Merge branch 'master' into windows-diff-file-handling-improvements

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

@@ -1,21 +1,12 @@
 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:

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_notification.py` for example
+Use `pytest` to validate/test, you can run the existing tests as `pytest tests/test_notifications.py` for example
 
 ```
 pip3 install -r requirements-dev

Dockerfile

@@ -64,7 +64,6 @@ EXPOSE 5000
 
 # The actual flask app
 COPY changedetectionio /app/changedetectionio
-
 # The eventlet server wrapper
 COPY changedetection.py /app/changedetection.py
 

changedetectionio/__init__.py

@@ -33,7 +33,7 @@ from flask_wtf import CSRFProtect
 from changedetectionio import html_tools
 from changedetectionio.api import api_v1
 
-__version__ = '0.39.20.4'
+__version__ = '0.39.20.3'
 
 datastore = None
 
@@ -194,9 +194,6 @@ 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})
-
 
 
 

changedetectionio/api/api_v1.py

@@ -122,37 +122,3 @@ 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:
-                # Use the system wide default
-                t = self.datastore.threshold_seconds
-
-            time_since_check = time.time() - watch.get('last_checked')
-
-            # Allow 5 minutes of grace time before we decide it's overdue
-            if time_since_check - (5 * 60) > t:
-                overdue_watches.append(uuid)
-
-        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

@@ -102,14 +102,6 @@ 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/fetch_site_status.py

@@ -65,9 +65,7 @@ class perform_site_check():
             request_headers['Accept-Encoding'] = request_headers['Accept-Encoding'].replace(', br', '')
 
         timeout = self.datastore.data['settings']['requests'].get('timeout')
-
+        url = watch.get('url')
-        url = watch.link
-
         request_body = self.datastore.data['watching'][uuid].get('body')
         request_method = self.datastore.data['watching'][uuid].get('method')
         ignore_status_codes = self.datastore.data['watching'][uuid].get('ignore_status_codes', False)

changedetectionio/model/Watch.py

@@ -87,16 +87,6 @@ class model(dict):
             print ("> Creating data dir {}".format(target_path))
             os.mkdir(target_path)
 
-    @property
-    def link(self):
-        url = self.get('url', '')
-        if '{%' in url or '{{' in url:
-            from jinja2 import Environment
-            # Jinja2 available in URLs along with https://pypi.org/project/jinja2-time/
-            jinja2_env = Environment(extensions=['jinja2_time.TimeExtension'])
-            return str(jinja2_env.from_string(url).render())
-        return url
-
     @property
     def label(self):
         # Used for sorting
@@ -128,10 +118,7 @@ class model(dict):
         if os.path.isfile(fname):
             logging.debug("Reading history index " + str(time.time()))
             with open(fname, "r") as f:
-                for i in f.readlines():
+                tmp_history = dict(i.strip().split(',', 2) 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]

changedetectionio/run_all_tests.sh

@@ -9,8 +9,6 @@
 # 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"
@@ -47,9 +45,7 @@ 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\.]+")
+pip3 install playwright~=1.24
-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.__data = App.model()
+        self.needs_write = False
         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={}))

changedetectionio/templates/edit.html

@@ -40,8 +40,7 @@
                 <fieldset>
                     <div class="pure-control-group">
                         {{ render_field(form.url, placeholder="https://...", required=true, class="m-d") }}
-                        <span class="pure-form-message-inline">Some sites use JavaScript to create the content, for this you should <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Fetching-pages-with-WebDriver">use the Chrome/WebDriver Fetcher</a></span><br/>
+                        <span class="pure-form-message-inline">Some sites use JavaScript to create the content, for this you should <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Fetching-pages-with-WebDriver">use the Chrome/WebDriver Fetcher</a></span>
-                        <span class="pure-form-message-inline">You can use variables in the URL, perfect for inserting the current date and other logic, <a href="https://github.com/dgtlmoon/changedetection.io/wiki/Handling-variables-in-the-watched-URL">help and examples here</a></span><br/>
                     </div>
                     <div class="pure-control-group">
                         {{ render_field(form.title, class="m-d") }}

changedetectionio/templates/watch-overview.html

@@ -87,7 +87,7 @@
                     <a class="state-{{'on' if watch.notification_muted}}" href="{{url_for('index', op='mute', uuid=watch.uuid, tag=active_tag)}}"><img src="{{url_for('static_content', group='images', filename='bell-off.svg')}}" alt="Mute notifications" title="Mute notifications"/></a>
                 </td>
                 <td class="title-col inline">{{watch.title if watch.title is not none and watch.title|length > 0 else watch.url}}
-                    <a class="external" target="_blank" rel="noopener" href="{{ watch.link.replace('source:','') }}"></a>
+                    <a class="external" target="_blank" rel="noopener" href="{{ watch.url.replace('source:','') }}"></a>
                     <a 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')}}" /></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')}}" />{% endif %}

changedetectionio/tests/test_api.py

@@ -147,16 +147,6 @@ 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_jinja2.py

@@ -1,33 +0,0 @@
-#!/usr/bin/python3
-
-import time
-from flask import url_for
-from .util import live_server_setup
-
-
-# If there was only a change in the whitespacing, then we shouldnt have a change detected
-def test_jinja2_in_url_query(client, live_server):
-    live_server_setup(live_server)
-
-    # Give the endpoint time to spin up
-    time.sleep(1)
-
-    # Add our URL to the import page
-    test_url = url_for('test_return_query', _external=True)
-
-    # because url_for() will URL-encode the var, but we dont here
-    full_url = "{}?{}".format(test_url,
-                              "date={% now 'Europe/Berlin', '%Y' %}.{% now 'Europe/Berlin', '%m' %}.{% now 'Europe/Berlin', '%d' %}", )
-    res = client.post(
-        url_for("form_quick_watch_add"),
-        data={"url": full_url, "tag": "test"},
-        follow_redirects=True
-    )
-    assert b"Watch added" in res.data
-    time.sleep(3)
-    # It should report nothing found (no new 'unviewed' class)
-    res = client.get(
-        url_for("preview_page", uuid="first"),
-        follow_redirects=True
-    )
-    assert b'date=2' in res.data

changedetectionio/tests/util.py

@@ -159,10 +159,5 @@ def live_server_setup(live_server):
         ret = " ".join([auth.username, auth.password, auth.type])
         return ret
 
-    # Just return some GET var
-    @live_server.app.route('/test-return-query', methods=['GET'])
-    def test_return_query():
-        return request.query_string
-
     live_server.start()
 

docker-compose.yml

@@ -45,9 +45,6 @@ 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
@@ -46,9 +46,4 @@ selenium ~= 4.1.0
 # need to revisit flask login versions
 werkzeug ~= 2.0.0
 
-# Templating, so far just in the URLs but in the future can be for the notifications also
-jinja2 ~= 3.1
-jinja2-time
-
 # playwright is installed at Dockerfile build time because it's not available on all platforms
-

x

@@ -0,0 +1,49 @@
+diff --git a/changedetectionio/__init__.py b/changedetectionio/__init__.py
+index c745dd3e..19873cce 100644
+--- a/changedetectionio/__init__.py
++++ b/changedetectionio/__init__.py
+@@ -819,8 +819,8 @@ def changedetection_app(config=None, datastore_o=None):
+         # 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, 'rb') as f:
+-                newest_version_file_contents = f.read().decode('utf-8')
++            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)
+ 
+@@ -832,8 +832,8 @@ def changedetection_app(config=None, datastore_o=None):
+             previous_file = history[dates[-2]]
+ 
+         try:
+-            with open(previous_file, 'rb') as f:
+-                previous_version_file_contents = f.read().decode('utf-8')
++            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)
+ 
+@@ -909,7 +909,7 @@ 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
+diff --git a/changedetectionio/model/Watch.py b/changedetectionio/model/Watch.py
+index 9a87ad71..566eb88e 100644
+--- a/changedetectionio/model/Watch.py
++++ b/changedetectionio/model/Watch.py
+@@ -158,7 +158,8 @@ class model(dict):
+ 
+         logging.debug("Saving history text {}".format(snapshot_fname))
+ 
+-        # in /diff/ we are going to assume for now that it's UTF-8 when reading
++        # 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()