Merge branch 'master' into extra-filters

changedetectionio/apprise_plugin/__init__.py

@@ -0,0 +1,78 @@
+# include the decorator
+from apprise.decorators import notify
+
+@notify(on="delete")
+@notify(on="deletes")
+@notify(on="get")
+@notify(on="gets")
+@notify(on="post")
+@notify(on="posts")
+@notify(on="put")
+@notify(on="puts")
+def apprise_custom_api_call_wrapper(body, title, notify_type, *args, **kwargs):
+    import requests
+    import json
+    from apprise.utils import parse_url as apprise_parse_url
+    from apprise import URLBase
+
+    url = kwargs['meta'].get('url')
+
+    if url.startswith('post'):
+        r = requests.post
+    elif url.startswith('get'):
+        r = requests.get
+    elif url.startswith('put'):
+        r = requests.put
+    elif url.startswith('delete'):
+        r = requests.delete
+
+    url = url.replace('post://', 'http://')
+    url = url.replace('posts://', 'https://')
+    url = url.replace('put://', 'http://')
+    url = url.replace('puts://', 'https://')
+    url = url.replace('get://', 'http://')
+    url = url.replace('gets://', 'https://')
+    url = url.replace('put://', 'http://')
+    url = url.replace('puts://', 'https://')
+    url = url.replace('delete://', 'http://')
+    url = url.replace('deletes://', 'https://')
+
+    headers = {}
+    params = {}
+    auth = None
+
+    # Convert /foobar?+some-header=hello to proper header dictionary
+    results = apprise_parse_url(url)
+    if results:
+        # Add our headers that the user can potentially over-ride if they wish
+        # to to our returned result set and tidy entries by unquoting them
+        headers = {URLBase.unquote(x): URLBase.unquote(y)
+                   for x, y in results['qsd+'].items()}
+
+        # https://github.com/caronc/apprise/wiki/Notify_Custom_JSON#get-parameter-manipulation
+        # In Apprise, it relies on prefixing each request arg with "-", because it uses say &method=update as a flag for apprise
+        # but here we are making straight requests, so we need todo convert this against apprise's logic
+        for k, v in results['qsd'].items():
+            if not k.strip('+-') in results['qsd+'].keys():
+                params[URLBase.unquote(k)] = URLBase.unquote(v)
+
+        # Determine Authentication
+        auth = ''
+        if results.get('user') and results.get('password'):
+            auth = (URLBase.unquote(results.get('user')), URLBase.unquote(results.get('user')))
+        elif results.get('user'):
+            auth = (URLBase.unquote(results.get('user')))
+
+    # Try to auto-guess if it's JSON
+    try:
+        json.loads(body)
+        headers['Content-Type'] = 'application/json; charset=utf-8'
+    except ValueError as e:
+        pass
+
+    r(results.get('url'),
+      auth=auth,
+      data=body.encode('utf-8') if type(body) is str else body,
+      headers=headers,
+      params=params
+      )

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -25,6 +25,7 @@ browser_step_ui_config = {'Choose one': '0 0',
                           'Click element if exists': '1 0',
                           'Click element': '1 0',
                           'Click element containing text': '0 1',
+                          'Click element containing text if exists': '0 1',
                           'Enter text in field': '1 1',
                           'Execute JS': '0 1',
 #                          'Extract text and use as filter': '1 0',
@@ -96,12 +97,24 @@ class steppable_browser_interface():
         return self.action_goto_url(value=self.start_url)
 
     def action_click_element_containing_text(self, selector=None, value=''):
+        logger.debug("Clicking element containing text")
         if not len(value.strip()):
             return
         elem = self.page.get_by_text(value)
         if elem.count():
             elem.first.click(delay=randint(200, 500), timeout=3000)
 
+    def action_click_element_containing_text_if_exists(self, selector=None, value=''):
+        logger.debug("Clicking element containing text if exists")
+        if not len(value.strip()):
+            return
+        elem = self.page.get_by_text(value)
+        logger.debug(f"Clicking element containing text - {elem.count()} elements found")
+        if elem.count():
+            elem.first.click(delay=randint(200, 500), timeout=3000)
+        else:
+            return
+
     def action_enter_text_in_field(self, selector, value):
         if not len(selector.strip()):
             return

changedetectionio/blueprint/tags/templates/edit-tag.html

@@ -58,9 +58,9 @@ xpath://body/div/span[contains(@class, 'example-class')]",
                         {% if '/text()' in  field %}
                           <span class="pure-form-message-inline"><strong>Note!: //text() function does not work where the &lt;element&gt; contains &lt;![CDATA[]]&gt;</strong></span><br>
                         {% endif %}
-                        <span class="pure-form-message-inline">One rule per line, <i>any</i> rules that matches will be used.<br>
-
-                    <ul>
+                        <span class="pure-form-message-inline">One CSS, xPath, JSON Path/JQ selector per line, <i>any</i> rules that matches will be used.<br>
+                    <div data-target="#advanced-help-selectors" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div>
+                    <ul id="advanced-help-selectors">
                         <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 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>

changedetectionio/content_fetchers/requests.py

@@ -1,8 +1,6 @@
 from loguru import logger
-import chardet
 import hashlib
 import os
-import requests
 from changedetectionio import strtobool
 from changedetectionio.content_fetchers.exceptions import BrowserStepsInUnsupportedFetcher, EmptyReply, Non200ErrorCodeReceived
 from changedetectionio.content_fetchers.base import Fetcher
@@ -28,6 +26,9 @@ class fetcher(Fetcher):
             is_binary=False,
             empty_pages_are_a_change=False):
 
+        import chardet
+        import requests
+
         if self.browser_steps_get_valid_steps():
             raise BrowserStepsInUnsupportedFetcher(url=url)
 

changedetectionio/flask_app.py

@@ -1,6 +1,8 @@
 #!/usr/bin/env python3
 
 import datetime
+import importlib
+
 import flask_login
 import locale
 import os
@@ -10,7 +12,9 @@ import threading
 import time
 import timeago
 
+from .content_fetchers.exceptions import ReplyWithContentButNoText
 from .processors import find_processors, get_parent_module, get_custom_watch_obj_for_processor
+from .processors.text_json_diff.processor import FilterNotFoundInResponse
 from .safe_jinja import render as jinja_render
 from changedetectionio.strtobool import strtobool
 from copy import deepcopy
@@ -537,7 +541,8 @@ def changedetection_app(config=None, datastore_o=None):
         import random
         from .apprise_asset import asset
         apobj = apprise.Apprise(asset=asset)
-
+        # so that the custom endpoints are registered
+        from changedetectionio.apprise_plugin import apprise_custom_api_call_wrapper
         is_global_settings_form = request.args.get('mode', '') == 'global-settings'
         is_group_settings_form = request.args.get('mode', '') == 'group-settings'
 
@@ -1395,6 +1400,57 @@ def changedetection_app(config=None, datastore_o=None):
         # Return a 500 error
         abort(500)
 
+    @app.route("/edit/<string:uuid>/preview-rendered", methods=['POST'])
+    @login_optionally_required
+    def watch_get_preview_rendered(uuid):
+        '''For when viewing the "preview" of the rendered text from inside of Edit'''
+        now = time.time()
+        import brotli
+        from . import forms
+
+        text_after_filter = ''
+        tmp_watch = deepcopy(datastore.data['watching'].get(uuid))
+
+        if tmp_watch and tmp_watch.history and os.path.isdir(tmp_watch.watch_data_dir):
+            # Splice in the temporary stuff from the form
+            form = forms.processor_text_json_diff_form(formdata=request.form if request.method == 'POST' else None,
+                                                       data=request.form
+                                                       )
+            # Only update vars that came in via the AJAX post
+            p = {k: v for k, v in form.data.items() if k in request.form.keys()}
+            tmp_watch.update(p)
+
+            latest_filename = next(reversed(tmp_watch.history))
+            html_fname = os.path.join(tmp_watch.watch_data_dir, f"{latest_filename}.html.br")
+            with open(html_fname, 'rb') as f:
+                decompressed_data = brotli.decompress(f.read()).decode('utf-8') if html_fname.endswith('.br') else f.read().decode('utf-8')
+
+                # Just like a normal change detection except provide a fake "watch" object and dont call .call_browser()
+                processor_module = importlib.import_module("changedetectionio.processors.text_json_diff.processor")
+                update_handler = processor_module.perform_site_check(datastore=datastore,
+                                                                     watch_uuid=uuid # probably not needed anymore anyway?
+                                                                     )
+                # Use the last loaded HTML as the input
+                update_handler.fetcher.content = decompressed_data
+                try:
+                    changed_detected, update_obj, contents, text_after_filter = update_handler.run_changedetection(
+                        watch=tmp_watch,
+                        skip_when_checksum_same=False,
+                    )
+                except FilterNotFoundInResponse as e:
+                    text_after_filter = f"Filter not found in HTML: {str(e)}"
+                except ReplyWithContentButNoText as e:
+                    text_after_filter = f"Filter found but no text (empty result)"
+                except Exception as e:
+                    text_after_filter = f"Error: {str(e)}"
+
+            if not text_after_filter.strip():
+                text_after_filter = 'Empty content'
+
+        logger.trace(f"Parsed in {time.time()-now:.3f}s")
+        return text_after_filter.strip()
+
+
     @app.route("/form/add/quickwatch", methods=['POST'])
     @login_optionally_required
     def form_quick_watch_add():

changedetectionio/forms.py

@@ -221,7 +221,8 @@ class ValidateAppRiseServers(object):
     def __call__(self, form, field):
         import apprise
         apobj = apprise.Apprise()
-
+        # so that the custom endpoints are registered
+        from changedetectionio.apprise_plugin import apprise_custom_api_call_wrapper
         for server_url in field.data:
             if not apobj.add(server_url):
                 message = field.gettext('\'%s\' is not a valid AppRise URL.' % (server_url))
@@ -479,8 +480,10 @@ class processor_text_json_diff_form(commonSettingsForm):
     body = TextAreaField('Request body', [validators.Optional()])
     method = SelectField('Request method', choices=valid_method, default=default_method)
     ignore_status_codes = BooleanField('Ignore status codes (process non-2xx status codes as normal)', default=False)
-    check_unique_lines = BooleanField('Only trigger when unique lines appear', default=False)
+    check_unique_lines = BooleanField('Only trigger when unique lines appear in all history', default=False)
+    remove_duplicate_lines = BooleanField('Remove duplicate lines of text', default=False)
     sort_text_alphabetically =  BooleanField('Sort text alphabetically', default=False)
+    trim_text_whitespace = BooleanField('Trim whitespace before and after text', default=False)
 
     filter_text_added = BooleanField('Added lines', default=True)
     filter_text_replaced = BooleanField('Replaced/changed lines', default=True)

changedetectionio/html_tools.py

@@ -1,10 +1,4 @@
-
-from bs4 import BeautifulSoup
-from inscriptis import get_text
-from jsonpath_ng.ext import parse
 from typing import List
-from inscriptis.model.config import ParserConfig
-from xml.sax.saxutils import escape as xml_escape
 import json
 import re
 
@@ -39,6 +33,7 @@ def perl_style_slash_enclosed_regex_to_options(regex):
 
 # Given a CSS Rule, and a blob of HTML, return the blob of HTML that matches
 def include_filters(include_filters, html_content, append_pretty_line_formatting=False):
+    from bs4 import BeautifulSoup
     soup = BeautifulSoup(html_content, "html.parser")
     html_block = ""
     r = soup.select(include_filters, separator="")
@@ -56,6 +51,7 @@ def include_filters(include_filters, html_content, append_pretty_line_formatting
     return html_block
 
 def subtractive_css_selector(css_selector, html_content):
+    from bs4 import BeautifulSoup
     soup = BeautifulSoup(html_content, "html.parser")
     for item in soup.select(css_selector):
         item.decompose()
@@ -181,6 +177,7 @@ def xpath1_filter(xpath_filter, html_content, append_pretty_line_formatting=Fals
 
 # Extract/find element
 def extract_element(find='title', html_content=''):
+    from bs4 import BeautifulSoup
 
     #Re #106, be sure to handle when its not found
     element_text = None
@@ -194,6 +191,8 @@ def extract_element(find='title', html_content=''):
 
 #
 def _parse_json(json_data, json_filter):
+    from jsonpath_ng.ext import parse
+
     if json_filter.startswith("json:"):
         jsonpath_expression = parse(json_filter.replace('json:', ''))
         match = jsonpath_expression.find(json_data)
@@ -242,6 +241,8 @@ def _get_stripped_text_from_json_match(match):
 # json_filter - ie json:$..price
 # ensure_is_ldjson_info_type - str "product", optional, "@type == product" (I dont know how to do that as a json selector)
 def extract_json_as_string(content, json_filter, ensure_is_ldjson_info_type=None):
+    from bs4 import BeautifulSoup
+
     stripped_text_from_html = False
 # https://github.com/dgtlmoon/changedetection.io/pull/2041#issuecomment-1848397161w
     # Try to parse/filter out the JSON, if we get some parser error, then maybe it's embedded within HTML tags
@@ -352,6 +353,7 @@ def strip_ignore_text(content, wordlist, mode="content"):
     return "\n".encode('utf8').join(output)
 
 def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False) -> str:
+    from xml.sax.saxutils import escape as xml_escape
     pattern = '<!\[CDATA\[(\s*(?:.(?<!\]\]>)\s*)*)\]\]>'
     def repl(m):
         text = m.group(1)
@@ -360,6 +362,9 @@ def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False
     return re.sub(pattern, repl, html_content)
 
 def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False) -> str:
+    from inscriptis import get_text
+    from inscriptis.model.config import ParserConfig
+
     """Converts html string to a string with just the text. If ignoring
     rendering anchor tag content is enable, anchor tag content are also
     included in the text

changedetectionio/model/__init__.py

@@ -59,6 +59,7 @@ class watch_base(dict):
             'time_between_check': {'weeks': None, 'days': None, 'hours': None, 'minutes': None, 'seconds': None},
             'time_between_check_use_default': True,
             'title': None,
+            'trim_text_whitespace': False,
             'track_ldjson_price_data': None,
             'trigger_text': [],  # List of text or regex to wait for until a change is detected
             'url': '',

changedetectionio/notification.py

@@ -1,9 +1,10 @@
-import apprise
+
 import time
 from apprise import NotifyFormat
-import json
+import apprise
 from loguru import logger
 
+
 valid_tokens = {
     'base_url': '',
     'current_snapshot': '',
@@ -34,86 +35,11 @@ valid_notification_formats = {
     default_notification_format_for_watch: default_notification_format_for_watch
 }
 
-# include the decorator
-from apprise.decorators import notify
-
-@notify(on="delete")
-@notify(on="deletes")
-@notify(on="get")
-@notify(on="gets")
-@notify(on="post")
-@notify(on="posts")
-@notify(on="put")
-@notify(on="puts")
-def apprise_custom_api_call_wrapper(body, title, notify_type, *args, **kwargs):
-    import requests
-    from apprise.utils import parse_url as apprise_parse_url
-    from apprise import URLBase
-
-    url = kwargs['meta'].get('url')
-
-    if url.startswith('post'):
-        r = requests.post
-    elif url.startswith('get'):
-        r = requests.get
-    elif url.startswith('put'):
-        r = requests.put
-    elif url.startswith('delete'):
-        r = requests.delete
-
-    url = url.replace('post://', 'http://')
-    url = url.replace('posts://', 'https://')
-    url = url.replace('put://', 'http://')
-    url = url.replace('puts://', 'https://')
-    url = url.replace('get://', 'http://')
-    url = url.replace('gets://', 'https://')
-    url = url.replace('put://', 'http://')
-    url = url.replace('puts://', 'https://')
-    url = url.replace('delete://', 'http://')
-    url = url.replace('deletes://', 'https://')
-
-    headers = {}
-    params = {}
-    auth = None
-
-    # Convert /foobar?+some-header=hello to proper header dictionary
-    results = apprise_parse_url(url)
-    if results:
-        # Add our headers that the user can potentially over-ride if they wish
-        # to to our returned result set and tidy entries by unquoting them
-        headers = {URLBase.unquote(x): URLBase.unquote(y)
-                   for x, y in results['qsd+'].items()}
-
-        # https://github.com/caronc/apprise/wiki/Notify_Custom_JSON#get-parameter-manipulation
-        # In Apprise, it relies on prefixing each request arg with "-", because it uses say &method=update as a flag for apprise
-        # but here we are making straight requests, so we need todo convert this against apprise's logic
-        for k, v in results['qsd'].items():
-            if not k.strip('+-') in results['qsd+'].keys():
-                params[URLBase.unquote(k)] = URLBase.unquote(v)
-
-        # Determine Authentication
-        auth = ''
-        if results.get('user') and results.get('password'):
-            auth = (URLBase.unquote(results.get('user')), URLBase.unquote(results.get('user')))
-        elif results.get('user'):
-            auth = (URLBase.unquote(results.get('user')))
-
-    # Try to auto-guess if it's JSON
-    try:
-        json.loads(body)
-        headers['Content-Type'] = 'application/json; charset=utf-8'
-    except ValueError as e:
-        pass
-
-    r(results.get('url'),
-      auth=auth,
-      data=body.encode('utf-8') if type(body) is str else body,
-      headers=headers,
-      params=params
-      )
 
 
 def process_notification(n_object, datastore):
+    # so that the custom endpoints are registered
+    from changedetectionio.apprise_plugin import apprise_custom_api_call_wrapper
 
     from .safe_jinja import render as jinja_render
     now = time.time()

changedetectionio/processors/__init__.py

@@ -1,4 +1,6 @@
 from abc import abstractmethod
+
+from changedetectionio.content_fetchers.base import Fetcher
 from changedetectionio.strtobool import strtobool
 
 from copy import deepcopy
@@ -23,10 +25,11 @@ class difference_detection_processor():
         super().__init__(*args, **kwargs)
         self.datastore = datastore
         self.watch = deepcopy(self.datastore.data['watching'].get(watch_uuid))
+        # Generic fetcher that should be extended (requests, playwright etc)
+        self.fetcher = Fetcher()
 
     def call_browser(self):
         from requests.structures import CaseInsensitiveDict
-        from changedetectionio.content_fetchers.exceptions import EmptyReply
 
         # Protect against file:// access
         if re.search(r'^file://', self.watch.get('url', '').strip(), re.IGNORECASE):
@@ -159,7 +162,7 @@ class difference_detection_processor():
         some_data = 'xxxxx'
         update_obj["previous_md5"] = hashlib.md5(some_data.encode('utf-8')).hexdigest()
         changed_detected = False
-        return changed_detected, update_obj, ''.encode('utf-8')
+        return changed_detected, update_obj, ''.encode('utf-8'), b''
 
 
 def find_sub_packages(package_name):

changedetectionio/processors/restock_diff/processor.py

@@ -2,8 +2,7 @@ from .. import difference_detection_processor
 from ..exceptions import ProcessorException
 from . import Restock
 from loguru import logger
-import hashlib
-import re
+
 import urllib3
 import time
 
@@ -27,6 +26,25 @@ def _search_prop_by_value(matches, value):
             if value in prop[0]:
                 return prop[1]  # Yield the desired value and exit the function
 
+def _deduplicate_prices(data):
+    seen = set()
+    unique_data = []
+
+    for datum in data:
+        # Convert 'value' to float if it can be a numeric string, otherwise leave it as is
+        try:
+            normalized_value = float(datum.value) if isinstance(datum.value, str) and datum.value.replace('.', '', 1).isdigit() else datum.value
+        except ValueError:
+            normalized_value = datum.value
+
+        # If the normalized value hasn't been seen yet, add it to unique data
+        if normalized_value not in seen:
+            unique_data.append(datum)
+            seen.add(normalized_value)
+    
+    return unique_data
+
+
 # should return Restock()
 # add casting?
 def get_itemprop_availability(html_content) -> Restock:
@@ -36,6 +54,7 @@ def get_itemprop_availability(html_content) -> Restock:
     """
     from jsonpath_ng import parse
 
+    import re
     now = time.time()
     import extruct
     logger.trace(f"Imported extruct module in {time.time() - now:.3f}s")
@@ -60,7 +79,7 @@ def get_itemprop_availability(html_content) -> Restock:
         pricecurrency_parse = parse('$..(pricecurrency|currency|priceCurrency )')
         availability_parse = parse('$..(availability|Availability)')
 
-        price_result = price_parse.find(data)
+        price_result = _deduplicate_prices(price_parse.find(data))
         if price_result:
             # Right now, we just support single product items, maybe we will store the whole actual metadata seperately in teh future and
             # parse that for the UI?
@@ -122,6 +141,10 @@ class perform_site_check(difference_detection_processor):
     xpath_data = None
 
     def run_changedetection(self, watch, skip_when_checksum_same=True):
+        import hashlib
+
+        from concurrent.futures import ProcessPoolExecutor
+        from functools import partial
         if not watch:
             raise Exception("Watch no longer exists.")
 
@@ -149,7 +172,11 @@ class perform_site_check(difference_detection_processor):
 
         itemprop_availability = {}
         try:
-            itemprop_availability = get_itemprop_availability(html_content=self.fetcher.content)
+            with ProcessPoolExecutor() as executor:
+                # Use functools.partial to create a callable with arguments
+                # anything using bs4/lxml etc is quite "leaky"
+                future = executor.submit(partial(get_itemprop_availability, self.fetcher.content))
+                itemprop_availability = future.result()
         except MoreThanOnePriceFound as e:
             # Add the real data
             raise ProcessorException(message="Cannot run, more than one price detected, this plugin is only for product pages with ONE product, try the content-change detection mode.",
@@ -263,4 +290,4 @@ class perform_site_check(difference_detection_processor):
         # Always record the new checksum
         update_obj["previous_md5"] = fetched_md5
 
-        return changed_detected, update_obj, snapshot_content.encode('utf-8').strip()
+        return changed_detected, update_obj, snapshot_content.encode('utf-8').strip(), b''

changedetectionio/processors/text_json_diff/processor.py

@@ -36,6 +36,9 @@ class PDFToHTMLToolNotFound(ValueError):
 class perform_site_check(difference_detection_processor):
 
     def run_changedetection(self, watch, skip_when_checksum_same=True):
+        from concurrent.futures import ProcessPoolExecutor
+        from functools import partial
+
         changed_detected = False
         html_content = ""
         screenshot = False  # as bytes
@@ -171,20 +174,30 @@ class perform_site_check(difference_detection_processor):
                     for filter_rule in include_filters_rule:
                         # For HTML/XML we offer xpath as an option, just start a regular xPath "/.."
                         if filter_rule[0] == '/' or filter_rule.startswith('xpath:'):
-                            html_content += html_tools.xpath_filter(xpath_filter=filter_rule.replace('xpath:', ''),
+                            with ProcessPoolExecutor() as executor:
+                                # Use functools.partial to create a callable with arguments - anything using bs4/lxml etc is quite "leaky"
+                                future = executor.submit(partial(html_tools.xpath_filter, xpath_filter=filter_rule.replace('xpath:', ''),
                                                                     html_content=self.fetcher.content,
                                                                     append_pretty_line_formatting=not watch.is_source_type_url,
-                                                                    is_rss=is_rss)
+                                                                    is_rss=is_rss))
+                                html_content += future.result()
+
                         elif filter_rule.startswith('xpath1:'):
-                            html_content += html_tools.xpath1_filter(xpath_filter=filter_rule.replace('xpath1:', ''),
+                            with ProcessPoolExecutor() as executor:
+                                # Use functools.partial to create a callable with arguments - anything using bs4/lxml etc is quite "leaky"
+                                future = executor.submit(partial(html_tools.xpath1_filter, xpath_filter=filter_rule.replace('xpath1:', ''),
                                                                     html_content=self.fetcher.content,
                                                                     append_pretty_line_formatting=not watch.is_source_type_url,
-                                                                    is_rss=is_rss)
+                                                                    is_rss=is_rss))
+                                html_content += future.result()
                         else:
-                            # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
-                            html_content += html_tools.include_filters(include_filters=filter_rule,
+                            with ProcessPoolExecutor() as executor:
+                                # Use functools.partial to create a callable with arguments - anything using bs4/lxml etc is quite "leaky"
+                                # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
+                                future = executor.submit(partial(html_tools.include_filters, include_filters=filter_rule,
                                                                        html_content=self.fetcher.content,
-                                                                       append_pretty_line_formatting=not watch.is_source_type_url)
+                                                                       append_pretty_line_formatting=not watch.is_source_type_url))
+                                html_content += future.result()
 
                     if not html_content.strip():
                         raise FilterNotFoundInResponse(msg=include_filters_rule, screenshot=self.fetcher.screenshot, xpath_data=self.fetcher.xpath_data)
@@ -197,18 +210,13 @@ class perform_site_check(difference_detection_processor):
                 else:
                     # extract text
                     do_anchor = self.datastore.data["settings"]["application"].get("render_anchor_tag_content", False)
-                    stripped_text_from_html = \
-                        html_tools.html_to_text(
-                            html_content=html_content,
+                    with ProcessPoolExecutor() as executor:
+                        # Use functools.partial to create a callable with arguments - anything using bs4/lxml etc is quite "leaky"
+                        # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
+                        future = executor.submit(partial(html_tools.html_to_text, html_content=html_content,
                             render_anchor_tag_content=do_anchor,
-                            is_rss=is_rss # #1874 activate the <title workaround hack
-                        )
-
-        if watch.get('sort_text_alphabetically') and stripped_text_from_html:
-            # Note: Because a <p>something</p> will add an extra line feed to signify the paragraph gap
-            # we end up with 'Some text\n\n', sorting will add all those extra \n at the start, so we remove them here.
-            stripped_text_from_html = stripped_text_from_html.replace('\n\n', '\n')
-            stripped_text_from_html = '\n'.join( sorted(stripped_text_from_html.splitlines(), key=lambda x: x.lower() ))
+                            is_rss=is_rss)) #1874 activate the <title workaround hack
+                        stripped_text_from_html = future.result()
 
         # Re #340 - return the content before the 'ignore text' was applied
         text_content_before_ignored_filter = stripped_text_from_html.encode('utf-8')
@@ -236,7 +244,7 @@ class perform_site_check(difference_detection_processor):
                 # We had some content, but no differences were found
                 # Store our new file as the MD5 so it will trigger in the future
                 c = hashlib.md5(text_content_before_ignored_filter.translate(None, b'\r\n\t ')).hexdigest()
-                return False, {'previous_md5': c}, stripped_text_from_html.encode('utf-8')
+                return False, {'previous_md5': c}, stripped_text_from_html.encode('utf-8'), stripped_text_from_html.encode('utf-8')
             else:
                 stripped_text_from_html = rendered_diff
 
@@ -298,6 +306,20 @@ class perform_site_check(difference_detection_processor):
                 stripped_text_from_html = b''.join(regex_matched_output)
                 text_content_before_ignored_filter = stripped_text_from_html
 
+
+        if watch.get('sort_text_alphabetically') and stripped_text_from_html:
+            # Note: Because a <p>something</p> will add an extra line feed to signify the paragraph gap
+            # we end up with 'Some text\n\n', sorting will add all those extra \n at the start, so we remove them here.
+            stripped_text_from_html = stripped_text_from_html.replace(b'\n\n', b'\n')
+            stripped_text_from_html = '\n'.join(sorted(stripped_text_from_html.decode('utf-8').splitlines(), key=lambda x: x.lower())).encode('utf-8')
+
+        #
+        if watch.get('trim_text_whitespace') and stripped_text_from_html:
+            stripped_text_from_html = '\n'.join(line.strip() for line in stripped_text_from_html.decode('utf-8').splitlines()).encode('utf-8')
+#
+        if watch.get('remove_duplicate_lines') and stripped_text_from_html:
+            stripped_text_from_html = '\n'.join(dict.fromkeys(line for line in stripped_text_from_html.decode('utf-8').splitlines())).encode('utf-8')
+
         # 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()
@@ -357,4 +379,4 @@ class perform_site_check(difference_detection_processor):
         if not watch.get('previous_md5'):
             watch['previous_md5'] = fetched_md5
 
-        return changed_detected, update_obj, text_content_before_ignored_filter
+        return changed_detected, update_obj, text_content_before_ignored_filter, stripped_text_from_html

changedetectionio/static/js/global-settings.js

@@ -18,9 +18,11 @@ $(document).ready(function () {
 
     });
 
-    $("#notification-token-toggle").click(function (e) {
+    $(".toggle-show").click(function (e) {
         e.preventDefault();
-        $('#notification-tokens-info').toggle();
+        let target = $(this).data('target');
+        $(target).toggle();
     });
+
 });
 

changedetectionio/static/js/watch-settings.js

@@ -12,6 +12,50 @@ function toggleOpacity(checkboxSelector, fieldSelector, inverted) {
     checkbox.addEventListener('change', updateOpacity);
 }
 
+(function($) {
+    // Object to store ongoing requests by namespace
+    const requests = {};
+
+    $.abortiveSingularAjax = function(options) {
+        const namespace = options.namespace || 'default';
+
+        // Abort the current request in this namespace if it's still ongoing
+        if (requests[namespace]) {
+            requests[namespace].abort();
+        }
+
+        // Start a new AJAX request and store its reference in the correct namespace
+        requests[namespace] = $.ajax(options);
+
+        // Return the current request in case it's needed
+        return requests[namespace];
+    };
+})(jQuery);
+
+function request_textpreview_update() {
+    const data = {};
+    $('textarea:visible, input:visible').each(function () {
+        const $element = $(this); // Cache the jQuery object for the current element
+        const name = $element.attr('name'); // Get the name attribute of the element
+        data[name] = $element.is(':checkbox') ? ($element.is(':checked') ? $element.val() : undefined) : $element.val();
+    });
+
+    $.abortiveSingularAjax({
+        type: "POST",
+        url: preview_text_edit_filters_url,
+        data: data,
+        namespace: 'watchEdit'
+    }).done(function (data) {
+        $('#filters-and-triggers #text-preview-inner').text(data);
+    }).fail(function (error) {
+        if (error.statusText === 'abort') {
+            console.log('Request was aborted due to a new request being fired.');
+        } else {
+            $('#filters-and-triggers #text-preview-inner').text('There was an error communicating with the server.');
+        }
+    })
+}
+
 $(document).ready(function () {
     $('#notification-setting-reset-to-default').click(function (e) {
         $('#notification_title').val('');
@@ -27,5 +71,22 @@ $(document).ready(function () {
 
     toggleOpacity('#time_between_check_use_default', '#time_between_check', false);
 
+    const vh = Math.max(document.documentElement.clientHeight || 0, window.innerHeight || 0);
+    $("#text-preview-inner").css('max-height', (vh-300)+"px");
+
+    $("#activate-text-preview").click(function (e) {
+        $(this).fadeOut();
+        $('body').toggleClass('preview-text-enabled')
+
+        request_textpreview_update();
+        $("#text-preview-refresh").click(function (e) {
+            request_textpreview_update();
+        });
+        $('textarea:visible, input:visible').on('keyup keypress blur change click', function (e) {
+            request_textpreview_update();
+        });
+
+    });
+
 });
 

changedetectionio/static/styles/scss/parts/_preview_text_filter.scss

@@ -0,0 +1,45 @@
+body.preview-text-enabled {
+  #filters-and-triggers > div {
+    display: flex; /* Establishes Flexbox layout */
+    gap: 20px; /* Adds space between the columns */
+    position: relative; /* Ensures the sticky positioning is relative to this parent */
+  }
+
+  /* layout of the page */
+  #edit-text-filter, #text-preview {
+    flex: 1; /* Each column takes an equal amount of available space */
+    align-self: flex-start; /* Aligns the right column to the start, allowing it to maintain its content height */
+  }
+
+  #edit-text-filter {
+    #pro-tips {
+      display: none;
+    }
+  }
+
+  #text-preview {
+    position: sticky;
+    top: 25px;
+    display: block !important;
+  }
+
+  /* actual preview area */
+  #text-preview-inner {
+    background: var(--color-grey-900);
+    border: 1px solid var(--color-grey-600);
+    padding: 1rem;
+    color: #333;
+    font-family: "Courier New", Courier, monospace; /* Sets the font to a monospace type */
+    font-size: 12px;
+    overflow-x: scroll;
+    white-space: pre-wrap; /* Preserves whitespace and line breaks like <pre> */
+    overflow-wrap: break-word; /* Allows long words to break and wrap to the next line */
+  }
+}
+
+#activate-text-preview {
+  right: 0;
+  position: absolute;
+  z-index: 0;
+  box-shadow: 1px 1px 4px var(--color-shadow-jump);
+}

changedetectionio/static/styles/scss/styles.scss

@@ -12,6 +12,7 @@
 @import "parts/_darkmode";
 @import "parts/_menu";
 @import "parts/_love";
+@import "parts/preview_text_filter";
 
 body {
   color: var(--color-text);

changedetectionio/static/styles/styles.css

@@ -411,6 +411,47 @@ html[data-darkmode="true"] #toggle-light-mode .icon-dark {
     fill: #ff0000 !important;
     transition: all ease 0.3s !important; }
 
+body.preview-text-enabled {
+  /* layout of the page */
+  /* actual preview area */ }
+  body.preview-text-enabled #filters-and-triggers > div {
+    display: flex;
+    /* Establishes Flexbox layout */
+    gap: 20px;
+    /* Adds space between the columns */
+    position: relative;
+    /* Ensures the sticky positioning is relative to this parent */ }
+  body.preview-text-enabled #edit-text-filter, body.preview-text-enabled #text-preview {
+    flex: 1;
+    /* Each column takes an equal amount of available space */
+    align-self: flex-start;
+    /* Aligns the right column to the start, allowing it to maintain its content height */ }
+  body.preview-text-enabled #edit-text-filter #pro-tips {
+    display: none; }
+  body.preview-text-enabled #text-preview {
+    position: sticky;
+    top: 25px;
+    display: block !important; }
+  body.preview-text-enabled #text-preview-inner {
+    background: var(--color-grey-900);
+    border: 1px solid var(--color-grey-600);
+    padding: 1rem;
+    color: #333;
+    font-family: "Courier New", Courier, monospace;
+    /* Sets the font to a monospace type */
+    font-size: 12px;
+    overflow-x: scroll;
+    white-space: pre-wrap;
+    /* Preserves whitespace and line breaks like <pre> */
+    overflow-wrap: break-word;
+    /* Allows long words to break and wrap to the next line */ }
+
+#activate-text-preview {
+  right: 0;
+  position: absolute;
+  z-index: 0;
+  box-shadow: 1px 1px 4px var(--color-shadow-jump); }
+
 body {
   color: var(--color-text);
   background: var(--color-background-page);
@@ -1194,11 +1235,9 @@ ul {
   color: #fff;
   opacity: 0.7; }
 
-
 .restock-label svg {
   vertical-align: middle; }
 
-
 #chrome-extension-link {
   padding: 9px;
   border: 1px solid var(--color-grey-800);

changedetectionio/store.py

@@ -11,7 +11,6 @@ from threading import Lock
 import json
 import os
 import re
-import requests
 import secrets
 import threading
 import time
@@ -270,6 +269,7 @@ class ChangeDetectionStore:
         self.needs_write_urgent = True
 
     def add_watch(self, url, tag='', extras=None, tag_uuids=None, write_to_disk_now=True):
+        import requests
 
         if extras is None:
             extras = {}

changedetectionio/templates/_common_fields.html

@@ -11,8 +11,11 @@
     class="notification-urls" )
                             }}
                             <div class="pure-form-message-inline">
-                              <ul>
-                                <li>Use <a target=_new href="https://github.com/caronc/apprise">AppRise URLs</a> for notification to just about any service! <i><a target=_new href="https://github.com/dgtlmoon/changedetection.io/wiki/Notification-configuration-notes">Please read the notification services wiki here for important configuration notes</a></i>.</li>
+                                <p>
+                                <strong>Tip:</strong> Use <a target=_new href="https://github.com/caronc/apprise">AppRise Notification URLs</a> for notification to just about any service! <i><a target=_new href="https://github.com/dgtlmoon/changedetection.io/wiki/Notification-configuration-notes">Please read the notification services wiki here for important configuration notes</a></i>.<br>
+</p>
+                                <div data-target="#advanced-help-notifications" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div>
+                              <ul style="display: none" id="advanced-help-notifications">
                                 <li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_discord">discord://</a></code> (or <code>https://discord.com/api/webhooks...</code>)) only supports a maximum <strong>2,000 characters</strong> of notification text, including the title.</li>
                                 <li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_telegram">tgram://</a></code> bots can't send messages to other bots, so you should specify chat ID of non-bot user.</li>
                                 <li><code><a target=_new href="https://github.com/caronc/apprise/wiki/Notify_telegram">tgram://</a></code> only supports very limited HTML and can fail when extra tags are sent, <a href="https://core.telegram.org/bots/api#html-style">read more here</a> (or use plaintext/markdown format)</li>
@@ -40,7 +43,7 @@
 
                             </div>
                             <div class="pure-controls">
-                                <div id="notification-token-toggle" class="pure-button button-tag button-xsmall">Show token/placeholders</div>
+                                <div data-target="#notification-tokens-info" class="toggle-show pure-button button-tag button-xsmall">Show token/placeholders</div>
                             </div>
                             <div class="pure-controls" style="display: none;" id="notification-tokens-info">
                                 <table class="pure-table" id="token-table">

changedetectionio/templates/base.html

@@ -33,7 +33,7 @@
     <script src="{{url_for('static_content', group='js', filename='csrf.js')}}" defer></script>
   </head>
 
-  <body>
+  <body class="">
     <div class="header">
       <div class="home-menu pure-menu pure-menu-horizontal pure-menu-fixed" id="nav-menu">
         {% if has_password and not current_user.is_authenticated %}

changedetectionio/templates/edit.html

@@ -4,6 +4,7 @@
 {% from '_common_fields.html' import render_common_settings_form %}
 <script src="{{url_for('static_content', group='js', filename='tabs.js')}}" defer></script>
 <script src="{{url_for('static_content', group='js', filename='vis.js')}}" defer></script>
+<script src="{{url_for('static_content', group='js', filename='global-settings.js')}}" defer></script>
 <script>
     const browser_steps_available_screenshots=JSON.parse('{{ watch.get_browsersteps_available_screenshots|tojson }}');
     const browser_steps_config=JSON.parse('{{ browser_steps_config|tojson }}');
@@ -253,7 +254,10 @@ User-Agent: wonderbra 1.0") }}
 
             {% if watch['processor'] == 'text_json_diff' %}
             <div class="tab-pane-inner" id="filters-and-triggers">
-                    <div class="pure-control-group">
+                <span id="activate-text-preview" class="pure-button pure-button-primary button-xsmall">Activate preview</span>
+              <div>
+              <div id="edit-text-filter">
+                    <div class="pure-control-group" id="pro-tips">
                             <strong>Pro-tips:</strong><br>
                             <ul>
                                 <li>
@@ -275,9 +279,9 @@ xpath://body/div/span[contains(@class, 'example-class')]",
                         {% if '/text()' in  field %}
                           <span class="pure-form-message-inline"><strong>Note!: //text() function does not work where the &lt;element&gt; contains &lt;![CDATA[]]&gt;</strong></span><br>
                         {% endif %}
-                        <span class="pure-form-message-inline">One rule per line, <i>any</i> rules that matches will be used.<br>
-
-                    <ul>
+                        <span class="pure-form-message-inline">One CSS, xPath, JSON Path/JQ selector per line, <i>any</i> rules that matches will be used.<br>
+<p><div data-target="#advanced-help-selectors" class="toggle-show pure-button button-tag button-xsmall">Show advanced help and tips</div><br></p>
+                    <ul id="advanced-help-selectors" style="display: none;">
                         <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 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>
@@ -297,9 +301,12 @@ xpath://body/div/span[contains(@class, 'example-class')]",
                                 <li>To use XPath1.0: Prefix with <code>xpath1:</code></li>
                             </ul>
                             </li>
-                    </ul>
-                    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
+                    <li>
+                        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>
+                    </li>
+                    </ul>
+
                 </span>
                     </div>
                 <fieldset class="pure-control-group">
@@ -326,14 +333,21 @@ nav
                     <span class="pure-form-message-inline">So it's always better to select <strong>Added</strong>+<strong>Replaced</strong> when you're interested in new content.</span><br>
                     <span class="pure-form-message-inline">When content is merely moved in a list, it will also trigger an <strong>addition</strong>, consider enabling <code><strong>Only trigger when unique lines appear</strong></code></span>
                 </fieldset>
-
+                <fieldset class="pure-control-group">
+                    {{ render_checkbox_field(form.check_unique_lines) }}
+                    <span class="pure-form-message-inline">Good for websites that just move the content around, and you want to know when NEW content is added, compares new lines against all history for this watch.</span>
+                </fieldset>
+                <fieldset class="pure-control-group">
+                    {{ render_checkbox_field(form.remove_duplicate_lines) }}
+                    <span class="pure-form-message-inline">Remove duplicate lines of text</span>
+                </fieldset>
                 <fieldset class="pure-control-group">
                     {{ render_checkbox_field(form.sort_text_alphabetically) }}
                     <span class="pure-form-message-inline">Helps reduce changes detected caused by sites shuffling lines around, combine with <i>check unique lines</i> below.</span>
                 </fieldset>
                 <fieldset class="pure-control-group">
-                    {{ render_checkbox_field(form.check_unique_lines) }}
-                    <span class="pure-form-message-inline">Good for websites that just move the content around, and you want to know when NEW content is added, compares new lines against all history for this watch.</span>
+                    {{ render_checkbox_field(form.trim_text_whitespace) }}
+                    <span class="pure-form-message-inline">Remove any whitespace before and after each line of text</span>
                 </fieldset>
                 <fieldset>
                     <div class="pure-control-group">
@@ -383,7 +397,7 @@ Unavailable") }}
                 </fieldset>
                 <fieldset>
                     <div class="pure-control-group">
-                        {{ render_field(form.extract_text, rows=5, placeholder="\d+ online") }}
+                        {{ render_field(form.extract_text, rows=5, placeholder="Example: /\d+ online/") }}
                         <span class="pure-form-message-inline">
                     <ul>
                         <li>Extracts text in the final output (line by line) after other filters using regular expressions or string match;
@@ -403,7 +417,19 @@ Unavailable") }}
                 </fieldset>
                 </div>
             </div>
-            {% endif %}
+              <div id="text-preview" style="display: none;" >
+                    <script>
+                        const preview_text_edit_filters_url="{{url_for('watch_get_preview_rendered', uuid=uuid)}}";
+                    </script>
+                    <span><strong>Preview of the text that is used for changedetection after all filters run.</strong></span><br>
+                    {#<div id="text-preview-controls"><span id="text-preview-refresh" class="pure-button button-xsmall">Refresh</span></div>#}
+                <p>
+                    <div id="text-preview-inner"></div>
+                </p>
+            </div>
+          </div>
+        </div>
+        {% endif %}
         {# rendered sub Template #}
         {% if extra_form_content %}
             <div class="tab-pane-inner" id="extras_tab">

changedetectionio/tests/restock/test_restock.py

@@ -2,7 +2,7 @@
 import os
 import time
 from flask import url_for
-from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
+from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client, wait_for_notification_endpoint_output
 from changedetectionio.notification import (
     default_notification_body,
     default_notification_format,
@@ -94,7 +94,7 @@ def test_restock_detection(client, live_server, measure_memory_usage):
     assert b'not-in-stock' not in res.data
 
     # We should have a notification
-    time.sleep(2)
+    wait_for_notification_endpoint_output()
     assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
     os.unlink("test-datastore/notification.txt")
 
@@ -103,6 +103,7 @@ def test_restock_detection(client, live_server, measure_memory_usage):
     set_original_response()
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
+    time.sleep(5)
     assert not os.path.isfile("test-datastore/notification.txt"), "No notification should have fired when it went OUT OF STOCK by default"
 
     # BUT we should see that it correctly shows "not in stock"

changedetectionio/tests/test_add_replace_remove_filter.py

@@ -2,7 +2,7 @@
 import os.path
 import time
 from flask import url_for
-from .util import live_server_setup, wait_for_all_checks
+from .util import live_server_setup, wait_for_all_checks, wait_for_notification_endpoint_output
 from changedetectionio import html_tools
 
 
@@ -165,7 +165,7 @@ def test_check_add_line_contains_trigger(client, live_server, measure_memory_usa
     assert b'unviewed' in res.data
 
     # Takes a moment for apprise to fire
-    time.sleep(3)
+    wait_for_notification_endpoint_output()
     assert os.path.isfile("test-datastore/notification.txt"), "Notification fired because I can see the output file"
     with open("test-datastore/notification.txt", 'rb') as f:
         response = f.read()

changedetectionio/tests/test_filter_exist_changes.py

@@ -4,7 +4,7 @@
 import os
 import time
 from flask import url_for
-from .util import set_original_response, live_server_setup
+from .util import set_original_response, live_server_setup, wait_for_notification_endpoint_output
 from changedetectionio.model import App
 
 
@@ -102,14 +102,15 @@ def test_filter_doesnt_exist_then_exists_should_get_notification(client, live_se
         follow_redirects=True
     )
     assert b"Updated watch." in res.data
-    time.sleep(3)
+    wait_for_notification_endpoint_output()
 
     # Shouldn't exist, shouldn't have fired
     assert not os.path.isfile("test-datastore/notification.txt")
     # Now the filter should exist
     set_response_with_filter()
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
-    time.sleep(3)
+
+    wait_for_notification_endpoint_output()
 
     assert os.path.isfile("test-datastore/notification.txt")
 

changedetectionio/tests/test_filter_failure_notification.py

@@ -1,7 +1,9 @@
 import os
 import time
+from loguru import logger
 from flask import url_for
-from .util import set_original_response, live_server_setup, extract_UUID_from_client, wait_for_all_checks
+from .util import set_original_response, live_server_setup, extract_UUID_from_client, wait_for_all_checks, \
+    wait_for_notification_endpoint_output
 from changedetectionio.model import App
 
 
@@ -26,6 +28,12 @@ def run_filter_test(client, live_server, content_filter):
     # Response WITHOUT the filter ID element
     set_original_response()
 
+    # Goto the edit page, add our ignore text
+    notification_url = url_for('test_notification_endpoint', _external=True).replace('http', 'json')
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+
     # cleanup for the next
     client.get(
         url_for("form_delete", uuid="all"),
@@ -34,83 +42,92 @@ def run_filter_test(client, live_server, content_filter):
     if os.path.isfile("test-datastore/notification.txt"):
         os.unlink("test-datastore/notification.txt")
 
-    # Add our URL to the import page
-    test_url = url_for('test_endpoint', _external=True)
     res = client.post(
-        url_for("form_quick_watch_add"),
-        data={"url": test_url, "tags": ''},
+        url_for("import_page"),
+        data={"urls": test_url},
         follow_redirects=True
     )
 
-    assert b"Watch added" in res.data
-
-    # Give the thread time to pick up the first version
+    assert b"1 Imported" in res.data
     wait_for_all_checks(client)
 
-    # Goto the edit page, add our ignore text
-    # Add our URL to the import page
-    url = url_for('test_notification_endpoint', _external=True)
-    notification_url = url.replace('http', 'json')
+    uuid = extract_UUID_from_client(client)
 
-    print(">>>> Notification URL: " + notification_url)
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid]['consecutive_filter_failures'] == 0, "No filter = No filter failure"
 
-    # Just a regular notification setting, this will be used by the special 'filter not found' notification
-    notification_form_data = {"notification_urls": notification_url,
-                              "notification_title": "New ChangeDetection.io Notification - {{watch_url}}",
-                              "notification_body": "BASE URL: {{base_url}}\n"
-                                                   "Watch URL: {{watch_url}}\n"
-                                                   "Watch UUID: {{watch_uuid}}\n"
-                                                   "Watch title: {{watch_title}}\n"
-                                                   "Watch tag: {{watch_tag}}\n"
-                                                   "Preview: {{preview_url}}\n"
-                                                   "Diff URL: {{diff_url}}\n"
-                                                   "Snapshot: {{current_snapshot}}\n"
-                                                   "Diff: {{diff}}\n"
-                                                   "Diff Full: {{diff_full}}\n"
-                                                   "Diff as Patch: {{diff_patch}}\n"
-                                                   ":-)",
-                              "notification_format": "Text"}
+    watch_data = {"notification_urls": notification_url,
+                  "notification_title": "New ChangeDetection.io Notification - {{watch_url}}",
+                  "notification_body": "BASE URL: {{base_url}}\n"
+                                       "Watch URL: {{watch_url}}\n"
+                                       "Watch UUID: {{watch_uuid}}\n"
+                                       "Watch title: {{watch_title}}\n"
+                                       "Watch tag: {{watch_tag}}\n"
+                                       "Preview: {{preview_url}}\n"
+                                       "Diff URL: {{diff_url}}\n"
+                                       "Snapshot: {{current_snapshot}}\n"
+                                       "Diff: {{diff}}\n"
+                                       "Diff Full: {{diff_full}}\n"
+                                       "Diff as Patch: {{diff_patch}}\n"
+                                       ":-)",
+                  "notification_format": "Text",
+                  "fetch_backend": "html_requests",
+                  "filter_failure_notification_send": 'y',
+                  "headers": "",
+                  "tags": "my tag",
+                  "title": "my title 123",
+                  "time_between_check-hours": 5,  # So that the queue runner doesnt also put it in
+                  "url": test_url,
+                  }
 
-    notification_form_data.update({
-        "url": test_url,
-        "tags": "my tag",
-        "title": "my title 123",
-        "headers": "",
-        "filter_failure_notification_send": 'y',
-        "include_filters": content_filter,
-        "fetch_backend": "html_requests"})
-
-    # A POST here will also reset the filter failure counter (filter_failure_notification_threshold_attempts)
     res = client.post(
-        url_for("edit_page", uuid="first"),
-        data=notification_form_data,
+        url_for("edit_page", uuid=uuid),
+        data=watch_data,
         follow_redirects=True
     )
-
     assert b"Updated watch." in res.data
     wait_for_all_checks(client)
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid]['consecutive_filter_failures'] == 0, "No filter = No filter failure"
 
-    # Now the notification should not exist, because we didnt reach the threshold
+    # Now add a filter, because recheck hours == 5, ONLY pressing of the [edit] or [recheck all] should trigger
+    watch_data['include_filters'] = content_filter
+    res = client.post(
+        url_for("edit_page", uuid=uuid),
+        data=watch_data,
+        follow_redirects=True
+    )
+    assert b"Updated watch." in res.data
+
+    # It should have checked once so far and given this error (because we hit SAVE)
+
+    wait_for_all_checks(client)
     assert not os.path.isfile("test-datastore/notification.txt")
 
+    # Hitting [save] would have triggered a recheck, and we have a filter, so this would be ONE failure
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid]['consecutive_filter_failures'] == 1, "Should have been checked once"
+
     # recheck it up to just before the threshold, including the fact that in the previous POST it would have rechecked (and incremented)
-    for i in range(0, App._FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT-2):
+    # Add 4 more checks
+    checked = 0
+    ATTEMPT_THRESHOLD_SETTING = live_server.app.config['DATASTORE'].data['settings']['application'].get('filter_failure_notification_threshold_attempts', 0)
+    for i in range(0, ATTEMPT_THRESHOLD_SETTING - 2):
+        checked += 1
         client.get(url_for("form_watch_checknow"), follow_redirects=True)
         wait_for_all_checks(client)
-        time.sleep(2) # delay for apprise to fire
-        assert not os.path.isfile("test-datastore/notification.txt"), f"test-datastore/notification.txt should not exist - Attempt {i} when threshold is {App._FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT}"
-
-    # We should see something in the frontend
-    res = client.get(url_for("index"))
-    assert b'Warning, no filters were found' in res.data
+        res = client.get(url_for("index"))
+        assert b'Warning, no filters were found' in res.data
+        assert not os.path.isfile("test-datastore/notification.txt")
+        time.sleep(1)
+        
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid]['consecutive_filter_failures'] == 5
 
+    time.sleep(2)
     # One more check should trigger the _FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT threshold
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
-    time.sleep(2)  # delay for apprise to fire
+    wait_for_notification_endpoint_output()
+
     # Now it should exist and contain our "filter not found" alert
     assert os.path.isfile("test-datastore/notification.txt")
-
     with open("test-datastore/notification.txt", 'r') as f:
         notification = f.read()
 
@@ -123,10 +140,11 @@ def run_filter_test(client, live_server, content_filter):
     set_response_with_filter()
 
     # Try several times, it should NOT have 'filter not found'
-    for i in range(0, App._FILTER_FAILURE_THRESHOLD_ATTEMPTS_DEFAULT):
+    for i in range(0, ATTEMPT_THRESHOLD_SETTING + 2):
         client.get(url_for("form_watch_checknow"), follow_redirects=True)
         wait_for_all_checks(client)
 
+    wait_for_notification_endpoint_output()
     # It should have sent a notification, but..
     assert os.path.isfile("test-datastore/notification.txt")
     # but it should not contain the info about a failed filter (because there was none in this case)
@@ -135,9 +153,6 @@ def run_filter_test(client, live_server, content_filter):
     assert not 'CSS/xPath filter was not present in the page' in notification
 
     # Re #1247 - All tokens got replaced correctly in the notification
-    res = client.get(url_for("index"))
-    uuid = extract_UUID_from_client(client)
-    # UUID is correct, but notification contains tag uuid as UUIID wtf
     assert uuid in notification
 
     # cleanup for the next
@@ -152,9 +167,11 @@ def test_setup(live_server):
     live_server_setup(live_server)
 
 def test_check_include_filters_failure_notification(client, live_server, measure_memory_usage):
+#    live_server_setup(live_server)
     run_filter_test(client, live_server,'#nope-doesnt-exist')
 
 def test_check_xpath_filter_failure_notification(client, live_server, measure_memory_usage):
+#    live_server_setup(live_server)
     run_filter_test(client, live_server, '//*[@id="nope-doesnt-exist"]')
 
 # Test that notification is never sent

changedetectionio/tests/test_nonrenderable_pages.py

@@ -2,6 +2,8 @@
 
 from flask import url_for
 from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks
+import time
+
 
 def set_nonrenderable_response():
     test_return_data = """<html>
@@ -11,17 +13,16 @@ def set_nonrenderable_response():
      </body>
      </html>
     """
-
     with open("test-datastore/endpoint-content.txt", "w") as f:
         f.write(test_return_data)
+    time.sleep(1)
 
     return None
 
 def set_zero_byte_response():
-
     with open("test-datastore/endpoint-content.txt", "w") as f:
         f.write("")
-
+    time.sleep(1)
     return None
 
 def test_check_basic_change_detection_functionality(client, live_server, measure_memory_usage):

changedetectionio/tests/test_restock_itemprop.py

@@ -3,7 +3,7 @@ import os
 import time
 
 from flask import url_for
-from .util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
+from .util import live_server_setup, wait_for_all_checks, extract_UUID_from_client, wait_for_notification_endpoint_output
 from ..notification import default_notification_format
 
 instock_props = [
@@ -146,14 +146,13 @@ def _run_test_minmax_limit(client, extra_watch_edit_form):
         data={"url": test_url, "tags": 'restock tests', 'processor': 'restock_diff'},
         follow_redirects=True
     )
-
-    # A change in price, should trigger a change by default
     wait_for_all_checks(client)
 
     data = {
         "tags": "",
         "url": test_url,
         "headers": "",
+        "time_between_check-hours": 5,
         'fetch_backend': "html_requests"
     }
     data.update(extra_watch_edit_form)
@@ -178,11 +177,9 @@ def _run_test_minmax_limit(client, extra_watch_edit_form):
     assert b'1,000.45' or b'1000.45' in res.data #depending on locale
     assert b'unviewed' not in res.data
 
-
     # price changed to something LESS than min (900), SHOULD be a change
     set_original_response(props_markup=instock_props[0], price='890.45')
-    # let previous runs wait
-    time.sleep(1)
+
     res = client.get(url_for("form_watch_checknow"), follow_redirects=True)
     assert b'1 watches queued for rechecking.' in res.data
     wait_for_all_checks(client)
@@ -197,7 +194,8 @@ def _run_test_minmax_limit(client, extra_watch_edit_form):
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
     res = client.get(url_for("index"))
-    assert b'1,890.45' or b'1890.45' in res.data
+    # Depending on the LOCALE it may be either of these (generally for US/default/etc)
+    assert b'1,890.45' in res.data or b'1890.45' in res.data
     assert b'unviewed' in res.data
 
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
@@ -362,7 +360,7 @@ def test_change_with_notification_values(client, live_server):
     set_original_response(props_markup=instock_props[0], price='1950.45')
     client.get(url_for("form_watch_checknow"))
     wait_for_all_checks(client)
-    time.sleep(3)
+    wait_for_notification_endpoint_output()
     assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
     with open("test-datastore/notification.txt", 'r') as f:
         notification = f.read()

changedetectionio/tests/util.py

@@ -76,6 +76,18 @@ def set_more_modified_response():
     return None
 
 
+def wait_for_notification_endpoint_output():
+    '''Apprise can take a few seconds to fire'''
+    #@todo - could check the apprise object directly instead of looking for this file
+    from os.path import isfile
+    for i in range(1, 20):
+        time.sleep(1)
+        if isfile("test-datastore/notification.txt"):
+            return True
+
+    return False
+
+
 # kinda funky, but works for now
 def extract_api_key_from_UI(client):
     import re

changedetectionio/update_worker.py

@@ -189,7 +189,9 @@ class update_worker(threading.Thread):
                 'screenshot': None
             })
             self.notification_q.put(n_object)
-            logger.error(f"Sent filter not found notification for {watch_uuid}")
+            logger.debug(f"Sent filter not found notification for {watch_uuid}")
+        else:
+            logger.debug(f"NOT sending filter not found notification for {watch_uuid} - no notification URLs")
 
     def send_step_failure_notification(self, watch_uuid, step_n):
         watch = self.datastore.data['watching'].get(watch_uuid, False)
@@ -276,7 +278,7 @@ class update_worker(threading.Thread):
 
                         update_handler.call_browser()
 
-                        changed_detected, update_obj, contents = update_handler.run_changedetection(
+                        changed_detected, update_obj, contents, content_after_filters = update_handler.run_changedetection(
                             watch=watch,
                             skip_when_checksum_same=skip_when_same_checksum,
                         )
@@ -364,18 +366,22 @@ class update_worker(threading.Thread):
 
                         # Only when enabled, send the notification
                         if watch.get('filter_failure_notification_send', False):
-                            c = watch.get('consecutive_filter_failures', 5)
+                            c = watch.get('consecutive_filter_failures', 0)
                             c += 1
                             # Send notification if we reached the threshold?
-                            threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts',
-                                                                                           0)
-                            logger.warning(f"Filter for {uuid} not found, consecutive_filter_failures: {c}")
-                            if threshold > 0 and c >= threshold:
+                            threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts', 0)
+                            logger.debug(f"Filter for {uuid} not found, consecutive_filter_failures: {c} of threshold {threshold}")
+                            if c >= threshold:
                                 if not watch.get('notification_muted'):
+                                    logger.debug(f"Sending filter failed notification for {uuid}")
                                     self.send_filter_failure_notification(uuid)
                                 c = 0
+                                logger.debug(f"Reset filter failure count back to zero")
 
                             self.datastore.update_watch(uuid=uuid, update_obj={'consecutive_filter_failures': c})
+                        else:
+                            logger.trace(f"{uuid} - filter_failure_notification_send not enabled, skipping")
+
 
                         process_changedetection_results = False
 
@@ -422,7 +428,7 @@ class update_worker(threading.Thread):
                                                     )
 
                         if watch.get('filter_failure_notification_send', False):
-                            c = watch.get('consecutive_filter_failures', 5)
+                            c = watch.get('consecutive_filter_failures', 0)
                             c += 1
                             # Send notification if we reached the threshold?
                             threshold = self.datastore.data['settings']['application'].get('filter_failure_notification_threshold_attempts',