(url=watch.link,
+                                            status_code=self.fetcher.get_last_status_code(),
+                                            screenshot=self.fetcher.screenshot,
+                                            html_content=self.fetcher.content,
+                                            xpath_data=self.fetcher.xpath_data
+                                            )
+
+        # Which restock settings to compare against?
+        restock_settings = watch.get('restock_settings', {})
+
+        # See if any tags have 'activate for individual watches in this tag/group?' enabled and use the first we find
+        for tag_uuid in watch.get('tags'):
+            tag = self.datastore.data['settings']['application']['tags'].get(tag_uuid, {})
+            if tag.get('overrides_watch'):
+                restock_settings = tag.get('restock_settings', {})
+                logger.info(f"Watch {watch.get('uuid')} - Tag '{tag.get('title')}' selected for restock settings override")
+                break
+
+
+        itemprop_availability = {}
+        try:
+            itemprop_availability = get_itemprop_availability(self.fetcher.content)
+        except MoreThanOnePriceFound as e:
+            # Add the real data
+            raise ProcessorException(message="Cannot run, more than one price detected, this plugin is only for product pages with ONE product, try the content-change detection mode.",
+                                     url=watch.get('url'),
+                                     status_code=self.fetcher.get_last_status_code(),
+                                     screenshot=self.fetcher.screenshot,
+                                     xpath_data=self.fetcher.xpath_data
+                                     )
+
+        # Something valid in get_itemprop_availability() by scraping metadata ?
+        if itemprop_availability.get('price') or itemprop_availability.get('availability'):
+            # Store for other usage
+            update_obj['restock'] = itemprop_availability
+
+            if itemprop_availability.get('availability'):
+                # @todo: Configurable?
+                if any(substring.lower() in itemprop_availability['availability'].lower() for substring in [
+                    'instock',
+                    'instoreonly',
+                    'limitedavailability',
+                    'onlineonly',
+                    'presale']
+                       ):
+                    update_obj['restock']['in_stock'] = True
+                else:
+                    update_obj['restock']['in_stock'] = False
+
+        # Main detection method
+        fetched_md5 = None
+
+        # store original price if not set
+        if itemprop_availability and itemprop_availability.get('price') and not itemprop_availability.get('original_price'):
+            itemprop_availability['original_price'] = itemprop_availability.get('price')
+            update_obj['restock']["original_price"] = itemprop_availability.get('price')
+
+        if not self.fetcher.instock_data and not itemprop_availability.get('availability') and not itemprop_availability.get('price'):
+            raise ProcessorException(
+                message=f"Unable to extract restock data for this page unfortunately. (Got code {self.fetcher.get_last_status_code()} from server), no embedded stock information was found and nothing interesting in the text, try using this watch with Chrome.",
+                url=watch.get('url'),
+                status_code=self.fetcher.get_last_status_code(),
+                screenshot=self.fetcher.screenshot,
+                xpath_data=self.fetcher.xpath_data
+                )
+
+        logger.debug(f"self.fetcher.instock_data is - '{self.fetcher.instock_data}' and itemprop_availability.get('availability') is {itemprop_availability.get('availability')}")
+        # Nothing automatic in microdata found, revert to scraping the page
+        if self.fetcher.instock_data and itemprop_availability.get('availability') is None:
+            # 'Possibly in stock' comes from stock-not-in-stock.js when no string found above the fold.
+            # Careful! this does not really come from chrome/js when the watch is set to plaintext
+            update_obj['restock']["in_stock"] = True if self.fetcher.instock_data == 'Possibly in stock' else False
+            logger.debug(f"Watch UUID {watch.get('uuid')} restock check returned instock_data - '{self.fetcher.instock_data}' from JS scraper.")
+
+        # Very often websites will lie about the 'availability' in the metadata, so if the scraped version says its NOT in stock, use that.
+        if self.fetcher.instock_data and self.fetcher.instock_data != 'Possibly in stock':
+            if update_obj['restock'].get('in_stock'):
+                logger.warning(
+                    f"Lie detected in the availability machine data!! when scraping said its not in stock!! itemprop was '{itemprop_availability}' and scraped from browser was '{self.fetcher.instock_data}' update obj was {update_obj['restock']} ")
+                logger.warning(f"Setting instock to FALSE, scraper found '{self.fetcher.instock_data}' in the body but metadata reported not-in-stock")
+                update_obj['restock']["in_stock"] = False
+
+        # What we store in the snapshot
+        price = update_obj.get('restock').get('price') if update_obj.get('restock').get('price') else ""
+        snapshot_content = f"In Stock: {update_obj.get('restock').get('in_stock')} - Price: {price}"
+
+        # Main detection method
+        fetched_md5 = hashlib.md5(snapshot_content.encode('utf-8')).hexdigest()
+
+        # The main thing that all this at the moment comes down to :)
+        changed_detected = False
+        logger.debug(f"Watch UUID {watch.get('uuid')} restock check - Previous MD5: {watch.get('previous_md5')}, Fetched MD5 {fetched_md5}")
+
+        # out of stock -> back in stock only?
+        if watch.get('restock') and watch['restock'].get('in_stock') != update_obj['restock'].get('in_stock'):
+            # Yes if we only care about it going to instock, AND we are in stock
+            if restock_settings.get('in_stock_processing') == 'in_stock_only' and update_obj['restock']['in_stock']:
+                changed_detected = True
+
+            if restock_settings.get('in_stock_processing') == 'all_changes':
+                # All cases
+                changed_detected = True
+
+        if restock_settings.get('follow_price_changes') and watch.get('restock') and update_obj.get('restock') and update_obj['restock'].get('price'):
+            price = float(update_obj['restock'].get('price'))
+            # Default to current price if no previous price found
+            if watch['restock'].get('original_price'):
+                previous_price = float(watch['restock'].get('original_price'))
+                # It was different, but negate it further down
+                if price != previous_price:
+                    changed_detected = True
+
+            # Minimum/maximum price limit
+            if update_obj.get('restock') and update_obj['restock'].get('price'):
+                logger.debug(
+                    f"{watch.get('uuid')} - Change was detected, 'price_change_max' is '{restock_settings.get('price_change_max', '')}' 'price_change_min' is '{restock_settings.get('price_change_min', '')}', price from website is '{update_obj['restock'].get('price', '')}'.")
+                if update_obj['restock'].get('price'):
+                    min_limit = float(restock_settings.get('price_change_min')) if restock_settings.get('price_change_min') else None
+                    max_limit = float(restock_settings.get('price_change_max')) if restock_settings.get('price_change_max') else None
+
+                    price = float(update_obj['restock'].get('price'))
+                    logger.debug(f"{watch.get('uuid')} after float conversion - Min limit: '{min_limit}' Max limit: '{max_limit}' Price: '{price}'")
+                    if min_limit or max_limit:
+                        if is_between(number=price, lower=min_limit, upper=max_limit):
+                            # Price was between min/max limit, so there was nothing todo in any case
+                            logger.trace(f"{watch.get('uuid')} {price} is between {min_limit} and {max_limit}, nothing to check, forcing changed_detected = False (was {changed_detected})")
+                            changed_detected = False
+                        else:
+                            logger.trace(f"{watch.get('uuid')} {price} is between {min_limit} and {max_limit}, continuing normal comparison")
+
+                    # Price comparison by %
+                    if watch['restock'].get('original_price') and changed_detected and restock_settings.get('price_change_threshold_percent'):
+                        previous_price = float(watch['restock'].get('original_price'))
+                        pc = float(restock_settings.get('price_change_threshold_percent'))
+                        change = abs((price - previous_price) / previous_price * 100)
+                        if change and change <= pc:
+                            logger.debug(f"{watch.get('uuid')} Override change-detected to FALSE because % threshold ({pc}%) was {change:.3f}%")
+                            changed_detected = False
+                        else:
+                            logger.debug(f"{watch.get('uuid')} Price change was {change:.3f}% , (threshold {pc}%)")
+
+        # Always record the new checksum
+        update_obj["previous_md5"] = fetched_md5
+
+        return changed_detected, update_obj, snapshot_content.strip()