Python 3.11, will it chooch?

.github/workflows/containers.yml

@@ -50,7 +50,6 @@ jobs:
           python -m pip install --upgrade pip
           pip install flake8 pytest
           if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
-          if [ -f requirements-dev.txt ]; then pip install -r requirements-dev.txt; fi
 
       - name: Create release metadata
         run: |

.github/workflows/pypi.yml

@@ -19,12 +19,6 @@ jobs:
         with:
           python-version: 3.9
 
-#      - name: Install dependencies
-#        run: |
-#          python -m pip install --upgrade pip
-#          pip install flake8 pytest
-#          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
-#          if [ -f requirements-dev.txt ]; then pip install -r requirements-dev.txt; fi
 
       - name: Test that pip builds without error
         run: |

.github/workflows/test-only.yml

@@ -8,32 +8,68 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v2
-      - name: Set up Python 3.9
+
+      # Mainly just for link/flake8
+      - name: Set up Python 3.10
         uses: actions/setup-python@v2
         with:
-          python-version: 3.9
-
-      - name: Install dependencies
-        run: |
-          python -m pip install --upgrade pip
-          pip install flake8 pytest
-          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
-          if [ -f requirements-dev.txt ]; then pip install -r requirements-dev.txt; fi
+          python-version: '3.10'
 
       - name: Lint with flake8
         run: |
+          pip3 install flake8
           # stop the build if there are Python syntax errors or undefined names
           flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
           # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
           flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
 
-      - name: Unit tests
+      - name: Spin up ancillary testable services
         run: |
-          python3 -m unittest changedetectionio.tests.unit.test_notification_diff
+          
+          docker network create changedet-network
 
-      - name: Test with pytest
+          # Selenium+browserless
+          docker run --network changedet-network -d --hostname selenium  -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome-debug:3.141.59
+          docker run --network changedet-network -d --hostname browserless -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm  -p 3000:3000  --shm-size="2g"  browserless/chrome:1.53-chrome-stable
+
+      - name: Build changedetection.io container for testing
+        run: |         
+          # Build a changedetection.io container and start testing inside
+          docker build . -t test-changedetectionio
+
+      - name: Test built container with pytest
         run: |
-          # Each test is totally isolated and performs its own cleanup/reset
-          cd changedetectionio; ./run_all_tests.sh
+          
+          # Unit tests
+          docker run test-changedetectionio  bash -c 'python3 -m unittest changedetectionio.tests.unit.test_notification_diff'
+          
+          # All tests
+          docker run --network changedet-network  test-changedetectionio  bash -c 'cd changedetectionio && ./run_basic_tests.sh'
 
+      - name: Test built container selenium+browserless/playwright
+        run: |
+          
+          # Selenium fetch
+          docker run -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'
+          
+          # Playwright/Browserless fetch
+          docker run -e "PLAYWRIGHT_DRIVER_URL=ws://browserless:3000" --network changedet-network  test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/fetchers/test_content.py && pytest tests/test_errorhandling.py && pytest tests/visualselector/test_fetch_data.py'
 
+      - name: Test proxy interaction
+        run: |
+          cd changedetectionio
+          ./run_proxy_tests.sh
+          cd ..
+          
+
+      - name: Test changedetection.io container starts+runs basically without error
+        run: |
+          docker run -p 5556:5000 -d test-changedetectionio
+          sleep 3
+          # Should return 0 (no error) when grep finds it
+          curl -s http://localhost:5556 |grep -q checkbox-uuid
+          curl -s http://localhost:5556/rss|grep -q rss-specification
+
+#export WEBDRIVER_URL=http://localhost:4444/wd/hub
+#pytest tests/fetchers/test_content.py
+#pytest tests/test_errorhandling.py

CONTRIBUTING.md

@@ -7,9 +7,3 @@ Otherwise, it's always best to PR into the `dev` branch.
 Please be sure that all new functionality has a matching test!
 
 Use `pytest` to validate/test, you can run the existing tests as `pytest tests/test_notification.py` for example
-
-```
-pip3 install -r requirements-dev
-```
-
-this is from https://github.com/dgtlmoon/changedetection.io/blob/master/requirements-dev.txt

Dockerfile

@@ -1,7 +1,7 @@
 # pip dependencies install stage
-FROM python:3.8-slim as builder
+FROM python:3.11-slim as builder
 
-# rustc compiler would be needed on ARM type devices but theres an issue with some deps not building..
+# See `cryptography` pin comment in requirements.txt
 ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -29,22 +29,16 @@ RUN pip install --target=/dependencies playwright~=1.27.1 \
     || echo "WARN: Failed to install Playwright. The application can still run, but the Playwright option will be disabled."
 
 # Final image stage
-FROM python:3.8-slim
+FROM python:3.11-slim
 
-# Actual packages needed at runtime, usually due to the notification (apprise) backend
-# rustc compiler would be needed on ARM type devices but theres an issue with some deps not building..
-ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1
-
-# Re #93, #73, excluding rustc (adds another 430Mb~)
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    g++ \
-    gcc \
-    libc-dev \
-    libffi-dev \
-    libjpeg-dev \
-    libssl-dev \
-    libxslt-dev \
-    zlib1g-dev
+    libssl1.1 \
+    libxslt1.1 \
+    # For pdftohtml
+    poppler-utils \
+    zlib1g \
+    && apt-get clean && rm -rf /var/lib/apt/lists/*
+
 
 # https://stackoverflow.com/questions/58701233/docker-logs-erroneously-appears-empty-until-container-stops
 ENV PYTHONUNBUFFERED=1

README.md

@@ -43,6 +43,7 @@ Requires Playwright to be enabled.
 
 - Products and services have a change in pricing
 - _Out of stock notification_ and _Back In stock notification_
+- Monitor and track PDF file changes, know when a PDF file has text changes.
 - Governmental department updates (changes are often only on their websites)
 - New software releases, security advisories when you're not on their mailing list.
 - Festivals with changes
@@ -68,6 +69,7 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
 - Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
 - Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JSONPath or jq
 - Switch between fast non-JS and Chrome JS based "fetchers"
+- Track changes in PDF files (Monitor text changed in the PDF, Also monitor PDF filesize and checksums)
 - Easily specify how often a site should be checked
 - Execute JS before extracting text (Good for logging in, see examples in the UI!)
 - Override Request Headers, Specify `POST` or `GET` and other methods
@@ -187,11 +189,29 @@ When you enable a `json:` or `jq:` filter, you can even automatically extract an
 <html>
 ...
 <script type="application/ld+json">
-  {"@context":"http://schema.org","@type":"Product","name":"Nan Optipro Stage 1 Baby Formula  800g","price": 23.50 }
+
+{
+   "@context":"http://schema.org/",
+   "@type":"Product",
+   "offers":{
+      "@type":"Offer",
+      "availability":"http://schema.org/InStock",
+      "price":"3949.99",
+      "priceCurrency":"USD",
+      "url":"https://www.newegg.com/p/3D5-000D-001T1"
+   },
+   "description":"Cobratype King Cobra Hero Desktop Gaming PC",
+   "name":"Cobratype King Cobra Hero Desktop Gaming PC",
+   "sku":"3D5-000D-001T1",
+   "itemCondition":"NewCondition"
+}
 </script>
 ```  
 
-`json:$.price` or `jq:.price` would give `23.50`, or you can extract the whole structure
+`json:$..price` or `jq:..price` would give `3949.99`, or you can extract the whole structure (use a JSONpath test website to validate with)
+
+The application also supports notifying you that it can follow this information automatically
+
 
 ## Proxy Configuration
 

changedetectionio/__init__.py

@@ -10,6 +10,7 @@ import threading
 import time
 import timeago
 
+from changedetectionio import queuedWatchMetaData
 from copy import deepcopy
 from distutils.util import strtobool
 from feedgen.feed import FeedGenerator
@@ -35,7 +36,7 @@ from flask_wtf import CSRFProtect
 from changedetectionio import html_tools
 from changedetectionio.api import api_v1
 
-__version__ = '0.40.0.1'
+__version__ = '0.40.0.3'
 
 datastore = None
 
@@ -404,7 +405,6 @@ def changedetection_app(config=None, datastore_o=None):
                 sorted_watches.append(watch)
 
         existing_tags = datastore.get_all_tags()
-
         form = forms.quickWatchForm(request.form)
         output = render_template("watch-overview.html",
                                  form=form,
@@ -416,7 +416,7 @@ def changedetection_app(config=None, datastore_o=None):
                                  # Don't link to hosting when we're on the hosting environment
                                  hosted_sticky=os.getenv("SALTED_PASS", False) == False,
                                  guid=datastore.data['app_guid'],
-                                 queued_uuids=[uuid for p,uuid in update_q.queue])
+                                 queued_uuids=[q_uuid.item['uuid'] for q_uuid in update_q.queue])
 
 
         if session.get('share-link'):
@@ -596,25 +596,16 @@ def changedetection_app(config=None, datastore_o=None):
                     using_default_check_time = False
                     break
 
-            # Use the default if its the same as system wide
+            # Use the default if it's the same as system-wide.
             if form.fetch_backend.data == datastore.data['settings']['application']['fetch_backend']:
                 extra_update_obj['fetch_backend'] = None
 
 
+
              # Ignore text
             form_ignore_text = form.ignore_text.data
             datastore.data['watching'][uuid]['ignore_text'] = form_ignore_text
 
-            # Reset the previous_md5 so we process a new snapshot including stripping ignore text.
-            if form_ignore_text:
-                if len(datastore.data['watching'][uuid].history):
-                    extra_update_obj['previous_md5'] = get_current_checksum_include_ignore_text(uuid=uuid)
-
-            # Reset the previous_md5 so we process a new snapshot including stripping ignore text.
-            if form.include_filters.data != datastore.data['watching'][uuid].get('include_filters', []):
-                if len(datastore.data['watching'][uuid].history):
-                    extra_update_obj['previous_md5'] = get_current_checksum_include_ignore_text(uuid=uuid)
-
             # Be sure proxy value is None
             if datastore.proxy_list is not None and form.data['proxy'] == '':
                 extra_update_obj['proxy'] = None
@@ -632,7 +623,7 @@ def changedetection_app(config=None, datastore_o=None):
             datastore.needs_write_urgent = True
 
             # Queue the watch for immediate recheck, with a higher priority
-            update_q.put((1, uuid))
+            update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
 
             # Diff page [edit] link should go back to diff page
             if request.args.get("next") and request.args.get("next") == 'diff':
@@ -773,7 +764,7 @@ def changedetection_app(config=None, datastore_o=None):
                 importer = import_url_list()
                 importer.run(data=request.values.get('urls'), flash=flash, datastore=datastore)
                 for uuid in importer.new_uuids:
-                    update_q.put((1, uuid))
+                    update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
 
                 if len(importer.remaining_data) == 0:
                     return redirect(url_for('index'))
@@ -786,7 +777,7 @@ def changedetection_app(config=None, datastore_o=None):
                 d_importer = import_distill_io_json()
                 d_importer.run(data=request.values.get('distill-io'), flash=flash, datastore=datastore)
                 for uuid in d_importer.new_uuids:
-                    update_q.put((1, uuid))
+                    update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
 
 
 
@@ -1151,7 +1142,7 @@ def changedetection_app(config=None, datastore_o=None):
 
         if not add_paused and new_uuid:
             # Straight into the queue.
-            update_q.put((1, new_uuid))
+            update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid}))
             flash("Watch added.")
 
         if add_paused:
@@ -1188,7 +1179,7 @@ def changedetection_app(config=None, datastore_o=None):
             uuid = list(datastore.data['watching'].keys()).pop()
 
         new_uuid = datastore.clone(uuid)
-        update_q.put((5, new_uuid))
+        update_q.put(queuedWatchMetaData.PrioritizedItem(priority=5, item={'uuid': new_uuid, 'skip_when_checksum_same': True}))
         flash('Cloned.')
 
         return redirect(url_for('index'))
@@ -1196,7 +1187,7 @@ def changedetection_app(config=None, datastore_o=None):
     @app.route("/api/checknow", methods=['GET'])
     @login_required
     def form_watch_checknow():
-
+        # Forced recheck will skip the 'skip if content is the same' rule (, 'reprocess_existing_data': True})))
         tag = request.args.get('tag')
         uuid = request.args.get('uuid')
         i = 0
@@ -1205,11 +1196,9 @@ def changedetection_app(config=None, datastore_o=None):
         for t in running_update_threads:
             running_uuids.append(t.current_uuid)
 
-        # @todo check thread is running and skip
-
         if uuid:
             if uuid not in running_uuids:
-                update_q.put((1, uuid))
+                update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
             i = 1
 
         elif tag != None:
@@ -1217,14 +1206,14 @@ def changedetection_app(config=None, datastore_o=None):
             for watch_uuid, watch in datastore.data['watching'].items():
                 if (tag != None and tag in watch['tag']):
                     if watch_uuid not in running_uuids and not datastore.data['watching'][watch_uuid]['paused']:
-                        update_q.put((1, watch_uuid))
+                        update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': watch_uuid, 'skip_when_checksum_same': False}))
                         i += 1
 
         else:
             # No tag, no uuid, add everything.
             for watch_uuid, watch in datastore.data['watching'].items():
                 if watch_uuid not in running_uuids and not datastore.data['watching'][watch_uuid]['paused']:
-                    update_q.put((1, watch_uuid))
+                    update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': watch_uuid, 'skip_when_checksum_same': False}))
                     i += 1
         flash("{} watches are queued for rechecking.".format(i))
         return redirect(url_for('index', tag=tag))
@@ -1271,6 +1260,14 @@ def changedetection_app(config=None, datastore_o=None):
                     datastore.data['watching'][uuid.strip()]['notification_muted'] = False
             flash("{} watches un-muted".format(len(uuids)))
 
+        elif (op == 'recheck'):
+            for uuid in uuids:
+                uuid = uuid.strip()
+                if datastore.data['watching'].get(uuid):
+                    # Recheck and require a full reprocessing
+                    update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
+
+            flash("{} watches un-muted".format(len(uuids)))
         elif (op == 'notification-default'):
             from changedetectionio.notification import (
                 default_notification_format_for_watch
@@ -1344,7 +1341,7 @@ def changedetection_app(config=None, datastore_o=None):
     app.register_blueprint(browser_steps.construct_blueprint(datastore), url_prefix='/browser-steps')
 
     import changedetectionio.blueprint.price_data_follower as price_data_follower
-    app.register_blueprint(price_data_follower.construct_blueprint(datastore), url_prefix='/price_data_follower')
+    app.register_blueprint(price_data_follower.construct_blueprint(datastore, update_q), url_prefix='/price_data_follower')
 
 
     # @todo handle ctrl break
@@ -1452,7 +1449,11 @@ def ticker_thread_check_time_launch_checks():
         watch_uuid_list = []
         while True:
             try:
-                watch_uuid_list = datastore.data['watching'].keys()
+                # Get a list of watches sorted by last_checked, [1] because it gets passed a tuple
+                # This is so we examine the most over-due first
+                for k in sorted(datastore.data['watching'].items(), key=lambda item: item[1].get('last_checked',0)):
+                    watch_uuid_list.append(k[0])
+
             except RuntimeError as e:
                 # RuntimeError: dictionary changed size during iteration
                 time.sleep(0.1)
@@ -1492,7 +1493,7 @@ def ticker_thread_check_time_launch_checks():
             seconds_since_last_recheck = now - watch['last_checked']
 
             if seconds_since_last_recheck >= (threshold + watch.jitter_seconds) and seconds_since_last_recheck >= recheck_time_minimum_seconds:
-                if not uuid in running_uuids and uuid not in [q_uuid for p,q_uuid in update_q.queue]:
+                if not uuid in running_uuids and uuid not in [q_uuid.item['uuid'] for q_uuid in update_q.queue]:
 
                     # Proxies can be set to have a limit on seconds between which they can be called
                     watch_proxy = datastore.get_preferred_proxy_for_watch(uuid=uuid)
@@ -1523,8 +1524,9 @@ def ticker_thread_check_time_launch_checks():
                             priority,
                             watch.jitter_seconds,
                             now - watch['last_checked']))
+
                     # Into the queue with you
-                    update_q.put((priority, uuid))
+                    update_q.put(queuedWatchMetaData.PrioritizedItem(priority=priority, item={'uuid': uuid, 'skip_when_checksum_same': True}))
 
                     # Reset for next time
                     watch.jitter_seconds = 0

changedetectionio/api/api_v1.py

@@ -1,3 +1,4 @@
+from changedetectionio import queuedWatchMetaData
 from flask_restful import abort, Resource
 from flask import request, make_response
 import validators
@@ -24,7 +25,7 @@ class Watch(Resource):
             abort(404, message='No watch exists with the UUID of {}'.format(uuid))
 
         if request.args.get('recheck'):
-            self.update_q.put((1, uuid))
+            self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
             return "OK", 200
 
         # Return without history, get that via another API call
@@ -100,7 +101,7 @@ class CreateWatch(Resource):
         extras = {'title': json_data['title'].strip()} if json_data.get('title') else {}
 
         new_uuid = self.datastore.add_watch(url=json_data['url'].strip(), tag=tag, extras=extras)
-        self.update_q.put((1, new_uuid))
+        self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': new_uuid, 'skip_when_checksum_same': True}))
         return {'uuid': new_uuid}, 201
 
     # Return concise list of available watches and some very basic info
@@ -118,7 +119,7 @@ class CreateWatch(Resource):
 
         if request.args.get('recheck_all'):
             for uuid in self.datastore.data['watching'].keys():
-                self.update_q.put((1, uuid))
+                self.update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': True}))
             return {'status': "OK"}, 200
 
         return list, 200

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -75,15 +75,13 @@ class steppable_browser_interface():
     def action_goto_url(self, url, optional_value):
         # self.page.set_viewport_size({"width": 1280, "height": 5000})
         now = time.time()
-        response = self.page.goto(url, timeout=0, wait_until='domcontentloaded')
-        print("Time to goto URL", time.time() - now)
+        response = self.page.goto(url, timeout=0, wait_until='commit')
 
         # Wait_until = commit
         # - `'commit'` - consider operation to be finished when network response is received and the document started loading.
         # Better to not use any smarts from Playwright and just wait an arbitrary number of seconds
         # This seemed to solve nearly all 'TimeoutErrors'
-        extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5))
-        self.page.wait_for_timeout(extra_wait * 1000)
+        print("Time to goto URL ", time.time() - now)
 
     def action_click_element_containing_text(self, selector=None, value=''):
         if not len(value.strip()):

changedetectionio/blueprint/price_data_follower/__init__.py

@@ -3,22 +3,28 @@ from distutils.util import strtobool
 from flask import Blueprint, flash, redirect, url_for
 from flask_login import login_required
 from changedetectionio.store import ChangeDetectionStore
+from changedetectionio import queuedWatchMetaData
+from queue import PriorityQueue
 
-def construct_blueprint(datastore: ChangeDetectionStore):
+PRICE_DATA_TRACK_ACCEPT = 'accepted'
+PRICE_DATA_TRACK_REJECT = 'rejected'
+
+def construct_blueprint(datastore: ChangeDetectionStore, update_q: PriorityQueue):
 
     price_data_follower_blueprint = Blueprint('price_data_follower', __name__)
 
     @login_required
     @price_data_follower_blueprint.route("/<string:uuid>/accept", methods=['GET'])
     def accept(uuid):
-        datastore.data['watching'][uuid]['track_ldjson_price_data'] = 'accepted'
+        datastore.data['watching'][uuid]['track_ldjson_price_data'] = PRICE_DATA_TRACK_ACCEPT
+        update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
         return redirect(url_for("form_watch_checknow", uuid=uuid))
 
 
     @login_required
     @price_data_follower_blueprint.route("/<string:uuid>/reject", methods=['GET'])
     def reject(uuid):
-        datastore.data['watching'][uuid]['track_ldjson_price_data'] = 'rejected'
+        datastore.data['watching'][uuid]['track_ldjson_price_data'] = PRICE_DATA_TRACK_REJECT
         return redirect(url_for("index"))
 
 

changedetectionio/content_fetcher.py

@@ -1,3 +1,4 @@
+import hashlib
 from abc import abstractmethod
 import chardet
 import json
@@ -23,6 +24,9 @@ class Non200ErrorCodeReceived(Exception):
             self.page_text = html_tools.html_to_text(page_html)
         return
 
+class checksumFromPreviousCheckWasTheSame(Exception):
+    def __init__(self):
+        return
 
 class JSActionExceptions(Exception):
     def __init__(self, status_code, url, screenshot, message=''):
@@ -39,7 +43,7 @@ class BrowserStepsStepTimout(Exception):
 
 
 class PageUnloadable(Exception):
-    def __init__(self, status_code, url, screenshot=False, message=False):
+    def __init__(self, status_code, url, message, screenshot=False):
         # Set this so we can use it in other parts of the app
         self.status_code = status_code
         self.url = url
@@ -113,7 +117,8 @@ class Fetcher():
             request_body,
             request_method,
             ignore_status_codes=False,
-            current_include_filters=None):
+            current_include_filters=None,
+            is_binary=False):
         # Should set self.error, self.status_code and self.content
         pass
 
@@ -238,6 +243,14 @@ class base_html_playwright(Fetcher):
         if proxy_override:
             self.proxy = {'server': proxy_override}
 
+        if self.proxy:
+            # Playwright needs separate username and password values
+            from urllib.parse import urlparse
+            parsed = urlparse(self.proxy.get('server'))
+            if parsed.username:
+                self.proxy['username'] = parsed.username
+                self.proxy['password'] = parsed.password
+
     def screenshot_step(self, step_n=''):
 
         # There's a bug where we need to do it twice or it doesnt take the whole page, dont know why.
@@ -264,7 +277,8 @@ class base_html_playwright(Fetcher):
             request_body,
             request_method,
             ignore_status_codes=False,
-            current_include_filters=None):
+            current_include_filters=None,
+            is_binary=False):
 
         from playwright.sync_api import sync_playwright
         import playwright._impl._api_types
@@ -286,6 +300,8 @@ class base_html_playwright(Fetcher):
                 proxy=self.proxy,
                 # This is needed to enable JavaScript execution on GitHub and others
                 bypass_csp=True,
+                # Can't think why we need the service workers for our use case?
+                service_workers='block',
                 # Should never be needed
                 accept_downloads=False
             )
@@ -294,24 +310,34 @@ class base_html_playwright(Fetcher):
             if len(request_headers):
                 context.set_extra_http_headers(request_headers)
 
-            try:
                 self.page.set_default_navigation_timeout(90000)
                 self.page.set_default_timeout(90000)
 
                 # Listen for all console events and handle errors
                 self.page.on("console", lambda msg: print(f"Playwright console: Watch URL: {url} {msg.type}: {msg.text} {msg.args}"))
 
-                # Bug - never set viewport size BEFORE page.goto
-
-
-                # Waits for the next navigation. Using Python context manager
-                # prevents a race condition between clicking and waiting for a navigation.
-                with self.page.expect_navigation():
-                    response = self.page.goto(url, wait_until='load')
+            # Goto page
+            try:
                 # Wait_until = commit
                 # - `'commit'` - consider operation to be finished when network response is received and the document started loading.
                 # Better to not use any smarts from Playwright and just wait an arbitrary number of seconds
                 # This seemed to solve nearly all 'TimeoutErrors'
+                response = self.page.goto(url, wait_until='commit')
+            except playwright._impl._api_types.Error as e:
+                # Retry once - https://github.com/browserless/chrome/issues/2485
+                # Sometimes errors related to invalid cert's and other can be random
+                print ("Content Fetcher > retrying request got error - ", str(e))
+                time.sleep(1)
+                response = self.page.goto(url, wait_until='commit')
+
+            except Exception as e:
+                print ("Content Fetcher > Other exception when page.goto", str(e))
+                context.close()
+                browser.close()
+                raise PageUnloadable(url=url, status_code=None, message=str(e))
+
+            # Execute any browser steps
+            try:
                 extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
                 self.page.wait_for_timeout(extra_wait * 1000)
 
@@ -324,17 +350,15 @@ class base_html_playwright(Fetcher):
                 # This can be ok, we will try to grab what we could retrieve
                 pass
             except Exception as e:
-                print ("other exception when page.goto")
-                print (str(e))
+                print ("Content Fetcher > Other exception when executing custom JS code", str(e))
                 context.close()
                 browser.close()
-                raise PageUnloadable(url=url, status_code=None)
-
+                raise PageUnloadable(url=url, status_code=None, message=str(e))
 
             if response is None:
                 context.close()
                 browser.close()
-                print ("response object was none")
+                print ("Content Fetcher > Response object was none")
                 raise EmptyReply(url=url, status_code=None)
 
             # Bug 2(?) Set the viewport size AFTER loading the page
@@ -353,8 +377,8 @@ class base_html_playwright(Fetcher):
             if len(self.page.content().strip()) == 0:
                 context.close()
                 browser.close()
-                print ("Content was empty")
-                raise EmptyReply(url=url, status_code=None)
+                print ("Content Fetcher > Content was empty")
+                raise EmptyReply(url=url, status_code=response.status)
 
             # Bug 2(?) Set the viewport size AFTER loading the page
             self.page.set_viewport_size({"width": 1280, "height": 1024})
@@ -440,7 +464,8 @@ class base_html_webdriver(Fetcher):
             request_body,
             request_method,
             ignore_status_codes=False,
-            current_include_filters=None):
+            current_include_filters=None,
+            is_binary=False):
 
         from selenium import webdriver
         from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
@@ -498,7 +523,7 @@ class base_html_webdriver(Fetcher):
             try:
                 self.driver.quit()
             except Exception as e:
-                print("Exception in chrome shutdown/quit" + str(e))
+                print("Content Fetcher > Exception in chrome shutdown/quit" + str(e))
 
 
 # "html_requests" is listed as the default fetcher in store.py!
@@ -515,7 +540,8 @@ class html_requests(Fetcher):
             request_body,
             request_method,
             ignore_status_codes=False,
-            current_include_filters=None):
+            current_include_filters=None,
+            is_binary=False):
 
         # Make requests use a more modern looking user-agent
         if not 'User-Agent' in request_headers:
@@ -545,10 +571,12 @@ class html_requests(Fetcher):
         # For example - some sites don't tell us it's utf-8, but return utf-8 content
         # This seems to not occur when using webdriver/selenium, it seems to detect the text encoding more reliably.
         # https://github.com/psf/requests/issues/1604 good info about requests encoding detection
-        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
+        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
 
         if not r.content or not len(r.content):
             raise EmptyReply(url=url, status_code=r.status_code)
@@ -560,8 +588,14 @@ class html_requests(Fetcher):
             raise Non200ErrorCodeReceived(url=url, status_code=r.status_code, page_html=r.text)
 
         self.status_code = r.status_code
-        self.content = r.text
+        if is_binary:
+            # Binary files just return their checksum until we add something smarter
+            self.content = hashlib.md5(r.content).hexdigest()
+        else:
+            self.content = r.text
+
         self.headers = r.headers
+        self.raw_content = r.content
 
 
 # Decide which is the 'real' HTML webdriver, this is more a system wide config

changedetectionio/dev-docker/Dockerfile

@@ -1,14 +0,0 @@
-FROM python:3.8-slim
-
-# https://stackoverflow.com/questions/58701233/docker-logs-erroneously-appears-empty-until-container-stops
-ENV PYTHONUNBUFFERED=1
-
-WORKDIR /app
-
-RUN [ ! -d "/datastore" ] && mkdir /datastore
-
-COPY sleep.py /
-CMD [ "python", "/sleep.py" ]
-
-
-

changedetectionio/dev-docker/sleep.py

@@ -1,7 +0,0 @@
-import time
-
-print ("Sleep loop, you should run your script from the console")
-
-while True: 
-    # Wait for 5 seconds
-    time.sleep(2)

changedetectionio/fetch_site_status.py

@@ -1,10 +1,13 @@
 import hashlib
+import json
 import logging
 import os
 import re
 import urllib3
 
 from changedetectionio import content_fetcher, html_tools
+from changedetectionio.blueprint.price_data_follower import PRICE_DATA_TRACK_ACCEPT, PRICE_DATA_TRACK_REJECT
+from copy import deepcopy
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
 
@@ -13,6 +16,10 @@ class FilterNotFoundInResponse(ValueError):
     def __init__(self, msg):
         ValueError.__init__(self, msg)
 
+class PDFToHTMLToolNotFound(ValueError):
+    def __init__(self, msg):
+        ValueError.__init__(self, msg)
+
 
 # Some common stuff here that can be moved to a base class
 # (set_proxy_from_list)
@@ -37,8 +44,7 @@ class perform_site_check():
 
         return regex
 
-    def run(self, uuid):
-        from copy import deepcopy
+    def run(self, uuid, skip_when_checksum_same=True):
         changed_detected = False
         screenshot = False  # as bytes
         stripped_text_from_html = ""
@@ -85,7 +91,7 @@ class perform_site_check():
             is_source = True
 
         # Pluggable content fetcher
-        prefer_backend = watch.get('fetch_backend')
+        prefer_backend = watch.get_fetch_backend
         if hasattr(content_fetcher, prefer_backend):
             klass = getattr(content_fetcher, prefer_backend)
         else:
@@ -115,12 +121,26 @@ class perform_site_check():
         if watch.get('webdriver_js_execute_code') is not None and watch.get('webdriver_js_execute_code').strip():
             fetcher.webdriver_js_execute_code = watch.get('webdriver_js_execute_code')
 
-        fetcher.run(url, timeout, request_headers, request_body, request_method, ignore_status_codes, watch.get('include_filters'))
+        # requests for PDF's, images etc should be passwd the is_binary flag
+        is_binary = watch.is_pdf
+
+        fetcher.run(url, timeout, request_headers, request_body, request_method, ignore_status_codes, watch.get('include_filters'), is_binary=is_binary)
         fetcher.quit()
 
         self.screenshot = fetcher.screenshot
         self.xpath_data = fetcher.xpath_data
 
+        # Track the content type
+        update_obj['content_type'] = fetcher.headers.get('Content-Type', '')
+
+        # Watches added automatically in the queue manager will skip if its the same checksum as the previous run
+        # Saves a lot of CPU
+        update_obj['previous_md5_before_filters'] = hashlib.md5(fetcher.content.encode('utf-8')).hexdigest()
+        if skip_when_checksum_same:
+            if update_obj['previous_md5_before_filters'] == watch.get('previous_md5_before_filters'):
+                raise content_fetcher.checksumFromPreviousCheckWasTheSame()
+
+
         # Fetching complete, now filters
         # @todo move to class / maybe inside of fetcher abstract base?
 
@@ -139,6 +159,31 @@ class perform_site_check():
             is_html = False
             is_json = False
 
+        if watch.is_pdf or 'application/pdf' in fetcher.headers.get('Content-Type', '').lower():
+            from shutil import which
+            tool = os.getenv("PDF_TO_HTML_TOOL", "pdftohtml")
+            if not which(tool):
+                raise PDFToHTMLToolNotFound("Command-line `{}` tool was not found in system PATH, was it installed?".format(tool))
+
+            import subprocess
+            proc = subprocess.Popen(
+                [tool, '-stdout', '-', '-s', 'out.pdf', '-i'],
+                stdout=subprocess.PIPE,
+                stdin=subprocess.PIPE)
+            proc.stdin.write(fetcher.raw_content)
+            proc.stdin.close()
+            fetcher.content = proc.stdout.read().decode('utf-8')
+            proc.wait(timeout=60)
+
+            # Add a little metadata so we know if the file changes (like if an image changes, but the text is the same
+            # @todo may cause problems with non-UTF8?
+            metadata = "<p>Added by changedetection.io: Document checksum - {} Filesize - {} bytes</p>".format(
+                hashlib.md5(fetcher.raw_content).hexdigest().upper(),
+                len(fetcher.content))
+
+            fetcher.content = fetcher.content.replace('</body>', metadata + '</body>')
+
+
         include_filters_rule = deepcopy(watch.get('include_filters', []))
         # include_filters_rule = watch['include_filters']
         subtractive_selectors = watch.get(
@@ -148,7 +193,7 @@ class perform_site_check():
         )
 
         # Inject a virtual LD+JSON price tracker rule
-        if watch.get('track_ldjson_price_data'):
+        if watch.get('track_ldjson_price_data', '') == PRICE_DATA_TRACK_ACCEPT:
             include_filters_rule.append(html_tools.LD_JSON_PRODUCT_OFFER_SELECTOR)
 
         has_filter_rule = include_filters_rule and len("".join(include_filters_rule).strip())
@@ -158,6 +203,14 @@ class perform_site_check():
             include_filters_rule.append("json:$")
             has_filter_rule = True
 
+        if is_json:
+            # Sort the JSON so we dont get false alerts when the content is just re-ordered
+            try:
+                fetcher.content = json.dumps(json.loads(fetcher.content), sort_keys=True)
+            except Exception as e:
+                # Might have just been a snippet, or otherwise bad JSON, continue
+                pass
+
         if has_filter_rule:
             json_filter_prefixes = ['json:', 'jq:']
             for filter in include_filters_rule:
@@ -165,6 +218,8 @@ class perform_site_check():
                     stripped_text_from_html += html_tools.extract_json_as_string(content=fetcher.content, json_filter=filter)
                     is_html = False
 
+
+
         if is_html or is_source:
 
             # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text

changedetectionio/forms.py

@@ -426,6 +426,13 @@ class watchForm(commonSettingsForm):
         return result
 
 
+class SingleExtraProxy(Form):
+
+    # maybe better to set some <script>var..
+    proxy_name = StringField('Name', [validators.Optional()], render_kw={"placeholder": "Name"})
+    proxy_url = StringField('Proxy URL', [validators.Optional()], render_kw={"placeholder": "http://user:pass@...:3128", "size":50})
+    # @todo do the validation here instead
+
 # datastore.data['settings']['requests']..
 class globalSettingsRequestForm(Form):
     time_between_check = FormField(TimeBetweenCheckForm)
@@ -433,6 +440,15 @@ class globalSettingsRequestForm(Form):
     jitter_seconds = IntegerField('Random jitter seconds ± check',
                                   render_kw={"style": "width: 5em;"},
                                   validators=[validators.NumberRange(min=0, message="Should contain zero or more seconds")])
+    extra_proxies = FieldList(FormField(SingleExtraProxy), min_entries=5)
+
+    def validate_extra_proxies(self, extra_validators=None):
+        for e in self.data['extra_proxies']:
+            if e.get('proxy_name') or e.get('proxy_url'):
+                if not e.get('proxy_name','').strip() or not e.get('proxy_url','').strip():
+                    self.extra_proxies.errors.append('Both a name, and a Proxy URL is required.')
+                    return False
+
 
 # datastore.data['settings']['application']..
 class globalSettingsApplicationForm(commonSettingsForm):

changedetectionio/model/App.py

@@ -15,11 +15,12 @@ class model(dict):
                 'headers': {
                 },
                 'requests': {
-                    'timeout': int(getenv("DEFAULT_SETTINGS_REQUESTS_TIMEOUT", "45")),  # Default 45 seconds
-                    'time_between_check': {'weeks': None, 'days': None, 'hours': 3, 'minutes': None, 'seconds': None},
+                    'extra_proxies': [], # Configurable extra proxies via the UI
                     'jitter_seconds': 0,
+                    'proxy': None, # Preferred proxy connection
+                    'time_between_check': {'weeks': None, 'days': None, 'hours': 3, 'minutes': None, 'seconds': None},
+                    'timeout': int(getenv("DEFAULT_SETTINGS_REQUESTS_TIMEOUT", "45")),  # Default 45 seconds
                     'workers': int(getenv("DEFAULT_SETTINGS_REQUESTS_WORKERS", "10")),  # Number of threads, lower is better for slow connections
-                    'proxy': None # Preferred proxy connection
                 },
                 'application': {
                     'api_access_token_enabled': True,
@@ -27,7 +28,6 @@ class model(dict):
                     'base_url' : None,
                     'extract_title_as_title': False,
                     'empty_pages_are_a_change': False,
-                    'css_dark_mode': False,
                     'fetch_backend': getenv("DEFAULT_FETCH_BACKEND", "html_requests"),
                     'filter_failure_notification_threshold_attempts': _FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT,
                     'global_ignore_text': [], # List of text to ignore when calculating the comparison checksum

changedetectionio/model/Watch.py

@@ -14,51 +14,52 @@ from changedetectionio.notification import (
 
 class model(dict):
     __newest_history_key = None
-    __history_n=0
+    __history_n = 0
     __base_config = {
-            #'history': {},  # Dict of timestamp and output stripped filename (removed)
-            #'newest_history_key': 0, (removed, taken from history.txt index)
-            'body': None,
-            'check_unique_lines': False, # On change-detected, compare against all history if its something new
-            'check_count': 0,
-            'consecutive_filter_failures': 0, # Every time the CSS/xPath filter cannot be located, reset when all is fine.
-            'extract_text': [],  # Extract text by regex after filters
-            'extract_title_as_title': False,
-            'fetch_backend': None,
-            'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
-            'has_ldjson_price_data': None,
-            'track_ldjson_price_data': None,
-            'headers': {},  # Extra headers to send
-            'ignore_text': [],  # List of text to ignore when calculating the comparison checksum
-            'include_filters': [],
-            'last_checked': 0,
-            'last_error': False,
-            'last_viewed': 0,  # history key value of the last viewed via the [diff] link
-            'method': 'GET',
-             # Custom notification content
-            'notification_body': None,
-            'notification_format': default_notification_format_for_watch,
-            'notification_muted': False,
-            'notification_title': None,
-            'notification_screenshot': False, # Include the latest screenshot if available and supported by the apprise URL
-            'notification_urls': [],  # List of URLs to add to the notification Queue (Usually AppRise)
-            'paused': False,
-            'previous_md5': False,
-            'proxy': None, # Preferred proxy connection
-            'subtractive_selectors': [],
-            'tag': None,
-            'text_should_not_be_present': [], # Text that should not present
-            # Re #110, so then if this is set to None, we know to use the default value instead
-            # Requires setting to None on submit if it's the same as the default
-            # Should be all None by default, so we use the system default in this case.
-            'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
-            'title': None,
-            'trigger_text': [],  # List of text or regex to wait for until a change is detected
-            'url': None,
-            'uuid': str(uuid.uuid4()),
-            'webdriver_delay': None,
-            'webdriver_js_execute_code': None, # Run before change-detection
-        }
+        # 'history': {},  # Dict of timestamp and output stripped filename (removed)
+        # 'newest_history_key': 0, (removed, taken from history.txt index)
+        'body': None,
+        'check_unique_lines': False,  # On change-detected, compare against all history if its something new
+        'check_count': 0,
+        'consecutive_filter_failures': 0,  # Every time the CSS/xPath filter cannot be located, reset when all is fine.
+        'extract_text': [],  # Extract text by regex after filters
+        'extract_title_as_title': False,
+        'fetch_backend': None,
+        'filter_failure_notification_send': strtobool(os.getenv('FILTER_FAILURE_NOTIFICATION_SEND_DEFAULT', 'True')),
+        'has_ldjson_price_data': None,
+        'track_ldjson_price_data': None,
+        'headers': {},  # Extra headers to send
+        'ignore_text': [],  # List of text to ignore when calculating the comparison checksum
+        'include_filters': [],
+        'last_checked': 0,
+        'last_error': False,
+        'last_viewed': 0,  # history key value of the last viewed via the [diff] link
+        'method': 'GET',
+        # Custom notification content
+        'notification_body': None,
+        'notification_format': default_notification_format_for_watch,
+        'notification_muted': False,
+        'notification_title': None,
+        'notification_screenshot': False,  # Include the latest screenshot if available and supported by the apprise URL
+        'notification_urls': [],  # List of URLs to add to the notification Queue (Usually AppRise)
+        'paused': False,
+        'previous_md5': False,
+        'previous_md5_before_filters': False,  # Used for skipping changedetection entirely
+        'proxy': None,  # Preferred proxy connection
+        'subtractive_selectors': [],
+        'tag': None,
+        'text_should_not_be_present': [],  # Text that should not present
+        # Re #110, so then if this is set to None, we know to use the default value instead
+        # Requires setting to None on submit if it's the same as the default
+        # Should be all None by default, so we use the system default in this case.
+        'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
+        'title': None,
+        'trigger_text': [],  # List of text or regex to wait for until a change is detected
+        'url': None,
+        'uuid': str(uuid.uuid4()),
+        'webdriver_delay': None,
+        'webdriver_js_execute_code': None,  # Run before change-detection
+    }
     jitter_seconds = 0
 
     def __init__(self, *arg, **kw):
@@ -113,6 +114,24 @@ class model(dict):
 
         return ready_url
 
+    @property
+    def get_fetch_backend(self):
+        """
+        Like just using the `fetch_backend` key but there could be some logic
+        :return:
+        """
+        # Maybe also if is_image etc?
+        # This is because chrome/playwright wont render the PDF in the browser and we will just fetch it and use pdf2html to see the text.
+        if self.is_pdf:
+            return 'html_requests'
+
+        return self.get('fetch_backend')
+
+    @property
+    def is_pdf(self):
+        # content_type field is set in the future
+        return '.pdf' in self.get('url', '').lower() or 'pdf' in self.get('content_type', '').lower()
+
     @property
     def label(self):
         # Used for sorting

changedetectionio/queuedWatchMetaData.py

@@ -0,0 +1,10 @@
+from dataclasses import dataclass, field
+from typing import Any
+
+# So that we can queue some metadata in `item`
+# https://docs.python.org/3/library/queue.html#queue.PriorityQueue
+#
+@dataclass(order=True)
+class PrioritizedItem:
+    priority: int
+    item: Any=field(compare=False)

changedetectionio/res/xpath_element_scraper.js

@@ -1,3 +1,6 @@
+// Copyright (C) 2021 Leigh Morresi (dgtlmoon@gmail.com)
+// All rights reserved.
+
 // @file Scrape the page looking for elements of concern (%ELEMENTS%)
 // http://matatk.agrip.org.uk/tests/position-and-width/
 // https://stackoverflow.com/questions/26813480/when-is-element-getboundingclientrect-guaranteed-to-be-updated-accurate
@@ -89,8 +92,8 @@ for (var i = 0; i < elements.length; i++) {
         continue
     }
 
-    // Forget really small ones
-    if (bbox['width'] < 10 && bbox['height'] < 10) {
+    // Skip really small ones, and where width or height ==0
+    if (bbox['width'] * bbox['height'] < 100) {
         continue;
     }
 
@@ -146,7 +149,6 @@ for (var i = 0; i < elements.length; i++) {
 
 }
 
-
 // Inject the current one set in the include_filters, which may be a CSS rule
 // used for displaying the current one in VisualSelector, where its not one we generated.
 if (include_filters.length) {
@@ -174,10 +176,23 @@ if (include_filters.length) {
         }
 
         if (q) {
-            bbox = q.getBoundingClientRect();
-            console.log("xpath_element_scraper: Got filter element, scroll from top was "+scroll_y)
-        } else {
-            console.log("xpath_element_scraper: filter element "+f+" was not found");
+            // #1231 - IN the case XPath attribute filter is applied, we will have to traverse up and find the element.
+            if (q.hasOwnProperty('getBoundingClientRect')) {
+                bbox = q.getBoundingClientRect();
+                console.log("xpath_element_scraper: Got filter element, scroll from top was " + scroll_y)
+            } else {
+                try {
+                    // Try and see we can find its ownerElement
+                    bbox = q.ownerElement.getBoundingClientRect();
+                    console.log("xpath_element_scraper: Got filter by ownerElement element, scroll from top was " + scroll_y)
+                } catch (e) {
+                    console.log("xpath_element_scraper: error looking up ownerElement")
+                }
+            }
+        }
+        
+        if(!q) {
+            console.log("xpath_element_scraper: filter element " + f + " was not found");
         }
 
         if (bbox && bbox['width'] > 0 && bbox['height'] > 0) {
@@ -192,5 +207,9 @@ if (include_filters.length) {
     }
 }
 
+// Sort the elements so we find the smallest one first, in other words, we find the smallest one matching in that area
+// so that we dont select the wrapping element by mistake and be unable to select what we want
+size_pos.sort((a, b) => (a.width*a.height > b.width*b.height) ? 1 : -1)
+
 // Window.width required for proper scaling in the frontend
 return {'size_pos': size_pos, 'browser_width': window.innerWidth};

changedetectionio/run_all_tests.sh

@@ -1,104 +0,0 @@
-#!/bin/bash
-
-
-# live_server will throw errors even with live_server_scope=function if I have the live_server setup in different functions
-# and I like to restart the server for each test (and have the test cleanup after each test)
-# merge request welcome :)
-
-
-# exit when any command fails
-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"
-  pytest $test_name
-done
-
-echo "RUNNING WITH BASE_URL SET"
-
-# Now re-run some tests with BASE_URL enabled
-# Re #65 - Ability to include a link back to the installation, in the notification.
-export BASE_URL="https://really-unique-domain.io"
-pytest tests/test_notification.py
-
-
-# Re-run with HIDE_REFERER set - could affect login
-export HIDE_REFERER=True
-pytest tests/test_access_control.py
-
-
-# Now for the selenium and playwright/browserless fetchers
-# Note - this is not UI functional tests - just checking that each one can fetch the content
-
-echo "TESTING WEBDRIVER FETCH > SELENIUM/WEBDRIVER..."
-docker run -d --name $$-test_selenium  -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome-debug:3.141.59
-# takes a while to spin up
-sleep 5
-export WEBDRIVER_URL=http://localhost:4444/wd/hub
-pytest tests/fetchers/test_content.py
-pytest tests/test_errorhandling.py
-unset WEBDRIVER_URL
-docker kill $$-test_selenium
-
-echo "TESTING WEBDRIVER FETCH > PLAYWRIGHT/BROWSERLESS..."
-# Not all platforms support playwright (not ARM/rPI), so it's not packaged in requirements.txt
-PLAYWRIGHT_VERSION=$(grep -i -E "RUN pip install.+" "$SCRIPT_DIR/../Dockerfile" | grep --only-matching -i -E "playwright[=><~+]+[0-9\.]+")
-echo "using $PLAYWRIGHT_VERSION"
-pip3 install "$PLAYWRIGHT_VERSION"
-docker run -d --name $$-test_browserless -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm  -p 3000:3000  --shm-size="2g"  browserless/chrome:1.53-chrome-stable
-# takes a while to spin up
-sleep 5
-export PLAYWRIGHT_DRIVER_URL=ws://127.0.0.1:3000
-pytest tests/fetchers/test_content.py
-pytest tests/test_errorhandling.py
-pytest tests/visualselector/test_fetch_data.py
-
-unset PLAYWRIGHT_DRIVER_URL
-docker kill $$-test_browserless
-
-# Test proxy list handling, starting two squids on different ports
-# Each squid adds a different header to the response, which is the main thing we test for.
-docker run -d --name $$-squid-one --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf -p 3128:3128 ubuntu/squid:4.13-21.10_edge
-docker run -d --name $$-squid-two --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf -p 3129:3128 ubuntu/squid:4.13-21.10_edge
-
-
-# So, basic HTTP as env var test
-export HTTP_PROXY=http://localhost:3128
-export HTTPS_PROXY=http://localhost:3128
-pytest tests/proxy_list/test_proxy.py
-docker logs $$-squid-one 2>/dev/null|grep one.changedetection.io
-if [ $? -ne 0 ]
-then
-  echo "Did not see a request to one.changedetection.io in the squid logs (while checking env vars HTTP_PROXY/HTTPS_PROXY)"
-fi
-unset HTTP_PROXY
-unset HTTPS_PROXY
-
-
-# 2nd test actually choose the preferred proxy from proxies.json
-cp tests/proxy_list/proxies.json-example ./test-datastore/proxies.json
-# Makes a watch use a preferred proxy
-pytest tests/proxy_list/test_multiple_proxy.py
-
-# Should be a request in the default "first" squid
-docker logs $$-squid-one 2>/dev/null|grep chosen.changedetection.io
-if [ $? -ne 0 ]
-then
-  echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy)"
-fi
-
-# And one in the 'second' squid (user selects this as preferred)
-docker logs $$-squid-two 2>/dev/null|grep chosen.changedetection.io
-if [ $? -ne 0 ]
-then
-  echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy)"
-fi
-
-# @todo - test system override proxy selection and watch defaults, setup a 3rd squid?
-docker kill $$-squid-one
-docker kill $$-squid-two
-
-

changedetectionio/run_basic_tests.sh

@@ -0,0 +1,30 @@
+#!/bin/bash
+
+
+# live_server will throw errors even with live_server_scope=function if I have the live_server setup in different functions
+# and I like to restart the server for each test (and have the test cleanup after each test)
+# merge request welcome :)
+
+
+# exit when any command fails
+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"
+  pytest $test_name
+done
+
+echo "RUNNING WITH BASE_URL SET"
+
+# Now re-run some tests with BASE_URL enabled
+# Re #65 - Ability to include a link back to the installation, in the notification.
+export BASE_URL="https://really-unique-domain.io"
+pytest tests/test_notification.py
+
+
+# Re-run with HIDE_REFERER set - could affect login
+export HIDE_REFERER=True
+pytest tests/test_access_control.py

changedetectionio/run_proxy_tests.sh

@@ -0,0 +1,59 @@
+#!/bin/bash
+
+# exit when any command fails
+set -e
+
+# Test proxy list handling, starting two squids on different ports
+# Each squid adds a different header to the response, which is the main thing we test for.
+docker run --network changedet-network -d --name squid-one --hostname squid-one --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf ubuntu/squid:4.13-21.10_edge
+docker run --network changedet-network -d --name squid-two --hostname squid-two --rm -v `pwd`/tests/proxy_list/squid.conf:/etc/squid/conf.d/debian.conf ubuntu/squid:4.13-21.10_edge
+
+# Used for configuring a custom proxy URL via the UI
+docker run --network changedet-network -d \
+  --name squid-custom \
+  --hostname squid-squid-custom \
+  --rm \
+  -v `pwd`/tests/proxy_list/squid-auth.conf:/etc/squid/conf.d/debian.conf \
+  -v `pwd`/tests/proxy_list/squid-passwords.txt:/etc/squid3/passwords \
+  ubuntu/squid:4.13-21.10_edge
+
+
+## 2nd test actually choose the preferred proxy from proxies.json
+
+docker run --network changedet-network \
+  -v `pwd`/tests/proxy_list/proxies.json-example:/app/changedetectionio/test-datastore/proxies.json \
+  test-changedetectionio \
+  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_multiple_proxy.py'
+
+
+## Should be a request in the default "first" squid
+docker logs squid-one 2>/dev/null|grep chosen.changedetection.io
+if [ $? -ne 0 ]
+then
+  echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy - squid one)"
+  exit 1
+fi
+
+# And one in the 'second' squid (user selects this as preferred)
+docker logs squid-two 2>/dev/null|grep chosen.changedetection.io
+if [ $? -ne 0 ]
+then
+  echo "Did not see a request to chosen.changedetection.io in the squid logs (while checking preferred proxy - squid two)"
+  exit 1
+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'
+
+
+# Should see a request for one.changedetection.io in there
+docker logs squid-custom 2>/dev/null|grep "TCP_TUNNEL.200.*changedetection.io"
+if [ $? -ne 0 ]
+then
+  echo "Did not see a valid request to changedetection.io in the squid logs (while checking preferred proxy - squid two)"
+  exit 1
+fi

changedetectionio/static/images/pdf-icon.svg

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg xmlns="http://www.w3.org/2000/svg" width="75.320129mm" height="92.604164mm" viewBox="0 0 75.320129 92.604164">
+  <g transform="translate(53.548057 -183.975276) scale(1.4843)">
+    <path fill="#ff2116" d="M-29.632812 123.94727c-3.551967 0-6.44336 2.89347-6.44336 6.44531v49.49804c0 3.55185 2.891393 6.44532 6.44336 6.44532H8.2167969c3.5519661 0 6.4433591-2.89335 6.4433591-6.44532v-40.70117s.101353-1.19181-.416015-2.35156c-.484969-1.08711-1.275391-1.84375-1.275391-1.84375a1.0584391 1.0584391 0 0 0-.0059-.008l-9.3906254-9.21094a1.0584391 1.0584391 0 0 0-.015625-.0156s-.8017392-.76344-1.9902344-1.27344c-1.39939552-.6005-2.8417968-.53711-2.8417968-.53711l.021484-.002z" color="#000" font-family="sans-serif" overflow="visible" paint-order="markers fill stroke" style="line-height:normal;font-variant-ligatures:normal;font-variant-position:normal;font-variant-caps:normal;font-variant-numeric:normal;font-variant-alternates:normal;font-feature-settings:normal;text-indent:0;text-align:start;text-decoration-line:none;text-decoration-style:solid;text-decoration-color:#000000;text-transform:none;text-orientation:mixed;white-space:normal;shape-padding:0;isolation:auto;mix-blend-mode:normal;solid-color:#000000;solid-opacity:1"/>
+    <path fill="#f5f5f5" d="M-29.632812 126.06445h28.3789058a1.0584391 1.0584391 0 0 0 .021484 0s1.13480448.011 1.96484378.36719c.79889772.34282 1.36536982.86176 1.36914062.86524.0000125.00001.00391.004.00391.004l9.3671868 9.18945s.564354.59582.837891 1.20899c.220779.49491.234375 1.40039.234375 1.40039a1.0584391 1.0584391 0 0 0-.002.0449v40.74609c0 2.41592-1.910258 4.32813-4.3261717 4.32813H-29.632812c-2.415914 0-4.326172-1.91209-4.326172-4.32813v-49.49804c0-2.41603 1.910258-4.32813 4.326172-4.32813z" color="#000" font-family="sans-serif" overflow="visible" paint-order="markers fill stroke" style="line-height:normal;font-variant-ligatures:normal;font-variant-position:normal;font-variant-caps:normal;font-variant-numeric:normal;font-variant-alternates:normal;font-feature-settings:normal;text-indent:0;text-align:start;text-decoration-line:none;text-decoration-style:solid;text-decoration-color:#000000;text-transform:none;text-orientation:mixed;white-space:normal;shape-padding:0;isolation:auto;mix-blend-mode:normal;solid-color:#000000;solid-opacity:1"/>
+    <path fill="#ff2116" d="M-23.40766 161.09299c-1.45669-1.45669.11934-3.45839 4.39648-5.58397l2.69124-1.33743 1.04845-2.29399c.57665-1.26169 1.43729-3.32036 1.91254-4.5748l.8641-2.28082-.59546-1.68793c-.73217-2.07547-.99326-5.19438-.52872-6.31588.62923-1.51909 2.69029-1.36323 3.50626.26515.63727 1.27176.57212 3.57488-.18329 6.47946l-.6193 2.38125.5455.92604c.30003.50932 1.1764 1.71867 1.9475 2.68743l1.44924 1.80272 1.8033728-.23533c5.72900399-.74758 7.6912472.523 7.6912472 2.34476 0 2.29921-4.4984914 2.48899-8.2760865-.16423-.8499666-.59698-1.4336605-1.19001-1.4336605-1.19001s-2.3665326.48178-3.531704.79583c-1.202707.32417-1.80274.52719-3.564509 1.12186 0 0-.61814.89767-1.02094 1.55026-1.49858 2.4279-3.24833 4.43998-4.49793 5.1723-1.3991.81993-2.86584.87582-3.60433.13733zm2.28605-.81668c.81883-.50607 2.47616-2.46625 3.62341-4.28553l.46449-.73658-2.11497 1.06339c-3.26655 1.64239-4.76093 3.19033-3.98386 4.12664.43653.52598.95874.48237 2.01093-.16792zm21.21809-5.95578c.80089-.56097.68463-1.69142-.22082-2.1472-.70466-.35471-1.2726074-.42759-3.1031574-.40057-1.1249.0767-2.9337647.3034-3.2403347.37237 0 0 .993716.68678 1.434896.93922.58731.33544 2.0145161.95811 3.0565161 1.27706 1.02785.31461 1.6224.28144 2.0729-.0409zm-8.53152-3.54594c-.4847-.50952-1.30889-1.57296-1.83152-2.3632-.68353-.89643-1.02629-1.52887-1.02629-1.52887s-.4996 1.60694-.90948 2.57394l-1.27876 3.16076-.37075.71695s1.971043-.64627 2.97389-.90822c1.0621668-.27744 3.21787-.70134 3.21787-.70134zm-2.74938-11.02573c.12363-1.0375.1761-2.07346-.15724-2.59587-.9246-1.01077-2.04057-.16787-1.85154 2.23517.0636.8084.26443 2.19033.53292 3.04209l.48817 1.54863.34358-1.16638c.18897-.64151.47882-2.02015.64411-3.06364z"/>
+    <path fill="#2c2c2c" d="M-20.930423 167.83862h2.364986q1.133514 0 1.840213.2169.706698.20991 1.189489.9446.482795.72769.482795 1.75625 0 .94459-.391832 1.6233-.391833.67871-1.056548.97958-.65772.30087-2.02913.30087h-.818651v3.72941h-1.581322zm1.581322 1.22447v3.33058h.783664q1.049552 0 1.44838-.39184.405826-.39183.405826-1.27345 0-.65772-.265887-1.06355-.265884-.41282-.587747-.50378-.314866-.098-1.000572-.098zm5.50664-1.22447h2.148082q1.560333 0 2.4909318.55276.9375993.55276 1.4133973 1.6443.482791 1.09153.482791 2.42096 0 1.3994-.4338151 2.49793-.4268149 1.09153-1.3154348 1.76324-.8816233.67172-2.5189212.67172h-2.267031zm1.581326 1.26645v7.018h.657715q1.378411 0 2.001144-.9516.6227329-.95858.6227329-2.5539 0-3.5125-2.6238769-3.5125zm6.4722254-1.26645h5.30372941v1.26645H-4.2075842v2.85478h2.9807225v1.26646h-2.9807225v4.16322h-1.5813254z" font-family="Franklin Gothic Medium Cond" letter-spacing="0" style="line-height:125%;-inkscape-font-specification:'Franklin Gothic Medium Cond'" word-spacing="4.26000023"/>
+  </g>
+</svg>

changedetectionio/static/js/visual-selector.js

@@ -1,4 +1,5 @@
-// Horrible proof of concept code :)
+// Copyright (C) 2021 Leigh Morresi (dgtlmoon@gmail.com)
+// All rights reserved.
 // yes - this is really a hack, if you are a front-ender and want to help, please get in touch!
 
 $(document).ready(function () {
@@ -177,9 +178,10 @@ $(document).ready(function () {
             // Basically, find the most 'deepest'
             var found = 0;
             ctx.fillStyle = 'rgba(205,0,0,0.35)';
-            for (var i = selector_data['size_pos'].length; i !== 0; i--) {
+            // Will be sorted by smallest width*height first
+            for (var i = 0; i <= selector_data['size_pos'].length; i++) {
                 // draw all of them? let them choose somehow?
-                var sel = selector_data['size_pos'][i - 1];
+                var sel = selector_data['size_pos'][i];
                 // If we are in a bounding-box
                 if (e.offsetY > sel.top * y_scale && e.offsetY < sel.top * y_scale + sel.height * y_scale
                     &&
@@ -195,7 +197,7 @@ $(document).ready(function () {
                     // no need to keep digging
                     // @todo or, O to go out/up, I to go in
                     // or double click to go up/out the selector?
-                    current_selected_i = i - 1;
+                    current_selected_i = i;
                     found += 1;
                     break;
                 }

changedetectionio/static/styles/.dockerignore

@@ -0,0 +1,3 @@
+node_modules
+package-lock.json
+

changedetectionio/static/styles/scss/parts/_extra_proxies.scss

@@ -0,0 +1,17 @@
+ul#requests-extra_proxies {
+  list-style: none;
+  /* tidy up the table to look more "inline" */
+  li {
+    > label {
+      display: none;
+    }
+
+  }
+  /* each proxy entry is a `table` */
+  table {
+    tr {
+      display: inline;
+    }
+  }
+}
+

changedetectionio/static/styles/scss/styles.scss

@@ -2,10 +2,11 @@
  * -- BASE STYLES --
  */
 
-@import "parts/_variables";
-@import "parts/_spinners";
-@import "parts/_browser-steps";
 @import "parts/_arrows";
+@import "parts/_browser-steps";
+@import "parts/_extra_proxies";
+@import "parts/_spinners";
+@import "parts/_variables";
 
 body {
   color: var(--color-text);
@@ -22,6 +23,13 @@ body {
   width: 1px;
 }
 
+// Row icons like chrome, pdf, share, etc
+.status-icon {
+  display: inline-block;
+  height: 1rem;
+  vertical-align: middle;
+}
+
 .pure-table-even {
   background: var(--color-background);
 }

changedetectionio/static/styles/styles.css

@@ -1,6 +1,139 @@
 /*
  * -- BASE STYLES --
  */
+.arrow {
+  border: solid #1b98f8;
+  border-width: 0 2px 2px 0;
+  display: inline-block;
+  padding: 3px; }
+  .arrow.right {
+    transform: rotate(-45deg);
+    -webkit-transform: rotate(-45deg); }
+  .arrow.left {
+    transform: rotate(135deg);
+    -webkit-transform: rotate(135deg); }
+  .arrow.up, .arrow.asc {
+    transform: rotate(-135deg);
+    -webkit-transform: rotate(-135deg); }
+  .arrow.down, .arrow.desc {
+    transform: rotate(45deg);
+    -webkit-transform: rotate(45deg); }
+
+#browser_steps {
+  /* convert rows to horizontal cells */ }
+  #browser_steps th {
+    display: none; }
+  #browser_steps li {
+    list-style: decimal;
+    padding: 5px; }
+    #browser_steps li:not(:first-child):hover {
+      opacity: 1.0; }
+    #browser_steps li .control {
+      padding-left: 5px;
+      padding-right: 5px; }
+      #browser_steps li .control a {
+        font-size: 70%; }
+    #browser_steps li.empty {
+      padding: 0px;
+      opacity: 0.35; }
+      #browser_steps li.empty .control {
+        display: none; }
+    #browser_steps li:hover {
+      background: #eee; }
+    #browser_steps li > label {
+      display: none; }
+
+#browser-steps-fieldlist {
+  height: 100%;
+  overflow-y: scroll; }
+
+#browser-steps .flex-wrapper {
+  display: flex;
+  flex-flow: row;
+  height: 600px;
+  /*@todo make this dynamic */ }
+
+/*  this is duplicate :( */
+#browsersteps-selector-wrapper {
+  height: 100%;
+  width: 100%;
+  overflow-y: scroll;
+  position: relative;
+  /* nice tall skinny one */ }
+  #browsersteps-selector-wrapper > img {
+    position: absolute;
+    max-width: 100%; }
+  #browsersteps-selector-wrapper > canvas {
+    position: relative;
+    max-width: 100%; }
+    #browsersteps-selector-wrapper > canvas:hover {
+      cursor: pointer; }
+  #browsersteps-selector-wrapper .loader {
+    position: absolute;
+    left: 50%;
+    top: 50%;
+    transform: translate(-50%, -50%);
+    margin-left: -40px;
+    z-index: 100;
+    max-width: 350px;
+    text-align: center; }
+  #browsersteps-selector-wrapper .spinner, #browsersteps-selector-wrapper .spinner:after {
+    width: 80px;
+    height: 80px;
+    font-size: 3px; }
+  #browsersteps-selector-wrapper #browsersteps-click-start {
+    color: var(--color-grey-400); }
+    #browsersteps-selector-wrapper #browsersteps-click-start:hover {
+      cursor: pointer; }
+
+ul#requests-extra_proxies {
+  list-style: none;
+  /* tidy up the table to look more "inline" */
+  /* each proxy entry is a `table` */ }
+  ul#requests-extra_proxies li > label {
+    display: none; }
+  ul#requests-extra_proxies table tr {
+    display: inline; }
+
+/* spinner */
+.spinner,
+.spinner:after {
+  border-radius: 50%;
+  width: 10px;
+  height: 10px; }
+
+.spinner {
+  margin: 0px auto;
+  font-size: 3px;
+  vertical-align: middle;
+  display: inline-block;
+  text-indent: -9999em;
+  border-top: 1.1em solid rgba(38, 104, 237, 0.2);
+  border-right: 1.1em solid rgba(38, 104, 237, 0.2);
+  border-bottom: 1.1em solid rgba(38, 104, 237, 0.2);
+  border-left: 1.1em solid #2668ed;
+  -webkit-transform: translateZ(0);
+  -ms-transform: translateZ(0);
+  transform: translateZ(0);
+  -webkit-animation: load8 1.1s infinite linear;
+  animation: load8 1.1s infinite linear; }
+
+@-webkit-keyframes load8 {
+  0% {
+    -webkit-transform: rotate(0deg);
+    transform: rotate(0deg); }
+  100% {
+    -webkit-transform: rotate(360deg);
+    transform: rotate(360deg); } }
+
+@keyframes load8 {
+  0% {
+    -webkit-transform: rotate(0deg);
+    transform: rotate(0deg); }
+  100% {
+    -webkit-transform: rotate(360deg);
+    transform: rotate(360deg); } }
+
 /**
  * CSS custom properties (aka variables).
  */
@@ -138,130 +271,6 @@ html[data-darkmode="true"] {
     html[data-darkmode="true"] .watch-table .unviewed.error {
       color: var(--color-watch-table-error); }
 
-/* spinner */
-.spinner,
-.spinner:after {
-  border-radius: 50%;
-  width: 10px;
-  height: 10px; }
-
-.spinner {
-  margin: 0px auto;
-  font-size: 3px;
-  vertical-align: middle;
-  display: inline-block;
-  text-indent: -9999em;
-  border-top: 1.1em solid rgba(38, 104, 237, 0.2);
-  border-right: 1.1em solid rgba(38, 104, 237, 0.2);
-  border-bottom: 1.1em solid rgba(38, 104, 237, 0.2);
-  border-left: 1.1em solid #2668ed;
-  -webkit-transform: translateZ(0);
-  -ms-transform: translateZ(0);
-  transform: translateZ(0);
-  -webkit-animation: load8 1.1s infinite linear;
-  animation: load8 1.1s infinite linear; }
-
-@-webkit-keyframes load8 {
-  0% {
-    -webkit-transform: rotate(0deg);
-    transform: rotate(0deg); }
-  100% {
-    -webkit-transform: rotate(360deg);
-    transform: rotate(360deg); } }
-
-@keyframes load8 {
-  0% {
-    -webkit-transform: rotate(0deg);
-    transform: rotate(0deg); }
-  100% {
-    -webkit-transform: rotate(360deg);
-    transform: rotate(360deg); } }
-
-#browser_steps {
-  /* convert rows to horizontal cells */ }
-  #browser_steps th {
-    display: none; }
-  #browser_steps li {
-    list-style: decimal;
-    padding: 5px; }
-    #browser_steps li:not(:first-child):hover {
-      opacity: 1.0; }
-    #browser_steps li .control {
-      padding-left: 5px;
-      padding-right: 5px; }
-      #browser_steps li .control a {
-        font-size: 70%; }
-    #browser_steps li.empty {
-      padding: 0px;
-      opacity: 0.35; }
-      #browser_steps li.empty .control {
-        display: none; }
-    #browser_steps li:hover {
-      background: #eee; }
-    #browser_steps li > label {
-      display: none; }
-
-#browser-steps-fieldlist {
-  height: 100%;
-  overflow-y: scroll; }
-
-#browser-steps .flex-wrapper {
-  display: flex;
-  flex-flow: row;
-  height: 600px;
-  /*@todo make this dynamic */ }
-
-/*  this is duplicate :( */
-#browsersteps-selector-wrapper {
-  height: 100%;
-  width: 100%;
-  overflow-y: scroll;
-  position: relative;
-  /* nice tall skinny one */ }
-  #browsersteps-selector-wrapper > img {
-    position: absolute;
-    max-width: 100%; }
-  #browsersteps-selector-wrapper > canvas {
-    position: relative;
-    max-width: 100%; }
-    #browsersteps-selector-wrapper > canvas:hover {
-      cursor: pointer; }
-  #browsersteps-selector-wrapper .loader {
-    position: absolute;
-    left: 50%;
-    top: 50%;
-    transform: translate(-50%, -50%);
-    margin-left: -40px;
-    z-index: 100;
-    max-width: 350px;
-    text-align: center; }
-  #browsersteps-selector-wrapper .spinner, #browsersteps-selector-wrapper .spinner:after {
-    width: 80px;
-    height: 80px;
-    font-size: 3px; }
-  #browsersteps-selector-wrapper #browsersteps-click-start {
-    color: var(--color-grey-400); }
-    #browsersteps-selector-wrapper #browsersteps-click-start:hover {
-      cursor: pointer; }
-
-.arrow {
-  border: solid #1b98f8;
-  border-width: 0 2px 2px 0;
-  display: inline-block;
-  padding: 3px; }
-  .arrow.right {
-    transform: rotate(-45deg);
-    -webkit-transform: rotate(-45deg); }
-  .arrow.left {
-    transform: rotate(135deg);
-    -webkit-transform: rotate(135deg); }
-  .arrow.up, .arrow.asc {
-    transform: rotate(-135deg);
-    -webkit-transform: rotate(-135deg); }
-  .arrow.down, .arrow.desc {
-    transform: rotate(45deg);
-    -webkit-transform: rotate(45deg); }
-
 body {
   color: var(--color-text);
   background: var(--color-background-page); }
@@ -275,6 +284,11 @@ body {
   white-space: nowrap;
   width: 1px; }
 
+.status-icon {
+  display: inline-block;
+  height: 1rem;
+  vertical-align: middle; }
+
 .pure-table-even {
   background: var(--color-background); }
 

changedetectionio/store.py

@@ -36,7 +36,6 @@ class ChangeDetectionStore:
         self.datastore_path = datastore_path
         self.json_store_path = "{}/url-watches.json".format(self.datastore_path)
         self.needs_write = False
-        self.proxy_list = None
         self.start_time = time.time()
         self.stop_thread = False
         # Base definition for all watchers
@@ -116,11 +115,6 @@ class ChangeDetectionStore:
             secret = secrets.token_hex(16)
             self.__data['settings']['application']['api_access_token'] = secret
 
-        # Proxy list support - available as a selection in settings when text file is imported
-        proxy_list_file = "{}/proxies.json".format(self.datastore_path)
-        if path.isfile(proxy_list_file):
-            self.import_proxy_list(proxy_list_file)
-
         # Bump the update version by running updates
         self.run_updates()
 
@@ -463,10 +457,30 @@ class ChangeDetectionStore:
                     print ("Removing",item)
                     unlink(item)
 
-    def import_proxy_list(self, filename):
-        with open(filename) as f:
-            self.proxy_list = json.load(f)
-            print ("Registered proxy list", list(self.proxy_list.keys()))
+    @property
+    def proxy_list(self):
+        proxy_list = {}
+        proxy_list_file = os.path.join(self.datastore_path, 'proxies.json')
+
+        # Load from external config file
+        if path.isfile(proxy_list_file):
+            with open("{}/proxies.json".format(self.datastore_path)) as f:
+                proxy_list = json.load(f)
+
+        # Mapping from UI config if available
+        extras = self.data['settings']['requests'].get('extra_proxies')
+        if extras:
+            i=0
+            for proxy in extras:
+                i += 0
+                if proxy.get('proxy_name') and proxy.get('proxy_url'):
+                    k = "ui-" + str(i) + proxy.get('proxy_name')
+                    proxy_list[k] = {'label': proxy.get('proxy_name'), 'url': proxy.get('proxy_url')}
+
+
+        return proxy_list if len(proxy_list) else None
+
+
 
 
     def get_preferred_proxy_for_watch(self, uuid):
@@ -476,11 +490,10 @@ class ChangeDetectionStore:
         :return: proxy "key" id
         """
 
-        proxy_id = None
         if self.proxy_list is None:
             return None
 
-        # If its a valid one
+        # If it's a valid one
         watch = self.data['watching'].get(uuid)
 
         if watch.get('proxy') and watch.get('proxy') in list(self.proxy_list.keys()):
@@ -493,8 +506,9 @@ class ChangeDetectionStore:
             if self.proxy_list.get(system_proxy_id):
                 return system_proxy_id
 
-        # Fallback - Did not resolve anything, use the first available
-        if system_proxy_id is None:
+
+        # Fallback - Did not resolve anything, or doesnt exist, use the first available
+        if system_proxy_id is None or not self.proxy_list.get(system_proxy_id):
             first_default = list(self.proxy_list)[0]
             return first_default
 

changedetectionio/templates/settings.html

@@ -21,6 +21,7 @@
             <li class="tab"><a href="#fetching">Fetching</a></li>
             <li class="tab"><a href="#filters">Global Filters</a></li>
             <li class="tab"><a href="#api">API</a></li>
+            <li class="tab"><a href="#proxies">CAPTCHA &amp; Proxies</a></li>
         </ul>
     </div>
     <div class="box-wrap inner">
@@ -170,7 +171,23 @@ nav
                     </div>
                 </div>
             </div>
+            <div class="tab-pane-inner" id="proxies">
 
+                <p><strong>Tip</strong>: You can connect to websites using <a href="https://brightdata.grsm.io/n0r16zf7eivq">BrightData</a> proxies, their service <strong>WebUnlocker</strong> will solve most CAPTCHAs, whilst their <strong>Residential Proxies</strong> may help to avoid CAPTCHA altogether. </p>
+                <p>It may be easier to try <strong>WebUnlocker</strong> first, WebUnlocker also supports country selection.</p>
+                <p>
+                    When you have <a href="https://brightdata.grsm.io/n0r16zf7eivq">registered</a>, enabled the required services, visit the <A href="https://brightdata.com/cp/api_example?">API example page</A>, then select <strong>Python</strong>, set the country you wish to use, then copy+paste the example URL below<br/>
+                    The Proxy URL with BrightData should start with <code>http://brd-customer...</code>
+                </p>
+
+                <p>When you sign up using <a href="https://brightdata.grsm.io/n0r16zf7eivq">https://brightdata.grsm.io/n0r16zf7eivq</a> BrightData will match any first deposit up to $150</p>
+
+
+                <div class="pure-control-group">
+                    {{ render_field(form.requests.form.extra_proxies) }}
+                    <div class="pure-form-message-inline">"Name" will be used for selecting the proxy in the Watch Edit settings</span>
+                </div>
+            </div>
             <div id="actions">
                 <div class="pure-control-group">
                     {{ render_button(form.save_button) }}

changedetectionio/templates/watch-overview.html

@@ -32,6 +32,7 @@
         <button class="pure-button button-secondary button-xsmall" style="font-size: 70%"  name="op" value="unpause">UnPause</button>
         <button class="pure-button button-secondary button-xsmall" style="font-size: 70%"  name="op" value="mute">Mute</button>
         <button class="pure-button button-secondary button-xsmall" style="font-size: 70%"  name="op" value="unmute">UnMute</button>
+        <button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="recheck">Recheck</button>
         <button class="pure-button button-secondary button-xsmall" style="font-size: 70%" name="op" value="notification-default">Use default notification</button>
         <button class="pure-button button-secondary button-xsmall" style="background: #dd4242; font-size: 70%" name="op" value="delete">Delete</button>
     </div>
@@ -88,10 +89,10 @@
                 </td>
                 <td class="title-col inline">{{watch.title if watch.title is not none and watch.title|length > 0 else watch.url}}
                     <a class="external" target="_blank" rel="noopener" href="{{ watch.link.replace('source:','') }}"></a>
-                    <a class="link-spread" href="{{url_for('form_share_put_watch', uuid=watch.uuid)}}"><img style="height: 1em;display:inline-block;" src="{{url_for('static_content', group='images', filename='spread.svg')}}" class="icon icon-spread" title="Create a link to share watch config with others" /></a>
-
-                    {%if watch.fetch_backend == "html_webdriver" %}<img style="height: 1em; display:inline-block;" src="{{url_for('static_content', group='images', filename='Google-Chrome-icon.png')}}" title="Using a chrome browser" />{% endif %}
+                    <a class="link-spread" href="{{url_for('form_share_put_watch', uuid=watch.uuid)}}"><img class="status-icon"  src="{{url_for('static_content', group='images', filename='spread.svg')}}" class="status-icon icon icon-spread" title="Create a link to share watch config with others" /></a>
 
+                    {%if watch.get_fetch_backend == "html_webdriver" %}<img class="status-icon" src="{{url_for('static_content', group='images', filename='Google-Chrome-icon.png')}}" title="Using a chrome browser" />{% endif %}
+                    {%if watch.is_pdf  %}<img class="status-icon" src="{{url_for('static_content', group='images', filename='pdf-icon.svg')}}" title="Converting PDF to text" />{% endif %}
                     {% if watch.last_error is defined and watch.last_error != False %}
                     <div class="fetch-error">{{ watch.last_error }}</div>
                     {% endif %}
@@ -102,7 +103,7 @@
                     <div class="ldjson-price-track-offer">Embedded price data detected, follow only price data? <a href="{{url_for('price_data_follower.accept', uuid=watch.uuid)}}" class="pure-button button-xsmall">Yes</a> <a href="{{url_for('price_data_follower.reject', uuid=watch.uuid)}}" class="">No</a></div>
                     {% endif %}
                     {% if watch['track_ldjson_price_data'] == 'accepted' %}
-                    <span class="tracking-ldjson-price-data" title="Automatically following embedded price information"><img src="{{url_for('static_content', group='images', filename='price-tag-icon.svg')}}"  class="price-follow-tag-icon"/> Price</span>
+                    <span class="tracking-ldjson-price-data" title="Automatically following embedded price information"><img src="{{url_for('static_content', group='images', filename='price-tag-icon.svg')}}"  class="status-icon price-follow-tag-icon"/> Price</span>
                     {% endif %}
                     {% if not active_tag %}
                     <span class="watch-tag-list">{{ watch.tag}}</span>

changedetectionio/tests/proxy_list/proxies.json-example

@@ -1,10 +1,10 @@
 {
   "proxy-one": {
-    "label": "One",
-    "url": "http://127.0.0.1:3128"
+    "label": "Proxy One",
+    "url": "http://squid-one:3128"
   },
   "proxy-two": {
-    "label": "two",
-    "url": "http://127.0.0.1:3129"
+    "label": "Proxy Two",
+    "url": "http://squid-two:3128"
   }
 }

changedetectionio/tests/proxy_list/squid-auth.conf

@@ -0,0 +1,48 @@
+acl localnet src 0.0.0.1-0.255.255.255  # RFC 1122 "this" network (LAN)
+acl localnet src 10.0.0.0/8             # RFC 1918 local private network (LAN)
+acl localnet src 100.64.0.0/10          # RFC 6598 shared address space (CGN)
+acl localnet src 169.254.0.0/16         # RFC 3927 link-local (directly plugged) machines
+acl localnet src 172.16.0.0/12          # RFC 1918 local private network (LAN)
+acl localnet src 192.168.0.0/16         # RFC 1918 local private network (LAN)
+acl localnet src fc00::/7               # RFC 4193 local private network range
+acl localnet src fe80::/10              # RFC 4291 link-local (directly plugged) machines
+acl localnet src 159.65.224.174
+acl SSL_ports port 443
+acl Safe_ports port 80          # http
+acl Safe_ports port 21          # ftp
+acl Safe_ports port 443         # https
+acl Safe_ports port 70          # gopher
+acl Safe_ports port 210         # wais
+acl Safe_ports port 1025-65535  # unregistered ports
+acl Safe_ports port 280         # http-mgmt
+acl Safe_ports port 488         # gss-http
+acl Safe_ports port 591         # filemaker
+acl Safe_ports port 777         # multiling http
+acl CONNECT method CONNECT
+
+http_access deny !Safe_ports
+http_access deny CONNECT !SSL_ports
+#http_access allow localhost manager
+http_access deny manager
+#http_access allow localhost
+#http_access allow localnet
+
+auth_param basic program /usr/lib/squid3/basic_ncsa_auth /etc/squid3/passwords
+auth_param basic realm proxy
+acl authenticated proxy_auth REQUIRED
+http_access allow authenticated
+http_access deny all
+
+
+http_port 3128
+coredump_dir /var/spool/squid
+refresh_pattern ^ftp:           1440    20%     10080
+refresh_pattern ^gopher:        1440    0%      1440
+refresh_pattern -i (/cgi-bin/|\?) 0     0%      0
+refresh_pattern \/(Packages|Sources)(|\.bz2|\.gz|\.xz)$ 0 0% 0 refresh-ims
+refresh_pattern \/Release(|\.gpg)$ 0 0% 0 refresh-ims
+refresh_pattern \/InRelease$ 0 0% 0 refresh-ims
+refresh_pattern \/(Translation-.*)(|\.bz2|\.gz|\.xz)$ 0 0% 0 refresh-ims
+refresh_pattern .               0       20%     4320
+logfile_rotate 0
+

changedetectionio/tests/proxy_list/squid-passwords.txt

@@ -0,0 +1 @@
+test:$apr1$xvhFolTA$E/kz5/Rw1ewcyaSUdwqZs.

changedetectionio/tests/proxy_list/test_select_custom_proxy.py

@@ -0,0 +1,51 @@
+#!/usr/bin/python3
+
+import time
+from flask import url_for
+from ..util import live_server_setup, wait_for_all_checks
+
+# just make a request, we will grep in the docker logs to see it actually got called
+def test_select_custom(client, live_server):
+    live_server_setup(live_server)
+
+    # Goto settings, add our custom one
+    res = client.post(
+        url_for("settings_page"),
+        data={
+            "requests-time_between_check-minutes": 180,
+            "application-ignore_whitespace": "y",
+            "application-fetch_backend": "html_requests",
+            "requests-extra_proxies-0-proxy_name": "custom-test-proxy",
+            # test:awesome is set in tests/proxy_list/squid-passwords.txt
+            "requests-extra_proxies-0-proxy_url": "http://test:awesome@squid-custom:3128",
+        },
+        follow_redirects=True
+    )
+
+    assert b"Settings updated." in res.data
+
+    res = client.post(
+        url_for("import_page"),
+        # Because a URL wont show in squid/proxy logs due it being SSLed
+        # Use plain HTTP or a specific domain-name here
+        data={"urls": "https://changedetection.io/CHANGELOG.txt"},
+        follow_redirects=True
+    )
+
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.get(url_for("index"))
+    assert b'Proxy Authentication Required' not in res.data
+
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    # We should see something via proxy
+    assert b'HEAD' in res.data
+
+    #
+    # Now we should see the request in the container logs for "squid-squid-custom" because it will be the only default
+

changedetectionio/tests/test_jsonpath_jq_selector.py

@@ -394,6 +394,48 @@ def check_json_ext_filter(json_filter, client, live_server):
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
 
+def test_ignore_json_order(client, live_server):
+    # A change in order shouldn't trigger a notification
+
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write('{"hello" : 123, "world": 123}')
+
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', content_type="application/json", _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+
+    time.sleep(2)
+
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write('{"world" : 123, "hello": 123}')
+
+    # Trigger a check
+    client.get(url_for("form_watch_checknow"), follow_redirects=True)
+    time.sleep(2)
+
+    res = client.get(url_for("index"))
+    assert b'unviewed' not in res.data
+
+    # Just to be sure it still works
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write('{"world" : 123, "hello": 124}')
+
+    # Trigger a check
+    client.get(url_for("form_watch_checknow"), follow_redirects=True)
+    time.sleep(2)
+
+    res = client.get(url_for("index"))
+    assert b'unviewed' in res.data
+
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
+
 def test_check_jsonpath_ext_filter(client, live_server):
     check_json_ext_filter('json:$[?(@.status==Sold)]', client, live_server)
 

changedetectionio/tests/test_pdf.py

@@ -0,0 +1,40 @@
+#!/usr/bin/python3
+
+import time
+from flask import url_for
+from .util import set_original_response, set_modified_response, live_server_setup
+
+sleep_time_for_fetch_thread = 3
+
+# `subtractive_selectors` should still work in `source:` type requests
+def test_fetch_pdf(client, live_server):
+    import shutil
+    shutil.copy("tests/test.pdf", "test-datastore/endpoint-test.pdf")
+
+    live_server_setup(live_server)
+    test_url = url_for('test_pdf_endpoint', _external=True)
+    # Add our URL to the import page
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+
+    assert b"1 Imported" in res.data
+
+    time.sleep(sleep_time_for_fetch_thread)
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+
+    assert b'PDF-1.5' not in res.data
+    assert b'hello world' in res.data
+
+    # So we know if the file changes in other ways
+    import hashlib
+    md5 = hashlib.md5(open("test-datastore/endpoint-test.pdf", 'rb').read()).hexdigest().upper()
+    # We should have one
+    assert len(md5) >0
+    # And it's going to be in the document
+    assert b'Document checksum - '+bytes(str(md5).encode('utf-8')) in res.data

changedetectionio/tests/util.py

@@ -168,5 +168,15 @@ def live_server_setup(live_server):
     def test_return_query():
         return request.query_string
 
+
+    @live_server.app.route('/endpoint-test.pdf')
+    def test_pdf_endpoint():
+
+        # Tried using a global var here but didn't seem to work, so reading from a file instead.
+        with open("test-datastore/endpoint-test.pdf", "rb") as f:
+            resp = make_response(f.read(), 200)
+            resp.headers['Content-Type'] = 'application/pdf'
+            return resp
+
     live_server.start()
 

changedetectionio/update_worker.py

@@ -4,6 +4,7 @@ import queue
 import time
 
 from changedetectionio import content_fetcher
+from changedetectionio import queuedWatchMetaData
 from changedetectionio.fetch_site_status import FilterNotFoundInResponse
 
 # A single update worker
@@ -157,11 +158,12 @@ class update_worker(threading.Thread):
         while not self.app.config.exit.is_set():
 
             try:
-                priority, uuid = self.q.get(block=False)
+                queued_item_data = self.q.get(block=False)
             except queue.Empty:
                 pass
 
             else:
+                uuid = queued_item_data.item.get('uuid')
                 self.current_uuid = uuid
 
                 if uuid in list(self.datastore.data['watching'].keys()):
@@ -171,11 +173,11 @@ class update_worker(threading.Thread):
                     update_obj= {}
                     xpath_data = False
                     process_changedetection_results = True
-                    print("> Processing UUID {} Priority {} URL {}".format(uuid, priority, self.datastore.data['watching'][uuid]['url']))
+                    print("> Processing UUID {} Priority {} URL {}".format(uuid, queued_item_data.priority, self.datastore.data['watching'][uuid]['url']))
                     now = time.time()
 
                     try:
-                        changed_detected, update_obj, contents = update_handler.run(uuid)
+                        changed_detected, update_obj, contents = update_handler.run(uuid, skip_when_checksum_same=queued_item_data.item.get('skip_when_checksum_same'))
                         # Re #342
                         # In Python 3, all strings are sequences of Unicode characters. There is a bytes type that holds raw bytes.
                         # We then convert/.decode('utf-8') for the notification etc
@@ -241,6 +243,10 @@ class update_worker(threading.Thread):
 
                         process_changedetection_results = True
 
+                    except content_fetcher.checksumFromPreviousCheckWasTheSame as e:
+                        # Yes fine, so nothing todo
+                        pass
+
                     except content_fetcher.BrowserStepsStepTimout as e:
 
                         if not self.datastore.data['watching'].get(uuid):

requirements-dev.txt

@@ -1,2 +0,0 @@
-pytest ~=6.2
-pytest-flask ~=1.2

requirements.txt

@@ -29,8 +29,9 @@ apprise~=1.2.0
 # apprise mqtt https://github.com/dgtlmoon/changedetection.io/issues/315
 paho-mqtt
 
-# Pinned version of cryptography otherwise
-# ERROR: Could not build wheels for cryptography which use PEP 517 and cannot be installed directly
+# This mainly affects some ARM builds, which unlike the other builds ignores "ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1"
+# so without this pinning, the newer versions on ARM will forcefully try to build rust, which results in "rust compiler not found"
+# (introduced once apprise became a dep)
 cryptography~=3.4
 
 # Used for CSS filtering
@@ -58,3 +59,7 @@ jq~=1.3 ;python_version >= "3.8" and sys_platform == "linux"
 # Any current modern version, required so far for screenshot PNG->JPEG conversion but will be used more in the future
 pillow
 # playwright is installed at Dockerfile build time because it's not available on all platforms
+
+# Include pytest, so if theres a support issue we can ask them to run these tests on their setup
+pytest ~=6.2
+pytest-flask ~=1.2