Python 3.14 test #3662

This reverts commit e2b407c6f3.

.github/actions/extract-memory-report/action.yml

@@ -0,0 +1,51 @@
+name: 'Extract Memory Test Report'
+description: 'Extracts and displays memory test report from a container'
+inputs:
+  container-name:
+    description: 'Name of the container to extract logs from'
+    required: true
+  python-version:
+    description: 'Python version for artifact naming'
+    required: true
+  output-dir:
+    description: 'Directory to store output logs'
+    required: false
+    default: 'output-logs'
+
+runs:
+  using: "composite"
+  steps:
+    - name: Create output directory
+      shell: bash
+      run: |
+        mkdir -p ${{ inputs.output-dir }}
+
+    - name: Dump container log
+      shell: bash
+      run: |
+        echo "Disabled for now"
+#        return
+#        docker logs ${{ inputs.container-name }} > ${{ inputs.output-dir }}/${{ inputs.container-name }}-stdout-${{ inputs.python-version }}.txt 2>&1 || echo "Could not get stdout"
+#        docker logs ${{ inputs.container-name }} 2> ${{ inputs.output-dir }}/${{ inputs.container-name }}-stderr-${{ inputs.python-version }}.txt || echo "Could not get stderr"
+
+    - name: Extract and display memory test report
+      shell: bash
+      run: |
+        echo "Disabled for now"
+#        echo "Extracting test-memory.log from container..."
+#        docker cp ${{ inputs.container-name }}:/app/changedetectionio/test-memory.log ${{ inputs.output-dir }}/test-memory-${{ inputs.python-version }}.log || echo "test-memory.log not found in container"
+#
+#        echo "=== Top 10 Highest Peak Memory Tests ==="
+#        if [ -f ${{ inputs.output-dir }}/test-memory-${{ inputs.python-version }}.log ]; then
+#          grep "Peak memory:" ${{ inputs.output-dir }}/test-memory-${{ inputs.python-version }}.log | \
+#            sed 's/.*Peak memory: //' | \
+#            paste -d'|' - <(grep "Peak memory:" ${{ inputs.output-dir }}/test-memory-${{ inputs.python-version }}.log) | \
+#            sort -t'|' -k1 -nr | \
+#            cut -d'|' -f2 | \
+#            head -10
+#          echo ""
+#          echo "=== Full Memory Test Report ==="
+#          cat ${{ inputs.output-dir }}/test-memory-${{ inputs.python-version }}.log
+#        else
+#          echo "No memory log available"
+#        fi

.github/workflows/codeql-analysis.yml

@@ -30,7 +30,7 @@ jobs:
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v5
+      uses: actions/checkout@v6
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL

.github/workflows/containers.yml

@@ -39,12 +39,20 @@ jobs:
     # Or if we are in a tagged release scenario.
     if: ${{ github.event.workflow_run.conclusion == 'success' }} || ${{ github.event.release.tag_name }} != ''
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       - name: Set up Python 3.11
         uses: actions/setup-python@v6
         with:
           python-version: 3.11
 
+      - name: Cache pip packages
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pip
+          key: ${{ runner.os }}-pip-${{ hashFiles('requirements.txt') }}
+          restore-keys: |
+            ${{ runner.os }}-pip-
+
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip

.github/workflows/pypi-release.yml

@@ -7,7 +7,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-    - uses: actions/checkout@v5
+    - uses: actions/checkout@v6
     - name: Set up Python
       uses: actions/setup-python@v6
       with:
@@ -21,7 +21,7 @@ jobs:
     - name: Build a binary wheel and a source tarball
       run: python3 -m build
     - name: Store the distribution packages
-      uses: actions/upload-artifact@v4
+      uses: actions/upload-artifact@v5
       with:
         name: python-package-distributions
         path: dist/
@@ -34,7 +34,7 @@ jobs:
     - build
     steps:
     - name: Download all the dists
-      uses: actions/download-artifact@v5
+      uses: actions/download-artifact@v6
       with:
         name: python-package-distributions
         path: dist/
@@ -93,7 +93,7 @@ jobs:
 
     steps:
     - name: Download all the dists
-      uses: actions/download-artifact@v5
+      uses: actions/download-artifact@v6
       with:
         name: python-package-distributions
         path: dist/

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

@@ -44,12 +44,20 @@ jobs:
           - platform: linux/arm64
             dockerfile: ./.github/test/Dockerfile-alpine
     steps:
-        - uses: actions/checkout@v5
+        - uses: actions/checkout@v6
         - name: Set up Python 3.11
           uses: actions/setup-python@v6
           with:
             python-version: 3.11
 
+        - name: Cache pip packages
+          uses: actions/cache@v4
+          with:
+            path: ~/.cache/pip
+            key: ${{ runner.os }}-pip-${{ hashFiles('requirements.txt') }}
+            restore-keys: |
+              ${{ runner.os }}-pip-
+
         # Just test that the build works, some libraries won't compile on ARM/rPi etc
         - name: Set up QEMU
           uses: docker/setup-qemu-action@v3
@@ -74,5 +82,5 @@ jobs:
             file: ${{ matrix.dockerfile }}
             platforms: ${{ matrix.platform }}
             cache-from: type=gha
-            cache-to: type=gha,mode=min
+            cache-to: type=gha,mode=max
 

.github/workflows/test-only.yml

@@ -7,7 +7,7 @@ jobs:
   lint-code:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       - name: Lint with Ruff
         run: |
           pip install ruff
@@ -21,6 +21,8 @@ jobs:
           python3 -c "from openapi_spec_validator import validate_spec; import yaml; validate_spec(yaml.safe_load(open('docs/api-spec.yaml')))"
 
   test-application-3-10:
+    # Only run on push to master (including PR merges)
+    if: github.event_name == 'push' && github.ref == 'refs/heads/master'
     needs: lint-code
     uses: ./.github/workflows/test-stack-reusable-workflow.yml
     with:
@@ -28,12 +30,15 @@ jobs:
 
 
   test-application-3-11:
+    # Always run
     needs: lint-code
     uses: ./.github/workflows/test-stack-reusable-workflow.yml
     with:
       python-version: '3.11'
 
   test-application-3-12:
+    # Only run on push to master (including PR merges)
+    if: github.event_name == 'push' && github.ref == 'refs/heads/master'
     needs: lint-code
     uses: ./.github/workflows/test-stack-reusable-workflow.yml
     with:
@@ -41,8 +46,19 @@ jobs:
       skip-pypuppeteer: true
 
   test-application-3-13:
+    # Only run on push to master (including PR merges)
+    if: github.event_name == 'push' && github.ref == 'refs/heads/master'
     needs: lint-code
     uses: ./.github/workflows/test-stack-reusable-workflow.yml
     with:
       python-version: '3.13'
-      skip-pypuppeteer: true
+      skip-pypuppeteer: true
+
+
+  test-application-3-14:
+    #if: github.event_name == 'push' && github.ref == 'refs/heads/master'
+    needs: lint-code
+    uses: ./.github/workflows/test-stack-reusable-workflow.yml
+    with:
+      python-version: '3.14'
+      skip-pypuppeteer: false

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

@@ -15,141 +15,294 @@ on:
         default: false
 
 jobs:
-  test-application:
+  # Build the Docker image once and share it with all test jobs
+  build:
     runs-on: ubuntu-latest
     env:
       PYTHON_VERSION: ${{ inputs.python-version }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
-      # Mainly just for link/flake8
       - name: Set up Python ${{ env.PYTHON_VERSION }}
         uses: actions/setup-python@v6
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
+      - name: Cache pip packages
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pip
+          key: ${{ runner.os }}-pip-py${{ env.PYTHON_VERSION }}-${{ hashFiles('requirements.txt') }}
+          restore-keys: |
+            ${{ runner.os }}-pip-py${{ env.PYTHON_VERSION }}-
+            ${{ runner.os }}-pip-
+
       - name: Build changedetection.io container for testing under Python ${{ env.PYTHON_VERSION }}
         run: |
           echo "---- Building for Python ${{ env.PYTHON_VERSION }} -----"
-          # Build a changedetection.io container and start testing inside
           docker build --build-arg PYTHON_VERSION=${{ env.PYTHON_VERSION }} --build-arg LOGGER_LEVEL=TRACE -t test-changedetectionio .
-          # Debug info
-          docker run test-changedetectionio  bash -c 'pip list'         
+          docker run test-changedetectionio bash -c 'pip list'
 
       - name: We should be Python ${{ env.PYTHON_VERSION }} ...
-        run: |         
-          docker run test-changedetectionio  bash -c 'python3 --version'
-
-      - name: Spin up ancillary testable services
         run: |
-          
-          docker network create changedet-network
-          
-          # Selenium
-          docker run --network changedet-network -d --hostname selenium  -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome:4
-          
-          # SocketPuppetBrowser + Extra for custom browser test
-          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser --hostname sockpuppetbrowser --rm -p 3000:3000 dgtlmoon/sockpuppetbrowser:latest                    
-          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser-custom-url --hostname sockpuppetbrowser-custom-url  -p 3001:3000 --rm dgtlmoon/sockpuppetbrowser:latest
+          docker run test-changedetectionio bash -c 'python3 --version'
 
-      - name: Spin up ancillary SMTP+Echo message test server
+      - name: Save Docker image
         run: |
-          # Debug SMTP server/echo message back server, telnet 11080 to it should immediately bounce back the most recent message that tried to send (then you can see if cdio tried to send, the format, etc)
-          # 11025 is the SMTP port for testing
-          # apprise example would be 'mailto://changedetection@localhost:11025/?to=fff@home.com  (it will also echo to STDOUT)
-          # telnet localhost 11080
-          docker run --network changedet-network -d -p 11025:11025 -p 11080:11080  --hostname mailserver test-changedetectionio  bash -c 'pip3 install aiosmtpd && python changedetectionio/tests/smtp/smtp-test-server.py'
-          docker ps
+          docker save test-changedetectionio -o /tmp/test-changedetectionio.tar
 
-      - name: Show docker container state and other debug info
+      - name: Upload Docker image artifact
+        uses: actions/upload-artifact@v5
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp/test-changedetectionio.tar
+          retention-days: 1
+
+  # Unit tests (lightweight, no ancillary services needed)
+  unit-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
         run: |
-          set -x
-          echo "Running processes in docker..."
-          docker ps
+          docker load -i /tmp/test-changedetectionio.tar
 
       - name: Run Unit Tests
         run: |
-          # Unit tests
-          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'
+          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)
+  # Basic pytest tests with ancillary services
+  basic-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 25
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
         run: |
-          # All tests
-          echo "run test with pytest"
-          # The default pytest logger_level is TRACE
-          # To change logger_level for pytest(test/conftest.py),
-          # append the docker option. e.g. '-e LOGGER_LEVEL=DEBUG'
-          docker run --name test-cdio-basic-tests --network changedet-network  test-changedetectionio  bash -c 'cd changedetectionio && ./run_basic_tests.sh'
+          docker load -i /tmp/test-changedetectionio.tar
 
-# PLAYWRIGHT/NODE-> CDP
-      - name: Playwright and SocketPuppetBrowser - Specific tests in built container
+      - name: Test built container with Pytest
         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  -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'
+          docker network inspect changedet-network >/dev/null 2>&1 || docker network create changedet-network
+          docker run --name test-cdio-basic-tests --network changedet-network test-changedetectionio bash -c 'cd changedetectionio && ./run_basic_tests.sh'
 
+      - name: Extract memory report and logs
+        if: always()
+        uses: ./.github/actions/extract-memory-report
+        with:
+          container-name: test-cdio-basic-tests
+          python-version: ${{ env.PYTHON_VERSION }}
 
-      - 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 'find .; cd changedetectionio; pytest --live-server-host=0.0.0.0  --live-server-port=5004 tests/test_request.py; pwd;find .'
+      - name: Store test artifacts
+        if: always()
+        uses: actions/upload-artifact@v5
+        with:
+          name: test-cdio-basic-tests-output-py${{ env.PYTHON_VERSION }}
+          path: output-logs
 
-      - name: Playwright and SocketPuppetBrowser - Restock detection
-        run: |                            
-          # restock detection via playwright - added name=changedet here so that playwright and sockpuppetbrowser can connect to it
-          docker run --rm --name "changedet" -e "FLASK_SERVER_NAME=changedet" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-port=5004 --live-server-host=0.0.0.0 tests/restock/test_restock.py'
+  # Playwright tests
+  playwright-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
 
-# STRAIGHT TO CDP
-      - name: Pyppeteer and SocketPuppetBrowser - Specific tests in built container
-        if: ${{ inputs.skip-pypuppeteer == false }}
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
         run: |
-          # Playwright via Sockpuppetbrowser fetch 
-          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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 load -i /tmp/test-changedetectionio.tar
 
-      - name: Pyppeteer and SocketPuppetBrowser - Headers and requests checks
-        if: ${{ inputs.skip-pypuppeteer == false }}
+      - name: Spin up ancillary services
         run: |
-          # Settings headers playwright tests - Call back in from Sockpuppetbrowser, check headers
-          docker run --name "changedet" --hostname changedet --rm  -e "FAST_PUPPETEER_CHROME_FETCHER=True" -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 network create changedet-network
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser --hostname sockpuppetbrowser --rm -p 3000:3000 dgtlmoon/sockpuppetbrowser:latest
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser-custom-url --hostname sockpuppetbrowser-custom-url -p 3001:3000 --rm dgtlmoon/sockpuppetbrowser:latest
 
-      - name: Pyppeteer and SocketPuppetBrowser - Restock detection
-        if: ${{ inputs.skip-pypuppeteer == false }}
-        run: |                            
-          # restock detection via playwright - added name=changedet here so that playwright and sockpuppetbrowser can connect to it
-          docker run --rm --name "changedet" -e "FLASK_SERVER_NAME=changedet"  -e "FAST_PUPPETEER_CHROME_FETCHER=True"  -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-port=5004 --live-server-host=0.0.0.0 tests/restock/test_restock.py'
+      - name: Playwright - Specific tests in built container
+        run: |
+          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 - Headers and requests
+        run: |
+          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 - Restock detection
+        run: |
+          docker run --rm --name "changedet" -e "FLASK_SERVER_NAME=changedet" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio bash -c 'cd changedetectionio;pytest --live-server-port=5004 --live-server-host=0.0.0.0 tests/restock/test_restock.py'
+
+  # Pyppeteer tests
+  pyppeteer-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    if: ${{ inputs.skip-pypuppeteer == false }}
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
+        run: |
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Spin up ancillary services
+        run: |
+          docker network create changedet-network
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser --hostname sockpuppetbrowser --rm -p 3000:3000 dgtlmoon/sockpuppetbrowser:latest
+
+      - name: Pyppeteer - Specific tests in built container
+        run: |
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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 "FAST_PUPPETEER_CHROME_FETCHER=True" -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'
+
+      - name: Pyppeteer - Headers and requests checks
+        run: |
+          docker run --name "changedet" --hostname changedet --rm -e "FAST_PUPPETEER_CHROME_FETCHER=True" -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'
+
+      - name: Pyppeteer - Restock detection
+        run: |
+          docker run --rm --name "changedet" -e "FLASK_SERVER_NAME=changedet" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio bash -c 'cd changedetectionio;pytest --live-server-port=5004 --live-server-host=0.0.0.0 tests/restock/test_restock.py'
+
+  # Selenium tests
+  selenium-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
+        run: |
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Spin up ancillary services
+        run: |
+          docker network create changedet-network
+          docker run --network changedet-network -d --hostname selenium -p 4444:4444 --rm --shm-size="2g" selenium/standalone-chrome:4
+          sleep 3
+
+      - name: Specific tests for headers and requests checks with Selenium
+        run: |
+
+          docker run --name "changedet" --hostname changedet --rm -e "FLASK_SERVER_NAME=changedet" -e "WEBDRIVER_URL=http://selenium:4444/wd/hub" --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'
 
-# SELENIUM
       - name: Specific tests in built container for Selenium
         run: |
-          # Selenium fetch
-          docker run --rm -e "WEBDRIVER_URL=http://selenium:4444/wd/hub" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/fetchers/test_content.py && pytest tests/test_errorhandling.py'
+          docker run --rm -e "WEBDRIVER_URL=http://selenium:4444/wd/hub" --network changedet-network test-changedetectionio bash -c 'cd changedetectionio;pytest tests/fetchers/test_content.py && pytest tests/test_errorhandling.py'
 
-      - name: Specific tests in built container for headers and requests checks with Selenium
+
+  # SMTP tests
+  smtp-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
         run: |
-          docker run --name "changedet" --hostname changedet --rm -e "FLASK_SERVER_NAME=changedet" -e "WEBDRIVER_URL=http://selenium:4444/wd/hub" --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 load -i /tmp/test-changedetectionio.tar
+
+      - name: Spin up SMTP test server
+        run: |
+          docker network create changedet-network
+          docker run --network changedet-network -d -p 11025:11025 -p 11080:11080 --hostname mailserver test-changedetectionio bash -c 'pip3 install aiosmtpd && python changedetectionio/tests/smtp/smtp-test-server.py'
 
-# OTHER STUFF
       - name: Test SMTP notification mime types
         run: |
-          # SMTP content types - needs the 'Debug SMTP server/echo message back server' container from above
-          # "mailserver" hostname defined above
           docker run --rm --network changedet-network test-changedetectionio bash -c 'cd changedetectionio;pytest tests/smtp/test_notification_smtp.py'
 
-      # @todo Add a test via playwright/puppeteer
-      # squid with auth is tested in run_proxy_tests.sh -> tests/proxy_list/test_select_custom_proxy.py
-      - name: Test proxy squid style interaction
+  # Proxy tests
+  proxy-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
+        run: |
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Spin up services
+        run: |
+          docker network create changedet-network
+          docker run --network changedet-network -d --hostname selenium -p 4444:4444 --rm --shm-size="2g" selenium/standalone-chrome:4
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser --hostname sockpuppetbrowser --rm -p 3000:3000 dgtlmoon/sockpuppetbrowser:latest
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser-custom-url --hostname sockpuppetbrowser-custom-url -p 3001:3000 --rm dgtlmoon/sockpuppetbrowser:latest
+
+      - name: Test proxy Squid style interaction
         run: |
           cd changedetectionio
           ./run_proxy_tests.sh
+          docker ps
           cd ..
 
       - name: Test proxy SOCKS5 style interaction
@@ -158,28 +311,65 @@ jobs:
           ./run_socks_proxy_tests.sh
           cd ..
 
+  # Custom browser URL tests
+  custom-browser-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
+        run: |
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Spin up ancillary services
+        run: |
+          docker network create changedet-network
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser --hostname sockpuppetbrowser --rm -p 3000:3000 dgtlmoon/sockpuppetbrowser:latest
+          docker run --network changedet-network -d -e "LOG_LEVEL=TRACE" --cap-add=SYS_ADMIN --name sockpuppetbrowser-custom-url --hostname sockpuppetbrowser-custom-url -p 3001:3000 --rm dgtlmoon/sockpuppetbrowser:latest
+
       - name: Test custom browser URL
         run: |
           cd changedetectionio
           ./run_custom_browser_url_tests.sh
-          cd ..
 
-      - name: Test changedetection.io container starts+runs basically without error
+  # Container startup tests
+  container-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
         run: |
-          docker run --name test-changedetectionio -p 5556:5000  -d test-changedetectionio
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Test container starts+runs basically without error
+        run: |
+          docker run --name test-changedetectionio -p 5556:5000 -d test-changedetectionio
           sleep 3
-          # Should return 0 (no error) when grep finds it
-          curl --retry-connrefused --retry 6  -s http://localhost:5556 |grep -q checkbox-uuid
-          
-          # and IPv6
-          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 came from STDOUT
+          curl --retry-connrefused --retry 6 -s http://localhost:5556 |grep -q checkbox-uuid
+          curl --retry-connrefused --retry 6 -s -g -6 "http://[::1]:5556"|grep -q checkbox-uuid
           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
@@ -187,102 +377,66 @@ jobs:
           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
+  # Signal tests
+  signal-tests:
+    runs-on: ubuntu-latest
+    needs: build
+    timeout-minutes: 10
+    env:
+      PYTHON_VERSION: ${{ inputs.python-version }}
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Download Docker image artifact
+        uses: actions/download-artifact@v6
+        with:
+          name: test-changedetectionio-${{ env.PYTHON_VERSION }}
+          path: /tmp
+
+      - name: Load Docker image
+        run: |
+          docker load -i /tmp/test-changedetectionio.tar
+
+      - name: Test SIGTERM and SIGINT signal shutdown
         run: |
-          
           echo SIGINT Shutdown request test
           docker run --name sig-test -d test-changedetectionio
           sleep 3
           echo ">>> Sending SIGINT to sig-test container"
           docker kill --signal=SIGINT sig-test
           sleep 3
-          # invert the check (it should be not 0/not running)
           docker ps
-          # check signal catch(STDERR) log. Because of
-          # changedetectionio/__init__.py: logger.add(sys.stderr, level=logger_level)
           docker logs sig-test 2>&1 | grep 'Shutdown: Got Signal - SIGINT' || exit 1
           test -z "`docker ps|grep sig-test`"
-          if [ $? -ne 0 ]
-          then
+          if [ $? -ne 0 ]; then
             echo "Looks like container was running when it shouldnt be"
             docker ps
             exit 1
           fi
-          
-          # @todo - scan the container log to see the right "graceful shutdown" text exists 
           docker rm sig-test
-          
+
           echo SIGTERM Shutdown request test
           docker run --name sig-test -d test-changedetectionio
           sleep 3
           echo ">>> Sending SIGTERM to sig-test container"
           docker kill --signal=SIGTERM sig-test
           sleep 3
-          # invert the check (it should be not 0/not running)
           docker ps
-          # check signal catch(STDERR) log. Because of
-          # changedetectionio/__init__.py: logger.add(sys.stderr, level=logger_level)
           docker logs sig-test 2>&1 | grep 'Shutdown: Got Signal - SIGTERM' || exit 1
           test -z "`docker ps|grep sig-test`"
-          if [ $? -ne 0 ]
-          then
+          if [ $? -ne 0 ]; then
             echo "Looks like container was running when it shouldnt be"
             docker ps
             exit 1
           fi
-          
-          # @todo - scan the container log to see the right "graceful shutdown" text exists           
           docker rm sig-test
-
-      - name: Dump container log
-        if: always()
-        run: |
-          mkdir output-logs
-          docker logs test-cdio-basic-tests > output-logs/test-cdio-basic-tests-stdout-${{ env.PYTHON_VERSION }}.txt
-          docker logs test-cdio-basic-tests 2> output-logs/test-cdio-basic-tests-stderr-${{ env.PYTHON_VERSION }}.txt
-
-      - name: Extract and display memory test report
-        if: always()
-        run: |
-          # Extract test-memory.log from the container
-          echo "Extracting test-memory.log from container..."
-          docker cp test-cdio-basic-tests:/app/changedetectionio/test-memory.log output-logs/test-memory-${{ env.PYTHON_VERSION }}.log || echo "test-memory.log not found in container"
-
-          # Display the memory log contents for immediate visibility in workflow output
-          echo "=== Top 10 Highest Peak Memory Tests ==="
-          if [ -f output-logs/test-memory-${{ env.PYTHON_VERSION }}.log ]; then
-            # Sort by peak memory value (extract number before MB and sort numerically, reverse order)
-            grep "Peak memory:" output-logs/test-memory-${{ env.PYTHON_VERSION }}.log | \
-              sed 's/.*Peak memory: //' | \
-              paste -d'|' - <(grep "Peak memory:" output-logs/test-memory-${{ env.PYTHON_VERSION }}.log) | \
-              sort -t'|' -k1 -nr | \
-              cut -d'|' -f2 | \
-              head -10
-            echo ""
-            echo "=== Full Memory Test Report ==="
-            cat output-logs/test-memory-${{ env.PYTHON_VERSION }}.log
-          else
-            echo "No memory log available"
-          fi
-
-      - 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: .

.gitignore

@@ -21,6 +21,7 @@ venv/
 # IDEs
 .idea
 .vscode/settings.json
+*~
 
 # Datastore files
 datastore/

Dockerfile

@@ -34,23 +34,27 @@ ENV OPENSSL_LIB_DIR="/usr/lib/arm-linux-gnueabihf"
 ENV OPENSSL_INCLUDE_DIR="/usr/include/openssl"
 # Additional environment variables for cryptography Rust build
 ENV CRYPTOGRAPHY_DONT_BUILD_RUST=1
-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
+RUN --mount=type=cache,id=pip,sharing=locked,target=/tmp/pip-cache \
+  pip install \
+  --prefer-binary \
+  --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 --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."
+RUN --mount=type=cache,id=pip,sharing=locked,target=/tmp/pip-cache \
+  pip install \
+  --prefer-binary \
+  --cache-dir=/tmp/pip-cache \
+  --target=/dependencies \
+  playwright~=1.56.0 \
+  || 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

README-pip.md

@@ -14,7 +14,7 @@ Ideal for monitoring price changes, content edits, conditional changes and more.
 [<img src="https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/screenshot.png" style="max-width:100%;" alt="Self-hosted web page change monitoring, list of websites with changes"  title="Self-hosted web page change monitoring, list of websites with changes"  />](https://changedetection.io)
 
 
-[**Don't have time? Try our extremely affordable subscription use our proxies and support!**](https://changedetection.io) 
+[**Don't have time? Try our extremely affordable subscription use our proxies and support!**](https://changedetection.io)
 
 
 
@@ -31,7 +31,7 @@ Available when connected to a <a href="https://github.com/dgtlmoon/changedetecti
 
 ### Perform interactive browser steps
 
-Fill in text boxes, click buttons and more, setup your changedetection scenario. 
+Fill in text boxes, click buttons and more, setup your changedetection scenario.
 
 Using the **Browser Steps** configuration, add basic steps before performing change detection, such as logging into websites, adding a product to a cart, accept cookie logins, entering dates and refining searches.
 
@@ -54,7 +54,7 @@ Requires Playwright to be enabled.
 - Know when your favourite whiskey is on sale, or other special deals are announced before anyone else
 - COVID related news from government websites
 - University/organisation news from their website
-- Detect and monitor changes in JSON API responses 
+- Detect and monitor changes in JSON API responses
 - JSON API monitoring and alerting
 - Changes in legal and other documents
 - Trigger API calls via notifications when text appears on a website
@@ -86,7 +86,7 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
 
 We [recommend and use Bright Data](https://brightdata.grsm.io/n0r16zf7eivq) global proxy services, Bright Data will match any first deposit up to $100 using our signup link.
 
-[Oxylabs](https://oxylabs.go2cloud.org/SH2d) is also an excellent proxy provider and well worth using, they offer Residental, ISP, Rotating and many other proxy types to suit your project. 
+[Oxylabs](https://oxylabs.go2cloud.org/SH2d) is also an excellent proxy provider and well worth using, they offer Residential, ISP, Rotating and many other proxy types to suit your project.
 
 Please :star: star :star: this project and help it grow! https://github.com/dgtlmoon/changedetection.io/
 
@@ -106,4 +106,3 @@ $ changedetection.io -d /path/to/empty/data/dir -p 5000
 Then visit http://127.0.0.1:5000 , You should now be able to access the UI.
 
 See https://changedetection.io for more information.
-

changedetectionio/PLUGIN_README.md

@@ -64,7 +64,7 @@ def count_words_in_history(watch):
             return 0
             
         latest_key = list(watch.history.keys())[-1]
-        latest_content = watch.get_history_snapshot(latest_key)
+        latest_content = watch.get_history_snapshot(timestamp=latest_key)
         return len(latest_content.split())
     except Exception as e:
         logger.error(f"Error counting words: {str(e)}")

changedetectionio/__init__.py

@@ -1,8 +1,8 @@
 #!/usr/bin/env python3
 
 # Read more https://github.com/dgtlmoon/changedetection.io/wiki
-
-__version__ = '0.50.29'
+# Semver means never use .01, or 00. Should be .1.
+__version__ = '0.51.4'
 
 from changedetectionio.strtobool import strtobool
 from json.decoder import JSONDecodeError
@@ -74,6 +74,12 @@ def main():
 
     datastore_path = None
     do_cleanup = False
+    # Optional URL to watch since start
+    default_url = None
+    # Set a default logger level
+    logger_level = 'DEBUG'
+    include_default_watches = True
+
     host = os.environ.get("LISTEN_HOST", "0.0.0.0").strip()
     port = int(os.environ.get('PORT', 5000))
     ssl_mode = False
@@ -87,15 +93,13 @@ def main():
         datastore_path = os.path.join(os.getcwd(), "../datastore")
 
     try:
-        opts, args = getopt.getopt(sys.argv[1:], "6Ccsd:h:p:l:", "port")
+        opts, args = getopt.getopt(sys.argv[1:], "6Ccsd:h:p:l:u:", "port")
     except getopt.GetoptError:
-        print('backend.py -s SSL enable -h [host] -p [port] -d [datastore path] -l [debug level - TRACE, DEBUG(default), INFO, SUCCESS, WARNING, ERROR, CRITICAL]')
+        print('backend.py -s SSL enable -h [host] -p [port] -d [datastore path] -u [default URL to watch] -l [debug level - TRACE, DEBUG(default), INFO, SUCCESS, WARNING, ERROR, CRITICAL]')
         sys.exit(2)
 
     create_datastore_dir = False
 
-    # Set a default logger level
-    logger_level = 'DEBUG'
     # Set a logger level via shell env variable
     # Used: Dockerfile for CICD
     # To set logger level for pytest, see the app function in tests/conftest.py
@@ -116,6 +120,10 @@ def main():
         if opt == '-d':
             datastore_path = arg
 
+        if opt == '-u':
+            default_url = arg
+            include_default_watches = False
+
         # Cleanup (remove text files that arent in the index)
         if opt == '-c':
             do_cleanup = True
@@ -172,13 +180,16 @@ def main():
             sys.exit(2)
 
     try:
-        datastore = store.ChangeDetectionStore(datastore_path=app_config['datastore_path'], version_tag=__version__)
+        datastore = store.ChangeDetectionStore(datastore_path=app_config['datastore_path'], version_tag=__version__, include_default_watches=include_default_watches)
     except JSONDecodeError as e:
         # Dont' start if the JSON DB looks corrupt
         logger.critical(f"ERROR: JSON DB or Proxy List JSON at '{app_config['datastore_path']}' appears to be corrupt, aborting.")
         logger.critical(str(e))
         return
 
+    if default_url:
+        datastore.add_watch(url = default_url)
+
     app = changedetection_app(app_config, datastore)
 
     # Get the SocketIO instance from the Flask app (created in flask_app.py)

changedetectionio/api/Import.py

@@ -1,9 +1,22 @@
-import os
 from changedetectionio.strtobool import strtobool
 from flask_restful import abort, Resource
 from flask import request
-import validators
+from functools import wraps
 from . import auth, validate_openapi_request
+from ..validate_url import is_safe_valid_url
+
+
+def default_content_type(content_type='text/plain'):
+    """Decorator to set a default Content-Type header if none is provided."""
+    def decorator(f):
+        @wraps(f)
+        def wrapper(*args, **kwargs):
+            if not request.content_type:
+                # Set default content type in the request environment
+                request.environ['CONTENT_TYPE'] = content_type
+            return f(*args, **kwargs)
+        return wrapper
+    return decorator
 
 
 class Import(Resource):
@@ -12,6 +25,7 @@ class Import(Resource):
         self.datastore = kwargs['datastore']
 
     @auth.check_token
+    @default_content_type('text/plain') #3547 #3542
     @validate_openapi_request('importWatches')
     def post(self):
         """Import a list of watched URLs."""
@@ -35,14 +49,13 @@ class Import(Resource):
 
         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):
+            if not is_safe_valid_url(url):
                 return f"Invalid or unsupported URL - {url}", 400
 
             if dedupe and self.datastore.url_exists(url):

changedetectionio/api/Watch.py

@@ -1,12 +1,12 @@
 import os
-from changedetectionio.strtobool import strtobool
+
+from changedetectionio.validate_url import is_safe_valid_url
 
 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, send_from_directory
-import validators
 from . import auth
 import copy
 
@@ -121,6 +121,10 @@ class Watch(Resource):
         if validation_error:
             return validation_error, 400
 
+        # XSS etc protection
+        if request.json.get('url') and not is_safe_valid_url(request.json.get('url')):
+            return "Invalid URL", 400
+
         watch.update(request.json)
 
         return "OK", 200
@@ -171,7 +175,7 @@ class WatchSingleHistory(Resource):
                 response = make_response("No content found", 404)
                 response.mimetype = "text/plain"
         else:
-            content = watch.get_history_snapshot(timestamp)
+            content = watch.get_history_snapshot(timestamp=timestamp)
             response = make_response(content, 200)
             response.mimetype = "text/plain"
 
@@ -226,9 +230,7 @@ class CreateWatch(Resource):
         json_data = request.get_json()
         url = json_data['url'].strip()
 
-        # If hosts that only contain alphanumerics are allowed ("localhost" for example)
-        allow_simplehost = not strtobool(os.getenv('BLOCK_SIMPLEHOSTS', 'False'))
-        if not validators.url(url, simple_host=allow_simplehost):
+        if not is_safe_valid_url(url):
             return "Invalid or unsupported URL", 400
 
         if json_data.get('proxy'):

changedetectionio/api/__init__.py

@@ -41,7 +41,7 @@ def get_openapi_spec():
         # Possibly for pip3 packages
         spec_path = os.path.join(os.path.dirname(__file__), '../docs/api-spec.yaml')
 
-    with open(spec_path, 'r') as f:
+    with open(spec_path, 'r', encoding='utf-8') as f:
         spec_dict = yaml.safe_load(f)
     _openapi_spec = OpenAPI.from_dict(spec_dict)
     return _openapi_spec

changedetectionio/api/api_schema.py

@@ -96,7 +96,10 @@ def build_watch_json_schema(d):
                                                            "enum": ["html_requests", "html_webdriver"]
                                                            })
 
-
+    schema['properties']['processor'] = {"anyOf": [
+        {"type": "string", "enum": ["restock_diff", "text_json_diff"]},
+        {"type": "null"}
+    ]}
 
     # All headers must be key/value type dict
     schema['properties']['headers'] = {

changedetectionio/async_update_worker.py

@@ -353,12 +353,15 @@ async def async_update_worker(worker_id, q, notification_q, app, datastore):
                 count = watch.get('check_count', 0) + 1
 
                 # Always record page title (used in notifications, and can change even when the content is the same)
-                try:
-                    page_title = html_tools.extract_title(data=update_handler.fetcher.content)
-                    logger.debug(f"UUID: {uuid} Page <title> is '{page_title}'")
-                    datastore.update_watch(uuid=uuid, update_obj={'page_title': page_title})
-                except Exception as e:
-                    logger.warning(f"UUID: {uuid} Exception when extracting <title> - {str(e)}")
+                if update_obj.get('content-type') and 'html' in update_obj.get('content-type'):
+                    try:
+                        page_title = html_tools.extract_title(data=update_handler.fetcher.content)
+                        if page_title:
+                            page_title = page_title.strip()[:2000]
+                            logger.debug(f"UUID: {uuid} Page <title> is '{page_title}'")
+                            datastore.update_watch(uuid=uuid, update_obj={'page_title': page_title})
+                    except Exception as e:
+                        logger.warning(f"UUID: {uuid} Exception when extracting <title> - {str(e)}")
 
                 # Record server header
                 try:

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -439,7 +439,7 @@ class browsersteps_live_ui(steppable_browser_interface):
             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()
+        xpath_element_js = importlib.resources.files("changedetectionio.content_fetchers.res").joinpath('xpath_element_scraper.js').read_text(encoding="utf-8")
 
         now = time.time()
         await self.page.wait_for_timeout(1 * 1000)

changedetectionio/blueprint/rss/__init__.py

@@ -1 +1,27 @@
-RSS_FORMAT_TYPES = [('plaintext', 'Plain text'), ('html', 'HTML Color')]
+from copy import deepcopy
+from loguru import logger
+
+from changedetectionio.model import USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
+from changedetectionio.notification import valid_notification_formats
+
+RSS_CONTENT_FORMAT_DEFAULT = 'text'
+
+# Some stuff not related
+RSS_FORMAT_TYPES = deepcopy(valid_notification_formats)
+if RSS_FORMAT_TYPES.get('markdown'):
+    del RSS_FORMAT_TYPES['markdown']
+
+if RSS_FORMAT_TYPES.get(USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH):
+    del RSS_FORMAT_TYPES[USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH]
+
+if not RSS_FORMAT_TYPES.get(RSS_CONTENT_FORMAT_DEFAULT):
+    logger.critical(f"RSS_CONTENT_FORMAT_DEFAULT not in the acceptable list {RSS_CONTENT_FORMAT_DEFAULT}")
+
+RSS_TEMPLATE_TYPE_OPTIONS = {'system_default': 'System default', 'notification_body': 'Notification body'}
+
+# @note: We use <pre> because nearly all RSS readers render only HTML (Thunderbird for example cant do just plaintext)
+RSS_TEMPLATE_PLAINTEXT_DEFAULT = "<pre>{{watch_label}} had a change.\n\n{{diff}}\n</pre>"
+
+# @todo add some [edit]/[history]/[goto] etc links
+# @todo need {{watch_edit_link}} + delete + history link token
+RSS_TEMPLATE_HTML_DEFAULT = "<html><body>\n<h4><a href=\"{{watch_url}}\">{{watch_label}}</a></h4>\n<p>{{diff}}</p>\n</body></html>\n"

changedetectionio/blueprint/rss/_util.py

@@ -0,0 +1,156 @@
+"""
+Utility functions for RSS feed generation.
+"""
+
+from changedetectionio.notification.handler import process_notification
+from changedetectionio.notification_service import NotificationContextData, _check_cascading_vars
+from loguru import logger
+import datetime
+import pytz
+import re
+
+
+BAD_CHARS_REGEX = r'[\x00-\x08\x0B\x0C\x0E-\x1F]'
+
+
+def scan_invalid_chars_in_rss(content):
+    """
+    Scan for invalid characters in RSS content.
+    Returns True if invalid characters are found.
+    """
+    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):
+    """
+    Remove invalid characters from RSS content.
+    """
+    cleaned = re.sub(BAD_CHARS_REGEX, '', content)
+    return cleaned
+
+
+def generate_watch_guid(watch, timestamp):
+    """
+    Generate a unique GUID for a watch RSS entry.
+
+    Args:
+        watch: The watch object
+        timestamp: The timestamp of the specific change this entry represents
+    """
+    return f"{watch['uuid']}/{timestamp}"
+
+
+def validate_rss_token(datastore, request):
+    """
+    Validate the RSS access token from the request.
+
+    Returns:
+        tuple: (is_valid, error_response) where error_response is None if valid
+    """
+    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 False, ("Access denied, bad token", 403)
+
+    return True, None
+
+
+def get_rss_template(datastore, watch, rss_content_format, default_html, default_plaintext):
+    """Get the appropriate template for RSS content."""
+    if datastore.data['settings']['application'].get('rss_template_type') == 'notification_body':
+        return _check_cascading_vars(datastore=datastore, var_name='notification_body', watch=watch)
+
+    override = datastore.data['settings']['application'].get('rss_template_override')
+    if override and override.strip():
+        return override
+    elif 'text' in rss_content_format:
+        return default_plaintext
+    else:
+        return default_html
+
+
+def get_watch_label(datastore, watch):
+    """Get the label for a watch based on settings."""
+    if datastore.data['settings']['application']['ui'].get('use_page_title_in_list') or watch.get('use_page_title_in_list'):
+        return watch.label
+    else:
+        return watch.get('url')
+
+
+def add_watch_categories(fe, watch, datastore):
+    """Add category tags to a feed entry based on watch tags."""
+    for tag_uuid in watch.get('tags', []):
+        tag = datastore.data['settings']['application'].get('tags', {}).get(tag_uuid)
+        if tag and tag.get('title'):
+            fe.category(term=tag.get('title'))
+
+
+def build_notification_context(watch, timestamp_from, timestamp_to, watch_label,
+                               n_body_template, rss_content_format):
+    """Build the notification context object."""
+    return NotificationContextData(initial_data={
+        'notification_urls': ['null://just-sending-a-null-test-for-the-render-in-RSS'],
+        'notification_body': n_body_template,
+        'timestamp_to': timestamp_to,
+        'timestamp_from': timestamp_from,
+        'watch_label': watch_label,
+        'notification_format': rss_content_format
+    })
+
+
+def render_notification(n_object, notification_service, watch, datastore,
+                       date_index_from=None, date_index_to=None):
+    """Process and render the notification content."""
+    kwargs = {'n_object': n_object, 'watch': watch}
+
+    if date_index_from is not None and date_index_to is not None:
+        kwargs['date_index_from'] = date_index_from
+        kwargs['date_index_to'] = date_index_to
+
+    n_object = notification_service.queue_notification_for_watch(**kwargs)
+    n_object['watch_mime_type'] = None
+
+    res = process_notification(n_object=n_object, datastore=datastore)
+    return res[0]
+
+
+def populate_feed_entry(fe, watch, content, guid, timestamp, link=None, title_suffix=None):
+    """Populate a feed entry with content and metadata."""
+    watch_label = watch.get('url')  # Already determined by caller
+
+    # Set link
+    if link:
+        fe.link(link=link)
+
+    # Set title
+    if title_suffix:
+        fe.title(title=f"{watch_label} - {title_suffix}")
+    else:
+        fe.title(title=watch_label)
+
+    # Clean and set content
+    if scan_invalid_chars_in_rss(content):
+        content = clean_entry_content(content)
+    fe.content(content=content, type='CDATA')
+
+    # Set GUID
+    fe.guid(guid, permalink=False)
+
+    # Set pubDate using the timestamp of this specific change
+    dt = datetime.datetime.fromtimestamp(int(timestamp))
+    dt = dt.replace(tzinfo=pytz.UTC)
+    fe.pubDate(dt)
+

changedetectionio/blueprint/rss/blueprint.py

@@ -1,155 +1,26 @@
 
-from changedetectionio.jinja2_custom import render as jinja_render
-from changedetectionio.notification.handler import apply_service_tweaks
 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
+from flask import Blueprint
 
-
-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
+from . import tag as tag_routes
+from . import main_feed
+from . import single_watch
 
 def construct_blueprint(datastore: ChangeDetectionStore):
+    """
+    Construct and configure the RSS blueprint with all routes.
+
+    Args:
+        datastore: The ChangeDetectionStore instance
+
+    Returns:
+        The configured Flask blueprint
+    """
     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)
-
-                # Same logic as watch-overview.html
-                if datastore.data['settings']['application']['ui'].get('use_page_title_in_list') or watch.get('use_page_title_in_list'):
-                    watch_label = watch.label
-                else:
-                    watch_label = watch.get('url')
-
-                fe.title(title=watch_label)
-                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>"
-                                                 )
-
-
-                    requested_output_format = 'htmlcolor' if html_colour_enable else 'html'
-                    html_diff = apply_service_tweaks(url='', n_body=html_diff, n_title=None, requested_output_format=requested_output_format)
-
-                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_label, 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
+    # Register all route modules
+    main_feed.construct_main_feed_routes(rss_blueprint, datastore)
+    single_watch.construct_single_watch_routes(rss_blueprint, datastore)
+    tag_routes.construct_tag_routes(rss_blueprint, datastore)
 
     return rss_blueprint

changedetectionio/blueprint/rss/main_feed.py

@@ -0,0 +1,105 @@
+from flask import make_response, request, url_for, redirect
+
+
+
+def construct_main_feed_routes(rss_blueprint, datastore):
+    """
+    Construct the main RSS feed routes.
+
+    Args:
+        rss_blueprint: The Flask blueprint to add routes to
+        datastore: The ChangeDetectionStore instance
+    """
+
+    # 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():
+        from feedgen.feed import FeedGenerator
+        from loguru import logger
+        import time
+
+        from . import RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT
+        from ._util import (validate_rss_token, generate_watch_guid, get_rss_template,
+                           get_watch_label, build_notification_context, render_notification,
+                           populate_feed_entry, add_watch_categories)
+        from ...notification_service import NotificationService
+
+        now = time.time()
+
+        # Validate token
+        is_valid, error = validate_rss_token(datastore, request)
+        if not is_valid:
+            return error
+
+        rss_content_format = datastore.data['settings']['application'].get('rss_content_format')
+
+        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
+            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')
+        notification_service = NotificationService(datastore=datastore, notification_q=False)
+
+        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)
+                watch_label = get_watch_label(datastore, watch)
+                timestamp_to = dates[-1]
+                timestamp_from = dates[-2]
+                guid = generate_watch_guid(watch, timestamp_to)
+                # 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)}
+
+                # Get template and build notification context
+                n_body_template = get_rss_template(datastore, watch, rss_content_format,
+                                                   RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT)
+
+                n_object = build_notification_context(watch, timestamp_from, timestamp_to,
+                                                     watch_label, n_body_template, rss_content_format)
+
+                # Render notification
+                res = render_notification(n_object, notification_service, watch, datastore)
+
+                # Create and populate feed entry
+                fe = fg.add_entry()
+                populate_feed_entry(fe, watch, res['body'], guid, timestamp_to, link=diff_link)
+                fe.title(title=watch_label)  # Override title to not include suffix
+                add_watch_categories(fe, watch, datastore)
+
+        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

changedetectionio/blueprint/rss/single_watch.py

@@ -0,0 +1,115 @@
+
+
+def construct_single_watch_routes(rss_blueprint, datastore):
+    """
+    Construct RSS feed routes for single watches.
+
+    Args:
+        rss_blueprint: The Flask blueprint to add routes to
+        datastore: The ChangeDetectionStore instance
+    """
+
+    @rss_blueprint.route("/watch/<string:uuid>", methods=['GET'])
+    def rss_single_watch(uuid):
+        import time
+
+        from flask import make_response, request
+        from feedgen.feed import FeedGenerator
+        from loguru import logger
+
+        from . import RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT
+        from ._util import (validate_rss_token, get_rss_template, get_watch_label,
+                           build_notification_context, render_notification,
+                           populate_feed_entry, add_watch_categories)
+        from ...notification_service import NotificationService
+
+        """
+        Display the most recent changes for a single watch as RSS feed.
+        Returns RSS XML with multiple entries showing diffs between consecutive snapshots.
+        The number of entries is controlled by the rss_diff_length setting.
+        """
+        now = time.time()
+
+        # Validate token
+        is_valid, error = validate_rss_token(datastore, request)
+        if not is_valid:
+            return error
+
+        rss_content_format = datastore.data['settings']['application'].get('rss_content_format')
+
+        # Get the watch by UUID
+        watch = datastore.data['watching'].get(uuid)
+        if not watch:
+            return f"Watch with UUID {uuid} not found", 404
+
+        # Check if watch has at least 2 history snapshots
+        dates = list(watch.history.keys())
+        if len(dates) < 2:
+            return f"Watch {uuid} does not have enough history snapshots to show changes (need at least 2)", 400
+
+        # Add uuid to watch for proper functioning
+        watch['uuid'] = uuid
+
+        # Get the number of diffs to include (default: 5)
+        rss_diff_length = datastore.data['settings']['application'].get('rss_diff_length', 5)
+
+        # Calculate how many diffs we can actually show (limited by available history)
+        # We need at least 2 snapshots to create 1 diff
+        max_possible_diffs = len(dates) - 1
+        num_diffs = min(rss_diff_length, max_possible_diffs) if rss_diff_length > 0 else max_possible_diffs
+
+        # Create RSS feed
+        fg = FeedGenerator()
+
+        # Set title: use "label (url)" if label differs from url, otherwise just url
+        watch_url = watch.get('url', '')
+        watch_label = get_watch_label(datastore, watch)
+
+        if watch_label != watch_url:
+            feed_title = f'changedetection.io - {watch_label} ({watch_url})'
+        else:
+            feed_title = f'changedetection.io - {watch_url}'
+
+        fg.title(feed_title)
+        fg.description('Changes')
+        fg.link(href='https://changedetection.io')
+
+        # Loop through history and create RSS entries for each diff
+        # Add entries in reverse order because feedgen reverses them
+        # This way, the newest change appears first in the final RSS
+
+        notification_service = NotificationService(datastore=datastore, notification_q=False)
+        for i in range(num_diffs - 1, -1, -1):
+            # Calculate indices for this diff (working backwards from newest)
+            # i=0: compare dates[-2] to dates[-1] (most recent change)
+            # i=1: compare dates[-3] to dates[-2] (previous change)
+            # etc.
+            date_index_to = -(i + 1)
+            date_index_from = -(i + 2)
+            timestamp_to = dates[date_index_to]
+            timestamp_from = dates[date_index_from]
+
+            # Get template and build notification context
+            n_body_template = get_rss_template(datastore, watch, rss_content_format,
+                                               RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT)
+
+            n_object = build_notification_context(watch, timestamp_from, timestamp_to,
+                                                 watch_label, n_body_template, rss_content_format)
+
+            # Render notification with date indices
+            res = render_notification(n_object, notification_service, watch, datastore,
+                                     date_index_from, date_index_to)
+
+            # Create and populate feed entry
+            guid = f"{watch['uuid']}/{timestamp_to}"
+            fe = fg.add_entry()
+            title_suffix = f"Change @ {res['original_context']['change_datetime']}"
+            populate_feed_entry(fe, watch, res.get('body', ''), guid, timestamp_to,
+                              link={'href': watch.get('url')}, title_suffix=title_suffix)
+            add_watch_categories(fe, watch, datastore)
+
+        response = make_response(fg.rss_str())
+        response.headers.set('Content-Type', 'application/rss+xml;charset=utf-8')
+        logger.debug(f"RSS Single watch built in {time.time()-now:.2f}s")
+
+        return response

changedetectionio/blueprint/rss/tag.py

@@ -0,0 +1,98 @@
+def construct_tag_routes(rss_blueprint, datastore):
+    """
+    Construct RSS feed routes for tags.
+
+    Args:
+        rss_blueprint: The Flask blueprint to add routes to
+        datastore: The ChangeDetectionStore instance
+    """
+
+    @rss_blueprint.route("/tag/<string:tag_uuid>", methods=['GET'])
+    def rss_tag_feed(tag_uuid):
+
+        from flask import make_response, request, url_for
+        from feedgen.feed import FeedGenerator
+
+        from . import RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT
+        from ._util import (validate_rss_token, generate_watch_guid, get_rss_template,
+                           get_watch_label, build_notification_context, render_notification,
+                           populate_feed_entry, add_watch_categories)
+        from ...notification_service import NotificationService
+
+        """
+        Display an RSS feed for all unviewed watches that belong to a specific tag.
+        Returns RSS XML with entries for each unviewed watch with sufficient history.
+        """
+        # Validate token
+        is_valid, error = validate_rss_token(datastore, request)
+        if not is_valid:
+            return error
+
+        rss_content_format = datastore.data['settings']['application'].get('rss_content_format')
+
+        # Verify tag exists
+        tag = datastore.data['settings']['application'].get('tags', {}).get(tag_uuid)
+        if not tag:
+            return f"Tag with UUID {tag_uuid} not found", 404
+
+        tag_title = tag.get('title', 'Unknown Tag')
+
+        # Create RSS feed
+        fg = FeedGenerator()
+        fg.title(f'changedetection.io - {tag_title}')
+        fg.description(f'Changes for watches tagged with {tag_title}')
+        fg.link(href='https://changedetection.io')
+        notification_service = NotificationService(datastore=datastore, notification_q=False)
+        # Find all watches with this tag
+        for uuid, watch in datastore.data['watching'].items():
+            #@todo  This is wrong, it needs to sort by most recently changed and then limit it  datastore.data['watching'].items().sorted(?)
+            # So get all watches in this tag then sort
+
+            # Skip if watch doesn't have this tag
+            if tag_uuid not in watch.get('tags', []):
+                continue
+
+            # Skip muted watches if configured
+            if datastore.data['settings']['application'].get('rss_hide_muted_watches') and watch.get('notification_muted'):
+                continue
+
+            # Check if watch has at least 2 history snapshots
+            dates = list(watch.history.keys())
+            if len(dates) < 2:
+                continue
+
+            # Only include unviewed watches
+            if not watch.viewed:
+                # Add uuid to watch for proper functioning
+                watch['uuid'] = uuid
+
+                # Include a link to the diff page
+                diff_link = {'href': url_for('ui.ui_views.diff_history_page', uuid=watch['uuid'], _external=True)}
+
+                # Get watch label
+                watch_label = get_watch_label(datastore, watch)
+
+                # Get template and build notification context
+                timestamp_to = dates[-1]
+                timestamp_from = dates[-2]
+
+                # Generate GUID for this entry
+                guid = generate_watch_guid(watch, timestamp_to)
+                n_body_template = get_rss_template(datastore, watch, rss_content_format,
+                                                   RSS_TEMPLATE_HTML_DEFAULT, RSS_TEMPLATE_PLAINTEXT_DEFAULT)
+
+                n_object = build_notification_context(watch, timestamp_from, timestamp_to,
+                                                     watch_label, n_body_template, rss_content_format)
+
+                # Render notification
+                res = render_notification(n_object, notification_service, watch, datastore)
+
+                # Create and populate feed entry
+                fe = fg.add_entry()
+                title_suffix = f"Change @ {res['original_context']['change_datetime']}"
+                populate_feed_entry(fe, watch, res['body'], guid, timestamp_to, link=diff_link, title_suffix=title_suffix)
+                add_watch_categories(fe, watch, datastore)
+
+        response = make_response(fg.rss_str())
+        response.headers.set('Content-Type', 'application/rss+xml;charset=utf-8')
+        return response

changedetectionio/blueprint/settings/templates/settings.html

@@ -2,7 +2,7 @@
 
 {% block content %}
 {% from '_helpers.html' import render_field, render_checkbox_field, render_button, render_time_schedule_form, render_ternary_field, render_fieldlist_with_inline_errors %}
-{% from '_common_fields.html' import render_common_settings_form %}
+{% from '_common_fields.html' import render_common_settings_form, show_token_placeholders %}
 <script>
     const notification_base_url="{{url_for('ui.ui_notification.ajax_callback_send_notification_test', mode="global-settings")}}";
 {% if emailprefix %}
@@ -24,6 +24,7 @@
             <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="#rss">RSS</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>
@@ -43,10 +44,6 @@
                                 </div>
                         </div>
                     </div>
-                    <div class="pure-control-group">
-                        {{ render_field(form.requests.form.jitter_seconds, class="jitter_seconds") }}
-                        <span class="pure-form-message-inline">Example - 3 seconds random jitter could trigger up to 3 seconds earlier or up to 3 seconds later</span>
-                    </div>
                     <div class="pure-control-group">
                         {{ render_field(form.application.form.filter_failure_notification_threshold_attempts, class="filter_failure_notification_threshold_attempts") }}
                         <span class="pure-form-message-inline">After this many consecutive times that the CSS/xPath filter is missing, send a notification
@@ -69,26 +66,13 @@
 
                     <div class="pure-control-group">
                         {{ render_checkbox_field(form.application.form.shared_diff_access, class="shared_diff_access") }}
-                        <span class="pure-form-message-inline">Allow access to view watch diff page when password is enabled (Good for sharing the diff page)
+                        <span class="pure-form-message-inline">Allow access to the watch change history page when password is enabled (Good for sharing the diff page)
                         </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 request returns no content, or the HTML does not contain any text, is this considered a change?</span>
                     </div>
-                    <div class="grey-form-border">
-                        <div class="pure-control-group">
-                            {{ render_checkbox_field(form.application.form.rss_hide_muted_watches) }}
-                        </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.rss_reader_mode) }}
-                            <span class="pure-form-message-inline">Transforms RSS/RDF feed watches into beautiful text only</span>
-                        </div>
-                    </div>
                 </fieldset>
             </div>
 
@@ -131,6 +115,10 @@
                     <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">
+                    {{ render_field(form.requests.form.jitter_seconds, class="jitter_seconds") }}
+                    <span class="pure-form-message-inline">Example - 3 seconds random jitter could trigger up to 3 seconds earlier or up to 3 seconds later</span>
+                </div>
                 <div class="pure-control-group">
                     {{ render_field(form.requests.form.timeout) }}
                     <span class="pure-form-message-inline">For regular plain requests (not chrome based), maximum number of seconds until timeout, 1-999.<br>
@@ -230,7 +218,37 @@ nav
                     </p>
                 </div>
             </div>
-            <div class="tab-pane-inner" id="timedate">
+            <div class="tab-pane-inner" id="rss">
+                <div class="pure-control-group">
+                    {{ render_checkbox_field(form.application.form.rss_hide_muted_watches) }}
+                </div>
+                <div class="pure-control-group">
+                    {{ render_field(form.application.form.rss_diff_length) }}
+                    <span class="pure-form-message-inline">Maximum number of history snapshots to include in the watch specific RSS feed.</span>
+                </div>
+                <div class="pure-control-group">
+                    {{ render_checkbox_field(form.application.form.rss_reader_mode) }}
+                    <span class="pure-form-message-inline">For watching other RSS feeds - When watching RSS/Atom feeds, convert them into clean text for better change detection.</span>
+                </div>
+                <div class="pure-control-group grey-form-border">
+                    <div class="pure-control-group">
+                        {{ render_field(form.application.form.rss_content_format) }}
+                        <span class="pure-form-message-inline">Does your reader support HTML? Set it here</span>
+                    </div>
+                    <div class="pure-control-group">
+                        {{ render_field(form.application.form.rss_template_type) }}
+                        <span class="pure-form-message-inline">'System default' for the same template for all items, or re-use your "Notification Body" as the template.</span>
+                    </div>
+                    <div>
+                       {{ render_field(form.application.form.rss_template_override) }}
+                        {{ show_token_placeholders(extra_notification_token_placeholder_info=extra_notification_token_placeholder_info, suffix="-rss") }}
+                    </div>
+                </div>
+                <br>
+
+
+            </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>
@@ -240,9 +258,7 @@ nav
                     <p>
                        {{ render_field(form.application.form.scheduler_timezone_default) }}
                         <datalist id="timezones" style="display: none;">
-                            {% for tz_name in available_timezones %}
-                                <option value="{{ tz_name }}">{{ tz_name }}</option>
-                            {% endfor %}
+                            {%- for timezone in available_timezones -%}<option value="{{ timezone }}">{{ timezone }}</option>{%- endfor -%}
                         </datalist>
                     </p>
                 </div>

changedetectionio/blueprint/tags/__init__.py

@@ -21,9 +21,10 @@ def construct_blueprint(datastore: ChangeDetectionStore):
         tag_count = Counter(tag for watch in datastore.data['watching'].values() if watch.get('tags') for tag in watch['tags'])
 
         output = render_template("groups-overview.html",
+                                 app_rss_token=datastore.data['settings']['application'].get('rss_access_token'),
                                  available_tags=sorted_tags,
                                  form=add_form,
-                                 tag_count=tag_count
+                                 tag_count=tag_count,
                                  )
 
         return output
@@ -149,9 +150,9 @@ def construct_blueprint(datastore: ChangeDetectionStore):
             included_content = template.render(**template_args)
 
         output = render_template("edit-tag.html",
-                                 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,
+                                 extra_tab_content=form.extra_tab_content() if form.extra_tab_content() else None,
+                                 settings_application=datastore.data['settings']['application'],
                                  **template_args
                                  )
 

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

@@ -52,6 +52,7 @@
                     <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>
                     <a class="pure-button pure-button-primary" href="{{ url_for('tags.unlink', uuid=uuid) }}" title="Keep the tag but unlink any watches">Unlink</a>
+                    <a href="{{ url_for('rss.rss_tag_feed', tag_uuid=uuid, token=app_rss_token)}}"><img alt="RSS Feed for this watch" style="padding-left: 1em;" src="{{url_for('static_content', group='images', filename='generic_feed-icon.svg')}}" height="15"></a>
                 </td>
             </tr>
             {% endfor %}

changedetectionio/blueprint/ui/__init__.py

@@ -76,14 +76,14 @@ def _handle_operations(op, uuids, datastore, worker_handler, update_q, queuedWat
 
     elif (op == 'notification-default'):
         from changedetectionio.notification import (
-            default_notification_format_for_watch
+            USE_SYSTEM_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
+                datastore.data['watching'][uuid]['notification_format'] = USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
         if emit_flash:
             flash(f"{len(uuids)} watches set to use default notification settings")
 

changedetectionio/blueprint/ui/edit.py

@@ -236,7 +236,7 @@ def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMe
 
             # Import the global plugin system
             from changedetectionio.pluggy_interface import collect_ui_edit_stats_extras
-            
+            app_rss_token = datastore.data['settings']['application'].get('rss_access_token'),
             template_args = {
                 'available_processors': processors.available_processors(),
                 'available_timezones': sorted(available_timezones()),
@@ -252,6 +252,11 @@ def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMe
                 'has_special_tag_options': _watch_has_tag_options_set(watch=watch),
                 'jq_support': jq_support,
                 'playwright_enabled': os.getenv('PLAYWRIGHT_DRIVER_URL', False),
+                'app_rss_token': app_rss_token,
+                'rss_uuid_feed' : {
+                    'label': watch.label,
+                    'url': url_for('rss.rss_single_watch', uuid=watch['uuid'], token=app_rss_token)
+                },
                 'settings_application': datastore.data['settings']['application'],
                 'system_has_playwright_configured': os.getenv('PLAYWRIGHT_DRIVER_URL'),
                 'system_has_webdriver_configured': os.getenv('WEBDRIVER_URL'),

changedetectionio/blueprint/ui/notification.py

@@ -2,7 +2,6 @@ from flask import Blueprint, request, make_response
 import random
 from loguru import logger
 
-from changedetectionio.notification_service import NotificationContextData
 from changedetectionio.store import ChangeDetectionStore
 from changedetectionio.auth_decorator import login_optionally_required
 
@@ -15,7 +14,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     @notification_blueprint.route("/notification/send-test/", methods=['POST'])
     @login_optionally_required
     def ajax_callback_send_notification_test(watch_uuid=None):
-
+        from changedetectionio.notification_service import NotificationContextData, set_basic_notification_vars
         # 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
@@ -39,11 +38,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
             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()
+        notification_urls = request.form.get('notification_urls','').strip().splitlines()
 
         if not notification_urls:
             logger.debug("Test notification - Trying by group/tag in the edit form if available")
@@ -81,6 +76,8 @@ def construct_blueprint(datastore: ChangeDetectionStore):
             # 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()
+            else:
+                n_object['notification_format'] = datastore.data['settings']['application'].get('notification_format')
 
             if 'notification_title' in request.form and request.form['notification_title'].strip():
                 n_object['notification_title'] = request.form.get('notification_title', '').strip()
@@ -97,7 +94,28 @@ def construct_blueprint(datastore: ChangeDetectionStore):
                 n_object['notification_body'] = "Test body"
 
             n_object['as_async'] = False
-            n_object.update(watch.extra_notification_token_values())
+
+            #  Same like in notification service, should be refactored
+            dates = list(watch.history.keys())
+            trigger_text = ''
+            snapshot_contents = ''
+
+            # 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(timestamp=dates[-2])
+                current_snapshot = watch.get_history_snapshot(timestamp=dates[-1])
+
+            n_object.update(set_basic_notification_vars(snapshot_contents=snapshot_contents,
+                                                        current_snapshot=current_snapshot,
+                                                        prev_snapshot=prev_snapshot,
+                                                        watch=watch,
+                                                        triggered_text=trigger_text,
+                                                        timestamp_changed=dates[-1] if dates else None))
+
+
             sent_obj = process_notification(n_object, datastore)
 
         except Exception as e:

changedetectionio/blueprint/ui/templates/edit.html

@@ -476,6 +476,7 @@ Math: {{ 1 + 1 }}") }}
                        class="pure-button button-error">Clear History</a>{% endif %}
                     <a href="{{url_for('ui.form_clone', uuid=uuid)}}"
                        class="pure-button">Clone &amp; Edit</a>
+                 <a href="{{ url_for('rss.rss_single_watch', uuid=uuid, token=app_rss_token)}}"><img alt="RSS Feed for this watch" style="padding: .5em 1em;" src="{{url_for('static_content', group='images', filename='generic_feed-icon.svg')}}" height="15"></a>
                 </div>
             </div>
         </form>

changedetectionio/blueprint/ui/views.py

@@ -47,7 +47,7 @@ def construct_blueprint(datastore: ChangeDetectionStore, update_q, queuedWatchMe
 
             try:
                 versions = list(watch.history.keys())
-                content = watch.get_history_snapshot(timestamp)
+                content = watch.get_history_snapshot(timestamp=timestamp)
 
                 triggered_line_numbers = html_tools.strip_ignore_text(content=content,
                                                                       wordlist=watch['trigger_text'],

changedetectionio/conditions/plugins/wordcount_plugin.py

@@ -14,7 +14,7 @@ def count_words_in_history(watch, incoming_text=None):
         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)
+            latest_content = watch.get_history_snapshot(timestamp=latest_key)
             return len(latest_content.split())
         return 0
     except Exception as e:

changedetectionio/content_fetchers/base.py

@@ -75,7 +75,6 @@ class Fetcher():
         self.screenshot = None
         self.xpath_data = None
         # Keep headers and status_code as they're small
-        logger.trace("Fetcher content cleared from memory")
 
     @abstractmethod
     def get_error(self):

changedetectionio/content_fetchers/playwright.py

@@ -139,7 +139,7 @@ class fetcher(Fetcher):
         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:
+        with open(destination, 'w', encoding='utf-8') as f:
             f.write(content)
 
     async def run(self,

changedetectionio/content_fetchers/requests.py

@@ -1,6 +1,7 @@
 from loguru import logger
 import hashlib
 import os
+import re
 import asyncio
 from changedetectionio import strtobool
 from changedetectionio.content_fetchers.exceptions import BrowserStepsInUnsupportedFetcher, EmptyReply, Non200ErrorCodeReceived
@@ -76,9 +77,22 @@ class fetcher(Fetcher):
         if not is_binary:
             # Don't run this for PDF (and requests identified as binary) takes a _long_ time
             if not r.headers.get('content-type') or not 'charset=' in r.headers.get('content-type'):
-                encoding = chardet.detect(r.content)['encoding']
-                if encoding:
-                    r.encoding = encoding
+                # For XML/RSS feeds, check the XML declaration for encoding attribute
+                # This is more reliable than chardet which can misdetect UTF-8 as MacRoman
+                content_type = r.headers.get('content-type', '').lower()
+                if 'xml' in content_type or 'rss' in content_type:
+                    # Look for <?xml version="1.0" encoding="UTF-8"?>
+                    xml_encoding_match = re.search(rb'<\?xml[^>]+encoding=["\']([^"\']+)["\']', r.content[:200])
+                    if xml_encoding_match:
+                        r.encoding = xml_encoding_match.group(1).decode('ascii')
+                    else:
+                        # Default to UTF-8 for XML if no encoding found
+                        r.encoding = 'utf-8'
+                else:
+                    # For other content types, use chardet
+                    encoding = chardet.detect(r.content)['encoding']
+                    if encoding:
+                        r.encoding = encoding
 
         self.headers = r.headers
 

changedetectionio/diff.py

@@ -1,21 +1,32 @@
 import difflib
 from typing import List, Iterator, Union
 
-HTML_REMOVED_STYLE = "background-color: #fadad7; color: #b30000;"
-HTML_ADDED_STYLE = "background-color: #eaf2c2; color: #406619;"
+# https://github.com/dgtlmoon/changedetection.io/issues/821#issuecomment-1241837050
+#HTML_ADDED_STYLE = "background-color: #d2f7c2; color: #255d00;"
+#HTML_CHANGED_INTO_STYLE = "background-color: #dafbe1; color: #116329;"
+#HTML_CHANGED_STYLE = "background-color: #ffd6cc; color: #7a2000;"
+#HTML_REMOVED_STYLE = "background-color: #ffebe9; color: #82071e;"
+
+# @todo - In the future we can make this configurable
+HTML_ADDED_STYLE = "background-color: #eaf2c2; color: #406619"
+HTML_REMOVED_STYLE = "background-color: #fadad7; color: #b30000"
+HTML_CHANGED_STYLE = HTML_REMOVED_STYLE
+HTML_CHANGED_INTO_STYLE = HTML_ADDED_STYLE
+
 
 # These get set to html or telegram type or discord compatible or whatever in handler.py
-REMOVED_PLACEMARKER_OPEN = '<<<removed_PLACEMARKER_OPEN'
-REMOVED_PLACEMARKER_CLOSED = '<<<removed_PLACEMARKER_CLOSED'
+# Something that cant get escaped to HTML by accident
+REMOVED_PLACEMARKER_OPEN = '@removed_PLACEMARKER_OPEN'
+REMOVED_PLACEMARKER_CLOSED = '@removed_PLACEMARKER_CLOSED'
 
-ADDED_PLACEMARKER_OPEN = '<<<added_PLACEMARKER_OPEN'
-ADDED_PLACEMARKER_CLOSED = '<<<added_PLACEMARKER_CLOSED'
+ADDED_PLACEMARKER_OPEN = '@added_PLACEMARKER_OPEN'
+ADDED_PLACEMARKER_CLOSED = '@added_PLACEMARKER_CLOSED'
 
-CHANGED_PLACEMARKER_OPEN = '<<<changed_PLACEMARKER_OPEN'
-CHANGED_PLACEMARKER_CLOSED = '<<<changed_PLACEMARKER_CLOSED'
+CHANGED_PLACEMARKER_OPEN = '@changed_PLACEMARKER_OPEN'
+CHANGED_PLACEMARKER_CLOSED = '@changed_PLACEMARKER_CLOSED'
 
-CHANGED_INTO_PLACEMARKER_OPEN = '<<<changed_into_PLACEMARKER_OPEN'
-CHANGED_INTO_PLACEMARKER_CLOSED = '<<<changed_into_PLACEMARKER_CLOSED'
+CHANGED_INTO_PLACEMARKER_OPEN = '@changed_into_PLACEMARKER_OPEN'
+CHANGED_INTO_PLACEMARKER_CLOSED = '@changed_into_PLACEMARKER_CLOSED'
 
 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."""

changedetectionio/flask_app.py

@@ -101,12 +101,12 @@ def init_app_secret(datastore_path):
     path = os.path.join(datastore_path, "secret.txt")
 
     try:
-        with open(path, "r") as f:
+        with open(path, "r", encoding='utf-8') as f:
             secret = f.read()
 
     except FileNotFoundError:
         import secrets
-        with open(path, "w") as f:
+        with open(path, "w", encoding='utf-8') as f:
             secret = secrets.token_hex(32)
             f.write(secret)
 
@@ -133,6 +133,11 @@ def get_socketio_path():
     # Socket.IO will be available at {prefix}/socket.io/
     return prefix
 
+@app.template_global('is_safe_valid_url')
+def _is_safe_valid_url(test_url):
+    from .validate_url import is_safe_valid_url
+    return is_safe_valid_url(test_url)
+
 
 @app.template_filter('format_number_locale')
 def _jinja2_filter_format_number_locale(value: float) -> str:
@@ -382,7 +387,7 @@ def changedetection_app(config=None, datastore_o=None):
             # We would sometimes get login loop errors on sites hosted in sub-paths
 
             # note for the future:
-            #            if not is_safe_url(next):
+            #            if not is_safe_valid_url(next):
             #                return flask.abort(400)
             return redirect(url_for('watchlist.index'))
 
@@ -789,15 +794,19 @@ def ticker_thread_check_time_launch_checks():
 
             # @todo - Maybe make this a hook?
             # Time schedule limit - Decide between watch or global settings
+            scheduler_source = None
             if watch.get('time_between_check_use_default'):
                 time_schedule_limit = datastore.data['settings']['requests'].get('time_schedule_limit', {})
-                logger.trace(f"{uuid} Time scheduler - Using system/global settings")
+                scheduler_source = 'system/global settings'
+
             else:
                 time_schedule_limit = watch.get('time_schedule_limit')
-                logger.trace(f"{uuid} Time scheduler - Using watch settings (not global settings)")
+                scheduler_source = 'watch'
+
             tz_name = datastore.data['settings']['application'].get('scheduler_timezone_default', os.getenv('TZ', 'UTC').strip())
 
             if time_schedule_limit and time_schedule_limit.get('enabled'):
+                logger.trace(f"{uuid} Time scheduler - Using scheduler settings from {scheduler_source}")
                 try:
                     result = is_within_schedule(time_schedule_limit=time_schedule_limit,
                                                 default_tz=tz_name
@@ -809,6 +818,7 @@ def ticker_thread_check_time_launch_checks():
                     logger.error(
                         f"{uuid} - Recheck scheduler, error handling timezone, check skipped - TZ name '{tz_name}' - {str(e)}")
                     return False
+
             # If they supplied an individual entry minutes to threshold.
             threshold = recheck_time_system_seconds if watch.get('time_between_check_use_default') else watch.threshold_seconds()
 

changedetectionio/forms.py

@@ -3,7 +3,7 @@ import re
 from loguru import logger
 from wtforms.widgets.core import TimeInput
 
-from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES
+from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES, RSS_TEMPLATE_TYPE_OPTIONS, RSS_TEMPLATE_HTML_DEFAULT
 from changedetectionio.conditions.form import ConditionFormRow
 from changedetectionio.notification_service import NotificationContextData
 from changedetectionio.strtobool import strtobool
@@ -28,11 +28,8 @@ from wtforms.utils import unset_value
 
 from wtforms.validators import ValidationError
 
-from validators.url import url as url_validator
-
 from changedetectionio.widgets import TernaryNoneBooleanField
 
-
 # default
 # each select <option data-enabled="enabled-0-0"
 from changedetectionio.blueprint.browser_steps.browser_steps import browser_step_ui_config
@@ -506,7 +503,9 @@ class ValidateJinja2Template(object):
             jinja2_env = create_jinja_env(loader=BaseLoader)
 
             # Add notification tokens for validation
-            jinja2_env.globals.update(NotificationContextData())
+            static_token_placeholders = NotificationContextData()
+            static_token_placeholders.set_random_for_validation()
+            jinja2_env.globals.update(static_token_placeholders)
             if hasattr(field, 'extra_notification_tokens'):
                 jinja2_env.globals.update(field.extra_notification_tokens)
 
@@ -541,19 +540,10 @@ class validateURL(object):
 
 
 def validate_url(test_url):
-    # If hosts that only contain alphanumerics are allowed ("localhost" for example)
-    try:
-        url_validator(test_url, simple_host=allow_simplehost)
-    except validators.ValidationError:
-        #@todo check for xss
-        message = f"'{test_url}' is not a valid URL."
+    from changedetectionio.validate_url import is_safe_valid_url
+    if not is_safe_valid_url(test_url):
         # This should be wtforms.validators.
-        raise ValidationError(message)
-
-    from .model.Watch import is_safe_url
-    if not is_safe_url(test_url):
-        # This should be wtforms.validators.
-        raise ValidationError('Watch protocol is not permitted by SAFE_PROTOCOL_REGEX or incorrect URL format')
+        raise ValidationError('Watch protocol is not permitted or invalid URL format')
 
 
 class ValidateSinglePythonRegexString(object):
@@ -741,7 +731,6 @@ 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
@@ -754,7 +743,7 @@ class commonSettingsForm(Form):
 
     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())
+    notification_format = SelectField('Notification format', choices=list(valid_notification_formats.items()))
     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")
@@ -1011,7 +1000,9 @@ class globalSettingsApplicationForm(commonSettingsForm):
                               validators=[validators.NumberRange(min=0,
                                                                  message="Should be atleast zero (disabled)")])
 
-    rss_content_format = SelectField('RSS Content format', choices=RSS_FORMAT_TYPES)
+    rss_content_format = SelectField('RSS Content format', choices=list(RSS_FORMAT_TYPES.items()))
+    rss_template_type = SelectField('RSS <description> body built from', choices=list(RSS_TEMPLATE_TYPE_OPTIONS.items()))
+    rss_template_override = TextAreaField('RSS "System default" template override', render_kw={"rows": "5", "placeholder": RSS_TEMPLATE_HTML_DEFAULT}, validators=[validators.Optional(), ValidateJinja2Template()])
 
     removepassword_button = SubmitField('Remove password', render_kw={"class": "pure-button pure-button-primary"})
     render_anchor_tag_content = BooleanField('Render anchor tag content', default=False)
@@ -1020,8 +1011,10 @@ class globalSettingsApplicationForm(commonSettingsForm):
     rss_hide_muted_watches = BooleanField('Hide muted watches from RSS feed', default=True,
                                       validators=[validators.Optional()])
 
-    rss_reader_mode = BooleanField('RSS reader mode ', default=False,
-                                      validators=[validators.Optional()])
+    rss_reader_mode = BooleanField('Enable RSS reader mode ', default=False, validators=[validators.Optional()])
+    rss_diff_length = IntegerField(label='Number of changes to show in watch RSS feed',
+                                   render_kw={"style": "width: 5em;"},
+                                   validators=[validators.NumberRange(min=0, message="Should contain zero or more attempts")])
 
     filter_failure_notification_threshold_attempts = IntegerField('Number of times the filter can be missing before sending a notification',
                                                                   render_kw={"style": "width: 5em;"},

changedetectionio/html_tools.py

@@ -1,3 +1,5 @@
+from functools import lru_cache
+
 from loguru import logger
 from typing import List
 import html
@@ -13,7 +15,6 @@ TITLE_RE = re.compile(r"<title[^>]*>(.*?)</title>", re.I | re.S)
 META_CS  = re.compile(r'<meta[^>]+charset=["\']?\s*([a-z0-9_\-:+.]+)', re.I)
 META_CT  = re.compile(r'<meta[^>]+http-equiv=["\']?content-type["\']?[^>]*content=["\'][^>]*charset=([a-z0-9_\-:+.]+)', re.I)
 
-
 # '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"]
@@ -22,9 +23,9 @@ class JSONNotFound(ValueError):
     def __init__(self, msg):
         ValueError.__init__(self, msg)
 
-
 # Doesn't look like python supports forward slash auto enclosure in re.findall
 # So convert it to inline flag "(?i)foobar" type configuration
+@lru_cache(maxsize=100)
 def perl_style_slash_enclosed_regex_to_options(regex):
 
     res = re.search(PERL_STYLE_REGEX, regex, re.IGNORECASE)
@@ -171,75 +172,131 @@ def elementpath_tostring(obj):
     return str(obj)
 
 # Return str Utf-8 of matched rules
-def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_rss=False):
+def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_xml=False):
+    """
+
+    :param xpath_filter:
+    :param html_content:
+    :param append_pretty_line_formatting:
+    :param is_xml: set to true if is XML or is RSS (RSS is XML)
+    :return:
+    """
     from lxml import etree, html
     import elementpath
     # xpath 2.0-3.1
     from elementpath.xpath3 import XPath3Parser
 
     parser = etree.HTMLParser()
-    if is_rss:
-        # So that we can keep CDATA for cdata_in_document_to_text() to process
-        parser = etree.XMLParser(strip_cdata=False)
-
-    tree = html.fromstring(bytes(html_content, encoding='utf-8'), parser=parser)
-    html_block = ""
-
-    r = elementpath.select(tree, xpath_filter.strip(), namespaces={'re': 'http://exslt.org/regular-expressions'}, parser=XPath3Parser)
-    #@note: //title/text() wont work where <title>CDATA..
-
-    if type(r) != list:
-        r = [r]
-
-    for element in r:
-        # When there's more than 1 match, then add the suffix to separate each line
-        # And where the matched result doesn't include something that will cause Inscriptis to add a newline
-        # (This way each 'match' reliably has a new-line in the diff)
-        # Divs are converted to 4 whitespaces by inscriptis
-        if append_pretty_line_formatting and len(html_block) and (not hasattr( element, 'tag' ) or not element.tag in (['br', 'hr', 'div', 'p'])):
-            html_block += TEXT_FILTER_LIST_LINE_SUFFIX
-
-        if type(element) == str:
-            html_block += element
-        elif issubclass(type(element), etree._Element) or issubclass(type(element), etree._ElementTree):
-            html_block += etree.tostring(element, pretty_print=True).decode('utf-8')
+    tree = None
+    try:
+        if is_xml:
+            # So that we can keep CDATA for cdata_in_document_to_text() to process
+            parser = etree.XMLParser(strip_cdata=False)
+            # For XML/RSS content, use etree.fromstring to properly handle XML declarations
+            tree = etree.fromstring(html_content.encode('utf-8') if isinstance(html_content, str) else html_content, parser=parser)
         else:
-            html_block += elementpath_tostring(element)
+            tree = html.fromstring(html_content, parser=parser)
+        html_block = ""
 
-    return html_block
+        # Build namespace map for XPath queries
+        namespaces = {'re': 'http://exslt.org/regular-expressions'}
+
+        # Handle default namespace in documents (common in RSS/Atom feeds, but can occur in any XML)
+        # XPath spec: unprefixed element names have no namespace, not the default namespace
+        # Solution: Register the default namespace with empty string prefix in elementpath
+        # This is primarily for RSS/Atom feeds but works for any XML with default namespace
+        if hasattr(tree, 'nsmap') and tree.nsmap and None in tree.nsmap:
+            # Register the default namespace with empty string prefix for elementpath
+            # This allows //title to match elements in the default namespace
+            namespaces[''] = tree.nsmap[None]
+
+        r = elementpath.select(tree, xpath_filter.strip(), namespaces=namespaces, parser=XPath3Parser)
+        #@note: //title/text() now works with default namespaces (fixed by registering '' prefix)
+        #@note: //title/text() wont work where <title>CDATA.. (use cdata_in_document_to_text first)
+
+        if type(r) != list:
+            r = [r]
+
+        for element in r:
+            # When there's more than 1 match, then add the suffix to separate each line
+            # And where the matched result doesn't include something that will cause Inscriptis to add a newline
+            # (This way each 'match' reliably has a new-line in the diff)
+            # Divs are converted to 4 whitespaces by inscriptis
+            if append_pretty_line_formatting and len(html_block) and (not hasattr( element, 'tag' ) or not element.tag in (['br', 'hr', 'div', 'p'])):
+                html_block += TEXT_FILTER_LIST_LINE_SUFFIX
+
+            if type(element) == str:
+                html_block += element
+            elif issubclass(type(element), etree._Element) or issubclass(type(element), etree._ElementTree):
+                # Use 'xml' method for RSS/XML content, 'html' for HTML content
+                # parser will be XMLParser if we detected XML content
+                method = 'xml' if (is_xml or isinstance(parser, etree.XMLParser)) else 'html'
+                html_block += etree.tostring(element, pretty_print=True, method=method, encoding='unicode')
+            else:
+                html_block += elementpath_tostring(element)
+
+        return html_block
+    finally:
+        # Explicitly clear the tree to free memory
+        # lxml trees can hold significant memory, especially with large documents
+        if tree is not None:
+            tree.clear()
 
 # Return str Utf-8 of matched rules
 # 'xpath1:'
-def xpath1_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_rss=False):
+def xpath1_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_xml=False):
     from lxml import etree, html
 
     parser = None
-    if is_rss:
-        # So that we can keep CDATA for cdata_in_document_to_text() to process
-        parser = etree.XMLParser(strip_cdata=False)
-
-    tree = html.fromstring(bytes(html_content, encoding='utf-8'), parser=parser)
-    html_block = ""
-
-    r = tree.xpath(xpath_filter.strip(), namespaces={'re': 'http://exslt.org/regular-expressions'})
-    #@note: //title/text() wont work where <title>CDATA..
-
-    for element in r:
-        # When there's more than 1 match, then add the suffix to separate each line
-        # And where the matched result doesn't include something that will cause Inscriptis to add a newline
-        # (This way each 'match' reliably has a new-line in the diff)
-        # Divs are converted to 4 whitespaces by inscriptis
-        if append_pretty_line_formatting and len(html_block) and (not hasattr(element, 'tag') or not element.tag in (['br', 'hr', 'div', 'p'])):
-            html_block += TEXT_FILTER_LIST_LINE_SUFFIX
-
-        # Some kind of text, UTF-8 or other
-        if isinstance(element, (str, bytes)):
-            html_block += element
+    tree = None
+    try:
+        if is_xml:
+            # So that we can keep CDATA for cdata_in_document_to_text() to process
+            parser = etree.XMLParser(strip_cdata=False)
+            # For XML/RSS content, use etree.fromstring to properly handle XML declarations
+            tree = etree.fromstring(html_content.encode('utf-8') if isinstance(html_content, str) else html_content, parser=parser)
         else:
-            # Return the HTML which will get parsed as text
-            html_block += etree.tostring(element, pretty_print=True).decode('utf-8')
+            tree = html.fromstring(html_content, parser=parser)
+        html_block = ""
 
-    return html_block
+        # Build namespace map for XPath queries
+        namespaces = {'re': 'http://exslt.org/regular-expressions'}
+
+        # NOTE: lxml's native xpath() does NOT support empty string prefix for default namespace
+        # For documents with default namespace (RSS/Atom feeds), users must use:
+        #   - local-name(): //*[local-name()='title']/text()
+        #   - Or use xpath_filter (not xpath1_filter) which supports default namespaces
+        # XPath spec: unprefixed element names have no namespace, not the default namespace
+
+        r = tree.xpath(xpath_filter.strip(), namespaces=namespaces)
+        #@note: xpath1 (lxml) does NOT automatically handle default namespaces
+        #@note: Use //*[local-name()='element'] or switch to xpath_filter for default namespace support
+        #@note: //title/text() wont work where <title>CDATA.. (use cdata_in_document_to_text first)
+
+        for element in r:
+            # When there's more than 1 match, then add the suffix to separate each line
+            # And where the matched result doesn't include something that will cause Inscriptis to add a newline
+            # (This way each 'match' reliably has a new-line in the diff)
+            # Divs are converted to 4 whitespaces by inscriptis
+            if append_pretty_line_formatting and len(html_block) and (not hasattr(element, 'tag') or not element.tag in (['br', 'hr', 'div', 'p'])):
+                html_block += TEXT_FILTER_LIST_LINE_SUFFIX
+
+            # Some kind of text, UTF-8 or other
+            if isinstance(element, (str, bytes)):
+                html_block += element
+            else:
+                # Return the HTML/XML which will get parsed as text
+                # Use 'xml' method for RSS/XML content, 'html' for HTML content
+                # parser will be XMLParser if we detected XML content
+                method = 'xml' if (is_xml or isinstance(parser, etree.XMLParser)) else 'html'
+                html_block += etree.tostring(element, pretty_print=True, method=method, encoding='unicode')
+
+        return html_block
+    finally:
+        # Explicitly clear the tree to free memory
+        # lxml trees can hold significant memory, especially with large documents
+        if tree is not None:
+            tree.clear()
 
 # Extract/find element
 def extract_element(find='title', html_content=''):

changedetectionio/jinja2_custom/__init__.py

@@ -9,6 +9,7 @@ from .safe_jinja import (
     JINJA2_MAX_RETURN_PAYLOAD_SIZE,
     DEFAULT_JINJA2_EXTENSIONS,
 )
+from .plugins.regex import regex_replace
 
 __all__ = [
     'TimeExtension',
@@ -17,4 +18,5 @@ __all__ = [
     'create_jinja_env',
     'JINJA2_MAX_RETURN_PAYLOAD_SIZE',
     'DEFAULT_JINJA2_EXTENSIONS',
+    'regex_replace',
 ]

changedetectionio/jinja2_custom/plugins/__init__.py

@@ -0,0 +1,6 @@
+"""
+Jinja2 custom filter plugins for changedetection.io
+"""
+from .regex import regex_replace
+
+__all__ = ['regex_replace']

changedetectionio/jinja2_custom/plugins/regex.py

@@ -0,0 +1,98 @@
+"""
+Regex filter plugin for Jinja2 templates.
+
+Provides regex_replace filter for pattern-based string replacements in templates.
+"""
+import re
+import signal
+from loguru import logger
+
+
+def regex_replace(value: str, pattern: str, replacement: str = '', count: int = 0) -> str:
+    """
+    Replace occurrences of a regex pattern in a string.
+
+    Security: Protected against ReDoS (Regular Expression Denial of Service) attacks:
+    - Limits input value size to prevent excessive processing
+    - Uses timeout mechanism to prevent runaway regex operations
+    - Validates pattern complexity to prevent catastrophic backtracking
+
+    Args:
+        value: The input string to perform replacements on
+        pattern: The regex pattern to search for
+        replacement: The replacement string (default: '')
+        count: Maximum number of replacements (0 = replace all, default: 0)
+
+    Returns:
+        String with replacements applied, or original value on error
+
+    Example:
+        {{ "hello world" | regex_replace("world", "universe") }}
+        {{ diff | regex_replace("<td>([^<]+)</td><td>([^<]+)</td>", "Label1: \\1\\nLabel2: \\2") }}
+
+    Security limits:
+        - Maximum input size: 10MB
+        - Maximum pattern length: 500 characters
+        - Operation timeout: 10 seconds
+        - Dangerous nested quantifier patterns are rejected
+    """
+    # Security limits
+    MAX_INPUT_SIZE = 1024 * 1024 * 10 # 10MB max input size
+    MAX_PATTERN_LENGTH = 500  # Maximum regex pattern length
+    REGEX_TIMEOUT_SECONDS = 10  # Maximum time for regex operation
+
+    # Validate input sizes
+    value_str = str(value)
+    if len(value_str) > MAX_INPUT_SIZE:
+        logger.warning(f"regex_replace: Input too large ({len(value_str)} bytes), truncating")
+        value_str = value_str[:MAX_INPUT_SIZE]
+
+    if len(pattern) > MAX_PATTERN_LENGTH:
+        logger.warning(f"regex_replace: Pattern too long ({len(pattern)} chars), rejecting")
+        return value_str
+
+    # Check for potentially dangerous patterns (basic checks)
+    # Nested quantifiers like (a+)+ can cause catastrophic backtracking
+    dangerous_patterns = [
+        r'\([^)]*\+[^)]*\)\+',  # (x+)+
+        r'\([^)]*\*[^)]*\)\+',  # (x*)+
+        r'\([^)]*\+[^)]*\)\*',  # (x+)*
+        r'\([^)]*\*[^)]*\)\*',  # (x*)*
+    ]
+
+    for dangerous in dangerous_patterns:
+        if re.search(dangerous, pattern):
+            logger.warning(f"regex_replace: Potentially dangerous pattern detected: {pattern}")
+            return value_str
+
+    def timeout_handler(signum, frame):
+        raise TimeoutError("Regex operation timed out")
+
+    try:
+        # Set up timeout for regex operation (Unix-like systems only)
+        # This prevents ReDoS attacks
+        old_handler = None
+        if hasattr(signal, 'SIGALRM'):
+            old_handler = signal.signal(signal.SIGALRM, timeout_handler)
+            signal.alarm(REGEX_TIMEOUT_SECONDS)
+
+        try:
+            result = re.sub(pattern, replacement, value_str, count=count)
+        finally:
+            # Cancel the alarm
+            if hasattr(signal, 'SIGALRM'):
+                signal.alarm(0)
+                if old_handler is not None:
+                    signal.signal(signal.SIGALRM, old_handler)
+
+        return result
+
+    except TimeoutError:
+        logger.error(f"regex_replace: Regex operation timed out - possible ReDoS attack. Pattern: {pattern}")
+        return value_str
+    except re.error as e:
+        logger.warning(f"regex_replace: Invalid regex pattern: {e}")
+        return value_str
+    except Exception as e:
+        logger.error(f"regex_replace: Unexpected error: {e}")
+        return value_str

changedetectionio/jinja2_custom/safe_jinja.py

@@ -8,13 +8,13 @@ import jinja2.sandbox
 import typing as t
 import os
 from .extensions.TimeExtension import TimeExtension
+from .plugins import regex_replace
 
 JINJA2_MAX_RETURN_PAYLOAD_SIZE = 1024 * int(os.getenv("JINJA2_MAX_RETURN_PAYLOAD_SIZE_KB", 1024 * 10))
 
 # Default extensions - can be overridden in create_jinja_env()
 DEFAULT_JINJA2_EXTENSIONS = [TimeExtension]
 
-
 def create_jinja_env(extensions=None, **kwargs) -> jinja2.sandbox.ImmutableSandboxedEnvironment:
     """
     Create a sandboxed Jinja2 environment with our custom extensions and default timezone.
@@ -38,6 +38,9 @@ def create_jinja_env(extensions=None, **kwargs) -> jinja2.sandbox.ImmutableSandb
     default_timezone = os.getenv('TZ', 'UTC').strip()
     jinja2_env.default_timezone = default_timezone
 
+    # Register custom filters
+    jinja2_env.filters['regex_replace'] = regex_replace
+
     return jinja2_env
 
 

changedetectionio/model/App.py

@@ -1,6 +1,7 @@
 from os import getenv
+from copy import deepcopy
 
-from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES
+from changedetectionio.blueprint.rss import RSS_FORMAT_TYPES, RSS_CONTENT_FORMAT_DEFAULT
 
 from changedetectionio.notification import (
     default_notification_body,
@@ -53,7 +54,10 @@ class model(dict):
                     'password': False,
                     'render_anchor_tag_content': False,
                     'rss_access_token': None,
-                    'rss_content_format': RSS_FORMAT_TYPES[0][0],
+                    'rss_content_format': RSS_CONTENT_FORMAT_DEFAULT,
+                    'rss_template_type': 'system_default',
+                    'rss_template_override': None,
+                    'rss_diff_length': 5,
                     'rss_hide_muted_watches': True,
                     'rss_reader_mode': False,
                     'scheduler_timezone_default': None,  # Default IANA timezone name
@@ -74,12 +78,13 @@ class model(dict):
 
     def __init__(self, *arg, **kw):
         super(model, self).__init__(*arg, **kw)
-        self.update(self.base_config)
+        # CRITICAL: deepcopy to avoid sharing mutable objects between instances
+        self.update(deepcopy(self.base_config))
 
 
 def parse_headers_from_text_file(filepath):
     headers = {}
-    with open(filepath, 'r') as f:
+    with open(filepath, 'r', encoding='utf-8') as f:
         for l in f.readlines():
             l = l.strip()
             if not l.startswith('#') and ':' in l:

changedetectionio/model/Watch.py

@@ -1,4 +1,5 @@
 from blinker import signal
+from changedetectionio.validate_url import is_safe_valid_url
 
 from changedetectionio.strtobool import strtobool
 from changedetectionio.jinja2_custom import render as jinja_render
@@ -12,32 +13,12 @@ from .. import jinja2_custom as safe_jinja
 from ..diff import ADDED_PLACEMARKER_OPEN
 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):'
 FAVICON_RESAVE_THRESHOLD_SECONDS=86400
 
 
 minimum_seconds_recheck_time = int(os.getenv('MINIMUM_SECONDS_RECHECK_TIME', 3))
 mtable = {'seconds': 1, 'minutes': 60, 'hours': 3600, 'days': 86400, 'weeks': 86400 * 7}
 
-
-def is_safe_url(test_url):
-    # See https://github.com/dgtlmoon/changedetection.io/issues/1358
-
-    # Remove 'source:' prefix so we dont get 'source:javascript:' etc
-    # 'source:' is a valid way to tell us to return the source
-
-    r = re.compile(re.escape('source:'), re.IGNORECASE)
-    test_url = r.sub('', test_url)
-
-    pattern = re.compile(os.getenv('SAFE_PROTOCOL_REGEX', SAFE_PROTOCOL_REGEX), re.IGNORECASE)
-    if not pattern.match(test_url.strip()):
-        return False
-
-    return True
-
-
 class model(watch_base):
     __newest_history_key = None
     __history_n = 0
@@ -80,7 +61,7 @@ class model(watch_base):
     def link(self):
 
         url = self.get('url', '')
-        if not is_safe_url(url):
+        if not is_safe_valid_url(url):
             return 'DISABLED'
 
         ready_url = url
@@ -101,7 +82,7 @@ class model(watch_base):
             ready_url=ready_url.replace('source:', '')
 
         # Also double check it after any Jinja2 formatting just incase
-        if not is_safe_url(ready_url):
+        if not is_safe_valid_url(ready_url):
             return 'DISABLED'
         return ready_url
 
@@ -207,7 +188,7 @@ class model(watch_base):
         fname = os.path.join(self.watch_data_dir, "history.txt")
         if os.path.isfile(fname):
             logger.debug(f"Reading watch history index for {self.get('uuid')}")
-            with open(fname, "r") as f:
+            with open(fname, "r", encoding='utf-8') as f:
                 for i in f.readlines():
                     if ',' in i:
                         k, v = i.strip().split(',', 2)
@@ -295,9 +276,17 @@ class model(watch_base):
         # When the 'last viewed' timestamp is less than the oldest snapshot, return oldest
         return sorted_keys[-1]
 
-    def get_history_snapshot(self, timestamp):
+    def get_history_snapshot(self, timestamp=None, filepath=None):
+        """
+        Accepts either timestamp or filepath
+        :param timestamp:
+        :param filepath:
+        :return:
+        """
         import brotli
-        filepath = self.history[timestamp]
+
+        if not filepath:
+            filepath = self.history[timestamp]
 
         # See if a brotli versions exists and switch to that
         if not filepath.endswith('.br') and os.path.isfile(f"{filepath}.br"):
@@ -401,7 +390,7 @@ class model(watch_base):
         # 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)
+            content = self.get_history_snapshot(filepath=v)
 
             if ignore_whitespace:
                 alist = set([line.translate(TRANSLATE_WHITESPACE_TABLE).lower() for line in content.splitlines()])
@@ -605,7 +594,7 @@ class model(watch_base):
         """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")
         if os.path.isfile(fname):
-            with open(fname, 'r') as f:
+            with open(fname, 'r', encoding='utf-8') as f:
                 return f.read()
         return False
 
@@ -658,7 +647,7 @@ class model(watch_base):
         for k, fname in self.history.items():
             if os.path.isfile(fname):
                 if True:
-                    contents = self.get_history_snapshot(k)
+                    contents = self.get_history_snapshot(timestamp=k)
                     res = re.findall(regex, contents, re.MULTILINE)
                     if res:
                         if not csv_writer:
@@ -751,7 +740,7 @@ class model(watch_base):
             # If a previous attempt doesnt yet exist, just snarf the previous snapshot instead
             dates = list(self.history.keys())
             if len(dates):
-                return self.get_history_snapshot(dates[-1])
+                return self.get_history_snapshot(timestamp=dates[-1])
             else:
                 return ''
 

changedetectionio/model/__init__.py

@@ -2,7 +2,7 @@ import os
 import uuid
 
 from changedetectionio import strtobool
-default_notification_format_for_watch = 'System default'
+USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH = 'System default'
 CONDITIONS_MATCH_LOGIC_DEFAULT = 'ALL'
 
 class watch_base(dict):
@@ -44,7 +44,7 @@ class watch_base(dict):
             'method': 'GET',
             'notification_alert_count': 0,
             'notification_body': None,
-            'notification_format': default_notification_format_for_watch,
+            'notification_format': USE_SYSTEM_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,

changedetectionio/notification/__init__.py

@@ -1,17 +1,16 @@
-from changedetectionio.model import default_notification_format_for_watch
+from changedetectionio.model import USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
 
-default_notification_format = 'HTML Color'
+default_notification_format = 'htmlcolor'
 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 = {
-    'Plain Text': 'text',
-    'HTML': 'html',
-    'HTML Color': 'htmlcolor',
-    'Markdown to HTML': 'markdown',
+    'text': 'Plain Text',
+    'html': 'HTML',
+    'htmlcolor': 'HTML Color',
+    'markdown': 'Markdown to HTML',
     # Used only for editing a watch (not for global)
-    default_notification_format_for_watch: default_notification_format_for_watch
+    USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH: USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
 }
-

changedetectionio/notification/apprise_plugin/custom_handlers.py

@@ -1,10 +1,61 @@
+"""
+Custom Apprise HTTP Handlers with format= Parameter Support
+
+IMPORTANT: This module works around a limitation in Apprise's @notify decorator.
+
+THE PROBLEM:
+-------------
+When using Apprise's @notify decorator to create custom notification handlers, the
+decorator creates a CustomNotifyPlugin that uses parse_url(..., simple=True) to parse
+URLs. This simple parsing mode does NOT extract the format= query parameter from the URL
+and set it as a top-level parameter that NotifyBase.__init__ can use to set notify_format.
+
+As a result:
+1. URL: post://example.com/webhook?format=html
+2. Apprise parses this and sees format=html in qsd (query string dictionary)
+3. But it does NOT extract it and pass it to NotifyBase.__init__
+4. NotifyBase defaults to notify_format=TEXT
+5. When you call apobj.notify(body="<html>...", body_format="html"):
+   - Apprise sees: input format = html, output format (notify_format) = text
+   - Apprise calls convert_between("html", "text", body)
+   - This strips all HTML tags, leaving only plain text
+6. Your custom handler receives stripped plain text instead of HTML
+
+THE SOLUTION:
+-------------
+Instead of using the @notify decorator directly, we:
+1. Manually register custom plugins using plugins.N_MGR.add()
+2. Create a CustomHTTPHandler class that extends CustomNotifyPlugin
+3. Override __init__ to extract format= from qsd and set it as kwargs['format']
+4. Call NotifyBase.__init__ which properly sets notify_format from kwargs['format']
+5. Set up _default_args like CustomNotifyPlugin does for compatibility
+
+This ensures that when format=html is in the URL:
+- notify_format is set to HTML
+- Apprise sees: input format = html, output format = html
+- No conversion happens (convert_between returns content unchanged)
+- Your custom handler receives the original HTML intact
+
+TESTING:
+--------
+To verify this works:
+>>> apobj = apprise.Apprise()
+>>> apobj.add('post://localhost:5005/test?format=html')
+>>> for server in apobj:
+...     print(server.notify_format)  # Should print: html (not text)
+>>> apobj.notify(body='<span>Test</span>', body_format='html')
+# Your handler should receive '<span>Test</span>' not 'Test'
+"""
+
 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 apprise import plugins
+from apprise.decorators.base import CustomNotifyPlugin
+from apprise.utils.parse import parse_url as apprise_parse_url, url_assembly
+from apprise.utils.logic import dict_full_update
 from loguru import logger
 from requests.structures import CaseInsensitiveDict
 
@@ -12,13 +63,66 @@ SUPPORTED_HTTP_METHODS = {"get", "post", "put", "delete", "patch", "head"}
 
 
 def notify_supported_methods(func):
+    """Register custom HTTP method handlers that properly support format= parameter."""
     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)
+        _register_http_handler(method, func)
+        _register_http_handler(f"{method}s", func)
     return func
 
 
+def _register_http_handler(schema, send_func):
+    """Register a custom HTTP handler that extracts format= from URL query parameters."""
+
+    # Parse base URL
+    base_url = f"{schema}://"
+    base_args = apprise_parse_url(base_url, default_schema=schema, verify_host=False, simple=True)
+
+    class CustomHTTPHandler(CustomNotifyPlugin):
+        secure_protocol = schema
+        service_name = f"Custom HTTP - {schema.upper()}"
+        _base_args = base_args
+
+        def __init__(self, **kwargs):
+            # Extract format from qsd and set it as a top-level kwarg
+            # This allows NotifyBase.__init__ to properly set notify_format
+            if 'qsd' in kwargs and 'format' in kwargs['qsd']:
+                kwargs['format'] = kwargs['qsd']['format']
+
+            # Call NotifyBase.__init__ (skip CustomNotifyPlugin.__init__)
+            super(CustomNotifyPlugin, self).__init__(**kwargs)
+
+            # Set up _default_args like CustomNotifyPlugin does
+            self._default_args = {}
+            kwargs.pop("secure", None)
+            dict_full_update(self._default_args, self._base_args)
+            dict_full_update(self._default_args, kwargs)
+            self._default_args["url"] = url_assembly(**self._default_args)
+
+        __send = staticmethod(send_func)
+
+        def send(self, body, title="", notify_type="info", *args, **kwargs):
+            """Call the custom send function."""
+            try:
+                result = self.__send(
+                    body, title, notify_type,
+                    *args,
+                    meta=self._default_args,
+                    **kwargs
+                )
+                return True if result is None else bool(result)
+            except Exception as e:
+                self.logger.warning(f"Exception in custom HTTP handler: {e}")
+                return False
+
+    # Register the plugin
+    plugins.N_MGR.add(
+        plugin=CustomHTTPHandler,
+        schemas=schema,
+        send_func=send_func,
+        url=base_url,
+    )
+
+
 def _get_auth(parsed_url: dict) -> str | tuple[str, str]:
     user: str | None = parsed_url.get("user")
     password: str | None = parsed_url.get("password")
@@ -74,6 +178,8 @@ def apprise_http_custom_handler(
     *args,
     **kwargs,
 ) -> bool:
+
+
     url: str = meta.get("url")
     schema: str = meta.get("schema")
     method: str = re.sub(r"s$", "", schema).upper()
@@ -89,25 +195,16 @@ def apprise_http_custom_handler(
 
     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 = 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()
+    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
+    logger.info(f"Successfully sent custom notification to {url}")
+    return True

changedetectionio/notification/email_helpers.py

@@ -0,0 +1,42 @@
+def as_monospaced_html_email(content: str, title: str) -> str:
+    """
+    Wraps `content` in a minimal, email-safe HTML template
+    that forces monospace rendering across Gmail, Hotmail, Apple Mail, etc.
+
+    Args:
+        content: The body text (plain text or HTML-like).
+        title: The title plaintext
+    Returns:
+        A complete HTML document string suitable for sending as an email body.
+    """
+
+    # All line feed types should be removed and then this function should only be fed <br>'s
+    # Then it works with our <pre> styling without double linefeeds
+    content = content.translate(str.maketrans('', '', '\r\n'))
+
+    if title:
+        import html
+        title = html.escape(title)
+    else:
+        title = ''
+    # 2. Full email-safe HTML
+    html_email = f"""<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8">
+  <meta name="x-apple-disable-message-reformatting">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <!--[if mso]>
+    <style>
+      body, div, pre, td {{ font-family: "Courier New", Courier, monospace !important; }}
+    </style>
+  <![endif]-->
+  <title>{title}</title>
+</head>
+<body style="-webkit-text-size-adjust:100%;-ms-text-size-adjust:100%;">
+  <pre role="article" aria-roledescription="email" lang="en"
+       style="font-family: monospace, 'Courier New', Courier; font-size: 0.9rem;
+              white-space: pre-wrap; word-break: break-word;">{content}</pre>
+</body>
+</html>"""
+    return html_email

changedetectionio/notification/handler.py

@@ -5,12 +5,16 @@ from apprise import NotifyFormat
 from loguru import logger
 from urllib.parse import urlparse
 from .apprise_plugin.assets import apprise_asset, APPRISE_AVATAR_URL
-from .apprise_plugin.custom_handlers import SUPPORTED_HTTP_METHODS
+from .email_helpers import as_monospaced_html_email
 from ..diff import HTML_REMOVED_STYLE, REMOVED_PLACEMARKER_OPEN, REMOVED_PLACEMARKER_CLOSED, ADDED_PLACEMARKER_OPEN, HTML_ADDED_STYLE, \
     ADDED_PLACEMARKER_CLOSED, CHANGED_INTO_PLACEMARKER_OPEN, CHANGED_INTO_PLACEMARKER_CLOSED, CHANGED_PLACEMARKER_OPEN, \
-    CHANGED_PLACEMARKER_CLOSED
+    CHANGED_PLACEMARKER_CLOSED, HTML_CHANGED_STYLE, HTML_CHANGED_INTO_STYLE
+import re
+
 from ..notification_service import NotificationContextData
 
+newline_re = re.compile(r'\r\n|\r|\n')
+
 
 def markup_text_links_to_html(body):
     """
@@ -61,13 +65,13 @@ def notification_format_align_with_apprise(n_format : str):
     :return:
     """
 
-    if n_format.lower().startswith('html'):
+    if n_format.startswith('html'):
         # Apprise only knows 'html' not 'htmlcolor' etc, which shouldnt matter here
         n_format = NotifyFormat.HTML.value
-    elif n_format.lower().startswith('markdown'):
+    elif n_format.startswith('markdown'):
         # probably the same but just to be safe
         n_format = NotifyFormat.MARKDOWN.value
-    elif n_format.lower().startswith('text'):
+    elif n_format.startswith('text'):
         # probably the same but just to be safe
         n_format = NotifyFormat.TEXT.value
     else:
@@ -75,9 +79,116 @@ def notification_format_align_with_apprise(n_format : str):
 
     return n_format
 
+def apply_discord_markdown_to_body(n_body):
+    """
+    Discord does not support <del> but it supports non-standard ~~strikethrough~~
+    :param n_body:
+    :return:
+    """
+    import re
+    # Define the mapping between your placeholders and markdown markers
+    replacements = [
+        (REMOVED_PLACEMARKER_OPEN, '~~', REMOVED_PLACEMARKER_CLOSED, '~~'),
+        (ADDED_PLACEMARKER_OPEN, '**', ADDED_PLACEMARKER_CLOSED, '**'),
+        (CHANGED_PLACEMARKER_OPEN, '~~', CHANGED_PLACEMARKER_CLOSED, '~~'),
+        (CHANGED_INTO_PLACEMARKER_OPEN, '**', CHANGED_INTO_PLACEMARKER_CLOSED, '**'),
+    ]
+    # So that the markdown gets added without any whitespace following it which would break it
+    for open_tag, open_md, close_tag, close_md in replacements:
+        # Regex: match opening tag, optional whitespace, capture the content, optional whitespace, then closing tag
+        pattern = re.compile(
+            re.escape(open_tag) + r'(\s*)(.*?)?(\s*)' + re.escape(close_tag),
+            flags=re.DOTALL
+        )
+        n_body = pattern.sub(lambda m: f"{m.group(1)}{open_md}{m.group(2)}{close_md}{m.group(3)}", n_body)
+    return n_body
+
+def apply_standard_markdown_to_body(n_body):
+    """
+    Apprise does not support ~~strikethrough~~ but it will convert <del> to HTML strikethrough.
+    :param n_body:
+    :return:
+    """
+    import re
+    # Define the mapping between your placeholders and markdown markers
+    replacements = [
+        (REMOVED_PLACEMARKER_OPEN, '<del>', REMOVED_PLACEMARKER_CLOSED, '</del>'),
+        (ADDED_PLACEMARKER_OPEN, '**', ADDED_PLACEMARKER_CLOSED, '**'),
+        (CHANGED_PLACEMARKER_OPEN, '<del>', CHANGED_PLACEMARKER_CLOSED, '</del>'),
+        (CHANGED_INTO_PLACEMARKER_OPEN, '**', CHANGED_INTO_PLACEMARKER_CLOSED, '**'),
+    ]
+
+    # So that the markdown gets added without any whitespace following it which would break it
+    for open_tag, open_md, close_tag, close_md in replacements:
+        # Regex: match opening tag, optional whitespace, capture the content, optional whitespace, then closing tag
+        pattern = re.compile(
+            re.escape(open_tag) + r'(\s*)(.*?)?(\s*)' + re.escape(close_tag),
+            flags=re.DOTALL
+        )
+        n_body = pattern.sub(lambda m: f"{m.group(1)}{open_md}{m.group(2)}{close_md}{m.group(3)}", n_body)
+    return n_body
+
+
+def replace_placemarkers_in_text(text, url, requested_output_format):
+    """
+    Replace diff placemarkers in text based on the URL service type and requested output format.
+    Used for both notification title and body to ensure consistent placeholder replacement.
+
+    :param text: The text to process
+    :param url: The notification URL (to detect service type)
+    :param requested_output_format: The output format (html, htmlcolor, markdown, text, etc.)
+    :return: Processed text with placemarkers replaced
+    """
+    if not text:
+        return text
+
+    if url.startswith('tgram://'):
+        # Telegram only supports a limited subset of HTML
+        # Use strikethrough for removed content, bold for added content
+        text = text.replace(REMOVED_PLACEMARKER_OPEN, '<s>')
+        text = text.replace(REMOVED_PLACEMARKER_CLOSED, '</s>')
+        text = text.replace(ADDED_PLACEMARKER_OPEN, '<b>')
+        text = text.replace(ADDED_PLACEMARKER_CLOSED, '</b>')
+        # Handle changed/replaced lines (old → new)
+        text = text.replace(CHANGED_PLACEMARKER_OPEN, '<s>')
+        text = text.replace(CHANGED_PLACEMARKER_CLOSED, '</s>')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_OPEN, '<b>')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_CLOSED, '</b>')
+    elif (url.startswith('discord://') or url.startswith('https://discordapp.com/api/webhooks')
+          or url.startswith('https://discord.com/api')) and requested_output_format == 'html':
+        # Discord doesn't support HTML, use Discord markdown
+        text = apply_discord_markdown_to_body(n_body=text)
+    elif requested_output_format == 'htmlcolor':
+        # https://github.com/dgtlmoon/changedetection.io/issues/821#issuecomment-1241837050
+        text = text.replace(REMOVED_PLACEMARKER_OPEN, f'<span style="{HTML_REMOVED_STYLE}" role="deletion" aria-label="Removed text" title="Removed text">')
+        text = text.replace(REMOVED_PLACEMARKER_CLOSED, f'</span>')
+        text = text.replace(ADDED_PLACEMARKER_OPEN, f'<span style="{HTML_ADDED_STYLE}" role="insertion" aria-label="Added text" title="Added text">')
+        text = text.replace(ADDED_PLACEMARKER_CLOSED, f'</span>')
+        # Handle changed/replaced lines (old → new)
+        text = text.replace(CHANGED_PLACEMARKER_OPEN, f'<span style="{HTML_CHANGED_STYLE}" role="note" aria-label="Changed text" title="Changed text">')
+        text = text.replace(CHANGED_PLACEMARKER_CLOSED, f'</span>')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_OPEN, f'<span style="{HTML_CHANGED_INTO_STYLE}" role="note" aria-label="Changed into" title="Changed into">')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_CLOSED, f'</span>')
+    elif requested_output_format == 'markdown':
+        # Markdown to HTML - Apprise will convert this to HTML
+        text = apply_standard_markdown_to_body(n_body=text)
+    else:
+        # plaintext, html, and default - use simple text markers
+        text = text.replace(REMOVED_PLACEMARKER_OPEN, '(removed) ')
+        text = text.replace(REMOVED_PLACEMARKER_CLOSED, '')
+        text = text.replace(ADDED_PLACEMARKER_OPEN, '(added) ')
+        text = text.replace(ADDED_PLACEMARKER_CLOSED, '')
+        text = text.replace(CHANGED_PLACEMARKER_OPEN, f'(changed) ')
+        text = text.replace(CHANGED_PLACEMARKER_CLOSED, f'')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_OPEN, f'(into) ')
+        text = text.replace(CHANGED_INTO_PLACEMARKER_CLOSED, f'')
+
+    return text
 
 def apply_service_tweaks(url, n_body, n_title, requested_output_format):
 
+    logger.debug(f"Applying markup in '{requested_output_format}' mode")
+
     # Re 323 - Limit discord length to their 2000 char limit total or it wont send.
     # Because different notifications may require different pre-processing, run each sequentially :(
     # 2000 bytes minus -
@@ -87,6 +198,12 @@ def apply_service_tweaks(url, n_body, n_title, requested_output_format):
     if not n_body or not n_body.strip():
         return url, n_body, n_title
 
+    # Normalize URL scheme to lowercase to prevent case-sensitivity issues
+    # e.g., "Discord://webhook" -> "discord://webhook", "TGRAM://bot123" -> "tgram://bot123"
+    scheme_separator_pos = url.find('://')
+    if scheme_separator_pos > 0:
+        url = url[:scheme_separator_pos].lower() + url[scheme_separator_pos:]
+
     # So if no avatar_url is specified, add one so it can be correctly calculated into the total payload
     parsed = urlparse(url)
     k = '?' if not parsed.query else '&'
@@ -98,23 +215,22 @@ def apply_service_tweaks(url, n_body, n_title, requested_output_format):
             and not url.startswith('put'):
         url += k + f"avatar_url={APPRISE_AVATAR_URL}"
 
+    # Replace placemarkers in title first (this was the missing piece causing the bug)
+    # Titles are ALWAYS plain text across all notification services (Discord embeds, Slack attachments,
+    # email Subject headers, etc.), so we always use 'text' format for title placemarker replacement
+    # Looking over apprise library it seems that all plugins only expect plain-text.
+    n_title = replace_placemarkers_in_text(n_title, url, 'text')
+
     if url.startswith('tgram://'):
         # Telegram only supports a limit subset of HTML, remove the '<br>' we place in.
         # re https://github.com/dgtlmoon/changedetection.io/issues/555
         # @todo re-use an existing library we have already imported to strip all non-allowed tags
         n_body = n_body.replace('<br>', '\n')
         n_body = n_body.replace('</br>', '\n')
+        n_body = newline_re.sub('\n', n_body)
 
-        # Use strikethrough for removed content, bold for added content
-        n_body = n_body.replace(REMOVED_PLACEMARKER_OPEN, '<s>')
-        n_body = n_body.replace(REMOVED_PLACEMARKER_CLOSED, '</s>')
-        n_body = n_body.replace(ADDED_PLACEMARKER_OPEN, '<b>')
-        n_body = n_body.replace(ADDED_PLACEMARKER_CLOSED, '</b>')
-        # Handle changed/replaced lines (old → new)
-        n_body = n_body.replace(CHANGED_PLACEMARKER_OPEN, '<s>')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_CLOSED, '</s>')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_OPEN, '<b>')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_CLOSED, '</b>')
+        # Replace placemarkers for body
+        n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
 
         # real limit is 4096, but minus some for extra metadata
         payload_max_size = 3600
@@ -128,6 +244,7 @@ def apply_service_tweaks(url, n_body, n_title, requested_output_format):
         # Discord doesn't support HTML, replace <br> with newlines
         n_body = n_body.strip().replace('<br>', '\n')
         n_body = n_body.replace('</br>', '\n')
+        n_body = newline_re.sub('\n', n_body)
 
         # Don't replace placeholders or truncate here - let the custom Discord plugin handle it
         # The plugin will use embeds (6000 char limit across all embeds) if placeholders are present,
@@ -137,15 +254,7 @@ def apply_service_tweaks(url, n_body, n_title, requested_output_format):
         if requested_output_format == 'html':
             # No diff placeholders, use Discord markdown for any other formatting
             # Use Discord markdown: strikethrough for removed, bold for added
-            n_body = n_body.replace(REMOVED_PLACEMARKER_OPEN, '~~')
-            n_body = n_body.replace(REMOVED_PLACEMARKER_CLOSED, '~~')
-            n_body = n_body.replace(ADDED_PLACEMARKER_OPEN, '**')
-            n_body = n_body.replace(ADDED_PLACEMARKER_CLOSED, '**')
-            # Handle changed/replaced lines (old → new)
-            n_body = n_body.replace(CHANGED_PLACEMARKER_OPEN, '~~')
-            n_body = n_body.replace(CHANGED_PLACEMARKER_CLOSED, '~~')
-            n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_OPEN, '**')
-            n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_CLOSED, '**')
+            n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
 
             # Apply 2000 char limit for plain content
             payload_max_size = 1700
@@ -156,57 +265,29 @@ def apply_service_tweaks(url, n_body, n_title, requested_output_format):
 
     # Is not discord/tgram and they want htmlcolor
     elif requested_output_format == 'htmlcolor':
-        n_body = n_body.replace(REMOVED_PLACEMARKER_OPEN, f'<span style="{HTML_REMOVED_STYLE}">')
-        n_body = n_body.replace(REMOVED_PLACEMARKER_CLOSED, f'</span>')
-        n_body = n_body.replace(ADDED_PLACEMARKER_OPEN, f'<span style="{HTML_ADDED_STYLE}">')
-        n_body = n_body.replace(ADDED_PLACEMARKER_CLOSED, f'</span>')
-        # Handle changed/replaced lines (old → new)
-        n_body = n_body.replace(CHANGED_PLACEMARKER_OPEN, f'<span style="{HTML_REMOVED_STYLE}">')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_CLOSED, f'</span>')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_OPEN, f'<span style="{HTML_ADDED_STYLE}">')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_CLOSED, f'</span>')
-        n_body = n_body.replace("\n", '<br>')
+        n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
+        n_body = newline_re.sub('<br>\n', n_body)
     elif requested_output_format == 'html':
-        n_body = n_body.replace(REMOVED_PLACEMARKER_OPEN, '(removed) ')
-        n_body = n_body.replace(REMOVED_PLACEMARKER_CLOSED, '')
-        n_body = n_body.replace(ADDED_PLACEMARKER_OPEN, '(added) ')
-        n_body = n_body.replace(ADDED_PLACEMARKER_CLOSED, '')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_OPEN, f'(changed) ')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_CLOSED, f'')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_OPEN, f'(into) ')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_CLOSED, f'')
-        n_body = n_body.replace("\n", '<br>')
+        n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
+        n_body = newline_re.sub('<br>\n', n_body)
+    elif requested_output_format == 'markdown':
+        # Markdown to HTML - Apprise will convert this to HTML
+        n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
 
     else: #plaintext etc default
-        n_body = n_body.replace(REMOVED_PLACEMARKER_OPEN, '(removed) ')
-        n_body = n_body.replace(REMOVED_PLACEMARKER_CLOSED, '')
-        n_body = n_body.replace(ADDED_PLACEMARKER_OPEN, '(added) ')
-        n_body = n_body.replace(ADDED_PLACEMARKER_CLOSED, '')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_OPEN, f'(changed) ')
-        n_body = n_body.replace(CHANGED_PLACEMARKER_CLOSED, f'')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_OPEN, f'(into) ')
-        n_body = n_body.replace(CHANGED_INTO_PLACEMARKER_CLOSED, f'')
+        n_body = replace_placemarkers_in_text(n_body, url, requested_output_format)
 
     return url, n_body, n_title
 
 
 def process_notification(n_object: NotificationContextData, datastore):
     from changedetectionio.jinja2_custom import render as jinja_render
-    from . import default_notification_format_for_watch, default_notification_format, valid_notification_formats
+    from . import USE_SYSTEM_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
     # Register custom Discord plugin
     from .apprise_plugin.discord import NotifyDiscordCustom
 
-    # Create list of custom handler protocols (both http and https versions)
-    custom_handler_protocols = [f"{method}://" for method in SUPPORTED_HTTP_METHODS]
-    custom_handler_protocols += [f"{method}s://" for method in SUPPORTED_HTTP_METHODS]
-
-    has_custom_handler = any(
-        url.startswith(tuple(custom_handler_protocols))
-        for url in n_object['notification_urls']
-    )
-
     if not isinstance(n_object, NotificationContextData):
         raise TypeError(f"Expected NotificationContextData, got {type(n_object)}")
 
@@ -217,27 +298,21 @@ def process_notification(n_object: NotificationContextData, datastore):
     # Insert variables into the notification content
     notification_parameters = create_notification_parameters(n_object, datastore)
 
-    requested_output_format = valid_notification_formats.get(
-        n_object.get('notification_format', default_notification_format),
-        valid_notification_formats[default_notification_format],
-    )
+    requested_output_format = n_object.get('notification_format', default_notification_format)
+    logger.debug(f"Requested notification output format: '{requested_output_format}'")
 
     # If we arrived with 'System default' then look it up
-    if requested_output_format == default_notification_format_for_watch and datastore.data['settings']['application'].get('notification_format') != default_notification_format_for_watch:
+    if requested_output_format == USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH:
         # Initially text or whatever
-        requested_output_format = datastore.data['settings']['application'].get('notification_format', valid_notification_formats[default_notification_format]).lower()
+        requested_output_format = datastore.data['settings']['application'].get('notification_format', default_notification_format)
 
     requested_output_format_original = requested_output_format
 
+    # Now clean it up so it fits perfectly with apprise
     requested_output_format = notification_format_align_with_apprise(n_format=requested_output_format)
 
     logger.trace(f"Complete notification body including Jinja and placeholders calculated in  {time.time() - now:.2f}s")
 
-    # If we have custom handlers, use invalid format to prevent conversion
-    # Otherwise use the proper format
-    if has_custom_handler:
-        input_format = 'raw-no-convert'
-
     # 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
@@ -258,74 +333,98 @@ def process_notification(n_object: NotificationContextData, datastore):
     if not n_object.get('notification_urls'):
         return None
 
-    with apprise.LogCapture(level=apprise.logging.DEBUG) as logs:
+    with (apprise.LogCapture(level=apprise.logging.DEBUG) as logs):
         for url in n_object['notification_urls']:
-            parsed_url = urlparse(url)
-            prefix_add_to_url = '?' if not parsed_url.query else '&'
 
-            # Get the notification body from datastore
             n_body = jinja_render(template_str=n_object.get('notification_body', ''), **notification_parameters)
-
-            if n_object.get('markup_text_to_html'):
-                n_body = markup_text_links_to_html(body=n_body)
-
-            # This actually means we request "Markdown to HTML"
-            if requested_output_format == NotifyFormat.MARKDOWN.value:
-                output_format = NotifyFormat.HTML.value
-                input_format = NotifyFormat.MARKDOWN.value
-                if not 'format=' in url.lower():
-                    url = f"{url}{prefix_add_to_url}format={output_format}"
-
-            # Deviation from apprise.
-            # No conversion, its like they want to send raw HTML but we add linebreaks
-            elif requested_output_format == NotifyFormat.HTML.value:
-                # same in and out means apprise wont try to convert
-                input_format = output_format = NotifyFormat.HTML.value
-                if not 'format=' in url.lower():
-                    url = f"{url}{prefix_add_to_url}format={output_format}"
-
-            else:
-                # Nothing to be done, leave it as plaintext
-                # `body_format` Tell apprise what format the INPUT is in
-                # &format= in URL Tell apprise what format the OUTPUT should be in (it can convert between)
-                input_format = output_format = NotifyFormat.TEXT.value
-                if not 'format=' in url.lower():
-                    url = f"{url}{prefix_add_to_url}format={output_format}"
-
-            if has_custom_handler:
-                input_format='raw-no-convert'
-
             n_title = jinja_render(template_str=n_object.get('notification_title', ''), **notification_parameters)
 
+            if n_object.get('markup_text_links_to_html_links'):
+                n_body = markup_text_links_to_html(body=n_body)
+
             url = url.strip()
-            if url.startswith('#'):
-                logger.trace(f"Skipping commented out notification URL - {url}")
+            if not url or url.startswith('#'):
+                logger.debug(f"Skipping commented out or empty notification URL - '{url}'")
                 continue
 
-            if not url:
-                logger.warning(f"Process Notification: skipping empty notification URL.")
-                continue
-
-            logger.info(f">> Process Notification: AppRise notifying {url}")
+            logger.info(f">> Process Notification: AppRise start notifying '{url}'")
             url = jinja_render(template_str=url, **notification_parameters)
 
+            # If it's a plaintext document, and they want HTML type email/alerts, so it needs to be escaped
+            watch_mime_type = n_object.get('watch_mime_type')
+            if watch_mime_type and 'text/' in watch_mime_type.lower() and not 'html' in watch_mime_type.lower():
+                if 'html' in requested_output_format:
+                    from markupsafe import escape
+                    n_body = str(escape(n_body))
+
+            if 'html' in requested_output_format:
+                # Since the n_body is always some kind of text from the 'diff' engine, attempt to preserve whitespaces that get sent to the HTML output
+                # But only where its more than 1 consecutive whitespace, otherwise "and this" becomes "and&nbsp;this" etc which is too much.
+                n_body = n_body.replace('  ', '&nbsp;&nbsp;')
+
             (url, n_body, n_title) = apply_service_tweaks(url=url, n_body=n_body, n_title=n_title, requested_output_format=requested_output_format_original)
 
-            apobj.add(url)
+            apprise_input_format = "NO-THANKS-WE-WILL-MANAGE-ALL-OF-THIS"
 
+            if not 'format=' in url:
+                parsed_url = urlparse(url)
+                prefix_add_to_url = '?' if not parsed_url.query else '&'
+
+                # THIS IS THE TRICK HOW TO DISABLE APPRISE DOING WEIRD AUTO-CONVERSION WITH BREAKING BR TAGS ETC
+                if 'html' in requested_output_format:
+                    url = f"{url}{prefix_add_to_url}format={NotifyFormat.HTML.value}"
+                    apprise_input_format = NotifyFormat.HTML.value
+                elif 'text' in requested_output_format:
+                    url = f"{url}{prefix_add_to_url}format={NotifyFormat.TEXT.value}"
+                    apprise_input_format = NotifyFormat.TEXT.value
+
+                elif requested_output_format == NotifyFormat.MARKDOWN.value:
+                    # Convert markdown to HTML ourselves since not all plugins do this
+                    from apprise.conversion import markdown_to_html
+                    # Make sure there are paragraph breaks around horizontal rules
+                    n_body = n_body.replace('---', '\n\n---\n\n')
+                    n_body = markdown_to_html(n_body)
+                    url = f"{url}{prefix_add_to_url}format={NotifyFormat.HTML.value}"
+                    requested_output_format = NotifyFormat.HTML.value
+                    apprise_input_format = NotifyFormat.HTML.value  # Changed from MARKDOWN to HTML
+
+            else:
+                # ?format was IN the apprise URL, they are kind of on their own here, we will try our best
+                if 'format=html' in url:
+                    n_body = newline_re.sub('<br>\r\n', n_body)
+                    # This will also prevent apprise from doing conversion
+                    apprise_input_format = NotifyFormat.HTML.value
+                    requested_output_format = NotifyFormat.HTML.value
+                elif 'format=text' in url:
+                    apprise_input_format = NotifyFormat.TEXT.value
+                    requested_output_format = NotifyFormat.TEXT.value
+
+#@todo on null:// (only if its a 1 url with null) probably doesnt need to actually .add/setup/etc
             sent_objs.append({'title': n_title,
                               'body': n_body,
-                              'url': url})
+                              'url': url,
+                              # So that we can do a null:// call and get back exactly what would have been sent
+                              'original_context': n_object })
 
-        apobj.notify(
-            title=n_title,
-            body=n_body,
-            # `body_format` Tell apprise what format the INPUT is in
-            # &format= in URL Tell apprise what format the OUTPUT should be in (it can convert between)
-            body_format=input_format,
-            # False is not an option for AppRise, must be type None
-            attach=n_object.get('screenshot', None)
-        )
+            if not url.startswith('null://'):
+                apobj.add(url)
+
+            # Since the output is always based on the plaintext of the 'diff' engine, wrap it nicely.
+            # It should always be similar to the 'history' part of the UI.
+            if url.startswith('mail') and 'html' in requested_output_format:
+                if not '<pre' in n_body and not '<body' in n_body: # No custom HTML-ish body was setup already
+                    n_body = as_monospaced_html_email(content=n_body, title=n_title)
+
+        if not url.startswith('null://'):
+            apobj.notify(
+                title=n_title,
+                body=n_body,
+                # `body_format` Tell apprise what format the INPUT is in, specify a wrong/bad type and it will force skip conversion in apprise
+                # &format= in URL Tell apprise what format the OUTPUT should be in (it can convert between)
+                body_format=apprise_input_format,
+                # False is not an option for AppRise, must be type None
+                attach=n_object.get('screenshot', None)
+            )
 
         # Returns empty string if nothing found, multi-line string otherwise
         log_value = logs.getvalue()
@@ -344,6 +443,8 @@ def create_notification_parameters(n_object: NotificationContextData, datastore)
     if not isinstance(n_object, NotificationContextData):
         raise TypeError(f"Expected NotificationContextData, got {type(n_object)}")
 
+    ext_base_url = datastore.data['settings']['application'].get('active_base_url').strip('/')+'/'
+
     watch = datastore.data['watching'].get(n_object['uuid'])
     if watch:
         watch_title = datastore.data['watching'][n_object['uuid']].label
@@ -357,20 +458,29 @@ def create_notification_parameters(n_object: NotificationContextData, datastore)
         watch_title = 'Change Detection'
         watch_tag = ''
 
-    # Create URLs to customise the notification with
-    # active_base_url - set in store.py data property
-    base_url = datastore.data['settings']['application'].get('active_base_url')
-
     watch_url = n_object['watch_url']
 
-    diff_url = "{}/diff/{}".format(base_url, n_object['uuid'])
-    preview_url = "{}/preview/{}".format(base_url, n_object['uuid'])
+    # Build URLs manually instead of using url_for() to avoid requiring a request context
+    # This allows notifications to be processed in background threads
+    uuid = n_object['uuid']
 
+    if n_object.get('timestamp_from') and n_object.get('timestamp_to'):
+        # Include a link to the diff page with specific versions
+        diff_url = f"{ext_base_url}diff/{uuid}?from_version={n_object['timestamp_from']}&to_version={n_object['timestamp_to']}"
+    else:
+        diff_url = f"{ext_base_url}diff/{uuid}"
+
+    preview_url = f"{ext_base_url}preview/{uuid}"
+    edit_url = f"{ext_base_url}edit/{uuid}"
+
+    # @todo test that preview_url is correct when running in not-null mode?
+    # if not, first time app loads i think it can set a flask context
     n_object.update(
         {
-            'base_url': base_url,
+            'base_url': ext_base_url,
             'diff_url': diff_url,
-            'preview_url': preview_url,
+            'preview_url': preview_url, #@todo include 'version='
+            'edit_url': edit_url, #@todo also pause, also mute link
             '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,

changedetectionio/notification_service.py

@@ -5,33 +5,83 @@ Notification Service Module
 Extracted from update_worker.py to provide standalone notification functionality
 for both sync and async workers
 """
+import datetime
 
+import pytz
 from loguru import logger
 import time
 
-from changedetectionio.notification import default_notification_format
+from changedetectionio.notification import default_notification_format, valid_notification_formats
+
+
+def _check_cascading_vars(datastore, var_name, watch):
+    """
+    Check notification variables in cascading priority:
+    Individual watch settings > Tag settings > Global settings
+    """
+    from changedetectionio.notification import (
+        USE_SYSTEM_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 == USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH:
+            return datastore.data['settings']['application'].get('notification_format')
+
+        return v
+
+    tags = 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 datastore.data['settings']['application'].get(var_name):
+        return datastore.data['settings']['application'].get(var_name)
+
+    # Otherwise could be defaults
+    if var_name == 'notification_format':
+        return USE_SYSTEM_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
+
 
 # What is passed around as notification context, also used as the complete list of valid {{ tokens }}
 class NotificationContextData(dict):
     def __init__(self, initial_data=None, **kwargs):
         super().__init__({
+            'base_url': None,
             'current_snapshot': None,
             'diff': None,
+            'diff_clean': None,
             'diff_added': None,
+            'diff_added_clean': None,
             'diff_full': None,
+            'diff_full_clean': None,
             'diff_patch': None,
             'diff_removed': None,
+            'diff_removed_clean': None,
+            'diff_url': None,
+            'markup_text_links_to_html_links': False, # If automatic conversion of plaintext to HTML should happen
             'notification_timestamp': time.time(),
+            'preview_url': None,
             'screenshot': None,
             'triggered_text': None,
+            'timestamp_from': None,
+            'timestamp_to': None,
             'uuid': 'XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX',  # Converted to 'watch_uuid' in create_notification_parameters
-            'watch_url': 'https://WATCH-PLACE-HOLDER/',
-            'base_url': None,
-            'diff_url': None,
-            'preview_url': None,
+            'watch_mime_type': None,
             'watch_tag': None,
             'watch_title': None,
-            'markup_text_to_html': False, # If automatic conversion of plaintext to HTML should happen
+            'watch_url': 'https://WATCH-PLACE-HOLDER/',
         })
 
         # Apply any initial data passed in
@@ -43,15 +93,79 @@ class NotificationContextData(dict):
         if kwargs:
             self.update(kwargs)
 
+        n_format = self.get('notification_format')
+        if n_format and not valid_notification_formats.get(n_format):
+            raise ValueError(f'Invalid notification format: "{n_format}"')
+
     def set_random_for_validation(self):
         import random, string
-        """Randomly fills all dict keys with random strings (for validation/testing)."""
+        """Randomly fills all dict keys with random strings (for validation/testing). 
+        So we can test the output in the notification body
+        """
         for key in self.keys():
             if key in ['uuid', 'time', 'watch_uuid']:
                 continue
             rand_str = 'RANDOM-PLACEHOLDER-'+''.join(random.choices(string.ascii_letters + string.digits, k=12))
             self[key] = rand_str
 
+    def __setitem__(self, key, value):
+        if key == 'notification_format' and isinstance(value, str) and not value.startswith('RANDOM-PLACEHOLDER-'):
+            if not valid_notification_formats.get(value):
+                raise ValueError(f'Invalid notification format: "{value}"')
+
+        super().__setitem__(key, value)
+
+
+def timestamp_to_localtime(timestamp):
+    # Format the date using locale-aware formatting with timezone
+    dt = datetime.datetime.fromtimestamp(int(timestamp))
+    dt = dt.replace(tzinfo=pytz.UTC)
+
+    # Get local timezone-aware datetime
+    local_tz = datetime.datetime.now().astimezone().tzinfo
+    local_dt = dt.astimezone(local_tz)
+
+    # Format date with timezone - using strftime for locale awareness
+    try:
+        formatted_date = local_dt.strftime('%Y-%m-%d %H:%M:%S %Z')
+    except:
+        # Fallback if locale issues
+        formatted_date = local_dt.isoformat()
+
+    return formatted_date
+
+def set_basic_notification_vars(snapshot_contents, current_snapshot, prev_snapshot, watch, triggered_text, timestamp_changed=None):
+    now = time.time()
+    from changedetectionio import diff
+
+    n_object = {
+        'current_snapshot': snapshot_contents,
+        'diff': diff.render_diff(prev_snapshot, current_snapshot),
+        'diff_clean': diff.render_diff(prev_snapshot, current_snapshot, include_change_type_prefix=False),
+        'diff_added': diff.render_diff(prev_snapshot, current_snapshot, include_removed=False),
+        'diff_added_clean': diff.render_diff(prev_snapshot, current_snapshot, include_removed=False, include_change_type_prefix=False),
+        'diff_full': diff.render_diff(prev_snapshot, current_snapshot, include_equal=True),
+        'diff_full_clean': diff.render_diff(prev_snapshot, current_snapshot, include_equal=True, include_change_type_prefix=False),
+        'diff_patch': diff.render_diff(prev_snapshot, current_snapshot, patch_format=True),
+        'diff_removed': diff.render_diff(prev_snapshot, current_snapshot, include_added=False),
+        'diff_removed_clean': diff.render_diff(prev_snapshot, current_snapshot, include_added=False, include_change_type_prefix=False),
+        'screenshot': watch.get_screenshot() if watch and watch.get('notification_screenshot') else None,
+        'change_datetime': timestamp_to_localtime(timestamp_changed) if timestamp_changed else None,
+        'triggered_text': triggered_text,
+        'uuid': watch.get('uuid') if watch else None,
+        'watch_url': watch.get('url') if watch else None,
+        'watch_uuid': watch.get('uuid') if watch else None,
+        'watch_mime_type': watch.get('content-type')
+    }
+
+    # The \n's in the content from the above will get converted to <br> etc depending on the notification format
+
+    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")
+    return n_object
+
 class NotificationService:
     """
     Standalone notification service that handles all notification functionality
@@ -62,12 +176,11 @@ class NotificationService:
         self.datastore = datastore
         self.notification_q = notification_q
     
-    def queue_notification_for_watch(self, n_object: NotificationContextData, watch):
+    def queue_notification_for_watch(self, n_object: NotificationContextData, watch, date_index_from=-2, date_index_to=-1):
         """
         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
+        from changedetectionio.notification import USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
 
         if not isinstance(n_object, NotificationContextData):
             raise TypeError(f"Expected NotificationContextData, got {type(n_object)}")
@@ -75,8 +188,6 @@ class NotificationService:
         dates = []
         trigger_text = ''
 
-        now = time.time()
-
         if watch:
             watch_history = watch.history
             dates = list(watch_history.keys())
@@ -84,100 +195,44 @@ class NotificationService:
 
         # Add text that was triggered
         if len(dates):
-            snapshot_contents = watch.get_history_snapshot(dates[-1])
+            snapshot_contents = watch.get_history_snapshot(timestamp=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:
+        if n_object.get('notification_format') == USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH:
             n_object['notification_format'] = self.datastore.data['settings']['application'].get('notification_format')
 
-        # 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)
-        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)
+                triggered_text = '\n'.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])
+            prev_snapshot = watch.get_history_snapshot(timestamp=dates[date_index_from])
+            current_snapshot = watch.get_history_snapshot(timestamp=dates[date_index_to])
 
-        n_object.update({
-            'current_snapshot': snapshot_contents,
-            'diff': diff.render_diff(prev_snapshot, current_snapshot, line_feed_sep=line_feed_sep),
-            '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),
-            '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),
-            '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,
-            'watch_uuid': watch.get('uuid') if watch else None,
-        })
 
-        if watch:
-            n_object.update(watch.extra_notification_token_values())
+        n_object.update(set_basic_notification_vars(snapshot_contents=snapshot_contents,
+                                                    current_snapshot=current_snapshot,
+                                                    prev_snapshot=prev_snapshot,
+                                                    watch=watch,
+                                                    triggered_text=triggered_text,
+                                                    timestamp_changed=dates[date_index_to]))
 
-        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
+        if self.notification_q:
+            logger.debug("Queued notification for sending")
+            self.notification_q.put(n_object)
+        else:
+            logger.debug("Not queued, no queue defined. Just returning processed data")
+            return n_object
 
     def send_content_changed_notification(self, watch_uuid):
         """
@@ -200,10 +255,11 @@ class NotificationService:
         # 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)
+        # this change probably not needed?
+        n_object['notification_urls'] = _check_cascading_vars(self.datastore, 'notification_urls', watch)
+        n_object['notification_title'] = _check_cascading_vars(self.datastore,'notification_title', watch)
+        n_object['notification_body'] = _check_cascading_vars(self.datastore,'notification_body', watch)
+        n_object['notification_format'] = _check_cascading_vars(self.datastore,'notification_format', watch)
 
         # (Individual watch) Only prepare to notify if the rules above matched
         queued = False
@@ -226,9 +282,8 @@ class NotificationService:
         if not watch:
             return
 
-        n_format = self.datastore.data['settings']['application'].get('notification_format', default_notification_format)
         filter_list = ", ".join(watch['include_filters'])
-        # @todo - This could be a markdown template on the disk, apprise will convert the markdown to HTML+Plaintext parts in the email, and then 'markup_text_to_html' is not needed
+        # @todo - This could be a markdown template on the disk, apprise will convert the markdown to HTML+Plaintext parts in the email, and then 'markup_text_links_to_html_links' is not needed
         body = f"""Hello,
 
 Your configured CSS/xPath filters of '{filter_list}' for {{{{watch_url}}}} did not appear on the page after {threshold} attempts.
@@ -243,9 +298,9 @@ Thanks - Your omniscient changedetection.io installation.
         n_object = NotificationContextData({
             'notification_title': 'Changedetection.io - Alert - CSS/xPath filter was not present in the page',
             'notification_body': body,
-            'notification_format': n_format,
-            'markup_text_to_html': n_format.lower().startswith('html')
+            'notification_format': _check_cascading_vars(self.datastore, 'notification_format', watch),
         })
+        n_object['markup_text_links_to_html_links'] = n_object.get('notification_format').startswith('html')
 
         if len(watch['notification_urls']):
             n_object['notification_urls'] = watch['notification_urls']
@@ -273,9 +328,9 @@ Thanks - Your omniscient changedetection.io installation.
         if not watch:
             return
         threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts')
-        n_format = self.datastore.data['settings']['application'].get('notification_format', default_notification_format).lower()
+
         step = step_n + 1
-        # @todo - This could be a markdown template on the disk, apprise will convert the markdown to HTML+Plaintext parts in the email, and then 'markup_text_to_html' is not needed
+        # @todo - This could be a markdown template on the disk, apprise will convert the markdown to HTML+Plaintext parts in the email, and then 'markup_text_links_to_html_links' is not needed
 
         # {{{{ }}}} because this will be Jinja2 {{ }} tokens
         body = f"""Hello,
@@ -292,9 +347,9 @@ Thanks - Your omniscient changedetection.io installation.
         n_object = NotificationContextData({
             'notification_title': f"Changedetection.io - Alert - Browser step at position {step} could not be run",
             'notification_body': body,
-            'notification_format': n_format,
-            'markup_text_to_html': n_format.lower().startswith('html')
+            'notification_format': self._check_cascading_vars('notification_format', watch),
         })
+        n_object['markup_text_links_to_html_links'] = n_object.get('notification_format').startswith('html')
 
         if len(watch['notification_urls']):
             n_object['notification_urls'] = watch['notification_urls']

changedetectionio/processors/__init__.py

@@ -91,6 +91,8 @@ class difference_detection_processor():
             else:
                 logger.debug("Skipping adding proxy data when custom Browser endpoint is specified. ")
 
+        logger.debug(f"Using proxy '{proxy_url}' for {self.watch['uuid']}")
+
         # 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)
         self.fetcher = fetcher_obj(proxy_override=proxy_url,

changedetectionio/processors/magic.py

@@ -88,7 +88,7 @@ class guess_stream_type():
                         magic_content_header = mime
 
         except Exception as e:
-            logger.error(f"Error getting a more precise mime type from 'puremagic' library ({str(e)}), using content-based detection")
+            logger.warning(f"Error getting a more precise mime type from 'puremagic' library ({str(e)}), using content-based detection")
 
         # Content-based detection (most reliable for text formats)
         # Check for HTML patterns first - if found, override magic's text/plain
@@ -103,15 +103,15 @@ class guess_stream_type():
             self.is_json = True
         elif 'pdf' in magic_content_header:
             self.is_pdf = True
-        elif has_html_patterns or http_content_header == 'text/html':
-            self.is_html = True
-        elif any(s in magic_content_header for s in JSON_CONTENT_TYPES):
-            self.is_json = True
         # magic will call a rss document 'xml'
         # Rarely do endpoints give the right header, usually just text/xml, so we check also for <rss
         # This also triggers the automatic CDATA text parser so the RSS goes back a nice content list
         elif '<rss' in test_content_normalized or '<feed' in test_content_normalized or any(s in magic_content_header for s in RSS_XML_CONTENT_TYPES) or '<rdf:' in test_content_normalized:
             self.is_rss = True
+        elif has_html_patterns or http_content_header == 'text/html':
+            self.is_html = True
+        elif any(s in magic_content_header for s in JSON_CONTENT_TYPES):
+            self.is_json = True
         elif any(s in http_content_header for s in XML_CONTENT_TYPES):
             # Only mark as generic XML if not already detected as RSS
             if not self.is_rss:

changedetectionio/processors/text_json_diff/__init__.py

@@ -32,7 +32,7 @@ def prepare_filter_prevew(datastore, watch_uuid, form_data):
     '''Used by @app.route("/edit/<string:uuid>/preview-rendered", methods=['POST'])'''
     from changedetectionio import forms, html_tools
     from changedetectionio.model.Watch import model as watch_model
-    from concurrent.futures import ProcessPoolExecutor
+    from concurrent.futures import ThreadPoolExecutor
     from copy import deepcopy
     from flask import request
     import brotli
@@ -76,13 +76,16 @@ def prepare_filter_prevew(datastore, watch_uuid, form_data):
             update_handler.fetcher.headers['content-type'] = tmp_watch.get('content-type')
 
             # Process our watch with filters and the HTML from disk, and also a blank watch with no filters but also with the same HTML from disk
-            # Do this as a parallel process because it could take some time
-            with ProcessPoolExecutor(max_workers=2) as executor:
-                future1 = executor.submit(_task, tmp_watch, update_handler)
-                future2 = executor.submit(_task, blank_watch_no_filters, update_handler)
+            # Do this as parallel threads (not processes) to avoid pickle issues with Lock objects
+            try:
+                with ThreadPoolExecutor(max_workers=2) as executor:
+                    future1 = executor.submit(_task, tmp_watch, update_handler)
+                    future2 = executor.submit(_task, blank_watch_no_filters, update_handler)
 
-                text_after_filter = future1.result()
-                text_before_filter = future2.result()
+                    text_after_filter = future1.result()
+                    text_before_filter = future2.result()
+            except Exception as e:
+                x=1
 
     try:
         trigger_line_numbers = html_tools.strip_ignore_text(content=text_after_filter,

changedetectionio/processors/text_json_diff/processor.py

@@ -280,7 +280,7 @@ class ContentProcessor:
 
         # Sort JSON to avoid false alerts from reordering
         try:
-            content = json.dumps(json.loads(content), sort_keys=True, indent=4)
+            content = json.dumps(json.loads(content), sort_keys=True, indent=2, ensure_ascii=False)
         except Exception:
             # Might be malformed JSON, continue anyway
             pass
@@ -298,7 +298,7 @@ class ContentProcessor:
                     xpath_filter=filter_rule.replace('xpath:', ''),
                     html_content=content,
                     append_pretty_line_formatting=not self.watch.is_source_type_url,
-                    is_rss=stream_content_type.is_rss
+                    is_xml=stream_content_type.is_rss or stream_content_type.is_xml
                 )
 
             # XPath1 filters (first match only)
@@ -307,7 +307,7 @@ class ContentProcessor:
                     xpath_filter=filter_rule.replace('xpath1:', ''),
                     html_content=content,
                     append_pretty_line_formatting=not self.watch.is_source_type_url,
-                    is_rss=stream_content_type.is_rss
+                    is_xml=stream_content_type.is_rss or stream_content_type.is_xml
                 )
 
             # JSON filters

changedetectionio/pytest.ini

@@ -1,5 +1,5 @@
 [pytest]
-addopts = --no-start-live-server --live-server-port=5005
+addopts = --no-start-live-server --live-server-port=0
 #testpaths = tests pytest_invenio
 #live_server_scope = function
 

changedetectionio/realtime/socket_server.py

@@ -37,18 +37,6 @@ class SignalHandler:
         notification_event_signal.connect(self.handle_notification_event, weak=False)
         logger.info("SignalHandler: Connected to notification_event signal")
 
-        # Create and start the queue update thread using standard threading
-        import threading
-        self.polling_emitter_thread = threading.Thread(
-            target=self.polling_emit_running_or_queued_watches_threaded,
-            daemon=True
-        )
-        self.polling_emitter_thread.start()
-        logger.info("Started polling thread using threading (eventlet-free)")
-
-        # Store the thread reference in socketio for clean shutdown
-        self.socketio_instance.polling_emitter_thread = self.polling_emitter_thread
-
     def handle_signal(self, *args, **kwargs):
         logger.trace(f"SignalHandler: Signal received with {len(args)} args and {len(kwargs)} kwargs")
         # Safely extract the watch UUID from kwargs
@@ -124,74 +112,6 @@ class SignalHandler:
         except Exception as e:
             logger.error(f"Socket.IO error in handle_notification_event: {str(e)}")
 
-    def polling_emit_running_or_queued_watches_threaded(self):
-        """Threading version of polling for Windows compatibility"""
-        import time
-        import threading
-        logger.info("Queue update thread started (threading mode)")
-
-        # Import here to avoid circular imports
-        from changedetectionio.flask_app import app
-        from changedetectionio import worker_handler
-        watch_check_update = signal('watch_check_update')
-
-        # Track previous state to avoid unnecessary emissions
-        previous_running_uuids = set()
-
-        # Run until app shutdown - check exit flag more frequently for fast shutdown
-        exit_event = getattr(app.config, 'exit', threading.Event())
-
-        while not exit_event.is_set():
-            try:
-                # Get current running UUIDs from async workers
-                running_uuids = set(worker_handler.get_running_uuids())
-
-                # Only send updates for UUIDs that changed state
-                newly_running = running_uuids - previous_running_uuids
-                no_longer_running = previous_running_uuids - running_uuids
-
-                # Send updates for newly running UUIDs (but exit fast if shutdown requested)
-                for uuid in newly_running:
-                    if exit_event.is_set():
-                        break
-                    logger.trace(f"Threading polling: UUID {uuid} started processing")
-                    with app.app_context():
-                        watch_check_update.send(app_context=app, watch_uuid=uuid)
-                    time.sleep(0.01)  # Small yield
-
-                # Send updates for UUIDs that finished processing (but exit fast if shutdown requested)
-                if not exit_event.is_set():
-                    for uuid in no_longer_running:
-                        if exit_event.is_set():
-                            break
-                        logger.trace(f"Threading polling: UUID {uuid} finished processing")
-                        with app.app_context():
-                            watch_check_update.send(app_context=app, watch_uuid=uuid)
-                        time.sleep(0.01)  # Small yield
-
-                # Update tracking for next iteration
-                previous_running_uuids = running_uuids
-
-                # Sleep between polling cycles, but check exit flag every 0.5 seconds for fast shutdown
-                for _ in range(20):  # 20 * 0.5 = 10 seconds total
-                    if exit_event.is_set():
-                        break
-                    time.sleep(0.5)
-
-            except Exception as e:
-                logger.error(f"Error in threading polling: {str(e)}")
-                # Even during error recovery, check for exit quickly
-                for _ in range(1):  # 1 * 0.5 = 0.5 seconds
-                    if exit_event.is_set():
-                        break
-                    time.sleep(0.5)
-
-        # 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("Queue update thread stopped (threading mode)")
 
 
 def handle_watch_update(socketio, **kwargs):
@@ -383,19 +303,6 @@ def init_socketio(app, datastore):
         """Shutdown the SocketIO server fast and aggressively"""
         try:
             logger.info("Socket.IO: Fast shutdown initiated...")
-
-            # For threading mode, give the thread a very short time to exit gracefully
-            if hasattr(socketio, 'polling_emitter_thread'):
-                if socketio.polling_emitter_thread.is_alive():
-                    logger.info("Socket.IO: Waiting 1 second for polling thread to stop...")
-                    socketio.polling_emitter_thread.join(timeout=1.0)  # Only 1 second timeout
-                    if socketio.polling_emitter_thread.is_alive():
-                        logger.info("Socket.IO: Polling thread still running after timeout - continuing with shutdown")
-                    else:
-                        logger.info("Socket.IO: Polling thread stopped quickly")
-                else:
-                    logger.info("Socket.IO: Polling thread already stopped")
-
             logger.info("Socket.IO: Fast shutdown complete")
         except Exception as e:
             logger.error(f"Socket.IO error during shutdown: {str(e)}")

changedetectionio/rss_tools.py

@@ -29,16 +29,135 @@ def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False
     return re.sub(pattern, repl, html_content)
 
 
+# Jinja2 template for formatting RSS/Atom feed entries
+# Covers all common feedparser entry fields including namespaced elements
+# Outputs HTML that will be converted to text via html_to_text
+# @todo - This could be a UI setting in the future
+RSS_ENTRY_TEMPLATE = """<article class="rss-item" id="{{ entry.id|replace('"', '')|replace(' ', '-') }}">{%- if entry.title -%}Title: {{ entry.title }}<br>{%- endif -%}
+{%- if entry.link -%}<strong>Link:</strong> <a href="{{ entry.link }}">{{ entry.link }}</a><br>
+{%- endif -%}
+{%- if entry.id -%}
+<strong>Guid:</strong> {{ entry.id }}<br>
+{%- endif -%}
+{%- if entry.published -%}
+<strong>PubDate:</strong> {{ entry.published }}<br>
+{%- endif -%}
+{%- if entry.updated and entry.updated != entry.published -%}
+<strong>Updated:</strong> {{ entry.updated }}<br>
+{%- endif -%}
+{%- if entry.author -%}
+<strong>Author:</strong> {{ entry.author }}<br>
+{%- elif entry.author_detail and entry.author_detail.name -%}
+<strong>Author:</strong> {{ entry.author_detail.name }}
+{%- if entry.author_detail.email %} ({{ entry.author_detail.email }}){% endif -%}
+<br>
+{%- endif -%}
+{%- if entry.contributors -%}
+<strong>Contributors:</strong> {% for contributor in entry.contributors -%}
+{{ contributor.name if contributor.name else contributor }}
+{%- if not loop.last %}, {% endif -%}
+{%- endfor %}<br>
+{%- endif -%}
+{%- if entry.publisher -%}
+<strong>Publisher:</strong> {{ entry.publisher }}<br>
+{%- endif -%}
+{%- if entry.rights -%}
+<strong>Rights:</strong> {{ entry.rights }}<br>
+{%- endif -%}
+{%- if entry.license -%}
+<strong>License:</strong> {{ entry.license }}<br>
+{%- endif -%}
+{%- if entry.language -%}
+<strong>Language:</strong> {{ entry.language }}<br>
+{%- endif -%}
+{%- if entry.tags -%}
+<strong>Tags:</strong> {% for tag in entry.tags -%}
+{{ tag.term if tag.term else tag }}
+{%- if not loop.last %}, {% endif -%}
+{%- endfor %}<br>
+{%- endif -%}
+{%- if entry.category -%}
+<strong>Category:</strong> {{ entry.category }}<br>
+{%- endif -%}
+{%- if entry.comments -%}
+<strong>Comments:</strong> <a href="{{ entry.comments }}">{{ entry.comments }}</a><br>
+{%- endif -%}
+{%- if entry.slash_comments -%}
+<strong>Comment Count:</strong> {{ entry.slash_comments }}<br>
+{%- endif -%}
+{%- if entry.enclosures -%}
+<strong>Enclosures:</strong><br>
+{%- for enclosure in entry.enclosures %}
+- <a href="{{ enclosure.href }}">{{ enclosure.href }}</a> ({{ enclosure.type if enclosure.type else 'unknown type' }}
+{%- if enclosure.length %}, {{ enclosure.length }} bytes{% endif -%}
+)<br>
+{%- endfor -%}
+{%- endif -%}
+{%- if entry.media_content -%}
+<strong>Media:</strong><br>
+{%- for media in entry.media_content %}
+- <a href="{{ media.url }}">{{ media.url }}</a>
+{%- if media.type %} ({{ media.type }}){% endif -%}
+{%- if media.width and media.height %} {{ media.width }}x{{ media.height }}{% endif -%}
+<br>
+{%- endfor -%}
+{%- endif -%}
+{%- if entry.media_thumbnail -%}
+<strong>Thumbnail:</strong> <a href="{{ entry.media_thumbnail[0].url if entry.media_thumbnail[0].url else entry.media_thumbnail[0] }}">{{ entry.media_thumbnail[0].url if entry.media_thumbnail[0].url else entry.media_thumbnail[0] }}</a><br>
+{%- endif -%}
+{%- if entry.media_description -%}
+<strong>Media Description:</strong> {{ entry.media_description }}<br>
+{%- endif -%}
+{%- if entry.itunes_duration -%}
+<strong>Duration:</strong> {{ entry.itunes_duration }}<br>
+{%- endif -%}
+{%- if entry.itunes_author -%}
+<strong>Podcast Author:</strong> {{ entry.itunes_author }}<br>
+{%- endif -%}
+{%- if entry.dc_identifier -%}
+<strong>Identifier:</strong> {{ entry.dc_identifier }}<br>
+{%- endif -%}
+{%- if entry.dc_source -%}
+<strong>DC Source:</strong> {{ entry.dc_source }}<br>
+{%- endif -%}
+{%- if entry.dc_type -%}
+<strong>Type:</strong> {{ entry.dc_type }}<br>
+{%- endif -%}
+{%- if entry.dc_format -%}
+<strong>Format:</strong> {{ entry.dc_format }}<br>
+{%- endif -%}
+{%- if entry.dc_relation -%}
+<strong>Related:</strong> {{ entry.dc_relation }}<br>
+{%- endif -%}
+{%- if entry.dc_coverage -%}
+<strong>Coverage:</strong> {{ entry.dc_coverage }}<br>
+{%- endif -%}
+{%- if entry.source and entry.source.title -%}
+<strong>Source:</strong> {{ entry.source.title }}
+{%- if entry.source.link %} (<a href="{{ entry.source.link }}">{{ entry.source.link }}</a>){% endif -%}
+<br>
+{%- endif -%}
+{%- if entry.dc_content -%}
+<strong>Content:</strong> {{ entry.dc_content | safe }}
+{%- elif entry.content and entry.content[0].value -%}
+<strong>Content:</strong> {{ entry.content[0].value | safe }}
+{%- elif entry.summary -%}
+<strong>Summary:</strong> {{ entry.summary | safe }}
+{%- endif -%}</article>
+"""
+
+
 def format_rss_items(rss_content: str, render_anchor_tag_content=False) -> str:
     """
-    Format RSS/Atom feed items in a readable text format using feedparser.
+    Format RSS/Atom feed items in a readable text format using feedparser and Jinja2.
 
-    Converts RSS <item> or Atom <entry> elements to formatted text with:
-    - <title> → <h1>Title</h1>
-    - <link> → Link: [url]
-    - <guid> → Guid: [id]
-    - <pubDate> → PubDate: [date]
-    - <description> or <content> → Raw HTML content (CDATA and entities automatically handled)
+    Converts RSS <item> or Atom <entry> elements to formatted text with all available fields:
+    - Basic fields: title, link, id/guid, published date, updated date
+    - Author fields: author, author_detail, contributors, publisher
+    - Content fields: content, summary, description
+    - Metadata: tags, category, rights, license
+    - Media: enclosures, media_content, media_thumbnail
+    - Dublin Core elements: dc:creator, dc:date, dc:publisher, etc. (mapped by feedparser)
 
     Args:
         rss_content: The RSS/Atom feed content
@@ -49,65 +168,19 @@ def format_rss_items(rss_content: str, render_anchor_tag_content=False) -> str:
     """
     try:
         import feedparser
-        from xml.sax.saxutils import escape as xml_escape
+        from changedetectionio.jinja2_custom import safe_jinja
 
         # Parse the feed - feedparser handles all RSS/Atom variants, CDATA, entity unescaping, etc.
         feed = feedparser.parse(rss_content)
 
-        formatted_items = []
-
-        # Determine feed type for appropriate labels when fields are missing
-        # feedparser sets feed.version to things like 'rss20', 'atom10', etc.
+        # Determine feed type for appropriate labels
         is_atom = feed.version and 'atom' in feed.version
 
+        formatted_items = []
         for entry in feed.entries:
-            item_parts = []
-
-            # Title - feedparser handles CDATA and entity unescaping automatically
-            if hasattr(entry, 'title') and entry.title:
-                item_parts.append(f'<h1>{xml_escape(entry.title)}</h1>')
-
-            # Link
-            if hasattr(entry, 'link') and entry.link:
-                item_parts.append(f'Link: {xml_escape(entry.link)}<br>')
-
-            # GUID/ID
-            if hasattr(entry, 'id') and entry.id:
-                item_parts.append(f'Guid: {xml_escape(entry.id)}<br>')
-
-            # Date - feedparser normalizes all date field names to 'published'
-            if hasattr(entry, 'published') and entry.published:
-                item_parts.append(f'PubDate: {xml_escape(entry.published)}<br>')
-
-            # Description/Content - feedparser handles CDATA and entity unescaping automatically
-            # Only add "Summary:" label for Atom <summary> tags
-            content = None
-            add_label = False
-
-            if hasattr(entry, 'content') and entry.content:
-                # Atom <content> - no label, just content
-                content = entry.content[0].value if entry.content[0].value else None
-            elif hasattr(entry, 'summary'):
-                # Could be RSS <description> or Atom <summary>
-                # feedparser maps both to entry.summary
-                content = entry.summary if entry.summary else None
-                # Only add "Summary:" label for Atom feeds (which use <summary> tag)
-                if is_atom:
-                    add_label = True
-
-            # Add content with or without label
-            if content:
-                if add_label:
-                    item_parts.append(f'Summary:<br>{content}')
-                else:
-                    item_parts.append(content)
-            else:
-                # No content - just show <none>
-                item_parts.append('&lt;none&gt;')
-
-            # Join all parts of this item
-            if item_parts:
-                formatted_items.append('\n'.join(item_parts))
+            # Render the entry using Jinja2 template
+            rendered = safe_jinja.render(RSS_ENTRY_TEMPLATE, entry=entry, is_atom=is_atom)
+            formatted_items.append(rendered.strip())
 
         # Wrap each item in a div with classes (first, last, item-N)
         items_html = []
@@ -122,7 +195,8 @@ def format_rss_items(rss_content: str, render_anchor_tag_content=False) -> str:
 
             class_str = ' '.join(classes)
             items_html.append(f'<div class="{class_str}">{item}</div>')
-        return '<html><body>\n'+"\n<br><br>".join(items_html)+'\n</body></html>'
+
+        return '<html><body>\n' + "\n<br>".join(items_html) + '\n</body></html>'
 
     except Exception as e:
         logger.warning(f"Error formatting RSS items: {str(e)}")

changedetectionio/run_basic_tests.sh

@@ -11,13 +11,60 @@ set -e
 
 SCRIPT_DIR=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )
 
-find tests/test_*py -type f|while read test_name
-do
-  echo "TEST RUNNING $test_name"
-  # REMOVE_REQUESTS_OLD_SCREENSHOTS disabled so that we can write a screenshot and send it in test_notifications.py without a real browser
-  REMOVE_REQUESTS_OLD_SCREENSHOTS=false pytest -vv -s --maxfail=1 --tb=long $test_name
-done
+# Since theres no curl installed lets roll with python3
+check_sanity() {
+  local port="$1"
+  if [ -z "$port" ]; then
+    echo "Usage: check_sanity <port>" >&2
+    return 1
+  fi
 
+  python3 - "$port" <<'PYCODE'
+import sys, time, urllib.request, socket
+
+port = sys.argv[1]
+url = f'http://localhost:{port}'
+ok = False
+
+for _ in range(6):  # --retry 6
+    try:
+        r = urllib.request.urlopen(url, timeout=3).read().decode()
+        if 'est-url-is-sanity' in r:
+            ok = True
+            break
+    except (urllib.error.URLError, ConnectionRefusedError, socket.error):
+        time.sleep(1)
+sys.exit(0 if ok else 1)
+PYCODE
+}
+
+data_sanity_test () {
+  # Restart data sanity test
+  cd ..
+  TMPDIR=$(mktemp -d)
+  PORT_N=$((5000 + RANDOM % (6501 - 5000)))
+  ./changedetection.py -p $PORT_N -d $TMPDIR -u "https://localhost?test-url-is-sanity=1" &
+  PID=$!
+  sleep 5
+  kill $PID
+  sleep 2
+  ./changedetection.py -p $PORT_N -d $TMPDIR &
+  PID=$!
+  sleep 5
+  # On a restart the URL should still be there
+  check_sanity $PORT_N || exit 1
+  kill $PID
+  cd $OLDPWD
+
+  # datastore looks alright, continue
+}
+
+data_sanity_test
+
+# REMOVE_REQUESTS_OLD_SCREENSHOTS disabled so that we can write a screenshot and send it in test_notifications.py without a real browser
+REMOVE_REQUESTS_OLD_SCREENSHOTS=false pytest -n 30 --dist load  tests/test_*.py
+
+#time pytest -n auto --dist loadfile -vv --tb=long tests/test_*.py
 echo "RUNNING WITH BASE_URL SET"
 
 # Now re-run some tests with BASE_URL enabled
@@ -43,4 +90,7 @@ FETCH_WORKERS=130 pytest  tests/test_history_consistency.py -v -l
 
 # Check file:// will pickup a file when enabled
 echo "Hello world" > /tmp/test-file.txt
-ALLOW_FILE_URI=yes pytest tests/test_security.py
+ALLOW_FILE_URI=yes pytest -vv -s  tests/test_security.py
+
+
+

changedetectionio/run_custom_browser_url_tests.sh

@@ -6,6 +6,8 @@
 
 # enable debug
 set -x
+docker network inspect changedet-network >/dev/null 2>&1 || docker network create changedet-network
+docker run --network changedet-network -d --hostname selenium  -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome:4
 
 # A extra browser is configured, but we never chose to use it, so it should NOT show in the logs
 docker run --rm -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/custom_browser_url/test_custom_browser_url.py::test_request_not_via_custom_browser_url'

changedetectionio/run_proxy_tests.sh

@@ -19,12 +19,13 @@ docker run --network changedet-network -d \
   -v `pwd`/tests/proxy_list/squid-passwords.txt:/etc/squid3/passwords \
   ubuntu/squid:4.13-21.10_edge
 
-
+sleep 5
 ## 2nd test actually choose the preferred proxy from proxies.json
+# This will force a request via "proxy-two"
 docker run --network changedet-network \
-  -v `pwd`/tests/proxy_list/proxies.json-example:/app/changedetectionio/test-datastore/proxies.json \
+  -v `pwd`/tests/proxy_list/proxies.json-example:/tmp/proxies.json \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_multiple_proxy.py'
+  bash -c 'cd changedetectionio && pytest -s tests/proxy_list/test_multiple_proxy.py --datastore-path /tmp'
 
 set +e
 echo "- Looking for chosen.changedetection.io request in squid-one - it should NOT be here"
@@ -48,8 +49,10 @@ fi
 # Test the UI configurable proxies
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_select_custom_proxy.py'
+  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_select_custom_proxy.py --datastore-path /tmp'
 
+# Give squid proxies a moment to flush their logs
+sleep 2
 
 # Should see a request for one.changedetection.io in there
 echo "- Looking for .changedetection.io request in squid-custom"
@@ -63,7 +66,10 @@ fi
 # Test "no-proxy" option
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_noproxy.py'
+  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_noproxy.py --datastore-path /tmp'
+
+# Give squid proxies a moment to flush their logs
+sleep 2
 
 # We need to handle grep returning 1
 set +e
@@ -80,6 +86,8 @@ for c in $(echo "squid-one squid-two squid-custom"); do
   fi
 done
 
+echo "docker ps output"
+docker ps
 
 docker kill squid-one squid-two squid-custom
 
@@ -88,19 +96,19 @@ docker kill squid-one squid-two squid-custom
 # Requests
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_proxy_noconnect.py'
+  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_proxy_noconnect.py --datastore-path /tmp'
 
 # Playwright
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py'
+  bash -c 'cd changedetectionio && PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py --datastore-path /tmp'
 
 # Puppeteer fast
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && FAST_PUPPETEER_CHROME_FETCHER=1 PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py'
+  bash -c 'cd changedetectionio && FAST_PUPPETEER_CHROME_FETCHER=1 PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py --datastore-path /tmp'
 
 # Selenium
 docker run --network changedet-network \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && WEBDRIVER_URL=http://selenium:4444/wd/hub pytest tests/proxy_list/test_proxy_noconnect.py'
+  bash -c 'cd changedetectionio && WEBDRIVER_URL=http://selenium:4444/wd/hub pytest tests/proxy_list/test_proxy_noconnect.py --datastore-path /tmp'

changedetectionio/run_socks_proxy_tests.sh

@@ -5,6 +5,7 @@ set -e
 # enable debug
 set -x
 
+docker network inspect changedet-network >/dev/null 2>&1 || docker network create changedet-network
 
 # SOCKS5 related - start simple Socks5 proxy server
 # SOCKSTEST=xyz should show in the logs of this service to confirm it fetched
@@ -14,13 +15,13 @@ docker run --network changedet-network -d --hostname socks5proxy-noauth --rm -p
 echo "---------------------------------- SOCKS5 -------------------"
 # SOCKS5 related - test from proxies.json
 docker run --network changedet-network \
-  -v `pwd`/tests/proxy_socks5/proxies.json-example:/app/changedetectionio/test-datastore/proxies.json \
+  -v `pwd`/tests/proxy_socks5/proxies.json-example:/tmp/proxies.json \
   --rm \
   -e "FLASK_SERVER_NAME=cdio" \
   --hostname cdio \
   -e "SOCKSTEST=proxiesjson" \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy_sources.py'
+  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy_sources.py  --datastore-path /tmp'
 
 # SOCKS5 related - by manually entering in UI
 docker run --network changedet-network \
@@ -29,18 +30,18 @@ docker run --network changedet-network \
   --hostname cdio \
   -e "SOCKSTEST=manual" \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy.py'
+  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy.py --datastore-path /tmp'
 
 # SOCKS5 related - test from proxies.json via playwright - NOTE- PLAYWRIGHT DOESNT SUPPORT AUTHENTICATING PROXY
 docker run --network changedet-network \
   -e "SOCKSTEST=manual-playwright" \
   --hostname cdio \
   -e "FLASK_SERVER_NAME=cdio" \
-  -v `pwd`/tests/proxy_socks5/proxies.json-example-noauth:/app/changedetectionio/test-datastore/proxies.json \
+  -v `pwd`/tests/proxy_socks5/proxies.json-example-noauth:/tmp/proxies.json \
   -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" \
   --rm \
   test-changedetectionio \
-  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy_sources.py'
+  bash -c 'cd changedetectionio && pytest --live-server-host=0.0.0.0 --live-server-port=5004  -s tests/proxy_socks5/test_socks5_proxy_sources.py --datastore-path /tmp'
 
 echo "socks5 server logs"
 docker logs socks5proxy

changedetectionio/static/js/notifications.js

@@ -14,10 +14,10 @@ $(document).ready(function () {
         e.preventDefault();
 
         data = {
-            notification_body: $('#notification_body').val(),
-            notification_format: $('#notification_format').val(),
-            notification_title: $('#notification_title').val(),
-            notification_urls: $('.notification-urls').val(),
+            notification_urls: $('textarea.notification-urls').val(),
+            notification_title: $('input.notification-title').val(),
+            notification_body: $('textarea.notification-body').val(),
+            notification_format: $('select.notification-format').val(),
             tags: $('#tags').val(),
             window_url: window.location.href,
         }

changedetectionio/static/styles/scss/styles.scss

@@ -329,12 +329,18 @@ a.pure-button-selected {
 .notifications-wrapper {
   padding-top: 0.5rem;
   #notification-test-log {
-    padding-top: 1rem;
+    margin-top: 1rem;
+    padding: 1rem;
     white-space: pre-wrap;
     word-break: break-word;
     overflow-wrap: break-word;
     max-width: 100%;
     box-sizing: border-box;
+    max-height: 12rem;
+    overflow-y: scroll;
+    border: 1px solid var(--color-border-notification);
+    border-radius: 5px;
+
   }
 }
 

changedetectionio/store.py

@@ -1,11 +1,14 @@
 from changedetectionio.strtobool import strtobool
 
+from changedetectionio.validate_url import is_safe_valid_url
+
 from flask import (
     flash
 )
 
+from .blueprint.rss import RSS_CONTENT_FORMAT_DEFAULT
 from .html_tools import TRANSLATE_WHITESPACE_TABLE
-from . model import App, Watch
+from .model import App, Watch, USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH
 from copy import deepcopy, copy
 from os import path, unlink
 from threading import Lock
@@ -20,6 +23,13 @@ import uuid as uuid_builder
 from loguru import logger
 from blinker import signal
 
+# Try to import orjson for faster JSON serialization
+try:
+    import orjson
+    HAS_ORJSON = True
+except ImportError:
+    HAS_ORJSON = False
+
 from .processors import get_custom_watch_obj_for_processor
 from .processors.restock_diff import Restock
 
@@ -35,22 +45,41 @@ class ChangeDetectionStore:
     lock = Lock()
     # For general updates/writes that can wait a few seconds
     needs_write = False
-
+    datastore_path = None
     # For when we edit, we should write to disk
     needs_write_urgent = False
 
     __version_check = True
+    save_data_thread = None
 
     def __init__(self, datastore_path="/datastore", include_default_watches=True, version_tag="0.0.0"):
         # Should only be active for docker
         # logging.basicConfig(filename='/dev/stdout', level=logging.INFO)
-        self.__data = App.model()
         self.datastore_path = datastore_path
-        self.json_store_path = os.path.join(self.datastore_path, "url-watches.json")
-        logger.info(f"Datastore path is '{self.json_store_path}'")
         self.needs_write = False
         self.start_time = time.time()
         self.stop_thread = False
+        self.save_version_copy_json_db(version_tag)
+        self.reload_state(datastore_path=datastore_path, include_default_watches=include_default_watches, version_tag=version_tag)
+
+    def save_version_copy_json_db(self, version_tag):
+        import re
+
+        version_text = re.sub(r'\D+', '-', version_tag)
+        db_path = os.path.join(self.datastore_path, "url-watches.json")
+        db_path_version_backup = os.path.join(self.datastore_path, f"url-watches-{version_text}.json")
+
+        if not os.path.isfile(db_path_version_backup) and os.path.isfile(db_path):
+            from shutil import copyfile
+            logger.info(f"Backing up JSON DB due to new version to '{db_path_version_backup}'.")
+            copyfile(db_path, db_path_version_backup)
+
+
+    def reload_state(self, datastore_path, include_default_watches, version_tag):
+        logger.info(f"Datastore path is '{datastore_path}'")
+
+        self.__data = App.model()
+        self.json_store_path = os.path.join(self.datastore_path, "url-watches.json")
         # Base definition for all watchers
         # deepcopy part of #569 - not sure why its needed exactly
         self.generic_definition = deepcopy(Watch.model(datastore_path = datastore_path, default={}))
@@ -62,37 +91,46 @@ class ChangeDetectionStore:
                 self.__data['build_sha'] = f.read()
 
         try:
-            # @todo retest with ", encoding='utf-8'"
-            with open(self.json_store_path) as json_file:
-                from_disk = json.load(json_file)
+            if HAS_ORJSON:
+                # orjson.loads() expects UTF-8 encoded bytes #3611
+                with open(self.json_store_path, 'rb') as json_file:
+                    from_disk = orjson.loads(json_file.read())
+            else:
+                with open(self.json_store_path, encoding='utf-8') as json_file:
+                    from_disk = json.load(json_file)
 
-                # @todo isnt there a way todo this dict.update recursively?
-                # Problem here is if the one on the disk is missing a sub-struct, it wont be present anymore.
-                if 'watching' in from_disk:
-                    self.__data['watching'].update(from_disk['watching'])
+            if not from_disk:
+                # No FileNotFound exception was thrown but somehow the JSON was empty - abort for safety.
+                logger.critical(f"JSON DB existed but was empty on load - empty JSON file? '{self.json_store_path}' Aborting")
+                raise Exception('JSON DB existed but was empty on load - Aborting')
 
-                if 'app_guid' in from_disk:
-                    self.__data['app_guid'] = from_disk['app_guid']
+            # @todo isnt there a way todo this dict.update recursively?
+            # Problem here is if the one on the disk is missing a sub-struct, it wont be present anymore.
+            if 'watching' in from_disk:
+                self.__data['watching'].update(from_disk['watching'])
 
-                if 'settings' in from_disk:
-                    if 'headers' in from_disk['settings']:
-                        self.__data['settings']['headers'].update(from_disk['settings']['headers'])
+            if 'app_guid' in from_disk:
+                self.__data['app_guid'] = from_disk['app_guid']
 
-                    if 'requests' in from_disk['settings']:
-                        self.__data['settings']['requests'].update(from_disk['settings']['requests'])
+            if 'settings' in from_disk:
+                if 'headers' in from_disk['settings']:
+                    self.__data['settings']['headers'].update(from_disk['settings']['headers'])
 
-                    if 'application' in from_disk['settings']:
-                        self.__data['settings']['application'].update(from_disk['settings']['application'])
+                if 'requests' in from_disk['settings']:
+                    self.__data['settings']['requests'].update(from_disk['settings']['requests'])
 
-                # Convert each existing watch back to the Watch.model object
-                for uuid, watch in self.__data['watching'].items():
-                    self.__data['watching'][uuid] = self.rehydrate_entity(uuid, watch)
-                    logger.info(f"Watching: {uuid} {watch['url']}")
+                if 'application' in from_disk['settings']:
+                    self.__data['settings']['application'].update(from_disk['settings']['application'])
 
-                # And for Tags also, should be Restock type because it has extra settings
-                for uuid, tag in self.__data['settings']['application']['tags'].items():
-                    self.__data['settings']['application']['tags'][uuid] = self.rehydrate_entity(uuid, tag, processor_override='restock_diff')
-                    logger.info(f"Tag: {uuid} {tag['title']}")
+            # Convert each existing watch back to the Watch.model object
+            for uuid, watch in self.__data['watching'].items():
+                self.__data['watching'][uuid] = self.rehydrate_entity(uuid, watch)
+                logger.info(f"Watching: {uuid} {watch['url']}")
+
+            # And for Tags also, should be Restock type because it has extra settings
+            for uuid, tag in self.__data['settings']['application']['tags'].items():
+                self.__data['settings']['application']['tags'][uuid] = self.rehydrate_entity(uuid, tag, processor_override='restock_diff')
+                logger.info(f"Tag: {uuid} {tag['title']}")
 
         # First time ran, Create the datastore.
         except (FileNotFoundError):
@@ -143,7 +181,10 @@ class ChangeDetectionStore:
         self.needs_write = True
 
         # Finally start the thread that will manage periodic data saves to JSON
-        save_data_thread = threading.Thread(target=self.save_datastore).start()
+        # Only start if thread is not already running (reload_state might be called multiple times)
+        if not self.save_data_thread or not self.save_data_thread.is_alive():
+            self.save_data_thread = threading.Thread(target=self.save_datastore)
+            self.save_data_thread.start()
 
     def rehydrate_entity(self, uuid, entity, processor_override=None):
         """Set the dict back to the dict Watch object"""
@@ -249,7 +290,8 @@ class ChangeDetectionStore:
                 self.__data['watching'] = {}
                 time.sleep(1) # Mainly used for testing to allow all items to flush before running next test
                 for uuid in self.data['watching']:
-                    path = pathlib.Path(os.path.join(self.datastore_path, uuid))
+                    path = pathlib.Path(
+                        os.path.join(self.datastore_path, uuid))
                     if os.path.exists(path):
                         self.delete(uuid)
 
@@ -340,9 +382,10 @@ class ChangeDetectionStore:
                 logger.error(f"Error fetching metadata for shared watch link {url} {str(e)}")
                 flash("Error fetching metadata for {}".format(url), 'error')
                 return False
-        from .model.Watch import is_safe_url
-        if not is_safe_url(url):
-            flash('Watch protocol is not permitted by SAFE_PROTOCOL_REGEX', 'error')
+
+        if not is_safe_valid_url(url):
+            flash('Watch protocol is not permitted or invalid URL format', 'error')
+
             return None
 
         if tag and type(tag) == str:
@@ -408,14 +451,19 @@ class ChangeDetectionStore:
             self.sync_to_json()
             return
         else:
-
             try:
                 # Re #286  - First write to a temp file, then confirm it looks OK and rename it
                 # This is a fairly basic strategy to deal with the case that the file is corrupted,
                 # system was out of memory, out of RAM etc
-                with open(self.json_store_path+".tmp", 'w') as json_file:
-                    # Use compact JSON in production for better performance
-                    json.dump(data, json_file, indent=2)
+                if HAS_ORJSON:
+                    # Use orjson for faster serialization
+                    # orjson.dumps() always returns UTF-8 encoded bytes #3611
+                    with open(self.json_store_path+".tmp", 'wb') as json_file:
+                        json_file.write(orjson.dumps(data, option=orjson.OPT_INDENT_2))
+                else:
+                    # Fallback to standard json module
+                    with open(self.json_store_path+".tmp", 'w', encoding='utf-8') as json_file:
+                        json.dump(data, json_file, indent=2, ensure_ascii=False)
                 os.replace(self.json_store_path+".tmp", self.json_store_path)
             except Exception as e:
                 logger.error(f"Error writing JSON!! (Main JSON file save was skipped) : {str(e)}")
@@ -438,7 +486,7 @@ class ChangeDetectionStore:
                 logger.remove()
                 logger.add(sys.stderr)
 
-                logger.critical("Shutting down datastore thread")
+                logger.info(f"Shutting down datastore '{self.datastore_path}' thread")
                 return
 
             if self.needs_write or self.needs_write_urgent:
@@ -477,8 +525,13 @@ class ChangeDetectionStore:
 
         # Load from external config file
         if path.isfile(proxy_list_file):
-            with open(os.path.join(self.datastore_path, "proxies.json")) as f:
-                proxy_list = json.load(f)
+            if HAS_ORJSON:
+                # orjson.loads() expects UTF-8 encoded bytes #3611
+                with open(os.path.join(self.datastore_path, "proxies.json"), 'rb') as f:
+                    proxy_list = orjson.loads(f.read())
+            else:
+                with open(os.path.join(self.datastore_path, "proxies.json"), encoding='utf-8') as f:
+                    proxy_list = json.load(f)
 
         # Mapping from UI config if available
         extras = self.data['settings']['requests'].get('extra_proxies')
@@ -723,6 +776,28 @@ class ChangeDetectionStore:
 
         return updates_available
 
+    def add_notification_url(self, notification_url):
+
+        logger.debug(f">>> Adding new notification_url - '{notification_url}'")
+
+        notification_urls = self.data['settings']['application'].get('notification_urls', [])
+
+        if notification_url in notification_urls:
+            return notification_url
+
+        with self.lock:
+            notification_urls = self.__data['settings']['application'].get('notification_urls', [])
+
+            if notification_url in notification_urls:
+                return notification_url
+
+            # Append and update the datastore
+            notification_urls.append(notification_url)
+            self.__data['settings']['application']['notification_urls'] = notification_urls
+            self.needs_write = True
+
+        return notification_url
+
     # Run all updates
     # IMPORTANT - Each update could be run even when they have a new install and the schema is correct
     #             So therefor - each `update_n` should be very careful about checking if it needs to actually run
@@ -735,7 +810,16 @@ class ChangeDetectionStore:
                 logger.critical(f"Applying update_{update_n}")
                 # Wont exist on fresh installs
                 if os.path.exists(self.json_store_path):
-                    shutil.copyfile(self.json_store_path, os.path.join(self.datastore_path, f"url-watches-before-{update_n}.json"))
+                    i = 0
+                    while True:
+                        i+=1
+                        dest = os.path.join(self.datastore_path, f"url-watches-before-{update_n}-{i}.json")
+                        if not os.path.exists(dest):
+                            logger.debug(f"Copying url-watches.json DB to '{dest}' backup.")
+                            shutil.copyfile(self.json_store_path, dest)
+                            break
+                        else:
+                            logger.warning(f"Backup of url-watches.json '{dest}', DB already exists, trying {i+1}.. ")
 
                 try:
                     update_method = getattr(self, f"update_{update_n}")()
@@ -987,29 +1071,54 @@ class ChangeDetectionStore:
             self.data['settings']['application']['ui']['use_page_title_in_list'] = self.data['settings']['application'].get('extract_title_as_title')
 
     def update_21(self):
-        self.data['settings']['application']['scheduler_timezone_default'] = self.data['settings']['application'].get('timezone')
-        del self.data['settings']['application']['timezone']
+        if self.data['settings']['application'].get('timezone'):
+            self.data['settings']['application']['scheduler_timezone_default'] = self.data['settings']['application'].get('timezone')
+            del self.data['settings']['application']['timezone']
 
 
-    def add_notification_url(self, notification_url):
-        
-        logger.debug(f">>> Adding new notification_url - '{notification_url}'")
+    # Some notification formats got the wrong name type
+    def update_23(self):
 
-        notification_urls = self.data['settings']['application'].get('notification_urls', [])
+        def re_run(formats):
+            sys_n_format = self.data['settings']['application'].get('notification_format')
+            key_exists_as_value = next((k for k, v in formats.items() if v == sys_n_format), None)
+            if key_exists_as_value:  # key of "Plain text"
+                logger.success(f"['settings']['application']['notification_format'] '{sys_n_format}' -> '{key_exists_as_value}'")
+                self.data['settings']['application']['notification_format'] = key_exists_as_value
 
-        if notification_url in notification_urls:
-            return notification_url
+            for uuid, watch in self.data['watching'].items():
+                n_format = self.data['watching'][uuid].get('notification_format')
+                key_exists_as_value = next((k for k, v in formats.items() if v == n_format), None)
+                if key_exists_as_value and key_exists_as_value != USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH:  # key of "Plain text"
+                    logger.success(f"['watching'][{uuid}]['notification_format'] '{n_format}' -> '{key_exists_as_value}'")
+                    self.data['watching'][uuid]['notification_format'] = key_exists_as_value  # should be 'text' or whatever
 
-        with self.lock:
-            notification_urls = self.__data['settings']['application'].get('notification_urls', [])
+            for uuid, tag in self.data['settings']['application']['tags'].items():
+                n_format = self.data['settings']['application']['tags'][uuid].get('notification_format')
+                key_exists_as_value = next((k for k, v in formats.items() if v == n_format), None)
+                if key_exists_as_value and key_exists_as_value != USE_SYSTEM_DEFAULT_NOTIFICATION_FORMAT_FOR_WATCH:  # key of "Plain text"
+                    logger.success(
+                        f"['settings']['application']['tags'][{uuid}]['notification_format'] '{n_format}' -> '{key_exists_as_value}'")
+                    self.data['settings']['application']['tags'][uuid][
+                        'notification_format'] = key_exists_as_value  # should be 'text' or whatever
 
-            if notification_url in notification_urls:
-                return notification_url
+        from .notification import valid_notification_formats
+        formats = deepcopy(valid_notification_formats)
+        re_run(formats)
+        # And in previous versions, it was "text" instead of Plain text, Markdown instead of "Markdown to HTML"
+        formats['text'] = 'Text'
+        formats['markdown'] = 'Markdown'
+        re_run(formats)
 
-            # Append and update the datastore
-            notification_urls.append(notification_url)
-            self.__data['settings']['application']['notification_urls'] = notification_urls
-            self.needs_write = True
-
-        return notification_url
 
+    # RSS types should be inline with the same names as notification types
+    def update_24(self):
+        rss_format = self.data['settings']['application'].get('rss_content_format')
+        if not rss_format or 'text' in rss_format:
+            # might have been 'plaintext, 'plain text' or something
+            self.data['settings']['application']['rss_content_format'] = RSS_CONTENT_FORMAT_DEFAULT
+        elif 'html' in rss_format:
+            self.data['settings']['application']['rss_content_format'] = 'htmlcolor'
+        else:
+            # safe fallback to text
+            self.data['settings']['application']['rss_content_format'] = RSS_CONTENT_FORMAT_DEFAULT

changedetectionio/templates/_common_fields.html

@@ -1,6 +1,118 @@
 
 {% from '_helpers.html' import render_field %}
 
+{% macro show_token_placeholders(extra_notification_token_placeholder_info, suffix="") %}
+
+
+    <div class="pure-controls">
+            <span class="pure-form-message-inline">
+        Body for all notifications &dash; You can use <a target="newwindow" href="https://jinja.palletsprojects.com/en/3.0.x/templates/">Jinja2</a> templating in the notification title, body and URL, and tokens from below.
+    </span><br>
+        <div data-target="#notification-tokens-info{{ suffix }}" class="toggle-show pure-button button-tag button-xsmall">Show
+            token/placeholders
+        </div>
+    </div>
+    <div class="pure-controls" style="display: none;" id="notification-tokens-info{{ suffix }}">
+        <table class="pure-table" id="token-table">
+            <thead>
+            <tr>
+                <th>Token</th>
+                <th>Description</th>
+            </tr>
+            </thead>
+            <tbody>
+            <tr>
+                <td><code>{{ '{{base_url}}' }}</code></td>
+                <td>The URL of the changedetection.io instance you are running.</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{watch_url}}' }}</code></td>
+                <td>The URL being watched.</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{watch_uuid}}' }}</code></td>
+                <td>The UUID of the watch.</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{watch_title}}' }}</code></td>
+                <td>The page title of the watch, uses &lt;title&gt; if not set, falls back to URL</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{watch_tag}}' }}</code></td>
+                <td>The watch group / tag</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{preview_url}}' }}</code></td>
+                <td>The URL of the preview page generated by changedetection.io.</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_url}}' }}</code></td>
+                <td>The URL of the diff output for the watch.</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff}}' }}</code></td>
+                <td>The diff output - only changes, additions, and removals</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_clean}}' }}</code></td>
+                <td>The diff output - only changes, additions, and removals &dash; <i>Without (added) prefix or colors</i>
+                </td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_added}}' }}</code></td>
+                <td>The diff output - only changes and additions</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_added_clean}}' }}</code></td>
+                <td>The diff output - only changes and additions &dash; <i>Without (added) prefix or colors</i></td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_removed}}' }}</code></td>
+                <td>The diff output - only changes and removals</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_removed_clean}}' }}</code></td>
+                <td>The diff output - only changes and removals &dash; <i>Without (added) prefix or colors</i></td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_full}}' }}</code></td>
+                <td>The diff output - full difference output</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_full_clean}}' }}</code></td>
+                <td>The diff output - full difference output &dash; <i>Without (added) prefix or colors</i></td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{diff_patch}}' }}</code></td>
+                <td>The diff output - patch in unified format</td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{current_snapshot}}' }}</code></td>
+                <td>The current snapshot text contents value, useful when combined with JSON or CSS filters
+                </td>
+            </tr>
+            <tr>
+                <td><code>{{ '{{triggered_text}}' }}</code></td>
+                <td>Text that tripped the trigger from filters</td>
+
+                {% if extra_notification_token_placeholder_info %}
+                    {% for token in extra_notification_token_placeholder_info %}
+                        <tr>
+                            <td><code>{{ '{{' }}{{ token[0] }}{{ '}}' }}</code></td>
+                            <td>{{ token[1] }}</td>
+                        </tr>
+                    {% endfor %}
+                {% endif %}
+            </tbody>
+        </table>
+
+        <span class="pure-form-message-inline">
+        Warning: Contents of <code>{{ '{{diff}}' }}</code>, <code>{{ '{{diff_removed}}' }}</code>, and <code>{{ '{{diff_added}}' }}</code> depend on how the difference algorithm perceives the change. <br>
+        For example, an addition or removal could be perceived as a change in some cases. <a target="newwindow" href="https://github.com/dgtlmoon/changedetection.io/wiki/Using-the-%7B%7Bdiff%7D%7D,-%7B%7Bdiff_added%7D%7D,-and-%7B%7Bdiff_removed%7D%7D-notification-tokens">More Here</a> <br>
+        </span>
+    </div>
+{%  endmacro %}
+
 {% macro render_common_settings_form(form, emailprefix, settings_application, extra_notification_token_placeholder_info) %}
                         <div class="pure-control-group">
                             {{ render_field(form.notification_urls, rows=5, placeholder="Examples:
@@ -40,103 +152,25 @@
                             </div>
                             <div class="pure-control-group">
                                 {{ render_field(form.notification_body , rows=5, class="notification-body", placeholder=settings_application['notification_body']) }}
-                                <span class="pure-form-message-inline">Body for all notifications &dash; You can use <a target="newwindow" href="https://jinja.palletsprojects.com/en/3.0.x/templates/">Jinja2</a> templating in the notification title, body and URL, and tokens from below.
-                                </span>
-
-                            </div>
-                            <div class="pure-controls">
-                                <div data-target="#notification-tokens-info" class="toggle-show pure-button button-tag button-xsmall">Show token/placeholders</div>
-                            </div>
-                            <div class="pure-controls" style="display: none;" id="notification-tokens-info">
-                                <table class="pure-table" id="token-table">
-                                    <thead>
-                                    <tr>
-                                        <th>Token</th>
-                                        <th>Description</th>
-                                    </tr>
-                                    </thead>
-                                    <tbody>
-                                    <tr>
-                                        <td><code>{{ '{{base_url}}' }}</code></td>
-                                        <td>The URL of the changedetection.io instance you are running.</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{watch_url}}' }}</code></td>
-                                        <td>The URL being watched.</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{watch_uuid}}' }}</code></td>
-                                        <td>The UUID of the watch.</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{watch_title}}' }}</code></td>
-                                        <td>The page title of the watch, uses &lt;title&gt; if not set, falls back to URL</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{watch_tag}}' }}</code></td>
-                                        <td>The watch group / tag</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{preview_url}}' }}</code></td>
-                                        <td>The URL of the preview page generated by changedetection.io.</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{diff_url}}' }}</code></td>
-                                        <td>The URL of the diff output for the watch.</td>
-                                    </tr>
-									<tr>
-                                        <td><code>{{ '{{diff}}' }}</code></td>
-                                        <td>The diff output - only changes, additions, and removals</td>
-                                    </tr>
-									<tr>
-                                        <td><code>{{ '{{diff_added}}' }}</code></td>
-                                        <td>The diff output - only changes and additions</td>
-                                    </tr>
-									<tr>
-                                        <td><code>{{ '{{diff_removed}}' }}</code></td>
-                                        <td>The diff output - only changes and removals</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{diff_full}}' }}</code></td>
-                                        <td>The diff output - full difference output</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{diff_patch}}' }}</code></td>
-                                        <td>The diff output - patch in unified format</td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{current_snapshot}}' }}</code></td>
-                                        <td>The current snapshot text contents value, useful when combined with JSON or CSS filters
-                                        </td>
-                                    </tr>
-                                    <tr>
-                                        <td><code>{{ '{{triggered_text}}' }}</code></td>
-                                        <td>Text that tripped the trigger from filters</td>
-
-                                        {% if extra_notification_token_placeholder_info %}
-                                            {% for token in extra_notification_token_placeholder_info %}
-                                                <tr>
-                                                    <td><code>{{ '{{' }}{{ token[0] }}{{ '}}' }}</code></td>
-                                                    <td>{{ token[1] }}</td>
-                                                </tr>
-                                            {% endfor %}
-                                        {% endif %}
-                                    </tbody>
-                                </table>
+                                {{ show_token_placeholders(extra_notification_token_placeholder_info=extra_notification_token_placeholder_info) }}
                                 <div class="pure-form-message-inline">
-                                    <p>
-									Warning: Contents of <code>{{ '{{diff}}' }}</code>, <code>{{ '{{diff_removed}}' }}</code>, and <code>{{ '{{diff_added}}' }}</code> depend on how the difference algorithm perceives the change. <br>
-                                    For example, an addition or removal could be perceived as a change in some cases. <a target="newwindow" href="https://github.com/dgtlmoon/changedetection.io/wiki/Using-the-%7B%7Bdiff%7D%7D,-%7B%7Bdiff_added%7D%7D,-and-%7B%7Bdiff_removed%7D%7D-notification-tokens">More Here</a> <br>
-                                    </p>
-                                    <p>
+                                    <ul>
+                                    <li><span class="pure-form-message-inline">
                                         For JSON payloads, use <strong>|tojson</strong> without quotes for automatic escaping, for example - <code>{ "name": {{ '{{ watch_title|tojson }}' }} }</code>
-                                    </p>
-                                    <p>
+                                    </span></li>
+                                    <li><span class="pure-form-message-inline">
                                         URL encoding, use <strong>|urlencode</strong>, for example - <code>gets://hook-website.com/test.php?title={{ '{{ watch_title|urlencode }}' }}</code>
-                                    </p>
-                                </div>
+                                    </span></li>
+                                    <li><span class="pure-form-message-inline">
+                                        Regular-expression replace, use <strong>|regex_replace</strong>, for example -   <code>{{ "{{ \"hello world 123\" | regex_replace('[0-9]+', 'no-more-numbers') }}" }}</code>
+                                    </span></li>
+                                    <li><span class="pure-form-message-inline">
+                                        For a complete reference of all Jinja2 built-in filters, users can refer to the <a href="https://jinja.palletsprojects.com/en/3.1.x/templates/#builtin-filters">https://jinja.palletsprojects.com/en/3.1.x/templates/#builtin-filters</a>
+                                    </span></li>
+                                    </ul>
+                                    <br>
                             </div>
-                            <div class="pure-control-group">
+                            <div class="">
                                 {{ render_field(form.notification_format , class="notification-format") }}
                                 <span class="pure-form-message-inline">Format for all notifications</span>
                             </div>

changedetectionio/templates/_helpers.html

@@ -266,9 +266,7 @@
             <li id="timezone-info">
                 {{ render_field(form.time_schedule_limit.timezone, placeholder=timezone_default_config) }} <span id="local-time-in-tz"></span>
                 <datalist id="timezones" style="display: none;">
-                    {% for timezone in available_timezones %}
-                        <option value="{{ timezone }}">{{ timezone }}</option>
-                    {% endfor %}
+                    {%- for timezone in available_timezones -%}<option value="{{ timezone }}">{{ timezone }}</option>{%- endfor -%}
                 </datalist>
             </li>
         </ul>

changedetectionio/templates/base.html

@@ -8,8 +8,13 @@
     <meta name="robots" content="noindex">
     <title>Change Detection{{extra_title}}</title>
     {% if app_rss_token %}
-      <link rel="alternate" type="application/rss+xml" title="Changedetection.io » Feed{% if active_tag_uuid %}- {{active_tag.title}}{% endif %}" href="{{ url_for('rss.feed', tag=active_tag_uuid , token=app_rss_token)}}" >
-    {% endif %}
+        <link rel="alternate" type="application/rss+xml" title="Changedetection.io » Feed{% if active_tag_uuid %}- {{active_tag.title}}{% endif %}" href="{{ url_for('rss.feed', tag=active_tag_uuid, token=app_rss_token, _external=True )}}" >
+
+        {% if rss_uuid_feed %}
+        <link rel="alternate" type="application/rss+xml" title="Feed » {{ rss_uuid_feed['label'] }}" href="{{ rss_uuid_feed['url'] }}" >
+
+        {%- endif -%}
+    {%- endif -%}
     <link rel="stylesheet" href="{{url_for('static_content', group='styles', filename='pure-min.css')}}" >
     <link rel="stylesheet" href="{{url_for('static_content', group='styles', filename='styles.css')}}?v={{ get_css_version() }}" >
     {% if extra_stylesheets %}
@@ -53,7 +58,7 @@
           <a class="pure-menu-heading" href="{{url_for('watchlist.index')}}">
             <strong>Change</strong>Detection.io</a>
         {% endif %}
-        {% if current_diff_url %}
+        {% if current_diff_url and is_safe_valid_url(current_diff_url) %}
           <a class="current-diff-url" href="{{ current_diff_url }}">
             <span style="max-width: 30%; overflow: hidden">{{ current_diff_url }}</span></a>
         {% else %}

changedetectionio/tests/conftest.py

@@ -11,6 +11,7 @@ import os
 import sys
 from loguru import logger
 
+from changedetectionio.flask_app import init_app_secret
 from changedetectionio.tests.util import live_server_setup, new_live_server_setup
 
 # https://github.com/pallets/flask/blob/1.1.2/examples/tutorial/tests/test_auth.py
@@ -87,7 +88,6 @@ def measure_memory_usage(request):
 def cleanup(datastore_path):
     import glob
     # Unlink test output files
-
     for g in ["*.txt", "*.json", "*.pdf"]:
         files = glob.glob(os.path.join(datastore_path, g))
         for f in files:
@@ -97,34 +97,121 @@ def cleanup(datastore_path):
             if os.path.isfile(f):
                 os.unlink(f)
 
-@pytest.fixture(scope='function', autouse=True)
-def prepare_test_function(live_server):
+def pytest_addoption(parser):
+    """Add custom command-line options for pytest.
 
+    Provides --datastore-path option for specifying custom datastore location.
+    Note: Cannot use -d short option as it's reserved by pytest for debug mode.
+    """
+    parser.addoption(
+        "--datastore-path",
+        action="store",
+        default=None,
+        help="Custom datastore path for tests"
+    )
+
+@pytest.fixture(scope='session')
+def datastore_path(tmp_path_factory, request):
+    """Provide datastore path unique to this worker.
+
+    Supports custom path via --datastore-path/-d flag (mirrors main app).
+
+    CRITICAL for xdist isolation:
+    - Each WORKER gets its own directory
+    - Tests on same worker run SEQUENTIALLY and cleanup between tests
+    - No subdirectories needed since tests don't overlap on same worker
+    - Example: /tmp/test-datastore-gw0/ for worker gw0
+    """
+    # Check for custom path first (mirrors main app's -d flag)
+    custom_path = request.config.getoption("--datastore-path")
+    if custom_path:
+        # Ensure the directory exists
+        os.makedirs(custom_path, exist_ok=True)
+        logger.info(f"Using custom datastore path: {custom_path}")
+        return custom_path
+
+    # Otherwise use default tmp_path_factory logic
+    worker_id = getattr(request.config, 'workerinput', {}).get('workerid', 'master')
+    if worker_id == 'master':
+        path = tmp_path_factory.mktemp("test-datastore")
+    else:
+        path = tmp_path_factory.mktemp(f"test-datastore-{worker_id}")
+    return str(path)
+
+
+@pytest.fixture(scope='function', autouse=True)
+def prepare_test_function(live_server, datastore_path):
+    """Prepare each test with complete isolation.
+
+    CRITICAL for xdist per-test isolation:
+    - Reuses the SAME datastore instance (so blueprint references stay valid)
+    - Clears all watches and state for a clean slate
+    - First watch will get uuid="first"
+    """
     routes = [rule.rule for rule in live_server.app.url_map.iter_rules()]
     if '/test-random-content-endpoint' not in routes:
         logger.debug("Setting up test URL routes")
         new_live_server_setup(live_server)
 
+    # CRITICAL: Point app to THIS test's unique datastore directory
+    live_server.app.config['TEST_DATASTORE_PATH'] = datastore_path
+
+    # CRITICAL: Get datastore and stop it from writing stale data
+    datastore = live_server.app.config.get('DATASTORE')
+
+    # Prevent background thread from writing during cleanup/reload
+    datastore.needs_write = False
+    datastore.needs_write_urgent = False
+
+    # CRITICAL: Clean up any files from previous tests
+    # This ensures a completely clean directory
+    cleanup(datastore_path)
+
+    # CRITICAL: Reload the EXISTING datastore instead of creating a new one
+    # This keeps blueprint references valid (they capture datastore at construction)
+    # reload_state() completely resets the datastore to a clean state
+
+    # Reload state with clean data (no default watches)
+    datastore.reload_state(
+        datastore_path=datastore_path,
+        include_default_watches=False,
+        version_tag=datastore.data.get('version_tag', '0.0.0')
+    )
+    live_server.app.secret_key = init_app_secret(datastore_path)
+    logger.debug(f"prepare_test_function: Reloaded datastore at {hex(id(datastore))}")
+    logger.debug(f"prepare_test_function: Path {datastore.datastore_path}")
 
     yield
-    # Then cleanup/shutdown
-    live_server.app.config['DATASTORE'].data['watching']={}
-    time.sleep(0.3)
-    live_server.app.config['DATASTORE'].data['watching']={}
+
+    # Cleanup: Clear watches again after test
+    try:
+        datastore.data['watching'] = {}
+        datastore.needs_write = True
+    except Exception as e:
+        logger.warning(f"Error during datastore cleanup: {e}")
+
+
+# So the app can also know which test name it was
+@pytest.fixture(autouse=True)
+def set_test_name(request):
+  """Automatically set TEST_NAME env var for every test"""
+  test_name = request.node.name
+  os.environ['PYTEST_CURRENT_TEST'] = test_name
+  yield
+  # Cleanup if needed
 
 
 @pytest.fixture(scope='session')
-def app(request):
-    """Create application for the tests."""
-    datastore_path = "./test-datastore"
+def app(request, datastore_path):
+    """Create application once per worker (session).
 
+    Note: Actual per-test isolation is handled by:
+    - prepare_test_function() recreates datastore and cleans directory
+    - All tests on same worker use same directory (cleaned between tests)
+    """
     # So they don't delay in fetching
     os.environ["MINIMUM_SECONDS_RECHECK_TIME"] = "0"
-    try:
-        os.mkdir(datastore_path)
-    except FileExistsError:
-        pass
-
+    logger.debug(f"Testing with datastore_path={datastore_path}")
     cleanup(datastore_path)
 
     app_config = {'datastore_path': datastore_path, 'disable_checkver' : True}
@@ -147,6 +234,8 @@ def app(request):
     # Disable CSRF while running tests
     app.config['WTF_CSRF_ENABLED'] = False
     app.config['STOP_THREADS'] = True
+    # Store datastore_path so Flask routes can access it
+    app.config['TEST_DATASTORE_PATH'] = datastore_path
 
     def teardown():
         # Stop all threads and services

changedetectionio/tests/custom_browser_url/test_custom_browser_url.py

@@ -73,13 +73,13 @@ def do_test(client, live_server, make_test_use_extra_browser=False):
 
 
 # Requires playwright to be installed
-def test_request_via_custom_browser_url(client, live_server, measure_memory_usage):
+def test_request_via_custom_browser_url(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     # We do this so we can grep the logs of the custom container and see if the request actually went through that container
     do_test(client, live_server, make_test_use_extra_browser=True)
 
 
-def test_request_not_via_custom_browser_url(client, live_server, measure_memory_usage):
+def test_request_not_via_custom_browser_url(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     # We do this so we can grep the logs of the custom container and see if the request actually went through that container
     do_test(client, live_server, make_test_use_extra_browser=False)

changedetectionio/tests/fetchers/test_content.py

@@ -8,7 +8,7 @@ import logging
 
 
 # Requires playwright to be installed
-def test_fetch_webdriver_content(client, live_server, measure_memory_usage):
+def test_fetch_webdriver_content(client, live_server, measure_memory_usage, datastore_path):
     #  live_server_setup(live_server) # Setup on conftest per function
 
     #####################

changedetectionio/tests/fetchers/test_custom_js_before_content.py

@@ -3,7 +3,7 @@ from flask import url_for
 from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
 
 
-def test_execute_custom_js(client, live_server, measure_memory_usage):
+def test_execute_custom_js(client, live_server, measure_memory_usage, datastore_path):
 
    #  live_server_setup(live_server) # Setup on conftest per function
     assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"

changedetectionio/tests/proxy_list/test_multiple_proxy.py

@@ -5,7 +5,7 @@ from flask import url_for
 from ..util import live_server_setup, wait_for_all_checks
 
 
-def test_preferred_proxy(client, live_server, measure_memory_usage):
+def test_preferred_proxy(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     url = "http://chosen.changedetection.io"
 

changedetectionio/tests/proxy_list/test_noproxy.py

@@ -5,7 +5,7 @@ from flask import url_for
 from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
 
 
-def test_noproxy_option(client, live_server, measure_memory_usage):
+def test_noproxy_option(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     # Run by run_proxy_tests.sh
     # Call this URL then scan the containers that it never went through them

changedetectionio/tests/proxy_list/test_proxy.py

@@ -5,7 +5,7 @@ from flask import url_for
 from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
 
 # just make a request, we will grep in the docker logs to see it actually got called
-def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage):
+def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     res = client.post(
         url_for("imports.import_page"),

changedetectionio/tests/proxy_list/test_proxy_noconnect.py

@@ -12,7 +12,7 @@ from ... import strtobool
 # FAST_PUPPETEER_CHROME_FETCHER=True PLAYWRIGHT_DRIVER_URL=ws://127.0.0.1:3000 pytest tests/proxy_list/test_proxy_noconnect.py
 # WEBDRIVER_URL=http://127.0.0.1:4444/wd/hub pytest tests/proxy_list/test_proxy_noconnect.py
 
-def test_proxy_noconnect_custom(client, live_server, measure_memory_usage):
+def test_proxy_noconnect_custom(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
     # Goto settings, add our custom one

changedetectionio/tests/proxy_list/test_select_custom_proxy.py

@@ -6,7 +6,7 @@ from ..util import live_server_setup, wait_for_all_checks
 import os
 
 # just make a request, we will grep in the docker logs to see it actually got called
-def test_select_custom(client, live_server, measure_memory_usage):
+def test_select_custom(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
     # Goto settings, add our custom one
@@ -50,7 +50,7 @@ def test_select_custom(client, live_server, measure_memory_usage):
     # Now we should see the request in the container logs for "squid-squid-custom" because it will be the only default
 
 
-def test_custom_proxy_validation(client, live_server, measure_memory_usage):
+def test_custom_proxy_validation(client, live_server, measure_memory_usage, datastore_path):
     #  live_server_setup(live_server) # Setup on conftest per function
 
     # Goto settings, add our custom one

changedetectionio/tests/proxy_socks5/test_socks5_proxy.py

@@ -5,7 +5,7 @@ from flask import url_for
 from changedetectionio.tests.util import live_server_setup, wait_for_all_checks, extract_UUID_from_client, delete_all_watches
 
 
-def set_response():
+def set_response(datastore_path):
     import time
     data = """<html>
        <body>
@@ -15,13 +15,13 @@ def set_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(data)
     time.sleep(1)
 
-def test_socks5(client, live_server, measure_memory_usage):
+def test_socks5(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
-    set_response()
+    set_response(datastore_path)
 
     # Setup a proxy
     res = client.post(

changedetectionio/tests/proxy_socks5/test_socks5_proxy_sources.py

@@ -4,7 +4,7 @@ from flask import url_for
 from changedetectionio.tests.util import live_server_setup, wait_for_all_checks
 
 
-def set_response():
+def set_response(datastore_path):
     import time
     data = """<html>
        <body>
@@ -14,15 +14,15 @@ def set_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(data)
     time.sleep(1)
 
 # should be proxies.json mounted from run_proxy_tests.sh already
 # -v `pwd`/tests/proxy_socks5/proxies.json-example:/app/changedetectionio/test-datastore/proxies.json
-def test_socks5_from_proxiesjson_file(client, live_server, measure_memory_usage):
+def test_socks5_from_proxiesjson_file(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
-    set_response()
+    set_response(datastore_path)
     # Because the socks server should connect back to us
     test_url = url_for('test_endpoint', _external=True) + f"?socks-test-tag={os.getenv('SOCKSTEST', '')}"
     test_url = test_url.replace('localhost.localdomain', 'cdio')

changedetectionio/tests/restock/test_restock.py

@@ -11,7 +11,7 @@ from changedetectionio.notification import (
 )
 
 
-def set_original_response():
+def set_original_response(datastore_path):
     test_return_data = """<html>
        <body>
        <section id=header style="padding: 50px; height: 350px">This is the header which should be ignored always - <span>add to cart</span></section>
@@ -26,13 +26,13 @@ def set_original_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
 
 
-def set_back_in_stock_response():
+def set_back_in_stock_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -45,14 +45,14 @@ def set_back_in_stock_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
 # Add a site in paused mode, add an invalid filter, we should still have visual selector data ready
-def test_restock_detection(client, live_server, measure_memory_usage):
+def test_restock_detection(client, live_server, measure_memory_usage, datastore_path):
 
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
     #assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
    #  live_server_setup(live_server) # Setup on conftest per function
     #####################
@@ -88,24 +88,25 @@ def test_restock_detection(client, live_server, measure_memory_usage):
     assert b'not-in-stock' in res.data # should be out of stock
 
     # Is it correctly shown as in stock
-    set_back_in_stock_response()
+    set_back_in_stock_response(datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
     assert b'not-in-stock' not in res.data
 
     # We should have a notification
-    wait_for_notification_endpoint_output()
-    assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
-    os.unlink("test-datastore/notification.txt")
+    notification_file = os.path.join(datastore_path, "notification.txt")
+    wait_for_notification_endpoint_output(datastore_path=datastore_path)
+    assert os.path.isfile(notification_file), "Notification received"
+    os.unlink(notification_file)
 
     # Default behaviour is to only fire notification when it goes OUT OF STOCK -> IN STOCK
     # So here there should be no file, because we go IN STOCK -> OUT OF STOCK
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     time.sleep(5)
-    assert not os.path.isfile("test-datastore/notification.txt"), "No notification should have fired when it went OUT OF STOCK by default"
+    assert not os.path.isfile(notification_file), "No notification should have fired when it went OUT OF STOCK by default"
 
     # BUT we should see that it correctly shows "not in stock"
     res = client.get(url_for("watchlist.index"))

changedetectionio/tests/smtp/smtp-test-server.py

@@ -1,51 +1,110 @@
 #!/usr/bin/env python3
-import asyncio
+import threading
+import time
 from aiosmtpd.controller import Controller
-from aiosmtpd.smtp import SMTP
+from flask import Flask, Response
+from email import message_from_bytes
+from email.policy import default
 
-# Accept a SMTP message and offer a way to retrieve the last message via TCP Socket
+# Accept a SMTP message and offer a way to retrieve the last message via HTTP
 
-last_received_message = b"Nothing"
+last_received_message = b"Nothing received yet."
+active_smtp_connections = 0
+smtp_lock = threading.Lock()
 
 
 class CustomSMTPHandler:
     async def handle_DATA(self, server, session, envelope):
-        global last_received_message
-        last_received_message = envelope.content
-        print('Receiving message from:', session.peer)
-        print('Message addressed from:', envelope.mail_from)
-        print('Message addressed to  :', envelope.rcpt_tos)
-        print('Message length        :', len(envelope.content))
-        print(envelope.content.decode('utf8'))
-        return '250 Message accepted for delivery'
+        global last_received_message, active_smtp_connections
+
+        with smtp_lock:
+            active_smtp_connections += 1
+
+        try:
+            last_received_message = envelope.content
+            print('Receiving message from:', session.peer)
+            print('Message addressed from:', envelope.mail_from)
+            print('Message addressed to  :', envelope.rcpt_tos)
+            print('Message length        :', len(envelope.content))
+            print('*******************************')
+            print(envelope.content.decode('utf8'))
+            print('*******************************')
+
+            # Parse the email message
+            msg = message_from_bytes(envelope.content, policy=default)
+            with open('/tmp/last.eml', 'wb') as f:
+                f.write(envelope.content)
+
+            # Write parts to files based on content type
+            if msg.is_multipart():
+                for part in msg.walk():
+                    content_type = part.get_content_type()
+                    payload = part.get_payload(decode=True)
+
+                    if payload:
+                        if content_type == 'text/plain':
+                            with open('/tmp/last.txt', 'wb') as f:
+                                f.write(payload)
+                            print(f'Written text/plain part to /tmp/last.txt')
+                        elif content_type == 'text/html':
+                            with open('/tmp/last.html', 'wb') as f:
+                                f.write(payload)
+                            print(f'Written text/html part to /tmp/last.html')
+            else:
+                # Single part message
+                content_type = msg.get_content_type()
+                payload = msg.get_payload(decode=True)
+
+                if payload:
+                    if content_type == 'text/plain' or content_type.startswith('text/'):
+                        with open('/tmp/last.txt', 'wb') as f:
+                            f.write(payload)
+                        print(f'Written single part message to /tmp/last.txt')
+
+            return '250 Message accepted for delivery'
+        finally:
+            with smtp_lock:
+                active_smtp_connections -= 1
 
 
-class EchoServerProtocol(asyncio.Protocol):
-    def connection_made(self, transport):
-        global last_received_message
-        self.transport = transport
-        peername = transport.get_extra_info('peername')
-        print('Incoming connection from {}'.format(peername))
-        self.transport.write(last_received_message)
-
-        last_received_message = b''
-        self.transport.close()
+# Simple Flask HTTP server to echo back the last SMTP message
+app = Flask(__name__)
 
 
-async def main():
+@app.route('/')
+def echo_last_message():
+    global last_received_message, active_smtp_connections
+
+    # Wait for any in-progress SMTP connections to complete
+    max_wait = 5  # Maximum 5 seconds
+    wait_interval = 0.05  # Check every 50ms
+    elapsed = 0
+
+    while elapsed < max_wait:
+        with smtp_lock:
+            if active_smtp_connections == 0:
+                break
+        time.sleep(wait_interval)
+        elapsed += wait_interval
+
+    return Response(last_received_message, mimetype='text/plain')
+
+
+def run_flask():
+    app.run(host='0.0.0.0', port=11080, debug=False, use_reloader=False)
+
+
+if __name__ == "__main__":
     # Start the SMTP server
     controller = Controller(CustomSMTPHandler(), hostname='0.0.0.0', port=11025)
     controller.start()
 
-    # Start the TCP Echo server
-    loop = asyncio.get_running_loop()
-    server = await loop.create_server(
-        lambda: EchoServerProtocol(),
-        '0.0.0.0', 11080
-    )
-    async with server:
-        await server.serve_forever()
+    # Start the HTTP server in a separate thread
+    flask_thread = threading.Thread(target=run_flask, daemon=True)
+    flask_thread.start()
 
-
-if __name__ == "__main__":
-    asyncio.run(main())
+    # Keep the main thread alive
+    try:
+        flask_thread.join()
+    except KeyboardInterrupt:
+        print("Shutting down...")

changedetectionio/tests/smtp/test_notification_smtp.py

@@ -3,7 +3,9 @@ from flask import url_for
 from email import message_from_string
 from email.policy import default as email_policy
 
-from changedetectionio.diff import HTML_REMOVED_STYLE, HTML_ADDED_STYLE
+from changedetectionio.diff import HTML_REMOVED_STYLE, HTML_ADDED_STYLE, HTML_CHANGED_STYLE, REMOVED_PLACEMARKER_OPEN, \
+    CHANGED_PLACEMARKER_OPEN, ADDED_PLACEMARKER_OPEN
+from changedetectionio.notification_service import NotificationContextData
 from changedetectionio.tests.util import set_original_response, set_modified_response, set_more_modified_response, live_server_setup, \
     wait_for_all_checks, \
     set_longer_modified_response, delete_all_watches
@@ -14,6 +16,8 @@ import logging
 # NOTE - RELIES ON mailserver as hostname running, see github build recipes
 smtp_test_server = 'mailserver'
 
+ALL_MARKUP_TOKENS = ''.join(f"TOKEN: '{t}'\n{{{{{t}}}}}\n" for t in NotificationContextData().keys())
+
 from changedetectionio.notification import (
     default_notification_body,
     default_notification_format,
@@ -24,24 +28,23 @@ from changedetectionio.notification import (
 
 
 def get_last_message_from_smtp_server():
-    import socket
-    port = 11080  # socket server port number
-
-    client_socket = socket.socket()  # instantiate
-    client_socket.connect((smtp_test_server, port))  # connect to the server
-
-    data = client_socket.recv(50024).decode()  # receive response
+    import requests
+    time.sleep(1) # wait for any smtp connects to die off
+    port = 11080  # HTTP server port number
+    # Make HTTP GET request to Flask server
+    response = requests.get(f'http://{smtp_test_server}:{port}/')
+    data = response.text
     logging.info("get_last_message_from_smtp_server..")
     logging.info(data)
-    client_socket.close()  # close the connection
     return data
 
 
 # Requires running the test SMTP server
 
-def test_check_notification_email_formats_default_HTML(client, live_server, measure_memory_usage):
+def test_check_notification_email_formats_default_HTML(client, live_server, measure_memory_usage, datastore_path):
     ##  live_server_setup(live_server) # Setup on conftest per function
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
+
 
     notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
 
@@ -52,7 +55,7 @@ def test_check_notification_email_formats_default_HTML(client, live_server, meas
         data={"application-notification_urls": notification_url,
               "application-notification_title": "fallback-title " + default_notification_title,
               "application-notification_body": "some text\nfallback-body<br> " + default_notification_body,
-              "application-notification_format": 'HTML',
+              "application-notification_format": 'html',
               "requests-time_between_check-minutes": 180,
               'application-fetch_backend': "html_requests"},
         follow_redirects=True
@@ -70,7 +73,7 @@ def test_check_notification_email_formats_default_HTML(client, live_server, meas
     assert b"Watch added" in res.data
 
     wait_for_all_checks(client)
-    set_longer_modified_response()
+    set_longer_modified_response(datastore_path=datastore_path)
     time.sleep(2)
 
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -109,8 +112,9 @@ def test_check_notification_email_formats_default_HTML(client, live_server, meas
     delete_all_watches(client)
 
 
-def test_check_notification_plaintext_format(client, live_server, measure_memory_usage):
-    set_original_response()
+def test_check_notification_plaintext_format(client, live_server, measure_memory_usage, datastore_path):
+    set_original_response(datastore_path=datastore_path)
+
 
     notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
 
@@ -119,9 +123,9 @@ def test_check_notification_plaintext_format(client, live_server, measure_memory
     res = client.post(
         url_for("settings.settings_page"),
         data={"application-notification_urls": notification_url,
-              "application-notification_title": "fallback-title " + default_notification_title,
-              "application-notification_body": "some text\n" + default_notification_body,
-              "application-notification_format": 'Plain Text',
+              "application-notification_title": "fallback-title {{watch_title}}  {{ diff_added.splitlines()[0] if diff_added else 'diff added didnt split' }}  " + default_notification_title,
+              "application-notification_body": f"some text\n" + default_notification_body + f"\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'text',
               "requests-time_between_check-minutes": 180,
               'application-fetch_backend': "html_requests"},
         follow_redirects=True
@@ -135,7 +139,7 @@ def test_check_notification_plaintext_format(client, live_server, measure_memory
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     time.sleep(2)
 
-    set_longer_modified_response()
+    set_longer_modified_response(datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
 
@@ -143,9 +147,18 @@ def test_check_notification_plaintext_format(client, live_server, measure_memory
 
     msg_raw = get_last_message_from_smtp_server()
     assert len(msg_raw) >= 1
-
+    #time.sleep(60)
     # Parse the email properly using Python's email library
     msg = message_from_string(msg_raw, policy=email_policy)
+    # Subject/title got marked up
+    subject = msg['subject']
+    # Subject should always be plaintext and never marked up to anything else
+    assert REMOVED_PLACEMARKER_OPEN not in subject
+    assert CHANGED_PLACEMARKER_OPEN not in subject
+    assert ADDED_PLACEMARKER_OPEN not in subject
+    assert 'diff added didnt split' not in subject
+    assert '(changed) Which is across' in subject
+    assert 'PLACEMARKER' not in subject
 
     # The email should be plain text only (not multipart)
     assert not msg.is_multipart()
@@ -161,8 +174,9 @@ def test_check_notification_plaintext_format(client, live_server, measure_memory
 
 
 
-def test_check_notification_html_color_format(client, live_server, measure_memory_usage):
-    set_original_response()
+def test_check_notification_html_color_format(client, live_server, measure_memory_usage, datastore_path):
+    set_original_response(datastore_path=datastore_path)
+
 
     notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
 
@@ -171,9 +185,9 @@ def test_check_notification_html_color_format(client, live_server, measure_memor
     res = client.post(
         url_for("settings.settings_page"),
         data={"application-notification_urls": notification_url,
-              "application-notification_title": "fallback-title " + default_notification_title,
-              "application-notification_body": "some text\n" + default_notification_body,
-              "application-notification_format": 'HTML Color',
+              "application-notification_title": "fallback-title {{watch_title}} - diff_added_lines_test : '{{ diff_added.splitlines()[0] if diff_added else 'diff added didnt split' }}' " + default_notification_title,
+              "application-notification_body": f"some text\n{default_notification_body}\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'htmlcolor',
               "requests-time_between_check-minutes": 180,
               'application-fetch_backend': "html_requests"},
         follow_redirects=True
@@ -192,7 +206,7 @@ def test_check_notification_html_color_format(client, live_server, measure_memor
     assert b"Watch added" in res.data
 
     wait_for_all_checks(client)
-    set_longer_modified_response()
+    set_longer_modified_response(datastore_path=datastore_path)
     time.sleep(2)
 
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -205,6 +219,19 @@ def test_check_notification_html_color_format(client, live_server, measure_memor
 
     # Parse the email properly using Python's email library
     msg = message_from_string(msg_raw, policy=email_policy)
+    # Subject/title got marked up
+    subject = msg['subject']
+    # Subject should always be plaintext and never marked up to anything else
+    assert REMOVED_PLACEMARKER_OPEN not in subject
+    assert CHANGED_PLACEMARKER_OPEN not in subject
+    assert ADDED_PLACEMARKER_OPEN not in subject
+    assert 'diff added didnt split' not in subject
+    assert '(changed) Which is across' in subject
+    assert 'PLACEMARKER' not in subject
+    assert 'head title' in subject
+    assert "span" not in subject
+    assert 'background-color' not in subject
+
 
     # The email should have two bodies (multipart/alternative with text/plain and text/html)
     assert msg.is_multipart()
@@ -225,14 +252,16 @@ def test_check_notification_html_color_format(client, live_server, measure_memor
     html_part = parts[1]
     assert html_part.get_content_type() == 'text/html'
     html_content = html_part.get_content()
-    assert HTML_REMOVED_STYLE in html_content
+    assert HTML_CHANGED_STYLE or HTML_REMOVED_STYLE in html_content
     assert HTML_ADDED_STYLE in html_content
+    assert '&lt;' not in html_content
 
     assert 'some text<br>' in html_content
     delete_all_watches(client)
 
-def test_check_notification_markdown_format(client, live_server, measure_memory_usage):
-    set_original_response()
+def test_check_notification_markdown_format(client, live_server, measure_memory_usage, datastore_path):
+    set_original_response(datastore_path=datastore_path)
+
 
     notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
 
@@ -241,9 +270,9 @@ def test_check_notification_markdown_format(client, live_server, measure_memory_
     res = client.post(
         url_for("settings.settings_page"),
         data={"application-notification_urls": notification_url,
-              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_title": "fallback-title  diff_added_lines_test : '{{ diff_added.splitlines()[0] if diff_added else 'diff added didnt split' }}' " + default_notification_title,
               "application-notification_body": "*header*\n\nsome text\n" + default_notification_body,
-              "application-notification_format": 'Markdown to HTML',
+              "application-notification_format": 'markdown',
               "requests-time_between_check-minutes": 180,
               'application-fetch_backend': "html_requests"},
         follow_redirects=True
@@ -262,7 +291,7 @@ def test_check_notification_markdown_format(client, live_server, measure_memory_
     assert b"Watch added" in res.data
 
     wait_for_all_checks(client)
-    set_longer_modified_response()
+    set_longer_modified_response(datastore_path=datastore_path)
     time.sleep(2)
 
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -279,6 +308,14 @@ def test_check_notification_markdown_format(client, live_server, measure_memory_
     # The email should have two bodies (multipart/alternative with text/plain and text/html)
     assert msg.is_multipart()
     assert msg.get_content_type() == 'multipart/alternative'
+    subject = msg['subject']
+    # Subject should always be plaintext and never marked up to anything else
+    assert REMOVED_PLACEMARKER_OPEN not in subject
+    assert CHANGED_PLACEMARKER_OPEN not in subject
+    assert ADDED_PLACEMARKER_OPEN not in subject
+    assert 'diff added didnt split' not in subject
+    assert '(changed) Which is across' in subject
+
 
     # Get the parts
     parts = list(msg.iter_parts())
@@ -288,7 +325,8 @@ def test_check_notification_markdown_format(client, live_server, measure_memory_
     text_part = parts[0]
     assert text_part.get_content_type() == 'text/plain'
     text_content = text_part.get_content()
-    assert '(added) So let\'s see what happens.\r\n' in text_content  # The plaintext part
+    # We wont see anything in the "FALLBACK" text but that's OK (no added/strikethrough etc)
+    assert 'So let\'s see what happens.\r\n' in text_content  # The plaintext part
 
 
     # Second part should be text/html and roughly converted from markdown to HTML
@@ -296,16 +334,20 @@ def test_check_notification_markdown_format(client, live_server, measure_memory_
     assert html_part.get_content_type() == 'text/html'
     html_content = html_part.get_content()
     assert '<p><em>header</em></p>' in html_content
-    assert '(added) So let\'s see what happens.<br' in html_content
+    assert '<strong>So let\'s see what happens.</strong><br />' in html_content # Additions are <strong> in markdown
+    # the '<br />' will come from apprises conversion, not from our code, we would rather use '<br>' correctly
+    # the '<br />' is actually a nice way to know if apprise done the conversion.
+
     delete_all_watches(client)
 
-
-def test_check_notification_email_formats_default_Text_override_HTML(client, live_server, measure_memory_usage):
+# Custom notification body with HTML, that is either sent as HTML or rendered to plaintext and sent
+def test_check_notification_email_formats_default_Text_override_HTML(client, live_server, measure_memory_usage, datastore_path):
 
     # HTML problems? see this
     # https://github.com/caronc/apprise/issues/633
 
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
+
     notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
     notification_body = f"""<!DOCTYPE html>
 <html lang="en">
@@ -326,7 +368,7 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
         data={"application-notification_urls": notification_url,
               "application-notification_title": "fallback-title " + default_notification_title,
               "application-notification_body": notification_body,
-              "application-notification_format": 'Plain Text',
+              "application-notification_format": 'text',
               "requests-time_between_check-minutes": 180,
               'application-fetch_backend': "html_requests"},
         follow_redirects=True
@@ -334,7 +376,7 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
     assert b"Settings updated." in res.data
 
     # Add a watch and trigger a HTTP POST
-    test_url = url_for('test_endpoint', _external=True)
+    test_url = url_for('test_endpoint',content_type="text/html", _external=True)
     res = client.post(
         url_for("ui.ui_views.form_quick_watch_add"),
         data={"url": test_url, "tags": 'nice one'},
@@ -343,8 +385,9 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
 
     assert b"Watch added" in res.data
 
+    #################################### FIRST SITUATION, PLAIN TEXT NOTIFICATION IS WANTED BUT WE HAVE HTML IN OUR TEMPLATE AND CONTENT ##########
     wait_for_all_checks(client)
-    set_longer_modified_response()
+    set_longer_modified_response(datastore_path=datastore_path)
     time.sleep(2)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
@@ -365,14 +408,18 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
     # Get the plain text content
     text_content = msg.get_content()
     assert '(added) So let\'s see what happens.\r\n' in text_content  # The plaintext part
+    assert '<!DOCTYPE html>' in text_content # even tho they added html, they selected plaintext so it should have not got converted
+
+
+    #################################### SECOND SITUATION, HTML IS CORRECTLY PASSED THROUGH TO THE EMAIL ####################
+    set_original_response(datastore_path=datastore_path)
 
-    set_original_response()
     # Now override as HTML format
     res = client.post(
         url_for("ui.ui_edit.edit_page", uuid="first"),
         data={
             "url": test_url,
-            "notification_format": 'HTML',
+            "notification_format": 'html',
             'fetch_backend': "html_requests",
             "time_between_check_use_default": "y"},
         follow_redirects=True
@@ -405,10 +452,346 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
     html_part = parts[1]
     assert html_part.get_content_type() == 'text/html'
     html_content = html_part.get_content()
-    assert '(removed) So let\'s see what happens.<br>' in html_content  # the html part
+    assert '(removed) So let\'s see what happens.' in html_content  # the html part
+    assert '&lt;!DOCTYPE html' not in html_content
+    assert '<!DOCTYPE html' in html_content # Our original template is working correctly
 
     # https://github.com/dgtlmoon/changedetection.io/issues/2103
     assert '<h1>Test</h1>' in html_content
     assert '&lt;' not in html_content
 
     delete_all_watches(client)
+
+def test_check_plaintext_document_plaintext_notification_smtp(client, live_server, measure_memory_usage, datastore_path):
+    """When following a plaintext document, notification in Plain Text format is sent correctly"""
+    import os
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("Some nice plain text\nwhich we add some extra data\nover here\n")
+
+    notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
+    notification_body = f"""{default_notification_body}"""
+
+    #####################
+    # Set this up for when we remove the notification from the watch, it should fallback with these details
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-notification_urls": notification_url,
+              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_body": f"{notification_body}\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'text',
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', content_type="text/plain", _external=True)
+    uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Change the content
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("Some nice plain text\nwhich we add some extra data\nAnd let's talk about <title> tags\nover here\n")
+
+
+    time.sleep(1)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Parse the email properly using Python's email library
+    msg = message_from_string(get_last_message_from_smtp_server(), policy=email_policy)
+
+    assert not msg.is_multipart()
+    assert msg.get_content_type() == 'text/plain'
+    body = msg.get_content()
+    # nothing is escaped, raw html stuff in text/plain
+    assert 'talk about <title> tags' in body
+    assert '(added)' in body
+    assert '<br' not in body
+    assert '&lt;' not in body
+    assert '<pre' not in body
+    delete_all_watches(client)
+
+def test_check_plaintext_document_html_notifications(client, live_server, measure_memory_usage, datastore_path):
+    """When following a plaintext document, notification in Plain Text format is sent correctly"""
+    import os
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("    Some nice plain text\nwhich we add some extra data\nover here\n")
+
+    notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
+    notification_body = f"""{default_notification_body}"""
+
+    #####################
+    # Set this up for when we remove the notification from the watch, it should fallback with these details
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-notification_urls": notification_url,
+              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_body": f"{notification_body}\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'html',
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', content_type="text/plain", _external=True)
+    uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Change the content
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("    Some nice plain text\nwhich we add some extra data\nAnd let's talk about <title> tags\nover here\n")
+
+
+    time.sleep(2)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Parse the email properly using Python's email library
+    msg = message_from_string(get_last_message_from_smtp_server(), policy=email_policy)
+
+
+    # The email should have two bodies (multipart/alternative)
+    assert msg.is_multipart()
+    assert msg.get_content_type() == 'multipart/alternative'
+
+    # Get the parts
+    parts = list(msg.iter_parts())
+    assert len(parts) == 2
+
+    # First part should be text/plain
+    text_part = parts[0]
+    assert text_part.get_content_type() == 'text/plain'
+    text_content = text_part.get_content()
+    html_part = parts[1]
+    assert html_part.get_content_type() == 'text/html'
+    html_content = html_part.get_content()
+
+
+    assert 'And let\'s talk about <title> tags\r\n' in text_content
+    assert '&lt;br' not in text_content
+    assert '<span' not in text_content
+
+
+    assert 'talk about <title>' not in html_content  # the html part, should have got marked up to &lt; etc
+    assert 'talk about &lt;title&gt;' in html_content
+    # Should be the HTML, but not HTML Color
+    assert 'background-color' not in html_content
+    assert '<br>(added) And let&#39;s talk about &lt;title&gt; tags<br>' in html_content
+    assert 'PLACEMARKER' not in html_content
+    assert '&lt;br' not in html_content
+    assert '<pre role="article"' in html_content # Should have got wrapped nicely in email_helpers.py
+
+    # And now for the whitespace retention
+    assert '&nbsp;&nbsp;&nbsp;&nbsp;Some nice plain text' in html_content
+    assert '(added) And let' in html_content # just to show a single whitespace didnt get touched
+    delete_all_watches(client)
+
+
+def test_check_plaintext_document_html_color_notifications(client, live_server, measure_memory_usage, datastore_path):
+    """When following a plaintext document, notification in Plain Text format is sent correctly"""
+    import os
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("Some nice plain text\nwhich we add some extra data\nover here\n")
+
+    notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
+    notification_body = f"""{default_notification_body}"""
+
+    #####################
+    # Set this up for when we remove the notification from the watch, it should fallback with these details
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-notification_urls": notification_url,
+              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_body": f"{notification_body}\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'htmlcolor',
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', content_type="text/plain", _external=True)
+    uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Change the content
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("Some nice plain text\nwhich we add some extra data\nAnd let's talk about <title> tags\nover here\n")
+
+    time.sleep(1)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    # Parse the email properly using Python's email library
+    msg = message_from_string(get_last_message_from_smtp_server(), policy=email_policy)
+
+    # The email should have two bodies (multipart/alternative)
+    assert msg.is_multipart()
+    assert msg.get_content_type() == 'multipart/alternative'
+
+    # Get the parts
+    parts = list(msg.iter_parts())
+    assert len(parts) == 2
+
+    # First part should be text/plain
+    text_part = parts[0]
+    assert text_part.get_content_type() == 'text/plain'
+    text_content = text_part.get_content()
+    html_part = parts[1]
+    assert html_part.get_content_type() == 'text/html'
+    html_content = html_part.get_content()
+
+
+    assert 'And let\'s talk about <title> tags\r\n' in text_content
+    assert '&lt;br' not in text_content
+    assert '<span' not in text_content
+
+    assert 'talk about <title>' not in html_content  # the html part, should have got marked up to &lt; etc
+    assert 'talk about &lt;title&gt;' in html_content
+    # Should be the HTML, but not HTML Color
+    assert 'background-color' in html_content
+    assert '(added) And let' not in html_content
+    assert '&lt;br' not in html_content
+    assert '<br>' in html_content
+    assert '<pre role="article"' in html_content # Should have got wrapped nicely in email_helpers.py
+    delete_all_watches(client)
+
+def test_check_html_document_plaintext_notification(client, live_server, measure_memory_usage, datastore_path):
+    """When following a HTML document, notification in Plain Text format is sent correctly"""
+    import os
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("<html><body>some stuff<br>and more stuff<br>and even more stuff<br></body></html>")
+
+    notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com'
+    notification_body = f"""{default_notification_body}"""
+
+    #####################
+    # Set this up for when we remove the notification from the watch, it should fallback with these details
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-notification_urls": notification_url,
+              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_body": f"{notification_body}\nMore output test\n{ALL_MARKUP_TOKENS}",
+              "application-notification_format": 'text',
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', content_type="text/html", _external=True)
+    uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write("<html><body>sxome stuff<br>and more stuff<br>lets slip this in<br>and this in<br>and even more stuff<br>&lt;tag&gt;</body></html>")
+
+    time.sleep(0.1)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+
+    # Parse the email properly using Python's email library
+    msg = message_from_string(get_last_message_from_smtp_server(), policy=email_policy)
+
+    assert not msg.is_multipart()
+    assert msg.get_content_type() == 'text/plain'
+    body = msg.get_content()
+
+    assert '<tag>' in body # Should have got converted from original HTML to plaintext
+    assert '(changed) some stuff\r\n' in body
+    assert 'PLACEMARKER' not in body
+    assert '(into) sxome stuff\r\n' in body
+    assert '(added) lets slip this in\r\n' in body
+    assert '(added) and this in\r\n' in body
+    assert '&nbsp;' not in body
+
+
+    delete_all_watches(client)
+
+
+def test_check_html_notification_with_apprise_format_is_html(client, live_server, measure_memory_usage, datastore_path):
+    ##  live_server_setup(live_server) # Setup on conftest per function
+    set_original_response(datastore_path=datastore_path)
+
+
+    notification_url = f'mailto://changedetection@{smtp_test_server}:11025/?to=fff@home.com&format=html'
+
+    #####################
+    # Set this up for when we remove the notification from the watch, it should fallback with these details
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-notification_urls": notification_url,
+              "application-notification_title": "fallback-title " + default_notification_title,
+              "application-notification_body": "some text\nfallback-body<br> " + default_notification_body,
+              "application-notification_format": 'html',
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+    assert b"Settings updated." in res.data
+
+    # Add a watch and trigger a HTTP POST
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("ui.ui_views.form_quick_watch_add"),
+        data={"url": test_url, "tags": 'nice one'},
+        follow_redirects=True
+    )
+
+    assert b"Watch added" in res.data
+
+    wait_for_all_checks(client)
+    set_longer_modified_response(datastore_path=datastore_path)
+    time.sleep(2)
+
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+
+    time.sleep(3)
+
+    msg_raw = get_last_message_from_smtp_server()
+    assert len(msg_raw) >= 1
+
+    # Parse the email properly using Python's email library
+    msg = message_from_string(msg_raw, policy=email_policy)
+
+    # The email should have two bodies (multipart/alternative with text/plain and text/html)
+    assert msg.is_multipart()
+    assert msg.get_content_type() == 'multipart/alternative'
+
+    # Get the parts
+    parts = list(msg.iter_parts())
+    assert len(parts) == 2
+
+    # First part should be text/plain (the auto-generated plaintext version)
+    text_part = parts[0]
+    assert text_part.get_content_type() == 'text/plain'
+    text_content = text_part.get_content()
+    assert '(added) So let\'s see what happens.\r\n' in text_content  # The plaintext part
+    assert 'fallback-body\r\n' in text_content  # The plaintext part
+
+    # Second part should be text/html
+    html_part = parts[1]
+    assert html_part.get_content_type() == 'text/html'
+    html_content = html_part.get_content()
+    assert 'some text<br>' in html_content  # We converted \n from the notification body
+    assert 'fallback-body<br>' in html_content  # kept the original <br>
+    assert '(added) So let\'s see what happens.<br>' in html_content  # the html part
+    delete_all_watches(client)

changedetectionio/tests/test_access_control.py

@@ -2,7 +2,7 @@ from .util import live_server_setup, wait_for_all_checks
 from flask import url_for
 import time
 
-def test_check_access_control(app, client, live_server, measure_memory_usage):
+def test_check_access_control(app, client, live_server, measure_memory_usage, datastore_path):
     # Still doesnt work, but this is closer.
    #  live_server_setup(live_server) # Setup on conftest per function
 

changedetectionio/tests/test_add_replace_remove_filter.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 
 import os.path
+import os
 
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks, wait_for_notification_endpoint_output, delete_all_watches
@@ -9,7 +10,7 @@ import time
 from ..diff import ADDED_PLACEMARKER_OPEN
 
 
-def set_original(excluding=None, add_line=None):
+def set_original(datastore_path, excluding=None, add_line=None):
     test_return_data = """<html>
      <body>
      <p>Some initial text</p>
@@ -35,16 +36,16 @@ def set_original(excluding=None, add_line=None):
 
         test_return_data = output
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
-# def test_setup(client, live_server, measure_memory_usage):
+# def test_setup(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
-def test_check_removed_line_contains_trigger(client, live_server, measure_memory_usage):
+def test_check_removed_line_contains_trigger(client, live_server, measure_memory_usage, datastore_path):
 
     # Give the endpoint time to spin up
-    set_original()
+    set_original(datastore_path=datastore_path)
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
     uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
@@ -64,9 +65,10 @@ def test_check_removed_line_contains_trigger(client, live_server, measure_memory
               "time_between_check_use_default": "y"},
         follow_redirects=True
     )
+
     assert b"Updated watch." in res.data
     wait_for_all_checks(client)
-    set_original(excluding='Something irrelevant')
+    set_original(excluding='Something irrelevant', datastore_path=datastore_path)
 
     # A line thats not the trigger should not trigger anything
     res = client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -77,7 +79,7 @@ def test_check_removed_line_contains_trigger(client, live_server, measure_memory
     assert b'has-unread-changes' not in res.data
 
     # The trigger line is REMOVED,  this should trigger
-    set_original(excluding='The golden line')
+    set_original(excluding='The golden line', datastore_path=datastore_path)
 
     # Check in the processor here what's going on, its triggering empty-reply and no change.
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -92,7 +94,7 @@ def test_check_removed_line_contains_trigger(client, live_server, measure_memory
     time.sleep(0.2)
 
     time.sleep(1)
-    set_original(excluding=None)
+    set_original(excluding=None, datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     time.sleep(1)
@@ -100,7 +102,7 @@ def test_check_removed_line_contains_trigger(client, live_server, measure_memory
     assert b'has-unread-changes' not in res.data
 
     # Remove it again, and we should get a trigger
-    set_original(excluding='The golden line')
+    set_original(excluding='The golden line', datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
@@ -109,7 +111,7 @@ def test_check_removed_line_contains_trigger(client, live_server, measure_memory
     delete_all_watches(client)
 
 
-def test_check_add_line_contains_trigger(client, live_server, measure_memory_usage):
+def test_check_add_line_contains_trigger(client, live_server, measure_memory_usage, datastore_path):
     
     delete_all_watches(client)
     time.sleep(1)
@@ -124,7 +126,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
               "application-notification_body": 'triggered text was -{{triggered_text}}- ### 网站监测 内容更新了 ####',
               # https://github.com/caronc/apprise/wiki/Notify_Custom_JSON#get-parameter-manipulation
               "application-notification_urls": test_notification_url,
-              "application-notification_format": 'Plain Text',
+              "application-notification_format": 'text',
               "application-minutes_between_check": 180,
               "application-fetch_backend": "html_requests"
               },
@@ -132,7 +134,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
     )
     assert b'Settings updated' in res.data
 
-    set_original()
+    set_original(datastore_path=datastore_path)
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
     uuid = client.application.config.get('DATASTORE').add_watch(url=test_url)
@@ -155,7 +157,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
     )
     assert b"Updated watch." in res.data
     wait_for_all_checks(client)
-    set_original(excluding='Something irrelevant')
+    set_original(excluding='Something irrelevant', datastore_path=datastore_path)
 
     # A line thats not the trigger should not trigger anything
     res = client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -166,7 +168,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
     assert b'has-unread-changes' not in res.data
 
     # The trigger line is ADDED,  this should trigger
-    set_original(add_line='<p>Oh yes please</p>')
+    set_original(add_line='<p>Oh yes please</p>', datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
@@ -174,9 +176,9 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
     assert b'has-unread-changes' in res.data
 
     # Takes a moment for apprise to fire
-    wait_for_notification_endpoint_output()
-    assert os.path.isfile("test-datastore/notification.txt"), "Notification fired because I can see the output file"
-    with open("test-datastore/notification.txt", 'rb') as f:
+    wait_for_notification_endpoint_output(datastore_path=datastore_path)
+    assert os.path.isfile(os.path.join(datastore_path, "notification.txt")), "Notification fired because I can see the output file"
+    with open(os.path.join(datastore_path, "notification.txt"), 'rb') as f:
         response = f.read()
         assert ADDED_PLACEMARKER_OPEN.encode('utf-8') not in response #  _apply_diff_filtering shouldnt add something here
         assert b'-Oh yes please' in response

changedetectionio/tests/test_api.py

@@ -3,12 +3,13 @@
 import time
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks, delete_all_watches
+import os
 
 import json
 import uuid
 
 
-def set_original_response():
+def set_original_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -21,12 +22,12 @@ def set_original_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
 
-def set_modified_response():
+def set_modified_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -39,7 +40,7 @@ def set_modified_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
     return None
@@ -52,17 +53,17 @@ def is_valid_uuid(val):
         return False
 
 
-# def test_setup(client, live_server, measure_memory_usage):
+# def test_setup(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
 
-def test_api_simple(client, live_server, measure_memory_usage):
+def test_api_simple(client, live_server, measure_memory_usage, datastore_path):
     
 
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
     # Create a watch
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
 
     # Validate bad URL
     test_url = url_for('test_endpoint', _external=True )
@@ -111,7 +112,7 @@ def test_api_simple(client, live_server, measure_memory_usage):
     time.sleep(1)
     wait_for_all_checks(client)
 
-    set_modified_response()
+    set_modified_response(datastore_path=datastore_path)
     # Trigger recheck of all ?recheck_all=1
     client.get(
         url_for("createwatch", recheck_all='1'),
@@ -244,7 +245,7 @@ def test_api_simple(client, live_server, measure_memory_usage):
     )
     assert len(res.json) == 0, "Watch list should be empty"
 
-def test_access_denied(client, live_server, measure_memory_usage):
+def test_access_denied(client, live_server, measure_memory_usage, datastore_path):
     # `config_api_token_enabled` Should be On by default
     res = client.get(
         url_for("createwatch")
@@ -289,11 +290,11 @@ def test_access_denied(client, live_server, measure_memory_usage):
     )
     assert b"Settings updated." in res.data
 
-def test_api_watch_PUT_update(client, live_server, measure_memory_usage):
+def test_api_watch_PUT_update(client, live_server, measure_memory_usage, datastore_path):
 
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
     # Create a watch
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
     test_url = url_for('test_endpoint', _external=True)
 
     # Create new
@@ -370,7 +371,7 @@ def test_api_watch_PUT_update(client, live_server, measure_memory_usage):
 
     ######################################################
 
-    # HTTP PUT try a field that doenst exist
+    # HTTP PUT try a field that doesn't exist
 
     # HTTP PUT an update
     res = client.put(
@@ -383,18 +384,30 @@ def test_api_watch_PUT_update(client, live_server, measure_memory_usage):
     # Message will come from `flask_expects_json`
     assert b'Additional properties are not allowed' in res.data
 
+
+    # Try a XSS URL
+    res = client.put(
+        url_for("watch", uuid=watch_uuid),
+        headers={'x-api-key': api_key, 'content-type': 'application/json'},
+        data=json.dumps({
+            'url': 'javascript:alert(document.domain)'
+        }),
+    )
+    assert res.status_code == 400
+
     # Cleanup everything
     delete_all_watches(client)
 
 
-def test_api_import(client, live_server, measure_memory_usage):
+def test_api_import(client, live_server, measure_memory_usage, datastore_path):
 
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
     res = client.post(
         url_for("import") + "?tag=import-test",
         data='https://website1.com\r\nhttps://website2.com',
-        headers={'x-api-key': api_key, 'content-type': 'text/plain'},
+        # We removed  'content-type': 'text/plain', the Import API should assume this if none is set #3547 #3542
+        headers={'x-api-key': api_key},
         follow_redirects=True
     )
 
@@ -408,7 +421,7 @@ def test_api_import(client, live_server, measure_memory_usage):
     res = client.get(url_for('tags.tags_overview_page'))
     assert b'import-test' in res.data
 
-def test_api_conflict_UI_password(client, live_server, measure_memory_usage):
+def test_api_conflict_UI_password(client, live_server, measure_memory_usage, datastore_path):
 
     
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
@@ -426,7 +439,7 @@ def test_api_conflict_UI_password(client, live_server, measure_memory_usage):
     assert b"Password protection enabled." in res.data
 
     # Create a watch
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
     test_url = url_for('test_endpoint', _external=True)
 
     # Create new

changedetectionio/tests/test_api_notifications.py

@@ -4,7 +4,7 @@ from flask import url_for
 from .util import live_server_setup
 import json
 
-def test_api_notifications_crud(client, live_server, measure_memory_usage):
+def test_api_notifications_crud(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 

changedetectionio/tests/test_api_openapi.py

@@ -12,7 +12,7 @@ from flask import url_for
 from .util import live_server_setup, wait_for_all_checks
 
 
-def test_openapi_validation_invalid_content_type_on_create_watch(client, live_server, measure_memory_usage):
+def test_openapi_validation_invalid_content_type_on_create_watch(client, live_server, measure_memory_usage, datastore_path):
     """Test that creating a watch with invalid content-type triggers OpenAPI validation error."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -29,7 +29,7 @@ def test_openapi_validation_invalid_content_type_on_create_watch(client, live_se
     assert b"OpenAPI validation failed" in res.data, "Should contain OpenAPI validation error message"
 
 
-def test_openapi_validation_missing_required_field_create_watch(client, live_server, measure_memory_usage):
+def test_openapi_validation_missing_required_field_create_watch(client, live_server, measure_memory_usage, datastore_path):
     """Test that creating a watch without required URL field triggers OpenAPI validation error."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -46,7 +46,7 @@ def test_openapi_validation_missing_required_field_create_watch(client, live_ser
     assert b"OpenAPI validation failed" in res.data, "Should contain OpenAPI validation error message"
 
 
-def test_openapi_validation_invalid_field_in_request_body(client, live_server, measure_memory_usage):
+def test_openapi_validation_invalid_field_in_request_body(client, live_server, measure_memory_usage, datastore_path):
     """Test that including invalid fields triggers OpenAPI validation error."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -83,7 +83,7 @@ def test_openapi_validation_invalid_field_in_request_body(client, live_server, m
     assert b"Additional properties are not allowed" in res.data, "Should contain validation error about additional properties"
 
 
-def test_openapi_validation_import_wrong_content_type(client, live_server, measure_memory_usage):
+def test_openapi_validation_import_wrong_content_type(client, live_server, measure_memory_usage, datastore_path):
     """Test that import endpoint with wrong content-type triggers OpenAPI validation error."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -100,7 +100,7 @@ def test_openapi_validation_import_wrong_content_type(client, live_server, measu
     assert b"OpenAPI validation failed" in res.data, "Should contain OpenAPI validation error message"
 
 
-def test_openapi_validation_import_correct_content_type_succeeds(client, live_server, measure_memory_usage):
+def test_openapi_validation_import_correct_content_type_succeeds(client, live_server, measure_memory_usage, datastore_path):
     """Test that import endpoint with correct content-type succeeds (positive test)."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -117,7 +117,7 @@ def test_openapi_validation_import_correct_content_type_succeeds(client, live_se
     assert len(res.json) == 2, "Should import 2 URLs"
 
 
-def test_openapi_validation_get_requests_bypass_validation(client, live_server, measure_memory_usage):
+def test_openapi_validation_get_requests_bypass_validation(client, live_server, measure_memory_usage, datastore_path):
     """Test that GET requests bypass OpenAPI validation entirely."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -141,7 +141,7 @@ def test_openapi_validation_get_requests_bypass_validation(client, live_server,
     assert isinstance(res.json, dict), "Should return JSON dictionary for watch list"
 
 
-def test_openapi_validation_create_tag_missing_required_title(client, live_server, measure_memory_usage):
+def test_openapi_validation_create_tag_missing_required_title(client, live_server, measure_memory_usage, datastore_path):
     """Test that creating a tag without required title triggers OpenAPI validation error."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 
@@ -158,7 +158,7 @@ def test_openapi_validation_create_tag_missing_required_title(client, live_serve
     assert b"OpenAPI validation failed" in res.data, "Should contain OpenAPI validation error message"
 
 
-def test_openapi_validation_watch_update_allows_partial_updates(client, live_server, measure_memory_usage):
+def test_openapi_validation_watch_update_allows_partial_updates(client, live_server, measure_memory_usage, datastore_path):
     """Test that watch updates allow partial updates without requiring all fields (positive test)."""
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 

changedetectionio/tests/test_api_search.py

@@ -6,7 +6,7 @@ import time
 from .util import live_server_setup, wait_for_all_checks
 
 
-def test_api_search(client, live_server, measure_memory_usage):
+def test_api_search(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
 

changedetectionio/tests/test_api_tags.py

@@ -5,13 +5,14 @@ from .util import live_server_setup, wait_for_all_checks, set_original_response
 import json
 import time
 
-def test_api_tags_listing(client, live_server, measure_memory_usage):
+def test_api_tags_listing(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
     api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
     tag_title = 'Test Tag'
 
 
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
+
 
     res = client.get(
         url_for("tags"),

changedetectionio/tests/test_auth.py

@@ -5,7 +5,7 @@ from flask import url_for
 from .util import live_server_setup, wait_for_all_checks
 
 # test pages with http://username@password:foobar.com/ work
-def test_basic_auth(client, live_server, measure_memory_usage):
+def test_basic_auth(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
 

changedetectionio/tests/test_automatic_follow_ldjson_price.py

@@ -3,9 +3,10 @@
 import time
 from flask import url_for
 from .util import live_server_setup, extract_UUID_from_client, wait_for_all_checks
+import os
 
 
-def set_response_with_ldjson():
+def set_response_with_ldjson(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -55,11 +56,11 @@ def set_response_with_ldjson():
      </html>
 """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
-def set_response_without_ldjson():
+def set_response_without_ldjson(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -72,17 +73,17 @@ def set_response_without_ldjson():
      </html>
 """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
-# def test_setup(client, live_server, measure_memory_usage):
+# def test_setup(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
 # actually only really used by the distll.io importer, but could be handy too
-def test_check_ldjson_price_autodetect(client, live_server, measure_memory_usage):
+def test_check_ldjson_price_autodetect(client, live_server, measure_memory_usage, datastore_path):
     
-    set_response_with_ldjson()
+    set_response_with_ldjson(datastore_path=datastore_path)
 
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
@@ -121,7 +122,7 @@ def test_check_ldjson_price_autodetect(client, live_server, measure_memory_usage
 
     ##########################################################################################
     # And we shouldnt see the offer
-    set_response_without_ldjson()
+    set_response_without_ldjson(datastore_path=datastore_path)
 
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
@@ -151,7 +152,7 @@ def _test_runner_check_bad_format_ignored(live_server, client, has_ldjson_price_
     client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
 
 
-def test_bad_ldjson_is_correctly_ignored(client, live_server, measure_memory_usage):
+def test_bad_ldjson_is_correctly_ignored(client, live_server, measure_memory_usage, datastore_path):
     
     test_return_data = """
             <html>
@@ -181,7 +182,7 @@ def test_bad_ldjson_is_correctly_ignored(client, live_server, measure_memory_usa
             <div class="yes">Some extra stuff</div>
             </body></html>
      """
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
     _test_runner_check_bad_format_ignored(live_server=live_server, client=client, has_ldjson_price_data=True)
@@ -215,7 +216,7 @@ def test_bad_ldjson_is_correctly_ignored(client, live_server, measure_memory_usa
     #         <div class="yes">Some extra stuff</div>
     #         </body></html>
     #  """
-    # with open("test-datastore/endpoint-content.txt", "w") as f:
+    # with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
     #     f.write(test_return_data)
     #
     # _test_runner_check_bad_format_ignored(live_server=live_server, client=client, has_ldjson_price_data=False)

changedetectionio/tests/test_backend.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+import os
 
 import time
 from flask import url_for
@@ -16,20 +17,11 @@ def test_inscriptus():
     assert stripped_text_from_html == 'test!\nok man'
 
 
-def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage):
-    set_original_response()
-   #  live_server_setup(live_server) # Setup on conftest per function
+def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage, datastore_path):
+    set_original_response(datastore_path=datastore_path)
 
-    # Add our URL to the import page
-    res = client.post(
-        url_for("imports.import_page"),
-        data={"urls": url_for('test_endpoint', _external=True)},
-        follow_redirects=True
-    )
+    uuid = client.application.config.get('DATASTORE').add_watch(url=url_for('test_endpoint', _external=True))
 
-    assert b"1 Imported" in res.data
-
-    wait_for_all_checks(client)
 
     # Do this a few times.. ensures we dont accidently set the status
     for n in range(3):
@@ -60,7 +52,7 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
     assert b'foobar-detection' not in res.data
 
     # Make a change
-    set_modified_response()
+    set_modified_response(datastore_path=datastore_path)
 
     # Force recheck
     res = client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -85,10 +77,9 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
     assert b'<rss' in res.data
 
     # re #16 should have the diff in here too
-    assert b'(into) which has this one new line' in res.data
+    assert b'which has this one new line' in res.data
     assert b'CDATA' in res.data
 
-    assert expected_url.encode('utf-8') in res.data
 #
     # Following the 'diff' link, it should no longer display as 'has-unread-changes' even after we recheck it a few times
     res = client.get(url_for("ui.ui_views.diff_history_page", uuid=uuid))
@@ -114,37 +105,19 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
         # It should report nothing found (no new 'has-unread-changes' class)
         res = client.get(url_for("watchlist.index"))
 
-
         assert b'has-unread-changes' not in res.data
         assert b'class="has-unread-changes' not in res.data
         assert b'head title' in res.data  # Should be ON by default
         assert b'test-endpoint' in res.data
 
     # Recheck it but only with a title change, content wasnt changed
-    set_original_response(extra_title=" and more")
+    set_original_response(datastore_path=datastore_path, extra_title=" and more")
 
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
     assert b'head title and more' in res.data
 
-    # disable <title> pickup
-    res = client.post(
-        url_for("settings.settings_page"),
-        data={"application-ui-use_page_title_in_list": "", "requests-time_between_check-minutes": 180,
-              'application-fetch_backend': "html_requests"},
-        follow_redirects=True
-    )
-
-    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
-    wait_for_all_checks(client)
-
-    res = client.get(url_for("watchlist.index"))
-    assert b'has-unread-changes' in res.data
-    assert b'class="has-unread-changes' in res.data
-    assert b'head title' not in res.data  # should now be off
-
-
     # Be sure the last_viewed is going to be greater than the last snapshot
     time.sleep(1)
 
@@ -165,9 +138,66 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
     # Cleanup everything
     delete_all_watches(client)
 
+def test_title_scraper(client, live_server, measure_memory_usage, datastore_path):
+
+    set_original_response(datastore_path=datastore_path)
+    uuid = client.application.config.get('DATASTORE').add_watch(url=url_for('test_endpoint', _external=True))
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks()
+
+    # It should report nothing found (no new 'has-unread-changes' class)
+    res = client.get(url_for("watchlist.index"))
+
+    assert b'head title' in res.data  # Should be ON by default
+
+    # Recheck it but only with a title change, content wasnt changed
+    set_original_response(datastore_path=datastore_path, extra_title=" and more")
+
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+    res = client.get(url_for("watchlist.index"))
+    assert b'head title and more' in res.data
+
+    # disable <title> pickup
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={"application-ui-use_page_title_in_list": "",
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+
+    set_original_response(datastore_path=datastore_path, extra_title=" SHOULD NOT APPEAR")
+
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks(client)
+    res = client.get(url_for("watchlist.index"))
+    assert b'SHOULD NOT APPEAR' not in res.data
+
+    delete_all_watches(client)
+
+def test_title_scraper_html_only(client, live_server, measure_memory_usage, datastore_path):
+
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
+        f.write('"My text document\nWhere I talk about <title>\nwhich should not get registered\n</title>')
+
+    test_url = url_for('test_endpoint', content_type="text/plain", _external=True)
+
+    uuid = client.application.config.get('DATASTORE').add_watch(test_url)
+    client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
+    wait_for_all_checks()
+
+    # It should report nothing found (no new 'has-unread-changes' class)
+    res = client.get(url_for("watchlist.index"))
+
+    assert b'which should not get registered' not in res.data  # Should be ON by default
+    assert not live_server.app.config['DATASTORE'].data['watching'][uuid].get('title')
+
+
+
 
 # Server says its plaintext, we should always treat it as plaintext, and then if they have a filter, try to apply that
-def test_requests_timeout(client, live_server, measure_memory_usage):
+def test_requests_timeout(client, live_server, measure_memory_usage, datastore_path):
     delay = 2
     test_url = url_for('test_endpoint', delay=delay, _external=True)
 
@@ -205,7 +235,7 @@ def test_requests_timeout(client, live_server, measure_memory_usage):
     res = client.get(url_for("watchlist.index"))
     assert b'Read timed out' not in res.data
 
-def test_non_text_mime_or_downloads(client, live_server, measure_memory_usage):
+def test_non_text_mime_or_downloads(client, live_server, measure_memory_usage, datastore_path):
     """
 
     https://github.com/dgtlmoon/changedetection.io/issues/3434
@@ -220,7 +250,7 @@ def test_non_text_mime_or_downloads(client, live_server, measure_memory_usage):
     :param measure_memory_usage:
     :return:
     """
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""some random text that should be split by line
 and not parsed with html_to_text
 this way we know that it correctly parsed as plain text
@@ -264,7 +294,7 @@ got it\r\n
     delete_all_watches(client)
 
 
-def test_standard_text_plain(client, live_server, measure_memory_usage):
+def test_standard_text_plain(client, live_server, measure_memory_usage, datastore_path):
     """
 
     https://github.com/dgtlmoon/changedetection.io/issues/3434
@@ -279,7 +309,7 @@ def test_standard_text_plain(client, live_server, measure_memory_usage):
     :param measure_memory_usage:
     :return:
     """
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""some random text that should be split by line
 and not parsed with html_to_text
 <title>Even this title should stay because we are just plain text</title>
@@ -325,9 +355,9 @@ got it\r\n
     delete_all_watches(client)
 
 # Server says its plaintext, we should always treat it as plaintext
-def test_plaintext_even_if_xml_content(client, live_server, measure_memory_usage):
+def test_plaintext_even_if_xml_content(client, live_server, measure_memory_usage, datastore_path):
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<?xml version="1.0" encoding="utf-8"?>
 <resources xmlns:tools="http://schemas.android.com/tools">
     <!--Activity and fragment titles-->
@@ -353,10 +383,10 @@ def test_plaintext_even_if_xml_content(client, live_server, measure_memory_usage
     delete_all_watches(client)
 
 # Server says its plaintext, we should always treat it as plaintext, and then if they have a filter, try to apply that
-def test_plaintext_even_if_xml_content_and_can_apply_filters(client, live_server, measure_memory_usage):
+def test_plaintext_even_if_xml_content_and_can_apply_filters(client, live_server, measure_memory_usage, datastore_path):
 
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<?xml version="1.0" encoding="utf-8"?>
 <resources xmlns:tools="http://schemas.android.com/tools">
     <!--Activity and fragment titles-->
@@ -375,7 +405,10 @@ def test_plaintext_even_if_xml_content_and_can_apply_filters(client, live_server
         follow_redirects=True
     )
 
-    assert b'&lt;string name=&#34;feed_update_receiver_name&#34;' in res.data
+    # Check that the string element with the correct name attribute is present
+    # Note: namespace declarations may be included when extracting elements, which is correct XML behavior
+    assert b'feed_update_receiver_name' in res.data
+    assert b'Abonnementen bijwerken' in res.data
     assert b'&lt;foobar' not in res.data
 
     res = client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)

changedetectionio/tests/test_backup.py

@@ -8,13 +8,11 @@ import re
 import time
 
 
-def test_backup(client, live_server, measure_memory_usage):
+def test_backup(client, live_server, measure_memory_usage, datastore_path):
    #  live_server_setup(live_server) # Setup on conftest per function
 
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
 
-    # Give the endpoint time to spin up
-    time.sleep(1)
 
     # Add our URL to the import page
     res = client.post(
@@ -31,7 +29,7 @@ def test_backup(client, live_server, measure_memory_usage):
         url_for("backups.request_backup"),
         follow_redirects=True
     )
-    time.sleep(2)
+    time.sleep(4)
 
     res = client.get(
         url_for("backups.index"),

changedetectionio/tests/test_basic_socketio.py

@@ -10,11 +10,12 @@ from .util import (
 )
 from loguru import logger
 
-def run_socketio_watch_update_test(client, live_server, password_mode=""):
+def run_socketio_watch_update_test(client, live_server, password_mode="", datastore_path=""):
     """Test that the socketio emits a watch update event when content changes"""
 
     # Set up the test server
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
+
 
     # Get the SocketIO instance from the app
     from changedetectionio.flask_app import app
@@ -47,7 +48,7 @@ def run_socketio_watch_update_test(client, live_server, password_mode=""):
     socketio_test_client.get_received()
 
     # Make a change to trigger an update
-    set_modified_response()
+    set_modified_response(datastore_path=datastore_path)
 
     # Force recheck
     res = client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -105,11 +106,11 @@ def run_socketio_watch_update_test(client, live_server, password_mode=""):
     # Clean up
     client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
 
-def test_everything(live_server, client):
+def test_everything(live_server, client, measure_memory_usage, datastore_path):
 
    #  live_server_setup(live_server) # Setup on conftest per function
 
-    run_socketio_watch_update_test(password_mode="", live_server=live_server, client=client)
+    run_socketio_watch_update_test(password_mode="", live_server=live_server, client=client, datastore_path=datastore_path)
 
     ############################ Password required auth check ##############################
 
@@ -124,7 +125,7 @@ def test_everything(live_server, client):
 
     assert b"Password protection enabled." in res.data
 
-    run_socketio_watch_update_test(password_mode="not logged in, should exit on connect", live_server=live_server, client=client)
+    run_socketio_watch_update_test(password_mode="not logged in, should exit on connect", live_server=live_server, client=client, datastore_path=datastore_path)
     res = client.post(
         url_for("login"),
         data={"password": "foobar"},
@@ -133,4 +134,4 @@ def test_everything(live_server, client):
 
     # Yes we are correctly logged in
     assert b"LOG OUT" in res.data
-    run_socketio_watch_update_test(password_mode="should be like normal", live_server=live_server, client=client)
+    run_socketio_watch_update_test(password_mode="should be like normal", live_server=live_server, client=client, datastore_path=datastore_path)

changedetectionio/tests/test_block_while_text_present.py

@@ -4,8 +4,9 @@ import time
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks, delete_all_watches
 from changedetectionio import html_tools
+import os
 
-def set_original_ignore_response():
+def set_original_ignore_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -17,11 +18,11 @@ def set_original_ignore_response():
 
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
 
-def set_modified_original_ignore_response():
+def set_modified_original_ignore_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some NEW nice initial text<br>
@@ -36,12 +37,12 @@ def set_modified_original_ignore_response():
 
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
 
 # Is the same but includes ZZZZZ, 'ZZZZZ' is the last line in ignore_text
-def set_modified_response_minus_block_text():
+def set_modified_response_minus_block_text(datastore_path):
     test_return_data = """<html>
        <body>
      Some NEW nice initial text<br>
@@ -56,16 +57,16 @@ def set_modified_response_minus_block_text():
 
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
 
-def test_check_block_changedetection_text_NOT_present(client, live_server, measure_memory_usage):
+def test_check_block_changedetection_text_NOT_present(client, live_server, measure_memory_usage, datastore_path):
 
    #  live_server_setup(live_server) # Setup on conftest per function
     # Use a mix of case in ZzZ to prove it works case-insensitive.
     ignore_text = "out of stoCk\r\nfoobar"
-    set_original_ignore_response()
+    set_original_ignore_response(datastore_path=datastore_path)
 
 
     # Add our URL to the import page
@@ -109,7 +110,7 @@ def test_check_block_changedetection_text_NOT_present(client, live_server, measu
     assert b'/test-endpoint' in res.data
 
     # The page changed, BUT the text is still there, just the rest of it changes, we should not see a change
-    set_modified_original_ignore_response()
+    set_modified_original_ignore_response(datastore_path=datastore_path)
 
     # Trigger a check
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -123,7 +124,7 @@ def test_check_block_changedetection_text_NOT_present(client, live_server, measu
 
     # 2548
     # Going back to the ORIGINAL should NOT trigger a change
-    set_original_ignore_response()
+    set_original_ignore_response(datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
@@ -131,10 +132,11 @@ def test_check_block_changedetection_text_NOT_present(client, live_server, measu
 
 
     # Now we set a change where the text is gone AND its different content, it should now trigger
-    set_modified_response_minus_block_text()
+    set_modified_response_minus_block_text(datastore_path=datastore_path)
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("watchlist.index"))
+
     assert b'has-unread-changes' in res.data
 
 

changedetectionio/tests/test_clone.py

@@ -3,12 +3,13 @@
 import time
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks
+import os
 
 
-def test_clone_functionality(client, live_server, measure_memory_usage):
+def test_clone_functionality(client, live_server, measure_memory_usage, datastore_path):
 
    #  live_server_setup(live_server) # Setup on conftest per function
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("<html><body>Some content</body></html>")
 
     test_url = url_for('test_endpoint', _external=True)

changedetectionio/tests/test_conditions.py

@@ -1,13 +1,14 @@
 #!/usr/bin/env python3
 import json
 import time
+import os
 
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks, delete_all_watches
 from ..model import CONDITIONS_MATCH_LOGIC_DEFAULT
 
 
-def set_original_response(number="50"):
+def set_original_response(datastore_path, number="50"):
     test_return_data = f"""<html>
        <body>
      <h1>Test Page for Conditions</h1>
@@ -17,10 +18,10 @@ def set_original_response(number="50"):
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
-def set_number_in_range_response(number="75"):
+def set_number_in_range_response(datastore_path, number="75"):
     test_return_data = f"""<html>
        <body>
      <h1>Test Page for Conditions</h1>
@@ -30,10 +31,10 @@ def set_number_in_range_response(number="75"):
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
-def set_number_out_of_range_response(number="150"):
+def set_number_out_of_range_response(datastore_path, number="150"):
     test_return_data = f"""<html>
        <body>
      <h1>Test Page for Conditions</h1>
@@ -43,18 +44,18 @@ def set_number_out_of_range_response(number="150"):
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
 
-# def test_setup(client, live_server, measure_memory_usage):
+# def test_setup(client, live_server, measure_memory_usage, datastore_path):
     """Test that both text and number conditions work together with AND logic."""
    #  live_server_setup(live_server) # Setup on conftest per function
 
-def test_conditions_with_text_and_number(client, live_server, measure_memory_usage):
+def test_conditions_with_text_and_number(client, live_server, measure_memory_usage, datastore_path):
     """Test that both text and number conditions work together with AND logic."""
     
-    set_original_response("50")
+    set_original_response(datastore_path=datastore_path, number="50")
     
 
     test_url = url_for('test_endpoint', _external=True)
@@ -114,7 +115,7 @@ def test_conditions_with_text_and_number(client, live_server, measure_memory_usa
     wait_for_all_checks(client)
 
     # Case 1
-    set_number_in_range_response("70.5")
+    set_number_in_range_response(datastore_path=datastore_path, number="70.5")
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
 
@@ -129,7 +130,7 @@ def test_conditions_with_text_and_number(client, live_server, measure_memory_usa
     client.get(url_for("ui.mark_all_viewed"), follow_redirects=True)
     time.sleep(0.2)
 
-    set_number_out_of_range_response("150.5")
+    set_number_out_of_range_response(datastore_path=datastore_path, number="150.5")
 
 
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -142,9 +143,9 @@ def test_conditions_with_text_and_number(client, live_server, measure_memory_usa
     delete_all_watches(client)
 
 # The 'validate' button next to each rule row
-def test_condition_validate_rule_row(client, live_server, measure_memory_usage):
+def test_condition_validate_rule_row(client, live_server, measure_memory_usage, datastore_path):
 
-    set_original_response("50")
+    set_original_response(datastore_path=datastore_path, number="50")
 
     test_url = url_for('test_endpoint', _external=True)
 
@@ -203,7 +204,7 @@ def test_condition_validate_rule_row(client, live_server, measure_memory_usage):
 
 
 # If there was only a change in the whitespacing, then we shouldnt have a change detected
-def test_wordcount_conditions_plugin(client, live_server, measure_memory_usage):
+def test_wordcount_conditions_plugin(client, live_server, measure_memory_usage, datastore_path):
     
 
     test_return_data = """<html>
@@ -216,7 +217,7 @@ def test_wordcount_conditions_plugin(client, live_server, measure_memory_usage):
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
     # Add our URL to the import page
@@ -242,10 +243,10 @@ def test_wordcount_conditions_plugin(client, live_server, measure_memory_usage):
     )
 
 # If there was only a change in the whitespacing, then we shouldnt have a change detected
-def test_lev_conditions_plugin(client, live_server, measure_memory_usage):
-    
+def test_lev_conditions_plugin(client, live_server, measure_memory_usage, datastore_path):
+    # This should break..
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<html>
        <body>
      Some initial text<br>
@@ -297,7 +298,7 @@ def test_lev_conditions_plugin(client, live_server, measure_memory_usage):
 
 
     ############### Now change it a LITTLE bit...
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<html>
        <body>
      Some initial text<br>
@@ -326,7 +327,7 @@ def test_lev_conditions_plugin(client, live_server, measure_memory_usage):
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     res = client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
     assert b'Queued 1 watch for rechecking.' in res.data

changedetectionio/tests/test_css_selector.py

@@ -3,12 +3,13 @@
 import time
 from flask import url_for
 from .util import live_server_setup, wait_for_all_checks
+import os
 
 from ..html_tools import *
 
 
 
-def set_original_response():
+def set_original_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -21,11 +22,11 @@ def set_original_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
     return None
 
-def set_modified_response():
+def set_modified_response(datastore_path):
     test_return_data = """<html>
        <body>
      Some initial text<br>
@@ -38,7 +39,7 @@ def set_modified_response():
      </html>
     """
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write(test_return_data)
 
     return None
@@ -69,12 +70,12 @@ def test_include_filters_output():
 
 
 # Tests the whole stack works with the CSS Filter
-def test_check_markup_include_filters_restriction(client, live_server, measure_memory_usage):
+def test_check_markup_include_filters_restriction(client, live_server, measure_memory_usage, datastore_path):
     sleep_time_for_fetch_thread = 3
 
     include_filters = "#sametext"
 
-    set_original_response()
+    set_original_response(datastore_path=datastore_path)
 
     # Give the endpoint time to spin up
     time.sleep(1)
@@ -105,7 +106,7 @@ def test_check_markup_include_filters_restriction(client, live_server, measure_m
     # Give the thread time to pick it up
     time.sleep(sleep_time_for_fetch_thread)
     #  Make a change
-    set_modified_response()
+    set_modified_response(datastore_path=datastore_path)
 
     # Trigger a check
     client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
@@ -119,11 +120,11 @@ def test_check_markup_include_filters_restriction(client, live_server, measure_m
 
 
 # Tests the whole stack works with the CSS Filter
-def test_check_multiple_filters(client, live_server, measure_memory_usage):
+def test_check_multiple_filters(client, live_server, measure_memory_usage, datastore_path):
     
     include_filters = "#blob-a\r\nxpath://*[contains(@id,'blob-b')]"
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<html><body>
      <div id="blob-a">Blob A</div>
      <div id="blob-b">Blob B</div>
@@ -168,12 +169,12 @@ def test_check_multiple_filters(client, live_server, measure_memory_usage):
 # The filter exists, but did not contain anything useful
 # Mainly used when the filter contains just an IMG, this can happen when someone selects an image in the visual-selector
 # Tests fetcher can throw a "ReplyWithContentButNoText" exception after applying filter and extracting text
-def test_filter_is_empty_help_suggestion(client, live_server, measure_memory_usage):
+def test_filter_is_empty_help_suggestion(client, live_server, measure_memory_usage, datastore_path):
     
 
     include_filters = "#blob-a"
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<html><body>
          <div id="blob-a">
            <img src="something.jpg">
@@ -216,7 +217,7 @@ def test_filter_is_empty_help_suggestion(client, live_server, measure_memory_usa
 
     ### Just an empty selector, no image
 
-    with open("test-datastore/endpoint-content.txt", "w") as f:
+    with open(os.path.join(datastore_path, "endpoint-content.txt"), "w") as f:
         f.write("""<html><body>
          <div id="blob-a">
            <!-- doo doo -->