Re #133 Option for ignoring whitespacing (#345)

* Global setting option to ignore whitespace when detecting a change

changedetectionio/__init__.py

@@ -405,7 +405,7 @@ def changedetection_app(config=None, datastore_o=None):
         # Get the most recent one
         newest_history_key = datastore.get_val(uuid, 'newest_history_key')
 
-        # 0 means that theres only one, so that there should be no 'unviewed' history availabe
+        # 0 means that theres only one, so that there should be no 'unviewed' history available
         if newest_history_key == 0:
             newest_history_key = list(datastore.data['watching'][uuid]['history'].keys())[0]
 
@@ -418,7 +418,11 @@ def changedetection_app(config=None, datastore_o=None):
                 stripped_content = handler.strip_ignore_text(raw_content,
                                                              datastore.data['watching'][uuid]['ignore_text'])
 
-                checksum = hashlib.md5(stripped_content).hexdigest()
+                if datastore.data['settings']['application'].get('ignore_whitespace', False):
+                    checksum = hashlib.md5(stripped_content.translate(None, b'\r\n\t ')).hexdigest()
+                else:
+                    checksum = hashlib.md5(stripped_content).hexdigest()
+
                 return checksum
 
         return datastore.data['watching'][uuid]['previous_md5']
@@ -553,6 +557,7 @@ def changedetection_app(config=None, datastore_o=None):
             form.minutes_between_check.data = int(datastore.data['settings']['requests']['minutes_between_check'])
             form.notification_urls.data = datastore.data['settings']['application']['notification_urls']
             form.global_ignore_text.data = datastore.data['settings']['application']['global_ignore_text']
+            form.ignore_whitespace.data = datastore.data['settings']['application']['ignore_whitespace']
             form.extract_title_as_title.data = datastore.data['settings']['application']['extract_title_as_title']
             form.fetch_backend.data = datastore.data['settings']['application']['fetch_backend']
             form.notification_title.data = datastore.data['settings']['application']['notification_title']
@@ -580,6 +585,7 @@ def changedetection_app(config=None, datastore_o=None):
             datastore.data['settings']['application']['notification_urls'] = form.notification_urls.data
             datastore.data['settings']['application']['base_url'] = form.base_url.data
             datastore.data['settings']['application']['global_ignore_text'] =  form.global_ignore_text.data
+            datastore.data['settings']['application']['ignore_whitespace'] = form.ignore_whitespace.data
 
             if form.trigger_check.data:
                 if len(form.notification_urls.data):

changedetectionio/fetch_site_status.py

@@ -58,8 +58,7 @@ class perform_site_check():
 
         watch = self.datastore.data['watching'][uuid]
 
-        update_obj = {'previous_md5': self.datastore.data['watching'][uuid]['previous_md5'],
+        update_obj = {
-                      'history': {},
                       "last_checked": timestamp
                       }
 
@@ -137,8 +136,16 @@ class perform_site_check():
             else:
                 stripped_text_from_html = stripped_text_from_html.encode('utf8')
 
+            # Re #133 - if we should strip whitespaces from triggering the change detected comparison
+            if self.datastore.data['settings']['application'].get('ignore_whitespace', False):
+                fetched_md5 = hashlib.md5(stripped_text_from_html.translate(None, b'\r\n\t ')).hexdigest()
+            else:
+                fetched_md5 = hashlib.md5(stripped_text_from_html).hexdigest()
 
-            fetched_md5 = hashlib.md5(stripped_text_from_html).hexdigest()
+            # On the first run of a site, watch['previous_md5'] will be an empty string, set it the current one.
+            if not len(watch['previous_md5']):
+                watch['previous_md5'] = fetched_md5
+                update_obj["previous_md5"] = fetched_md5
 
             blocked_by_not_found_trigger_text = False
 
@@ -160,16 +167,12 @@ class perform_site_check():
                         break
 
 
-            # could be None or False depending on JSON type
+
-            # On the first run of a site, watch['previous_md5'] will be an empty string
             if not blocked_by_not_found_trigger_text and watch['previous_md5'] != fetched_md5:
                 changed_detected = True
-
-                # Don't confuse people by updating as last-changed, when it actually just changed from None..
-                if self.datastore.get_val(uuid, 'previous_md5'):
-                    update_obj["last_changed"] = timestamp
-
                 update_obj["previous_md5"] = fetched_md5
+                update_obj["last_changed"] = timestamp
+
 
             # Extract title as title
             if is_html:

changedetectionio/forms.py

@@ -259,3 +259,4 @@ class globalSettingsForm(commonSettingsForm):
     extract_title_as_title = BooleanField('Extract <title> from document and use as watch title')
     base_url = StringField('Base URL', validators=[validators.Optional()])
     global_ignore_text = StringListField('Ignore Text', [ValidateListRegex()])
+    ignore_whitespace = BooleanField('Ignore whitespace')

changedetectionio/store.py

@@ -46,6 +46,7 @@ class ChangeDetectionStore:
                     'extract_title_as_title': False,
                     'fetch_backend': 'html_requests',
                     'global_ignore_text': [], # List of text to ignore when calculating the comparison checksum
+                    'ignore_whitespace': False,
                     'notification_urls': [], # Apprise URL list
                     # Custom notification content
                     'notification_title': None,

changedetectionio/templates/settings.html

@@ -69,15 +69,24 @@
 
 
             <div class="tab-pane-inner" id="filters">
-                <span class="pure-form-message-inline">Note: This is applied globally in addition to the per-watch rules.</span>
+
-                            <fieldset class="pure-group">
+                    <fieldset class="pure-group">
+                    {{ render_field(form.ignore_whitespace) }}
+                    <span class="pure-form-message-inline">Ignore whitespace, tabs and new-lines/line-feeds when considering if a change was detected.<br/>
+                    <i>Note:</i> Changing this will change the status of your existing watches, possibily trigger alerts etc.
+                    </span>
+                    </fieldset>
+
+
+                    <fieldset class="pure-group">
                     {{ render_field(form.global_ignore_text, rows=5, placeholder="Some text to ignore in a line
 /some.regex\d{2}/ for case-INsensitive regex
                     ") }}
-                    <span class="pure-form-message-inline">
+                    <span class="pure-form-message-inline">Note: This is applied globally in addition to the per-watch rules.</span><br/>
-                    Each line processed separately, any line matching will be ignored.<br/>
+                    <span class="pure-form-message-inline">Each line processed separately, any line matching will be ignored.<br/>
                     Regular Expression support, wrap the line in forward slash <b>/regex/</b>.
-                </span>
+                     </span>
+                    </fieldset>
            </div>
 
             <div id="actions">

changedetectionio/tests/conftest.py

@@ -18,7 +18,8 @@ def cleanup(datastore_path):
              'url-watches.json',
              'notification.txt',
              'count.txt',
-             'endpoint-content.txt']
+             'endpoint-content.txt'
+                 ]
     for file in files:
         try:
             os.unlink("{}/{}".format(datastore_path, file))

changedetectionio/tests/test_ignorewhitespace.py

@@ -0,0 +1,96 @@
+#!/usr/bin/python3
+
+import time
+from flask import url_for
+from . util import live_server_setup
+
+def test_setup(live_server):
+    live_server_setup(live_server)
+
+
+# Should be the same as set_original_ignore_response() but with a little more whitespacing
+def set_original_ignore_response_but_with_whitespace():
+    test_return_data = """<html>
+       <body>
+     Some initial text</br>
+     <p>
+
+
+     Which is across multiple lines</p>
+     <br>
+     </br>
+
+         So let's see what happens.  </br>
+
+
+     </body>
+     </html>
+
+    """
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write(test_return_data)
+
+
+def set_original_ignore_response():
+    test_return_data = """<html>
+       <body>
+     Some initial text</br>
+     <p>Which is across multiple lines</p>
+     </br>
+     So let's see what happens.  </br>
+     </body>
+     </html>
+
+    """
+
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write(test_return_data)
+
+
+
+# If there was only a change in the whitespacing, then we shouldnt have a change detected
+def test_check_ignore_whitespace(client, live_server):
+    sleep_time_for_fetch_thread = 3
+
+    # Give the endpoint time to spin up
+    time.sleep(1)
+
+    set_original_ignore_response()
+
+    # Goto the settings page, add our ignore text
+    res = client.post(
+        url_for("settings_page"),
+        data={
+            "minutes_between_check": 180,
+            "ignore_whitespace": "y",
+            'fetch_backend': "html_requests"
+        },
+        follow_redirects=True
+    )
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _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(sleep_time_for_fetch_thread)
+    # Trigger a check
+    client.get(url_for("api_watch_checknow"), follow_redirects=True)
+
+    set_original_ignore_response_but_with_whitespace()
+    time.sleep(sleep_time_for_fetch_thread)
+    # Trigger a check
+    client.get(url_for("api_watch_checknow"), follow_redirects=True)
+
+    # Give the thread time to pick it up
+    time.sleep(sleep_time_for_fetch_thread)
+
+    # It should report nothing found (no new 'unviewed' class)
+    res = client.get(url_for("index"))
+    assert b'unviewed' not in res.data
+    assert b'/test-endpoint' in res.data

changedetectionio/update_worker.py

@@ -64,74 +64,77 @@ class update_worker(threading.Thread):
                         self.datastore.update_watch(uuid=uuid, update_obj={'last_error': str(e)})
 
                     else:
-                        if update_obj:
+                        try:
-                            try:
+                            watch = self.datastore.data['watching'][uuid]
-                                self.datastore.update_watch(uuid=uuid, update_obj=update_obj)
+
-                                if changed_detected:
+                            # For the FIRST time we check a site, or a change detected, save the snapshot.
-                                    n_object = {}
+                            if changed_detected or not watch['last_checked']:
-                                    # A change was detected
+                                # A change was detected
-                                    fname = self.datastore.save_history_text(watch_uuid=uuid, contents=contents)
+                                fname = self.datastore.save_history_text(watch_uuid=uuid, contents=contents)
-
+                                # Should always be keyed by string(timestamp)
-                                    # Update history with the stripped text for future reference, this will also mean we save the first
+                                self.datastore.update_watch(uuid, {"history": {str(update_obj["last_checked"]): fname}})
-                                    # Should always be keyed by string(timestamp)
+
-                                    self.datastore.update_watch(uuid, {"history": {str(update_obj["last_checked"]): fname}})
+                            # Generally update anything interesting returned
-
+                            self.datastore.update_watch(uuid=uuid, update_obj=update_obj)
-                                    watch = self.datastore.data['watching'][uuid]
+
-
+                            # A change was detected
-                                    print (">> Change detected in UUID {} - {}".format(uuid, watch['url']))
+                            if changed_detected:
-
+                                n_object = {}
-                                    # Notifications should only trigger on the second time (first time, we gather the initial snapshot)
+                                print (">> Change detected in UUID {} - {}".format(uuid, watch['url']))
-                                    if len(watch['history']) > 1:
+
-
+                                # Notifications should only trigger on the second time (first time, we gather the initial snapshot)
-                                        dates = list(watch['history'].keys())
+                                if len(watch['history']) > 1:
-                                        # Convert to int, sort and back to str again
+
-                                        # @todo replace datastore getter that does this automatically
+                                    dates = list(watch['history'].keys())
-                                        dates = [int(i) for i in dates]
+                                    # Convert to int, sort and back to str again
-                                        dates.sort(reverse=True)
+                                    # @todo replace datastore getter that does this automatically
-                                        dates = [str(i) for i in dates]
+                                    dates = [int(i) for i in dates]
-
+                                    dates.sort(reverse=True)
-                                        prev_fname = watch['history'][dates[1]]
+                                    dates = [str(i) for i in dates]
-
+
-
+                                    prev_fname = watch['history'][dates[1]]
-                                        # Did it have any notification alerts to hit?
+
-                                        if len(watch['notification_urls']):
+
-                                            print(">>> Notifications queued for UUID from watch {}".format(uuid))
+                                    # Did it have any notification alerts to hit?
-                                            n_object['notification_urls'] = watch['notification_urls']
+                                    if len(watch['notification_urls']):
-                                            n_object['notification_title'] = watch['notification_title']
+                                        print(">>> Notifications queued for UUID from watch {}".format(uuid))
-                                            n_object['notification_body'] = watch['notification_body']
+                                        n_object['notification_urls'] = watch['notification_urls']
-                                            n_object['notification_format'] = watch['notification_format']
+                                        n_object['notification_title'] = watch['notification_title']
-
+                                        n_object['notification_body'] = watch['notification_body']
-                                        # No? maybe theres a global setting, queue them all
+                                        n_object['notification_format'] = watch['notification_format']
-                                        elif len(self.datastore.data['settings']['application']['notification_urls']):
+
-                                            print(">>> Watch notification URLs were empty, using GLOBAL notifications for UUID: {}".format(uuid))
+                                    # No? maybe theres a global setting, queue them all
-                                            n_object['notification_urls'] = self.datastore.data['settings']['application']['notification_urls']
+                                    elif len(self.datastore.data['settings']['application']['notification_urls']):
-                                            n_object['notification_title'] = self.datastore.data['settings']['application']['notification_title']
+                                        print(">>> Watch notification URLs were empty, using GLOBAL notifications for UUID: {}".format(uuid))
-                                            n_object['notification_body'] = self.datastore.data['settings']['application']['notification_body']
+                                        n_object['notification_urls'] = self.datastore.data['settings']['application']['notification_urls']
-                                            n_object['notification_format'] = self.datastore.data['settings']['application']['notification_format']
+                                        n_object['notification_title'] = self.datastore.data['settings']['application']['notification_title']
+                                        n_object['notification_body'] = self.datastore.data['settings']['application']['notification_body']
+                                        n_object['notification_format'] = self.datastore.data['settings']['application']['notification_format']
+                                    else:
+                                        print(">>> NO notifications queued, watch and global notification URLs were empty.")
+
+                                    # Only prepare to notify if the rules above matched
+                                    if 'notification_urls' in n_object:
+                                        # HTML needs linebreak, but MarkDown and Text can use a linefeed
+                                        if n_object['notification_format'] == 'HTML':
+                                            line_feed_sep = "</br>"
                                         else:
-                                            print(">>> NO notifications queued, watch and global notification URLs were empty.")
+                                            line_feed_sep = "\n"
-
+
-                                        # Only prepare to notify if the rules above matched
+                                        from changedetectionio import diff
-                                        if 'notification_urls' in n_object:
+                                        n_object.update({
-                                            # HTML needs linebreak, but MarkDown and Text can use a linefeed
+                                            'watch_url': watch['url'],
-                                            if n_object['notification_format'] == 'HTML':
+                                            'uuid': uuid,
-                                                line_feed_sep = "</br>"
+                                            'current_snapshot': contents.decode('utf-8'),
-                                            else:
+                                            'diff_full': diff.render_diff(prev_fname, fname, line_feed_sep=line_feed_sep),
-                                                line_feed_sep = "\n"
+                                            'diff': diff.render_diff(prev_fname, fname, True, line_feed_sep=line_feed_sep)
-
+                                        })
-                                            from changedetectionio import diff
+
-                                            n_object.update({
+                                        self.notification_q.put(n_object)
-                                                'watch_url': watch['url'],
+
-                                                'uuid': uuid,
+                        except Exception as e:
-                                                'current_snapshot': contents.decode('utf-8'),
+                            # Catch everything possible here, so that if a worker crashes, we don't lose it until restart!
-                                                'diff_full': diff.render_diff(prev_fname, fname, line_feed_sep=line_feed_sep),
+                            print("!!!! Exception in update_worker !!!\n", e)
-                                                'diff': diff.render_diff(prev_fname, fname, True, line_feed_sep=line_feed_sep)
-                                            })
-
-                                            self.notification_q.put(n_object)
-
-                            except Exception as e:
-                                print("!!!! Exception in update_worker !!!\n", e)
 
                 self.current_uuid = None  # Done
                 self.q.task_done()