Merge branch 'master' into add-system-info-api

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

@@ -0,0 +1,55 @@
+name: ChangeDetection.io Container Build Test
+
+# Triggers the workflow on push or pull request events
+
+# This line doesnt work, even tho it is the documented one
+#on: [push, pull_request]
+
+on:
+  push:
+    paths:
+      - requirements.txt
+      - Dockerfile
+
+  pull_request:
+    paths:
+      - requirements.txt
+      - Dockerfile
+
+  # Changes to requirements.txt packages and Dockerfile may or may not always be compatible with arm etc, so worth testing
+  # @todo: some kind of path filter for requirements.txt and Dockerfile
+jobs:
+  test-container-build:
+    runs-on: ubuntu-latest
+    steps:
+        - uses: actions/checkout@v2
+        - name: Set up Python 3.9
+          uses: actions/setup-python@v2
+          with:
+            python-version: 3.9
+
+        # Just test that the build works, some libraries won't compile on ARM/rPi etc
+        - name: Set up QEMU
+          uses: docker/setup-qemu-action@v1
+          with:
+            image: tonistiigi/binfmt:latest
+            platforms: all
+
+        - name: Set up Docker Buildx
+          id: buildx
+          uses: docker/setup-buildx-action@v1
+          with:
+            install: true
+            version: latest
+            driver-opts: image=moby/buildkit:master
+
+        - name: Test that the docker containers can build
+          id: docker_build
+          uses: docker/build-push-action@v2
+          # https://github.com/docker/build-push-action#customizing
+          with:
+            context: ./
+            file: ./Dockerfile
+            platforms: linux/arm/v7,linux/arm/v6,linux/amd64,linux/arm64,
+            cache-from: type=local,src=/tmp/.buildx-cache
+            cache-to: type=local,dest=/tmp/.buildx-cache

.github/workflows/test-only.yml

@@ -1,28 +1,25 @@
-name: ChangeDetection.io Test
+name: ChangeDetection.io App Test
 
 # Triggers the workflow on push or pull request events
 on: [push, pull_request]
 
 jobs:
-  test-build:
+  test-application:
     runs-on: ubuntu-latest
     steps:
-
       - uses: actions/checkout@v2
       - name: Set up Python 3.9
         uses: actions/setup-python@v2
         with:
           python-version: 3.9
 
-      - name: Show env vars
-        run: set
-
       - 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: Lint with flake8
         run: |
           # stop the build if there are Python syntax errors or undefined names
@@ -39,7 +36,4 @@ jobs:
           # Each test is totally isolated and performs its own cleanup/reset
           cd changedetectionio; ./run_all_tests.sh
 
-      # https://github.com/docker/build-push-action/blob/master/docs/advanced/test-before-push.md ?
-      # https://github.com/docker/buildx/issues/59 ? Needs to be one platform?
 
-      # https://github.com/docker/buildx/issues/495#issuecomment-918925854

CONTRIBUTING.md

@@ -6,7 +6,7 @@ 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_notifications.py` for example
+Use `pytest` to validate/test, you can run the existing tests as `pytest tests/test_notification.py` for example
 
 ```
 pip3 install -r requirements-dev

Dockerfile

@@ -5,13 +5,14 @@ FROM python:3.8-slim as builder
 ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1
 
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    libssl-dev \
-    libffi-dev \
+    g++ \
     gcc \
     libc-dev \
+    libffi-dev \
+    libssl-dev \
     libxslt-dev \
-    zlib1g-dev \
-    g++
+    make \
+    zlib1g-dev
 
 RUN mkdir /install
 WORKDIR /install
@@ -22,9 +23,14 @@ RUN pip install --target=/dependencies -r /requirements.txt
 
 # Playwright is an alternative to Selenium
 # Excluded this package from requirements.txt to prevent arm/v6 and arm/v7 builds from failing
-RUN pip install --target=/dependencies playwright~=1.25 \
+RUN pip install --target=/dependencies playwright~=1.26 \
     || echo "WARN: Failed to install Playwright. The application can still run, but the Playwright option will be disabled."
 
+
+RUN pip install --target=/dependencies jq~=1.3 \
+    || echo "WARN: Failed to install JQ. The application can still run, but the Jq: filter option will be disabled."
+
+
 # Final image stage
 FROM python:3.8-slim
 
@@ -58,6 +64,7 @@ EXPOSE 5000
 
 # The actual flask app
 COPY changedetectionio /app/changedetectionio
+
 # The eventlet server wrapper
 COPY changedetection.py /app/changedetection.py
 

MANIFEST.in

@@ -2,6 +2,7 @@ recursive-include changedetectionio/api *
 recursive-include changedetectionio/templates *
 recursive-include changedetectionio/static *
 recursive-include changedetectionio/model *
+recursive-include changedetectionio/tests *
 include changedetection.py
 global-exclude *.pyc
 global-exclude node_modules

README-pip.md

@@ -33,7 +33,7 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
 #### Key Features
 
 - Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
-- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JsonPath rules
+- 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"
 - Easily specify how often a site should be checked
 - Execute JS before extracting text (Good for logging in, see examples in the UI!)

README.md

@@ -47,13 +47,15 @@ _Need an actual Chrome runner with Javascript support? We support fetching via W
 #### Key Features
 
 - Lots of trigger filters, such as "Trigger on text", "Remove text by selector", "Ignore text", "Extract text", also using regular-expressions!
-- Target elements with xPath and CSS Selectors, Easily monitor complex JSON with JsonPath rules
+- 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"
 - 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
 - Use the "Visual Selector" to help target specific elements
+- Configurable [proxy per watch](https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration)
 
+We [recommend and use Bright Data](https://brightdata.grsm.io/n0r16zf7eivq) global proxy services, Bright Data will match any first deposit up to $100 using our signup link.
 
 ## Screenshots
 
@@ -119,8 +121,8 @@ See the wiki for more information https://github.com/dgtlmoon/changedetection.io
 
 
 ## Filters
-XPath, JSONPath and CSS support comes baked in! You can be as specific as you need, use XPath exported from various XPath element query creation tools.
 
+XPath, JSONPath, jq, and CSS support comes baked in! You can be as specific as you need, use XPath exported from various XPath element query creation tools. 
 (We support LXML `re:test`, `re:math` and `re:replace`.)
 
 ## Notifications
@@ -149,7 +151,7 @@ Now you can also customise your notification content!
 
 ## JSON API Monitoring
 
-Detect changes and monitor data in JSON API's by using the built-in JSONPath selectors as a filter / selector.
+Detect changes and monitor data in JSON API's by using either JSONPath or jq to filter, parse, and restructure JSON as needed.
 
 ![image](https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/json-filter-field-example.png)
 
@@ -157,9 +159,20 @@ This will re-parse the JSON and apply formatting to the text, making it super ea
 
 ![image](https://raw.githubusercontent.com/dgtlmoon/changedetection.io/master/docs/json-diff-example.png)
 
+### JSONPath or jq?
+
+For more complex parsing, filtering, and modifying of JSON data, jq is recommended due to the built-in operators and functions. Refer to the [documentation](https://stedolan.github.io/jq/manual/) for more specifc information on jq.
+
+One big advantage of `jq` is that you can use logic in your JSON filter, such as filters to only show items that have a value greater than/less than etc.
+
+See the wiki https://github.com/dgtlmoon/changedetection.io/wiki/JSON-Selector-Filter-help for more information and examples
+
+Note: `jq` library must be added separately (`pip3 install jq`)
+
+
 ### Parse JSON embedded in HTML!
 
-When you enable a `json:` filter, you can even automatically extract and parse embedded JSON inside a HTML page! Amazingly handy for sites that build content based on JSON, such as many e-commerce websites. 
+When you enable a `json:` or `jq:` filter, you can even automatically extract and parse embedded JSON inside a HTML page! Amazingly handy for sites that build content based on JSON, such as many e-commerce websites. 
 
 ```
 <html>
@@ -169,7 +182,7 @@ When you enable a `json:` filter, you can even automatically extract and parse e
 </script>
 ```  
 
-`json:$.price` would give `23.50`, or you can extract the whole structure
+`json:$.price` or `jq:.price` would give `23.50`, or you can extract the whole structure
 
 ## Proxy configuration
 

changedetectionio/__init__.py

@@ -33,7 +33,7 @@ from flask_wtf import CSRFProtect
 from changedetectionio import html_tools
 from changedetectionio.api import api_v1
 
-__version__ = '0.39.19.1'
+__version__ = '0.39.20.4'
 
 datastore = None
 
@@ -194,6 +194,9 @@ def changedetection_app(config=None, datastore_o=None):
     watch_api.add_resource(api_v1.Watch, '/api/v1/watch/<string:uuid>',
                            resource_class_kwargs={'datastore': datastore, 'update_q': update_q})
 
+    watch_api.add_resource(api_v1.SystemInfo, '/api/v1/systeminfo',
+                           resource_class_kwargs={'datastore': datastore, 'update_q': update_q})
+
 
 
 
@@ -636,20 +639,27 @@ def changedetection_app(config=None, datastore_o=None):
             # Only works reliably with Playwright
             visualselector_enabled = os.getenv('PLAYWRIGHT_DRIVER_URL', False) and default['fetch_backend'] == 'html_webdriver'
 
+            # JQ is difficult to install on windows and must be manually added (outside requirements.txt)
+            jq_support = True
+            try:
+                import jq
+            except ModuleNotFoundError:
+                jq_support = False
 
             output = render_template("edit.html",
-                                     uuid=uuid,
-                                     watch=datastore.data['watching'][uuid],
-                                     form=form,
-                                     has_empty_checktime=using_default_check_time,
-                                     has_default_notification_urls=True if len(datastore.data['settings']['application']['notification_urls']) else False,
-                                     using_global_webdriver_wait=default['webdriver_delay'] is None,
                                      current_base_url=datastore.data['settings']['application']['base_url'],
                                      emailprefix=os.getenv('NOTIFICATION_MAIL_BUTTON_PREFIX', False),
+                                     form=form,
+                                     has_default_notification_urls=True if len(datastore.data['settings']['application']['notification_urls']) else False,
+                                     has_empty_checktime=using_default_check_time,
+                                     jq_support=jq_support,
+                                     playwright_enabled=os.getenv('PLAYWRIGHT_DRIVER_URL', False),
                                      settings_application=datastore.data['settings']['application'],
+                                     using_global_webdriver_wait=default['webdriver_delay'] is None,
+                                     uuid=uuid,
                                      visualselector_data_is_ready=visualselector_data_is_ready,
                                      visualselector_enabled=visualselector_enabled,
-                                     playwright_enabled=os.getenv('PLAYWRIGHT_DRIVER_URL', False)
+                                     watch=datastore.data['watching'][uuid],
                                      )
 
         return output
@@ -809,8 +819,10 @@ def changedetection_app(config=None, datastore_o=None):
 
         newest_file = history[dates[-1]]
 
+        # Read as binary and force decode as UTF-8
+        # Windows may fail decode in python if we just use 'r' mode (chardet decode exception)
         try:
-            with open(newest_file, 'r') as f:
+            with open(newest_file, 'r', encoding='utf-8', errors='ignore') as f:
                 newest_version_file_contents = f.read()
         except Exception as e:
             newest_version_file_contents = "Unable to read {}.\n".format(newest_file)
@@ -823,7 +835,7 @@ def changedetection_app(config=None, datastore_o=None):
             previous_file = history[dates[-2]]
 
         try:
-            with open(previous_file, 'r') as f:
+            with open(previous_file, 'r', encoding='utf-8', errors='ignore') as f:
                 previous_version_file_contents = f.read()
         except Exception as e:
             previous_version_file_contents = "Unable to read {}.\n".format(previous_file)
@@ -900,7 +912,7 @@ def changedetection_app(config=None, datastore_o=None):
         timestamp = list(watch.history.keys())[-1]
         filename = watch.history[timestamp]
         try:
-            with open(filename, 'r') as f:
+            with open(filename, 'r', encoding='utf-8', errors='ignore') as f:
                 tmp = f.readlines()
 
                 # Get what needs to be highlighted

changedetectionio/api/api_v1.py

@@ -122,3 +122,33 @@ class CreateWatch(Resource):
             return {'status': "OK"}, 200
 
         return list, 200
+
+class SystemInfo(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+        self.update_q = kwargs['update_q']
+
+    @auth.check_token
+    def get(self):
+        import time
+        overdue_watches = []
+
+        # Check all watches and report which have not been checked but should have been
+
+        for uuid, watch in self.datastore.data.get('watching', {}).items():
+            # see if now - last_checked is greater than the time that should have been
+            # this is not super accurate (maybe they just edited it) but better than nothing
+            t = watch.threshold_seconds()
+            if not t:
+                t = self.datastore.threshold_seconds
+            time_since_check = time.time() - watch.get('last_checked')
+            if time_since_check > t:
+                overdue_watches.append(uuid)
+
+        return {
+                   'queue_size': self.update_q.qsize(),
+                   'overdue_watches': overdue_watches,
+                   'uptime': round(time.time() - self.datastore.start_time, 2),
+                   'watch_count': len(self.datastore.data.get('watching', {}))
+               }, 200

changedetectionio/changedetection.py

@@ -102,6 +102,14 @@ def main():
                     has_password=datastore.data['settings']['application']['password'] != False
                     )
 
+    # Monitored websites will not receive a Referer header
+    # when a user clicks on an outgoing link.
+    @app.after_request
+    def hide_referrer(response):
+        if os.getenv("HIDE_REFERER", False):
+            response.headers["Referrer-Policy"] = "no-referrer"
+        return response
+
     # Proxy sub-directory support
     # Set environment var USE_X_SETTINGS=1 on this script
     # And then in your proxy_pass settings

changedetectionio/content_fetcher.py

@@ -316,6 +316,7 @@ class base_html_playwright(Fetcher):
         import playwright._impl._api_types
         from playwright._impl._api_types import Error, TimeoutError
         response = None
+
         with sync_playwright() as p:
             browser_type = getattr(p, self.browser_type)
 
@@ -373,8 +374,11 @@ class base_html_playwright(Fetcher):
                 print("response object was none")
                 raise EmptyReply(url=url, status_code=None)
 
-            # Bug 2(?) Set the viewport size AFTER loading the page
-            page.set_viewport_size({"width": 1280, "height": 1024})
+
+            # Removed browser-set-size, seemed to be needed to make screenshots work reliably in older playwright versions
+            # Was causing exceptions like 'waiting for page but content is changing' etc
+            # https://www.browserstack.com/docs/automate/playwright/change-browser-window-size 1280x720 should be the default
+                        
             extra_wait = int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay
             time.sleep(extra_wait)
 
@@ -398,6 +402,13 @@ class base_html_playwright(Fetcher):
 
                     raise JSActionExceptions(status_code=response.status, screenshot=error_screenshot, message=str(e), url=url)
 
+                else:
+                    # JS eval was run, now we also wait some time if possible to let the page settle
+                    if self.render_extract_delay:
+                        page.wait_for_timeout(self.render_extract_delay * 1000)
+
+            page.wait_for_timeout(500)
+
             self.content = page.content()
             self.status_code = response.status
             self.headers = response.all_headers()
@@ -514,8 +525,6 @@ class base_html_webdriver(Fetcher):
             # Selenium doesn't automatically wait for actions as good as Playwright, so wait again
             self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
 
-        self.screenshot = self.driver.get_screenshot_as_png()
-
         # @todo - how to check this? is it possible?
         self.status_code = 200
         # @todo somehow we should try to get this working for WebDriver
@@ -526,6 +535,8 @@ class base_html_webdriver(Fetcher):
         self.content = self.driver.page_source
         self.headers = {}
 
+        self.screenshot = self.driver.get_screenshot_as_png()
+
     # Does the connection to the webdriver work? run a test connection.
     def is_ready(self):
         from selenium import webdriver
@@ -564,6 +575,11 @@ class html_requests(Fetcher):
             ignore_status_codes=False,
             current_css_filter=None):
 
+        # Make requests use a more modern looking user-agent
+        if not 'User-Agent' in request_headers:
+            request_headers['User-Agent'] = os.getenv("DEFAULT_SETTINGS_HEADERS_USERAGENT",
+                                                      'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.66 Safari/537.36')
+
         proxies = {}
 
         # Allows override the proxy on a per-request basis

changedetectionio/fetch_site_status.py

@@ -141,8 +141,9 @@ class perform_site_check():
             has_filter_rule = True
 
         if has_filter_rule:
-            if 'json:' in css_filter_rule:
-                stripped_text_from_html = html_tools.extract_json_as_string(content=fetcher.content, jsonpath_filter=css_filter_rule)
+            json_filter_prefixes = ['json:', 'jq:']
+            if any(prefix in css_filter_rule for prefix in json_filter_prefixes):
+                stripped_text_from_html = html_tools.extract_json_as_string(content=fetcher.content, json_filter=css_filter_rule)
                 is_html = False
 
         if is_html or is_source:

changedetectionio/forms.py

@@ -303,6 +303,25 @@ class ValidateCSSJSONXPATHInput(object):
 
                 # Re #265 - maybe in the future fetch the page and offer a
                 # warning/notice that its possible the rule doesnt yet match anything?
+                if not self.allow_json:
+                    raise ValidationError("jq not permitted in this field!")
+
+            if 'jq:' in line:
+                try:
+                    import jq
+                except ModuleNotFoundError:
+                    # `jq` requires full compilation in windows and so isn't generally available
+                    raise ValidationError("jq not support not found")
+
+                input = line.replace('jq:', '')
+
+                try:
+                    jq.compile(input)
+                except (ValueError) as e:
+                    message = field.gettext('\'%s\' is not a valid jq expression. (%s)')
+                    raise ValidationError(message % (input, str(e)))
+                except:
+                    raise ValidationError("A system-error occurred when validating your jq expression")
 
 
 class quickWatchForm(Form):

changedetectionio/html_tools.py

@@ -1,11 +1,11 @@
-import json
-from typing import List
 
 from bs4 import BeautifulSoup
-from jsonpath_ng.ext import parse
-import re
 from inscriptis import get_text
 from inscriptis.model.config import ParserConfig
+from jsonpath_ng.ext import parse
+from typing import List
+import json
+import re
 
 class FilterNotFoundInResponse(ValueError):
     def __init__(self, msg):
@@ -79,19 +79,35 @@ def extract_element(find='title', html_content=''):
     return element_text
 
 #
-def _parse_json(json_data, jsonpath_filter):
-    s=[]
-    jsonpath_expression = parse(jsonpath_filter.replace('json:', ''))
-    match = jsonpath_expression.find(json_data)
+def _parse_json(json_data, json_filter):
+    if 'json:' in json_filter:
+        jsonpath_expression = parse(json_filter.replace('json:', ''))
+        match = jsonpath_expression.find(json_data)
+        return _get_stripped_text_from_json_match(match)
 
+    if 'jq:' in json_filter:
+
+        try:
+            import jq
+        except ModuleNotFoundError:
+            # `jq` requires full compilation in windows and so isn't generally available
+            raise Exception("jq not support not found")
+
+        jq_expression = jq.compile(json_filter.replace('jq:', ''))
+        match = jq_expression.input(json_data).all()
+
+        return _get_stripped_text_from_json_match(match)
+
+def _get_stripped_text_from_json_match(match):
+    s = []
     # More than one result, we will return it as a JSON list.
     if len(match) > 1:
         for i in match:
-            s.append(i.value)
+            s.append(i.value if hasattr(i, 'value') else i)
 
     # Single value, use just the value, as it could be later used in a token in notifications.
     if len(match) == 1:
-        s = match[0].value
+        s = match[0].value if hasattr(match[0], 'value') else match[0]
 
     # Re #257 - Better handling where it does not exist, in the case the original 's' value was False..
     if not match:
@@ -103,16 +119,16 @@ def _parse_json(json_data, jsonpath_filter):
 
     return stripped_text_from_html
 
-def extract_json_as_string(content, jsonpath_filter):
+def extract_json_as_string(content, json_filter):
 
     stripped_text_from_html = False
 
     # Try to parse/filter out the JSON, if we get some parser error, then maybe it's embedded <script type=ldjson>
     try:
-        stripped_text_from_html = _parse_json(json.loads(content), jsonpath_filter)
+        stripped_text_from_html = _parse_json(json.loads(content), json_filter)
     except json.JSONDecodeError:
 
-        # Foreach <script json></script> blob.. just return the first that matches jsonpath_filter
+        # Foreach <script json></script> blob.. just return the first that matches json_filter
         s = []
         soup = BeautifulSoup(content, 'html.parser')
         bs_result = soup.findAll('script')
@@ -131,7 +147,7 @@ def extract_json_as_string(content, jsonpath_filter):
                 # Just skip it
                 continue
             else:
-                stripped_text_from_html = _parse_json(json_data, jsonpath_filter)
+                stripped_text_from_html = _parse_json(json_data, json_filter)
                 if stripped_text_from_html:
                     break
 

changedetectionio/model/App.py

@@ -13,10 +13,6 @@ class model(dict):
             'watching': {},
             'settings': {
                 'headers': {
-                    'User-Agent': getenv("DEFAULT_SETTINGS_HEADERS_USERAGENT", 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.66 Safari/537.36'),
-                    'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9',
-                    'Accept-Encoding': 'gzip, deflate',  # No support for brolti in python requests yet.
-                    'Accept-Language': 'en-GB,en-US;q=0.9,en;'
                 },
                 'requests': {
                     'timeout': int(getenv("DEFAULT_SETTINGS_REQUESTS_TIMEOUT", "45")),  # Default 45 seconds

changedetectionio/model/Watch.py

@@ -118,7 +118,10 @@ class model(dict):
         if os.path.isfile(fname):
             logging.debug("Reading history index " + str(time.time()))
             with open(fname, "r") as f:
-                tmp_history = dict(i.strip().split(',', 2) for i in f.readlines())
+                for i in f.readlines():
+                    if ',' in i:
+                        k, v = i.strip().split(',', 2)
+                        tmp_history[k] = v
 
         if len(tmp_history):
             self.__newest_history_key = list(tmp_history.keys())[-1]
@@ -151,28 +154,30 @@ class model(dict):
         import uuid
         import logging
 
-        output_path = "{}/{}".format(self.__datastore_path, self['uuid'])
+        output_path = os.path.join(self.__datastore_path, self['uuid'])
 
         self.ensure_data_dir_exists()
+        snapshot_fname = os.path.join(output_path, str(uuid.uuid4()))
 
-        snapshot_fname = "{}/{}.stripped.txt".format(output_path, uuid.uuid4())
         logging.debug("Saving history text {}".format(snapshot_fname))
 
+        # in /diff/ and /preview/ we are going to assume for now that it's UTF-8 when reading
+        # most sites are utf-8 and some are even broken utf-8
         with open(snapshot_fname, 'wb') as f:
             f.write(contents)
             f.close()
 
         # Append to index
         # @todo check last char was \n
-        index_fname = "{}/history.txt".format(output_path)
+        index_fname = os.path.join(output_path, "history.txt")
         with open(index_fname, 'a') as f:
             f.write("{},{}\n".format(timestamp, snapshot_fname))
             f.close()
 
         self.__newest_history_key = timestamp
-        self.__history_n+=1
+        self.__history_n += 1
 
-        #@todo bump static cache of the last timestamp so we dont need to examine the file to set a proper ''viewed'' status
+        # @todo bump static cache of the last timestamp so we dont need to examine the file to set a proper ''viewed'' status
         return snapshot_fname
 
     @property

changedetectionio/run_all_tests.sh

@@ -9,6 +9,8 @@
 # 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"
@@ -23,6 +25,13 @@ export BASE_URL="https://really-unique-domain.io"
 pytest tests/test_notification.py
 
 
+## JQ + JSON: filter test
+# jq is not available on windows and we should just test it when the package is installed
+# this will re-test with jq support
+pip3 install jq~=1.3
+pytest tests/test_jsonpath_jq_selector.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
 
@@ -38,7 +47,9 @@ 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
-pip3 install playwright~=1.24
+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

changedetectionio/store.py

@@ -30,14 +30,14 @@ class ChangeDetectionStore:
     def __init__(self, datastore_path="/datastore", include_default_watches=True, version_tag="0.0.0"):
         # Should only be active for docker
         # logging.basicConfig(filename='/dev/stdout', level=logging.INFO)
-        self.needs_write = False
+        self.__data = App.model()
         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
 
-        self.__data = App.model()
-
         # Base definition for all watchers
         # deepcopy part of #569 - not sure why its needed exactly
         self.generic_definition = deepcopy(Watch.model(datastore_path = datastore_path, default={}))
@@ -81,8 +81,6 @@ class ChangeDetectionStore:
         except (FileNotFoundError, json.decoder.JSONDecodeError):
             if include_default_watches:
                 print("Creating JSON store at", self.datastore_path)
-
-                self.add_watch(url='http://www.quotationspage.com/random.php', tag='test')
                 self.add_watch(url='https://news.ycombinator.com/', tag='Tech news')
                 self.add_watch(url='https://changedetection.io/CHANGELOG.txt', tag='changedetection.io')
 
@@ -577,3 +575,11 @@ class ChangeDetectionStore:
                 continue
         return
 
+
+    # We incorrectly used common header overrides that should only apply to Requests
+    # These are now handled in content_fetcher::html_requests and shouldnt be passed to Playwright/Selenium
+    def update_7(self):
+        # These were hard-coded in early versions
+        for v in ['User-Agent', 'Accept', 'Accept-Encoding', 'Accept-Language']:
+            if self.data['settings']['headers'].get(v):
+                del self.data['settings']['headers'][v]

changedetectionio/templates/edit.html

@@ -77,6 +77,7 @@
                         <span class="pure-form-message-inline">
                             <p>Use the <strong>Basic</strong> method (default) where your watched site doesn't need Javascript to render.</p>
                             <p>The <strong>Chrome/Javascript</strong> method requires a network connection to a running WebDriver+Chrome server, set by the ENV var 'WEBDRIVER_URL'. </p>
+                            Tip: <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration#brightdata-proxy-support">Connect using BrightData Proxies, find out more here.</a>
                         </span>
                     </div>
                 {% if form.proxy %}
@@ -183,8 +184,16 @@ User-Agent: wonderbra 1.0") }}
                         <span class="pure-form-message-inline">
                     <ul>
                         <li>CSS - Limit text to this CSS rule, only text matching this CSS rule is included.</li>
-                        <li>JSON - Limit text to this JSON rule, using <a href="https://pypi.org/project/jsonpath-ng/">JSONPath</a>, prefix with <code>"json:"</code>, use <code>json:$</code> to force re-formatting if required,  <a
-                                href="https://jsonpath.com/" target="new">test your JSONPath here</a></li>
+                        <li>JSON - Limit text to this JSON rule, using either <a href="https://pypi.org/project/jsonpath-ng/" target="new">JSONPath</a> or <a href="https://stedolan.github.io/jq/" target="new">jq</a> (if installed).
+                            <ul>
+                                <li>JSONPath: Prefix with <code>json:</code>, use <code>json:$</code> to force re-formatting if required,  <a href="https://jsonpath.com/" target="new">test your JSONPath here</a>.</li>
+                                {% if jq_support %}
+                                <li>jq: Prefix with <code>jq:</code> and <a href="https://jqplay.org/" target="new">test your jq here</a>. Using <a href="https://stedolan.github.io/jq/" target="new">jq</a> allows for complex filtering and processing of JSON data with built-in functions, regex, filtering, and more. See examples and documentation <a href="https://stedolan.github.io/jq/manual/" target="new">here</a>.</li>
+                                {% else %}
+                                <li>jq support not installed</li>
+                                {% endif %}
+                            </ul>
+                        </li>
                         <li>XPath - Limit text to this XPath rule, simply start with a forward-slash,
                             <ul>
                                 <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath://*[contains(@class, 'sametext')]</code>, <a
@@ -193,7 +202,7 @@ User-Agent: wonderbra 1.0") }}
                             </ul>
                             </li>
                     </ul>
-                    Please be sure that you thoroughly understand how to write CSS or JSONPath, XPath selector rules before filing an issue on GitHub! <a
+                    Please be sure that you thoroughly understand how to write CSS, JSONPath, XPath{% if jq_support %}, or jq selector{%endif%} rules before filing an issue on GitHub! <a
                                 href="https://github.com/dgtlmoon/changedetection.io/wiki/CSS-Selector-help">here for more CSS selector help</a>.<br/>
                 </span>
                     </div>

changedetectionio/templates/settings.html

@@ -99,6 +99,8 @@
                         <p>Use the <strong>Basic</strong> method (default) where your watched sites don't need Javascript to render.</p>
                         <p>The <strong>Chrome/Javascript</strong> method requires a network connection to a running WebDriver+Chrome server, set by the ENV var 'WEBDRIVER_URL'. </p>
                     </span>
+                    <br/>
+                    Tip: <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Proxy-configuration#brightdata-proxy-support">Connect using BrightData Proxies, find out more here.</a>
                 </div>
                 <fieldset class="pure-group" id="webdriver-override-options">
                     <div class="pure-form-message-inline">

changedetectionio/tests/test_api.py

@@ -147,6 +147,16 @@ def test_api_simple(client, live_server):
     # @todo how to handle None/default global values?
     assert watch['history_n'] == 2, "Found replacement history section, which is in its own API"
 
+    # basic systeminfo check
+    res = client.get(
+        url_for("systeminfo"),
+        headers={'x-api-key': api_key},
+    )
+    info = json.loads(res.data)
+    assert info.get('watch_count') == 1
+    assert info.get('uptime') > 0.5
+
+
     # Finally delete the watch
     res = client.delete(
         url_for("watch", uuid=watch_uuid),

changedetectionio/tests/test_jsonpath_jq_selector.py

@@ -2,10 +2,15 @@
 # coding=utf-8
 
 import time
-from flask import url_for
+from flask import url_for, escape
 from . util import live_server_setup
 import pytest
+jq_support = True
 
+try:
+    import jq
+except ModuleNotFoundError:
+    jq_support = False
 
 def test_setup(live_server):
     live_server_setup(live_server)
@@ -36,16 +41,28 @@ and it can also be repeated
     from .. import html_tools
 
     # See that we can find the second <script> one, which is not broken, and matches our filter
-    text = html_tools.extract_json_as_string(content, "$.offers.price")
+    text = html_tools.extract_json_as_string(content, "json:$.offers.price")
     assert text == "23.5"
 
-    text = html_tools.extract_json_as_string('{"id":5}', "$.id")
+    # also check for jq
+    if jq_support:
+        text = html_tools.extract_json_as_string(content, "jq:.offers.price")
+        assert text == "23.5"
+
+        text = html_tools.extract_json_as_string('{"id":5}', "jq:.id")
+        assert text == "5"
+
+    text = html_tools.extract_json_as_string('{"id":5}', "json:$.id")
     assert text == "5"
 
     # When nothing at all is found, it should throw JSONNOTFound
     # Which is caught and shown to the user in the watch-overview table
     with pytest.raises(html_tools.JSONNotFound) as e_info:
-        html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "$.id")
+        html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "json:$.id")
+
+    if jq_support:
+        with pytest.raises(html_tools.JSONNotFound) as e_info:
+            html_tools.extract_json_as_string('COMPLETE GIBBERISH, NO JSON!', "jq:.id")
 
 def set_original_ext_response():
     data = """
@@ -66,6 +83,7 @@ def set_original_ext_response():
 
     with open("test-datastore/endpoint-content.txt", "w") as f:
         f.write(data)
+    return None
 
 def set_modified_ext_response():
     data = """
@@ -86,6 +104,7 @@ def set_modified_ext_response():
 
     with open("test-datastore/endpoint-content.txt", "w") as f:
         f.write(data)
+    return None
 
 def set_original_response():
     test_return_data = """
@@ -184,10 +203,10 @@ def test_check_json_without_filter(client, live_server):
     assert b'&#34;&lt;b&gt;' in res.data
     assert res.data.count(b'{\n') >= 2
 
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
 
-def test_check_json_filter(client, live_server):
-    json_filter = 'json:boss.name'
-
+def check_json_filter(json_filter, client, live_server):
     set_original_response()
 
     # Give the endpoint time to spin up
@@ -226,7 +245,7 @@ def test_check_json_filter(client, live_server):
     res = client.get(
         url_for("edit_page", uuid="first"),
     )
-    assert bytes(json_filter.encode('utf-8')) in res.data
+    assert bytes(escape(json_filter).encode('utf-8')) in res.data
 
     # Trigger a check
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
@@ -252,10 +271,17 @@ def test_check_json_filter(client, live_server):
     # And #462 - check we see the proper utf-8 string there
     assert "Örnsköldsvik".encode('utf-8') in res.data
 
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
 
-def test_check_json_filter_bool_val(client, live_server):
-    json_filter = "json:$['available']"
+def test_check_jsonpath_filter(client, live_server):
+    check_json_filter('json:boss.name', client, live_server)
 
+def test_check_jq_filter(client, live_server):
+    if jq_support:
+        check_json_filter('jq:.boss.name', client, live_server)
+
+def check_json_filter_bool_val(json_filter, client, live_server):
     set_original_response()
 
     # Give the endpoint time to spin up
@@ -304,14 +330,22 @@ def test_check_json_filter_bool_val(client, live_server):
     # But the change should be there, tho its hard to test the change was detected because it will show old and new versions
     assert b'false' 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_filter_bool_val(client, live_server):
+    check_json_filter_bool_val("json:$['available']", client, live_server)
+
+def test_check_jq_filter_bool_val(client, live_server):
+    if jq_support:
+        check_json_filter_bool_val("jq:.available", client, live_server)
+
 # Re #265 - Extended JSON selector test
 # Stuff to consider here
 # - Selector should be allowed to return empty when it doesnt match (people might wait for some condition)
 # - The 'diff' tab could show the old and new content
 # - Form should let us enter a selector that doesnt (yet) match anything
-def test_check_json_ext_filter(client, live_server):
-    json_filter = 'json:$[?(@.status==Sold)]'
-
+def check_json_ext_filter(json_filter, client, live_server):
     set_original_ext_response()
 
     # Give the endpoint time to spin up
@@ -350,7 +384,7 @@ def test_check_json_ext_filter(client, live_server):
     res = client.get(
         url_for("edit_page", uuid="first"),
     )
-    assert bytes(json_filter.encode('utf-8')) in res.data
+    assert bytes(escape(json_filter).encode('utf-8')) in res.data
 
     # Trigger a check
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
@@ -376,3 +410,12 @@ def test_check_json_ext_filter(client, live_server):
     assert b'ForSale' not in res.data
     assert b'Sold' 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)
+
+def test_check_jq_ext_filter(client, live_server):
+    if jq_support:
+        check_json_ext_filter('jq:.[] | select(.status | contains("Sold"))', client, live_server)

changedetectionio/tests/visualselector/test_fetch_data.py

@@ -13,9 +13,9 @@ def test_visual_selector_content_ready(client, live_server):
     live_server_setup(live_server)
     time.sleep(1)
 
-    # Add our URL to the import page, maybe better to use something we control?
-    # We use an external URL because the docker container is too difficult to setup to connect back to the pytest socket
-    test_url = 'https://news.ycombinator.com'
+    # Add our URL to the import page, because the docker container (playwright/selenium) wont be able to connect to our usual test url
+    test_url = "https://changedetection.io/ci-test/test-runjs.html"
+
     res = client.post(
         url_for("form_quick_watch_add"),
         data={"url": test_url, "tag": '', 'edit_and_watch_submit_button': 'Edit > Watch'},
@@ -25,13 +25,27 @@ def test_visual_selector_content_ready(client, live_server):
 
     res = client.post(
         url_for("edit_page", uuid="first", unpause_on_save=1),
-        data={"css_filter": ".does-not-exist", "url": test_url, "tag": "", "headers": "", 'fetch_backend': "html_webdriver"},
+        data={
+              "url": test_url,
+              "tag": "",
+              "headers": "",
+              'fetch_backend': "html_webdriver",
+              'webdriver_js_execute_code': 'document.querySelector("button[name=test-button]").click();'
+        },
         follow_redirects=True
     )
     assert b"unpaused" in res.data
     time.sleep(1)
     wait_for_all_checks(client)
     uuid = extract_UUID_from_client(client)
+
+    # Check the JS execute code before extract worked
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b'I smell JavaScript' in res.data
+
     assert os.path.isfile(os.path.join('test-datastore', uuid, 'last-screenshot.png')), "last-screenshot.png should exist"
     assert os.path.isfile(os.path.join('test-datastore', uuid, 'elements.json')), "xpath elements.json data should exist"
 

docker-compose.yml

@@ -45,6 +45,9 @@ services:
   #        Respect proxy_pass type settings, `proxy_set_header Host "localhost";` and `proxy_set_header X-Forwarded-Prefix /app;`
   #        More here https://github.com/dgtlmoon/changedetection.io/wiki/Running-changedetection.io-behind-a-reverse-proxy-sub-directory
   #      - USE_X_SETTINGS=1
+  #
+  #        Hides the `Referer` header so that monitored websites can't see the changedetection.io hostname.
+  #      - HIDE_REFERER=true
 
       # Comment out ports: when using behind a reverse proxy , enable networks: etc.
       ports:

requirements.txt

@@ -1,8 +1,8 @@
-flask~= 2.0
+flask ~= 2.0
 flask_wtf
-eventlet>=0.31.0
+eventlet >= 0.31.0
 validators
-timeago ~=1.0
+timeago ~= 1.0
 inscriptis ~= 2.2
 feedgen ~= 0.9
 flask-login ~= 0.5
@@ -10,15 +10,20 @@ flask_restful
 pytz
 
 # Set these versions together to avoid a RequestsDependencyWarning
-requests[socks] ~= 2.26
+# >= 2.26 also adds Brotli support if brotli is installed
+brotli ~= 1.0
+requests[socks] ~= 2.28
+
 urllib3 > 1.26
 chardet > 2.3.0
 
 wtforms ~= 3.0
 jsonpath-ng ~= 1.5.3
 
+# jq not available on Windows so must be installed manually
+
 # Notification library
-apprise ~= 1.0.0
+apprise ~= 1.1.0
 
 # apprise mqtt https://github.com/dgtlmoon/changedetection.io/issues/315
 paho-mqtt
@@ -42,3 +47,4 @@ selenium ~= 4.1.0
 werkzeug ~= 2.0.0
 
 # playwright is installed at Dockerfile build time because it's not available on all platforms
+