Adding setting

* improving unique key fix

* also bump timestamp along 1 sec

changedetectionio/content_fetchers/base.py

@@ -28,7 +28,7 @@ def manage_user_agent(headers, current_ua=''):
     :return:
     """
     # Ask it what the user agent is, if its obviously ChromeHeadless, switch it to the default
-    ua_in_custom_headers = next((v for k, v in headers.items() if k.lower() == "user-agent"), None)
+    ua_in_custom_headers = headers.get('User-Agent')
     if ua_in_custom_headers:
         return ua_in_custom_headers
 

changedetectionio/content_fetchers/puppeteer.py

@@ -115,12 +115,11 @@ class fetcher(Fetcher):
 
         # This user agent is similar to what was used when tweaking the evasions in inject_evasions_into_page(..)
         user_agent = None
-        if request_headers:
-            user_agent = next((value for key, value in request_headers.items() if key.lower().strip() == 'user-agent'), None)
-            if user_agent:
-                await self.page.setUserAgent(user_agent)
-                # Remove it so it's not sent again with headers after
-                [request_headers.pop(key) for key in list(request_headers) if key.lower().strip() == 'user-agent'.lower().strip()]
+        if request_headers and request_headers.get('User-Agent'):
+            # Request_headers should now be CaaseInsensitiveDict
+            # Remove it so it's not sent again with headers after
+            user_agent = request_headers.pop('User-Agent').strip()
+            await self.page.setUserAgent(user_agent)
 
         if not user_agent:
             # Attempt to strip 'HeadlessChrome' etc

changedetectionio/flask_app.py

@@ -339,7 +339,7 @@ def changedetection_app(config=None, datastore_o=None):
         # @todo needs a .itemsWithTag() or something - then we can use that in Jinaj2 and throw this away
         for uuid, watch in datastore.data['watching'].items():
             # @todo tag notification_muted skip also (improve Watch model)
-            if watch.get('notification_muted'):
+            if datastore.data['settings']['application'].get('rss_hide_muted_watches') and watch.get('notification_muted'):
                 continue
             if limit_tag and not limit_tag in watch['tags']:
                 continue
@@ -472,7 +472,7 @@ def changedetection_app(config=None, datastore_o=None):
                                  # Don't link to hosting when we're on the hosting environment
                                  active_tag=active_tag,
                                  active_tag_uuid=active_tag_uuid,
-                                 app_rss_token=datastore.data['settings']['application']['rss_access_token'],
+                                 app_rss_token=datastore.data['settings']['application'].get('rss_access_token'),
                                  datastore=datastore,
                                  errored_count=errored_count,
                                  form=form,

changedetectionio/forms.py

@@ -572,6 +572,8 @@ class globalSettingsApplicationForm(commonSettingsForm):
     removepassword_button = SubmitField('Remove password', render_kw={"class": "pure-button pure-button-primary"})
     render_anchor_tag_content = BooleanField('Render anchor tag content', default=False)
     shared_diff_access = BooleanField('Allow access to view diff page when password is enabled', default=False, validators=[validators.Optional()])
+    rss_hide_muted_watches = BooleanField('Hide muted watches from RSS feed', default=True,
+                                      validators=[validators.Optional()])
     filter_failure_notification_threshold_attempts = IntegerField('Number of times the filter can be missing before sending a notification',
                                                                   render_kw={"style": "width: 5em;"},
                                                                   validators=[validators.NumberRange(min=0,

changedetectionio/model/App.py

@@ -46,6 +46,8 @@ class model(dict):
                     'pager_size': 50,
                     'password': False,
                     'render_anchor_tag_content': False,
+                    'rss_access_token': None,
+                    'rss_hide_muted_watches': True,
                     'schema_version' : 0,
                     'shared_diff_access': False,
                     'webdriver_delay': None , # Extra delay in seconds before extracting text

changedetectionio/model/Watch.py

@@ -333,7 +333,9 @@ class model(dict):
         # Small hack so that we sleep just enough to allow 1 second  between history snapshots
         # this is because history.txt indexes/keys snapshots by epoch seconds and we dont want dupe keys
         if self.__newest_history_key and int(timestamp) == int(self.__newest_history_key):
-            time.sleep(timestamp - self.__newest_history_key)
+            logger.warning(f"Timestamp {timestamp} already exists, waiting 1 seconds so we have a unique key in history.txt")
+            timestamp = str(int(timestamp) + 1)
+            time.sleep(1)
 
         threshold = int(os.getenv('SNAPSHOT_BROTLI_COMPRESSION_THRESHOLD', 1024))
         skip_brotli = strtobool(os.getenv('DISABLE_BROTLI_TEXT_SNAPSHOT', 'False'))

changedetectionio/processors/__init__.py

@@ -1,10 +1,10 @@
 from abc import abstractmethod
-import os
-import hashlib
-import re
-from copy import deepcopy
 from changedetectionio.strtobool import strtobool
+from copy import deepcopy
 from loguru import logger
+import hashlib
+import os
+import re
 
 class difference_detection_processor():
 
@@ -21,7 +21,7 @@ class difference_detection_processor():
         self.watch = deepcopy(self.datastore.data['watching'].get(watch_uuid))
 
     def call_browser(self):
-
+        from requests.structures import CaseInsensitiveDict
         # Protect against file:// access
         if re.search(r'^file://', self.watch.get('url', '').strip(), re.IGNORECASE):
             if not strtobool(os.getenv('ALLOW_FILE_URI', 'false')):
@@ -93,14 +93,16 @@ class difference_detection_processor():
             self.fetcher.browser_steps_screenshot_path = os.path.join(self.datastore.datastore_path, self.watch.get('uuid'))
 
         # Tweak the base config with the per-watch ones
-        request_headers = self.watch.get('headers', [])
-        request_headers.update(self.datastore.get_all_base_headers())
-        request_headers.update(self.datastore.get_all_headers_in_textfile_for_watch(uuid=self.watch.get('uuid')))
+        request_headers = CaseInsensitiveDict()
 
         ua = self.datastore.data['settings']['requests'].get('default_ua')
         if ua and ua.get(prefer_fetch_backend):
             request_headers.update({'User-Agent': ua.get(prefer_fetch_backend)})
 
+        request_headers.update(self.watch.get('headers', {}))
+        request_headers.update(self.datastore.get_all_base_headers())
+        request_headers.update(self.datastore.get_all_headers_in_textfile_for_watch(uuid=self.watch.get('uuid')))
+
         # https://github.com/psf/requests/issues/4525
         # Requests doesnt yet support brotli encoding, so don't put 'br' here, be totally sure that the user cannot
         # do this by accident.

changedetectionio/store.py

@@ -124,12 +124,12 @@ class ChangeDetectionStore:
                 self.__data['app_guid'] = str(uuid_builder.uuid4())
 
         # Generate the URL access token for RSS feeds
-        if not 'rss_access_token' in self.__data['settings']['application']:
+        if not self.__data['settings']['application'].get('rss_access_token'):
             secret = secrets.token_hex(16)
             self.__data['settings']['application']['rss_access_token'] = secret
 
         # Generate the API access token
-        if not 'api_access_token' in self.__data['settings']['application']:
+        if not self.__data['settings']['application'].get('api_access_token'):
             secret = secrets.token_hex(16)
             self.__data['settings']['application']['api_access_token'] = secret
 
@@ -178,7 +178,7 @@ class ChangeDetectionStore:
     @property
     def has_unviewed(self):
         for uuid, watch in self.__data['watching'].items():
-            if watch.viewed == False:
+            if watch.history_n >= 2 and watch.viewed == False:
                 return True
         return False
 

changedetectionio/templates/settings.html

@@ -62,6 +62,9 @@
                         <span class="pure-form-message-inline">Allow access to view watch diff page when password is enabled (Good for sharing the diff page)
                         </span>
                     </div>
+                    <div class="pure-control-group">
+                        {{ render_checkbox_field(form.application.form.rss_hide_muted_watches) }}
+                    </div>
                     <div class="pure-control-group">
                         {{ render_field(form.application.form.pager_size) }}
                         <span class="pure-form-message-inline">Number of items per page in the watch overview list, 0 to disable.</span>

changedetectionio/tests/smtp/test_notification_smtp.py

@@ -71,6 +71,8 @@ def test_check_notification_email_formats_default_HTML(client, live_server):
 
     wait_for_all_checks(client)
     set_longer_modified_response()
+    time.sleep(2)
+
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
 
@@ -135,6 +137,7 @@ def test_check_notification_email_formats_default_Text_override_HTML(client, liv
 
     wait_for_all_checks(client)
     set_longer_modified_response()
+    time.sleep(2)
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
 

changedetectionio/tests/test_automatic_follow_ldjson_price.py

@@ -105,6 +105,7 @@ def test_check_ldjson_price_autodetect(client, live_server):
     wait_for_all_checks(client)
 
     # Trigger a check
+    time.sleep(1)
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     # Offer should be gone

changedetectionio/tests/test_backend.py

@@ -135,6 +135,9 @@ def test_check_basic_change_detection_functionality(client, live_server):
     # It should have picked up the <title>
     assert b'head title' in res.data
 
+    # Be sure the last_viewed is going to be greater than the last snapshot
+    time.sleep(1)
+
     # hit the mark all viewed link
     res = client.get(url_for("mark_all_viewed"), follow_redirects=True)
 

changedetectionio/tests/test_notification_errors.py

@@ -9,9 +9,6 @@ def test_check_notification_error_handling(client, live_server):
     live_server_setup(live_server)
     set_original_response()
 
-    # Give the endpoint time to spin up
-    time.sleep(1)
-
     # Set a URL and fetch it, then set a notification URL which is going to give errors
     test_url = url_for('test_endpoint', _external=True)
     res = client.post(

changedetectionio/tests/test_request.py

@@ -253,6 +253,62 @@ def test_method_in_request(client, live_server):
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
 
+# Re #2408 - user-agent override test, also should handle case-insensitive header deduplication
+def test_ua_global_override(client, live_server):
+    # live_server_setup(live_server)
+    test_url = url_for('test_headers', _external=True)
+
+    res = client.post(
+        url_for("settings_page"),
+        data={
+            "application-fetch_backend": "html_requests",
+            "application-minutes_between_check": 180,
+            "requests-default_ua-html_requests": "html-requests-user-agent"
+        },
+        follow_redirects=True
+    )
+    assert b'Settings updated' in res.data
+
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+
+    wait_for_all_checks(client)
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+
+    assert b"html-requests-user-agent" in res.data
+    # default user-agent should have shown by now
+    # now add a custom one in the headers
+
+
+    # Add some headers to a request
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={
+            "url": test_url,
+            "tags": "testtag",
+            "fetch_backend": 'html_requests',
+            # Important - also test case-insensitive
+            "headers": "User-AGent: agent-from-watch"},
+        follow_redirects=True
+    )
+    assert b"Updated watch." in res.data
+    wait_for_all_checks(client)
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b"agent-from-watch" in res.data
+    assert b"html-requests-user-agent" not in res.data
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
+
 def test_headers_textfile_in_request(client, live_server):
     #live_server_setup(live_server)
     # Add our URL to the import page
@@ -333,7 +389,7 @@ def test_headers_textfile_in_request(client, live_server):
     # Not needed anymore
     os.unlink('test-datastore/headers.txt')
     os.unlink('test-datastore/headers-testtag.txt')
-    os.unlink('test-datastore/' + extract_UUID_from_client(client) + '/headers.txt')
+
     # The service should echo back the request verb
     res = client.get(
         url_for("preview_page", uuid="first"),