Updating pyppeteer-ng (mainly newer pillow release)

.github/workflows/test-only.yml

@@ -72,7 +72,11 @@ jobs:
         run: |
           # Playwright via Sockpuppetbrowser fetch
           # tests/visualselector/test_fetch_data.py will do browser steps  
-          docker run --rm -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/fetchers/test_content.py && pytest tests/test_errorhandling.py && pytest tests/visualselector/test_fetch_data.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_content.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/test_errorhandling.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/visualselector/test_fetch_data.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_custom_js_before_content.py'
+
 
       - name: Playwright and SocketPuppetBrowser - Headers and requests
         run: |       
@@ -87,8 +91,11 @@ jobs:
 # STRAIGHT TO CDP
       - name: Pyppeteer and SocketPuppetBrowser - Specific tests in built container
         run: |
-          # Playwright via Sockpuppetbrowser fetch
-          docker run --rm -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" -e "FAST_PUPPETEER_CHROME_FETCHER=True" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/fetchers/test_content.py && pytest tests/test_errorhandling.py && pytest tests/visualselector/test_fetch_data.py'
+          # Playwright via Sockpuppetbrowser fetch 
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_content.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/test_errorhandling.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/visualselector/test_fetch_data.py'
+          docker run --rm -e "FLASK_SERVER_NAME=cdio" -e "FAST_PUPPETEER_CHROME_FETCHER=True" -e "PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000" --network changedet-network --hostname=cdio test-changedetectionio  bash -c 'cd changedetectionio;pytest --live-server-host=0.0.0.0 --live-server-port=5004 tests/fetchers/test_custom_js_before_content.py'
 
       - name: Pyppeteer and SocketPuppetBrowser - Headers and requests checks
         run: |       

MANIFEST.in

@@ -1,8 +1,8 @@
 recursive-include changedetectionio/api *
 recursive-include changedetectionio/blueprint *
+recursive-include changedetectionio/content_fetchers *
 recursive-include changedetectionio/model *
 recursive-include changedetectionio/processors *
-recursive-include changedetectionio/res *
 recursive-include changedetectionio/static *
 recursive-include changedetectionio/templates *
 recursive-include changedetectionio/tests *

README.md

@@ -91,6 +91,14 @@ We [recommend and use Bright Data](https://brightdata.grsm.io/n0r16zf7eivq) glob
 
 Please :star: star :star: this project and help it grow! https://github.com/dgtlmoon/changedetection.io/
 
+### We have a Chrome extension!
+
+Easily add the current web page to your changedetection.io tool, simply install the extension and click "Sync" to connect it to your existing changedetection.io install.
+
+[<img src="./docs/chrome-extension-screenshot.png" style="max-width:80%;" alt="Chrome Extension to easily add the current web-page to detect a change."  title="Chrome Extension to easily add the current web-page to detect a change."  />](https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop)
+
+[Goto the Chrome Webstore to download the extension.](https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop)
+
 ## Installation
 
 ### Docker

changedetectionio/__init__.py

@@ -2,7 +2,7 @@
 
 # Read more https://github.com/dgtlmoon/changedetection.io/wiki
 
-__version__ = '0.45.14'
+__version__ = '0.45.16'
 
 from distutils.util import strtobool
 from json.decoder import JSONDecodeError

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -6,6 +6,8 @@ import re
 from random import randint
 from loguru import logger
 
+from changedetectionio.content_fetchers.base import manage_user_agent
+
 # Two flags, tell the JS which of the "Selector" or "Value" field should be enabled in the front end
 # 0- off, 1- on
 browser_step_ui_config = {'Choose one': '0 0',
@@ -178,6 +180,7 @@ class browsersteps_live_ui(steppable_browser_interface):
     stale = False
     # bump and kill this if idle after X sec
     age_start = 0
+    headers = {}
 
     # use a special driver, maybe locally etc
     command_executor = os.getenv(
@@ -192,7 +195,8 @@ class browsersteps_live_ui(steppable_browser_interface):
 
     browser_type = os.getenv("PLAYWRIGHT_BROWSER_TYPE", 'chromium').strip('"')
 
-    def __init__(self, playwright_browser, proxy=None):
+    def __init__(self, playwright_browser, proxy=None, headers=None):
+        self.headers = headers or {}
         self.age_start = time.time()
         self.playwright_browser = playwright_browser
         if self.context is None:
@@ -206,16 +210,17 @@ class browsersteps_live_ui(steppable_browser_interface):
 
         # @todo handle multiple contexts, bind a unique id from the browser on each req?
         self.context = self.playwright_browser.new_context(
-            # @todo
-            #                user_agent=request_headers['User-Agent'] if request_headers.get('User-Agent') else 'Mozilla/5.0',
-            #               proxy=self.proxy,
-            # This is needed to enable JavaScript execution on GitHub and others
-            bypass_csp=True,
-            # Should never be needed
-            accept_downloads=False,
-            proxy=proxy
+            accept_downloads=False,  # Should never be needed
+            bypass_csp=True,  # This is needed to enable JavaScript execution on GitHub and others
+            extra_http_headers=self.headers,
+            ignore_https_errors=True,
+            proxy=proxy,
+            service_workers=os.getenv('PLAYWRIGHT_SERVICE_WORKERS', 'allow'),
+            # Should be `allow` or `block` - sites like YouTube can transmit large amounts of data via Service Workers
+            user_agent=manage_user_agent(headers=self.headers),
         )
 
+
         self.page = self.context.new_page()
 
         # self.page.set_default_navigation_timeout(keep_open)

changedetectionio/blueprint/browser_steps/nonContext.py

@@ -1,5 +1,4 @@
 from playwright.sync_api import PlaywrightContextManager
-import asyncio
 
 # So playwright wants to run as a context manager, but we do something horrible and hacky
 # we are holding the session open for as long as possible, then shutting it down, and opening a new one

changedetectionio/blueprint/tags/__init__.py

@@ -11,9 +11,10 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     def tags_overview_page():
         from .form import SingleTag
         add_form = SingleTag(request.form)
+        sorted_tags = sorted(datastore.data['settings']['application'].get('tags').items(), key=lambda x: x[1]['title'])
         output = render_template("groups-overview.html",
                                  form=add_form,
-                                 available_tags=datastore.data['settings']['application'].get('tags', {}),
+                                 available_tags=sorted_tags,
                                  )
 
         return output

changedetectionio/blueprint/tags/templates/groups-overview.html

@@ -40,7 +40,7 @@
                 <td colspan="3">No website organisational tags/groups configured</td>
             </tr>
             {% endif %}
-            {% for uuid, tag in available_tags.items()  %}
+            {% for uuid, tag in available_tags  %}
             <tr id="{{ uuid }}" class="{{ loop.cycle('pure-table-odd', 'pure-table-even') }}">
                 <td class="watch-controls">
                     <a class="link-mute state-{{'on' if tag.notification_muted else 'off'}}" href="{{url_for('tags.mute', uuid=tag.uuid)}}"><img src="{{url_for('static_content', group='images', filename='bell-off.svg')}}" alt="Mute notifications" title="Mute notifications" class="icon icon-mute" ></a>

changedetectionio/content_fetchers/__init__.py

@@ -1,10 +1,10 @@
 import sys
 from distutils.util import strtobool
-
+from loguru import logger
 from changedetectionio.content_fetchers.exceptions import BrowserStepsStepException
 import os
 
-visualselector_xpath_selectors = 'div,span,form,table,tbody,tr,td,a,p,ul,li,h1,h2,h3,h4, header, footer, section, article, aside, details, main, nav, section, summary'
+visualselector_xpath_selectors = 'div,span,form,table,tbody,tr,td,a,p,ul,li,h1,h2,h3,h4,header,footer,section,article,aside,details,main,nav,section,summary'
 
 # available_fetchers() will scan this implementation looking for anything starting with html_
 # this information is used in the form selections
@@ -29,10 +29,15 @@ def available_fetchers():
 # rather than site-specific.
 use_playwright_as_chrome_fetcher = os.getenv('PLAYWRIGHT_DRIVER_URL', False)
 if use_playwright_as_chrome_fetcher:
+    # @note - For now, browser steps always uses playwright
     if not strtobool(os.getenv('FAST_PUPPETEER_CHROME_FETCHER', 'False')):
+        logger.debug('Using Playwright library as fetcher')
         from .playwright import fetcher as html_webdriver
     else:
+        logger.debug('Using direct Python Puppeteer library as fetcher')
         from .puppeteer import fetcher as html_webdriver
 
 else:
+    logger.debug("Falling back to selenium as fetcher")
     from .webdriver_selenium import fetcher as html_webdriver
+

changedetectionio/content_fetchers/base.py

@@ -5,6 +5,40 @@ from loguru import logger
 from changedetectionio.content_fetchers import BrowserStepsStepException
 
 
+def manage_user_agent(headers, current_ua=''):
+    """
+    Basic setting of user-agent
+
+    NOTE!!!!!! The service that does the actual Chrome fetching should handle any anti-robot techniques
+    THERE ARE MANY WAYS THAT IT CAN BE DETECTED AS A ROBOT!!
+    This does not take care of
+    - Scraping of 'navigator' (platform, productSub, vendor, oscpu etc etc) browser object (navigator.appVersion) etc
+    - TCP/IP fingerprint JA3 etc
+    - Graphic rendering fingerprinting
+    - Your IP being obviously in a pool of bad actors
+    - Too many requests
+    - Scraping of SCH-UA browser replies (thanks google!!)
+    - Scraping of ServiceWorker, new window calls etc
+
+    See https://filipvitas.medium.com/how-to-set-user-agent-header-with-puppeteer-js-and-not-fail-28c7a02165da
+    Puppeteer requests https://github.com/dgtlmoon/pyppeteerstealth
+
+    :param page:
+    :param headers:
+    :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)
+    if ua_in_custom_headers:
+        return ua_in_custom_headers
+
+    if not ua_in_custom_headers and current_ua:
+        current_ua = current_ua.replace('HeadlessChrome', 'Chrome')
+        return current_ua
+
+    return None
+
+
 class Fetcher():
     browser_connection_is_custom = None
     browser_connection_url = None

changedetectionio/content_fetchers/playwright.py

@@ -3,7 +3,8 @@ import os
 from urllib.parse import urlparse
 
 from loguru import logger
-from changedetectionio.content_fetchers.base import Fetcher
+
+from changedetectionio.content_fetchers.base import Fetcher, manage_user_agent
 from changedetectionio.content_fetchers.exceptions import PageUnloadable, Non200ErrorCodeReceived, EmptyReply, ScreenshotUnavailable
 
 class fetcher(Fetcher):
@@ -102,19 +103,16 @@ class fetcher(Fetcher):
             # Set user agent to prevent Cloudflare from blocking the browser
             # Use the default one configured in the App.py model that's passed from fetch_site_status.py
             context = browser.new_context(
-                user_agent={k.lower(): v for k, v in request_headers.items()}.get('user-agent', None),
+                accept_downloads=False,  # Should never be needed
+                bypass_csp=True,  # This is needed to enable JavaScript execution on GitHub and others
+                extra_http_headers=request_headers,
+                ignore_https_errors=True,
                 proxy=self.proxy,
-                # This is needed to enable JavaScript execution on GitHub and others
-                bypass_csp=True,
-                # Should be `allow` or `block` - sites like YouTube can transmit large amounts of data via Service Workers
-                service_workers=os.getenv('PLAYWRIGHT_SERVICE_WORKERS', 'allow'),
-                # Should never be needed
-                accept_downloads=False
+                service_workers=os.getenv('PLAYWRIGHT_SERVICE_WORKERS', 'allow'), # Should be `allow` or `block` - sites like YouTube can transmit large amounts of data via Service Workers
+                user_agent=manage_user_agent(headers=request_headers),
             )
 
             self.page = context.new_page()
-            if len(request_headers):
-                context.set_extra_http_headers(request_headers)
 
             # Listen for all console events and handle errors
             self.page.on("console", lambda msg: print(f"Playwright console: Watch URL: {url} {msg.type}: {msg.text} {msg.args}"))

changedetectionio/content_fetchers/puppeteer.py

@@ -5,7 +5,8 @@ import websockets.exceptions
 from urllib.parse import urlparse
 
 from loguru import logger
-from changedetectionio.content_fetchers.base import Fetcher
+
+from changedetectionio.content_fetchers.base import Fetcher, manage_user_agent
 from changedetectionio.content_fetchers.exceptions import PageUnloadable, Non200ErrorCodeReceived, EmptyReply, BrowserFetchTimedOut, BrowserConnectError
 
 
@@ -100,10 +101,11 @@ class fetcher(Fetcher):
         else:
             self.page = await browser.newPage()
 
+        await self.page.setUserAgent(manage_user_agent(headers=request_headers, current_ua=await self.page.evaluate('navigator.userAgent')))
+
         await self.page.setBypassCSP(True)
         if request_headers:
             await self.page.setExtraHTTPHeaders(request_headers)
-            # @todo check user-agent worked
 
         # SOCKS5 with authentication is not supported (yet)
         # https://github.com/microsoft/playwright/issues/10567
@@ -212,8 +214,12 @@ class fetcher(Fetcher):
                 logger.error('ERROR: Failed to get viewport-only reduced screenshot :(')
                 pass
         finally:
+            # It's good to log here in the case that the browser crashes on shutting down but we still get the data we need
+            logger.success(f"Fetching '{url}' complete, closing page")
             await self.page.close()
+            logger.success(f"Fetching '{url}' complete, closing browser")
             await browser.close()
+        logger.success(f"Fetching '{url}' complete, exiting puppeteer fetch.")
 
     async def main(self, **kwargs):
         await self.fetch_page(**kwargs)

changedetectionio/content_fetchers/res/stock-not-in-stock.js

@@ -10,15 +10,16 @@ function isItemInStock() {
     const outOfStockTexts = [
         ' أخبرني عندما يتوفر',
         '0 in stock',
-        'actuellement indisponible',        
+        'actuellement indisponible',
         'agotado',
         'article épuisé',
         'artikel zurzeit vergriffen',
         'as soon as stock is available',
         'ausverkauft', // sold out
         'available for back order',
-        'back-order or out of stock',
+        'awaiting stock',
         'back in stock soon',
+        'back-order or out of stock',
         'backordered',
         'benachrichtigt mich', // notify me
         'brak na stanie',
@@ -45,9 +46,9 @@ function isItemInStock() {
         'no tickets available',
         'not available',
         'not currently available',
-        'not in stock',        
+        'not in stock',
         'notify me when available',
-        'notify when available',            
+        'notify when available',
         'não estamos a aceitar encomendas',
         'out of stock',
         'out-of-stock',
@@ -57,18 +58,26 @@ function isItemInStock() {
         'sold-out',
         'temporarily out of stock',
         'temporarily unavailable',
+        'there were no search results for',
+        'this item is currently unavailable',
         'tickets unavailable',
         'tijdelijk uitverkocht',
         'unavailable tickets',
+        'vorbestellung ist bald möglich',
+        'we couldn\'t find any products that match',
         'we do not currently have an estimate of when this product will be back in stock.',
         'we don\'t know when or if this item will be back in stock.',
+        'we were not able to find a match',
         'zur zeit nicht an lager',
         '品切れ',
+        '已售',
         '已售完',
         '품절'
     ];
 
+
     const vh = Math.max(document.documentElement.clientHeight || 0, window.innerHeight || 0);
+
     function getElementBaseText(element) {
         // .textContent can include text from children which may give the wrong results
         // scan only immediate TEXT_NODEs, which will be a child of the element
@@ -79,29 +88,69 @@ function isItemInStock() {
         return text.toLowerCase().trim();
     }
 
-    const negateOutOfStockRegex = new RegExp('([0-9] in stock|add to cart)', 'ig');
+    const negateOutOfStockRegex = new RegExp('^([0-9] in stock|add to cart|in stock)', 'ig');
 
     // The out-of-stock or in-stock-text is generally always above-the-fold
     // and often below-the-fold is a list of related products that may or may not contain trigger text
     // so it's good to filter to just the 'above the fold' elements
     // and it should be atleast 100px from the top to ignore items in the toolbar, sometimes menu items like "Coming soon" exist
-    const elementsToScan = Array.from(document.getElementsByTagName('*')).filter(element => element.getBoundingClientRect().top + window.scrollY <= vh && element.getBoundingClientRect().top + window.scrollY >= 100);
+
+
+// @todo - if it's SVG or IMG, go into image diff mode
+// %ELEMENTS% replaced at injection time because different interfaces use it with different settings
+
+    console.log("Scanning %ELEMENTS%");
+
+    function collectVisibleElements(parent, visibleElements) {
+        if (!parent) return; // Base case: if parent is null or undefined, return
+
+        // Add the parent itself to the visible elements array if it's of the specified types
+        visibleElements.push(parent);
+
+        // Iterate over the parent's children
+        const children = parent.children;
+        for (let i = 0; i < children.length; i++) {
+            const child = children[i];
+            if (
+                child.nodeType === Node.ELEMENT_NODE &&
+                window.getComputedStyle(child).display !== 'none' &&
+                window.getComputedStyle(child).visibility !== 'hidden' &&
+                child.offsetWidth >= 0 &&
+                child.offsetHeight >= 0 &&
+                window.getComputedStyle(child).contentVisibility !== 'hidden'
+            ) {
+                // If the child is an element and is visible, recursively collect visible elements
+                collectVisibleElements(child, visibleElements);
+            }
+        }
+    }
+
+    const elementsToScan = [];
+    collectVisibleElements(document.body, elementsToScan);
 
     var elementText = "";
 
     // REGEXS THAT REALLY MEAN IT'S IN STOCK
     for (let i = elementsToScan.length - 1; i >= 0; i--) {
         const element = elementsToScan[i];
+
+        // outside the 'fold' or some weird text in the heading area
+        // .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
+        if (element.getBoundingClientRect().top + window.scrollY >= vh || element.getBoundingClientRect().top + window.scrollY <= 100) {
+            continue
+        }
+
         elementText = "";
         if (element.tagName.toLowerCase() === "input") {
-            elementText = element.value.toLowerCase();
+            elementText = element.value.toLowerCase().trim();
         } else {
             elementText = getElementBaseText(element);
         }
 
         if (elementText.length) {
             // try which ones could mean its in stock
-            if (negateOutOfStockRegex.test(elementText)) {
+            if (negateOutOfStockRegex.test(elementText) && !elementText.includes('(0 products)')) {
+                console.log(`Negating/overriding 'Out of Stock' back to "Possibly in stock" found "${elementText}"`)
                 return 'Possibly in stock';
             }
         }
@@ -110,28 +159,34 @@ function isItemInStock() {
     // OTHER STUFF THAT COULD BE THAT IT'S OUT OF STOCK
     for (let i = elementsToScan.length - 1; i >= 0; i--) {
         const element = elementsToScan[i];
-        if (element.offsetWidth > 0 || element.offsetHeight > 0 || element.getClientRects().length > 0) {
-            elementText = "";
-            if (element.tagName.toLowerCase() === "input") {
-                elementText = element.value.toLowerCase();
-            } else {
-                elementText = getElementBaseText(element);
-            }
+        // outside the 'fold' or some weird text in the heading area
+        // .getBoundingClientRect() was causing a crash in chrome 119, can only be run on contentVisibility != hidden
+        if (element.getBoundingClientRect().top + window.scrollY >= vh + 150 || element.getBoundingClientRect().top + window.scrollY <= 100) {
+            continue
+        }
+        elementText = "";
+        if (element.tagName.toLowerCase() === "input") {
+            elementText = element.value.toLowerCase().trim();
+        } else {
+            elementText = getElementBaseText(element);
+        }
 
-            if (elementText.length) {
-                // and these mean its out of stock
-                for (const outOfStockText of outOfStockTexts) {
-                    if (elementText.includes(outOfStockText)) {
-                        return outOfStockText; // item is out of stock
-                    }
+        if (elementText.length) {
+            // and these mean its out of stock
+            for (const outOfStockText of outOfStockTexts) {
+                if (elementText.includes(outOfStockText)) {
+                    console.log(`Selected 'Out of Stock' - found text "${outOfStockText}" - "${elementText}"`)
+                    return outOfStockText; // item is out of stock
                 }
             }
         }
     }
 
+    console.log(`Returning 'Possibly in stock' - cant' find any useful matching text`)
     return 'Possibly in stock'; // possibly in stock, cant decide otherwise.
 }
 
 // returns the element text that makes it think it's out of stock
 return isItemInStock().trim()
 
+

changedetectionio/content_fetchers/res/xpath_element_scraper.js

@@ -16,24 +16,23 @@ try {
 }
 
 
-
 // Include the getXpath script directly, easier than fetching
 function getxpath(e) {
-        var n = e;
-        if (n && n.id) return '//*[@id="' + n.id + '"]';
-        for (var o = []; n && Node.ELEMENT_NODE === n.nodeType;) {
-            for (var i = 0, r = !1, d = n.previousSibling; d;) d.nodeType !== Node.DOCUMENT_TYPE_NODE && d.nodeName === n.nodeName && i++, d = d.previousSibling;
-            for (d = n.nextSibling; d;) {
-                if (d.nodeName === n.nodeName) {
-                    r = !0;
-                    break
-                }
-                d = d.nextSibling
+    var n = e;
+    if (n && n.id) return '//*[@id="' + n.id + '"]';
+    for (var o = []; n && Node.ELEMENT_NODE === n.nodeType;) {
+        for (var i = 0, r = !1, d = n.previousSibling; d;) d.nodeType !== Node.DOCUMENT_TYPE_NODE && d.nodeName === n.nodeName && i++, d = d.previousSibling;
+        for (d = n.nextSibling; d;) {
+            if (d.nodeName === n.nodeName) {
+                r = !0;
+                break
             }
-            o.push((n.prefix ? n.prefix + ":" : "") + n.localName + (i || r ? "[" + (i + 1) + "]" : "")), n = n.parentNode
+            d = d.nextSibling
         }
-        return o.length ? "/" + o.reverse().join("/") : ""
+        o.push((n.prefix ? n.prefix + ":" : "") + n.localName + (i || r ? "[" + (i + 1) + "]" : "")), n = n.parentNode
     }
+    return o.length ? "/" + o.reverse().join("/") : ""
+}
 
 const findUpTag = (el) => {
     let r = el
@@ -59,14 +58,14 @@ const findUpTag = (el) => {
 
     // Strategy 2: Keep going up until we hit an ID tag, imagine it's like  #list-widget div h4
     while (r.parentNode) {
-        if (depth == 5) {
+        if (depth === 5) {
             break;
         }
         if ('' !== r.id) {
             chained_css.unshift("#" + CSS.escape(r.id));
             final_selector = chained_css.join(' > ');
             // Be sure theres only one, some sites have multiples of the same ID tag :-(
-            if (window.document.querySelectorAll(final_selector).length == 1) {
+            if (window.document.querySelectorAll(final_selector).length === 1) {
                 return final_selector;
             }
             return null;
@@ -82,30 +81,60 @@ const findUpTag = (el) => {
 
 // @todo - if it's SVG or IMG, go into image diff mode
 // %ELEMENTS% replaced at injection time because different interfaces use it with different settings
-var elements = window.document.querySelectorAll("%ELEMENTS%");
+
 var size_pos = [];
 // after page fetch, inject this JS
 // build a map of all elements and their positions (maybe that only include text?)
 var bbox;
-for (var i = 0; i < elements.length; i++) {
-    bbox = elements[i].getBoundingClientRect();
+console.log("Scanning %ELEMENTS%");
 
-    // Exclude items that are not interactable or visible
-    if(elements[i].style.opacity === "0") {
-        continue
+function collectVisibleElements(parent, visibleElements) {
+    if (!parent) return; // Base case: if parent is null or undefined, return
+
+
+    // Add the parent itself to the visible elements array if it's of the specified types
+    const tagName = parent.tagName.toLowerCase();
+    if ("%ELEMENTS%".split(',').includes(tagName)) {
+        visibleElements.push(parent);
     }
-    if(elements[i].style.display === "none" || elements[i].style.pointerEvents === "none" ) {
-        continue
+
+    // Iterate over the parent's children
+    const children = parent.children;
+    for (let i = 0; i < children.length; i++) {
+        const child = children[i];
+        if (
+            child.nodeType === Node.ELEMENT_NODE &&
+            window.getComputedStyle(child).display !== 'none' &&
+            window.getComputedStyle(child).visibility !== 'hidden' &&
+            child.offsetWidth >= 0 &&
+            child.offsetHeight >= 0 &&
+            window.getComputedStyle(child).contentVisibility !== 'hidden'
+        ) {
+            // If the child is an element and is visible, recursively collect visible elements
+            collectVisibleElements(child, visibleElements);
+        }
     }
+}
+
+// Create an array to hold the visible elements
+const visibleElementsArray = [];
+
+// Call collectVisibleElements with the starting parent element
+collectVisibleElements(document.body, visibleElementsArray);
+
+
+visibleElementsArray.forEach(function (element) {
+
+    bbox = element.getBoundingClientRect();
 
     // Skip really small ones, and where width or height ==0
-    if (bbox['width'] * bbox['height'] < 100) {
-        continue;
+    if (bbox['width'] * bbox['height'] < 10) {
+        return
     }
 
     // Don't include elements that are offset from canvas
-    if (bbox['top']+scroll_y < 0 || bbox['left'] < 0) {
-        continue;
+    if (bbox['top'] + scroll_y < 0 || bbox['left'] < 0) {
+        return
     }
 
     // @todo the getXpath kind of sucks, it doesnt know when there is for example just one ID sometimes
@@ -114,46 +143,41 @@ for (var i = 0; i < elements.length; i++) {
 
     // 1st primitive - if it has class, try joining it all and select, if theres only one.. well thats us.
     xpath_result = false;
-
     try {
-        var d = findUpTag(elements[i]);
+        var d = findUpTag(element);
         if (d) {
             xpath_result = d;
         }
     } catch (e) {
         console.log(e);
     }
-
     // You could swap it and default to getXpath and then try the smarter one
     // default back to the less intelligent one
     if (!xpath_result) {
         try {
             // I've seen on FB and eBay that this doesnt work
             // ReferenceError: getXPath is not defined at eval (eval at evaluate (:152:29), <anonymous>:67:20) at UtilityScript.evaluate (<anonymous>:159:18) at UtilityScript.<anonymous> (<anonymous>:1:44)
-            xpath_result = getxpath(elements[i]);
+            xpath_result = getxpath(element);
         } catch (e) {
             console.log(e);
-            continue;
+            return
         }
     }
 
-    if (window.getComputedStyle(elements[i]).visibility === "hidden") {
-        continue;
-    }
 
-    // @todo Possible to ONLY list where it's clickable to save JSON xfer size
     size_pos.push({
         xpath: xpath_result,
         width: Math.round(bbox['width']),
         height: Math.round(bbox['height']),
         left: Math.floor(bbox['left']),
-        top: Math.floor(bbox['top'])+scroll_y,
-        tagName: (elements[i].tagName) ? elements[i].tagName.toLowerCase() : '',
-        tagtype: (elements[i].tagName == 'INPUT' && elements[i].type) ? elements[i].type.toLowerCase() : '',
-        isClickable: (elements[i].onclick) || window.getComputedStyle(elements[i]).cursor == "pointer"
+        top: Math.floor(bbox['top']) + scroll_y,
+        tagName: (element.tagName) ? element.tagName.toLowerCase() : '',
+        tagtype: (element.tagName.toLowerCase() === 'input' && element.type) ? element.type.toLowerCase() : '',
+        isClickable: window.getComputedStyle(element).cursor == "pointer"
     });
 
-}
+});
+
 
 // Inject the current one set in the include_filters, which may be a CSS rule
 // used for displaying the current one in VisualSelector, where its not one we generated.
@@ -180,7 +204,7 @@ if (include_filters.length) {
             }
         } catch (e) {
             // Maybe catch DOMException and alert?
-            console.log("xpath_element_scraper: Exception selecting element from filter "+f);
+            console.log("xpath_element_scraper: Exception selecting element from filter " + f);
             console.log(e);
         }
 
@@ -210,8 +234,8 @@ if (include_filters.length) {
                 }
             }
         }
-        
-        if(!q) {
+
+        if (!q) {
             console.log("xpath_element_scraper: filter element " + f + " was not found");
         }
 
@@ -221,7 +245,7 @@ if (include_filters.length) {
                 width: parseInt(bbox['width']),
                 height: parseInt(bbox['height']),
                 left: parseInt(bbox['left']),
-                top: parseInt(bbox['top'])+scroll_y
+                top: parseInt(bbox['top']) + scroll_y
             });
         }
     }
@@ -229,7 +253,7 @@ if (include_filters.length) {
 
 // Sort the elements so we find the smallest one first, in other words, we find the smallest one matching in that area
 // so that we dont select the wrapping element by mistake and be unable to select what we want
-size_pos.sort((a, b) => (a.width*a.height > b.width*b.height) ? 1 : -1)
+size_pos.sort((a, b) => (a.width * a.height > b.width * b.height) ? 1 : -1)
 
 // Window.width required for proper scaling in the frontend
 return {'size_pos': size_pos, 'browser_width': window.innerWidth};

changedetectionio/flask_app.py

@@ -30,6 +30,7 @@ from flask_compress import Compress as FlaskCompress
 from flask_login import current_user
 from flask_paginate import Pagination, get_page_parameter
 from flask_restful import abort, Api
+from flask_cors import CORS
 from flask_wtf import CSRFProtect
 from loguru import logger
 
@@ -53,6 +54,9 @@ app = Flask(__name__,
             static_folder="static",
             template_folder="templates")
 
+# Enable CORS, especially useful for the Chrome extension to operate from anywhere
+CORS(app)
+
 # Super handy for compressing large BrowserSteps responses and others
 FlaskCompress(app)
 
@@ -404,17 +408,21 @@ def changedetection_app(config=None, datastore_o=None):
         global datastore
         from changedetectionio import forms
 
-        limit_tag = request.args.get('tag', '').lower().strip()
+        active_tag_req = request.args.get('tag', '').lower().strip()
+        active_tag_uuid = active_tag = None
 
         # Be sure limit_tag is a uuid
-        for uuid, tag in datastore.data['settings']['application'].get('tags', {}).items():
-            if limit_tag == tag.get('title', '').lower().strip():
-                limit_tag = uuid
+        if active_tag_req:
+            for uuid, tag in datastore.data['settings']['application'].get('tags', {}).items():
+                if active_tag_req == tag.get('title', '').lower().strip() or active_tag_req == uuid:
+                    active_tag = tag
+                    active_tag_uuid = uuid
+                    break
 
 
         # Redirect for the old rss path which used the /?rss=true
         if request.args.get('rss'):
-            return redirect(url_for('rss', tag=limit_tag))
+            return redirect(url_for('rss', tag=active_tag_uuid))
 
         op = request.args.get('op')
         if op:
@@ -425,7 +433,7 @@ def changedetection_app(config=None, datastore_o=None):
                 datastore.data['watching'][uuid].toggle_mute()
 
             datastore.needs_write = True
-            return redirect(url_for('index', tag = limit_tag))
+            return redirect(url_for('index', tag = active_tag_uuid))
 
         # Sort by last_changed and add the uuid which is usually the key..
         sorted_watches = []
@@ -436,7 +444,7 @@ def changedetection_app(config=None, datastore_o=None):
             if with_errors and not watch.get('last_error'):
                 continue
 
-            if limit_tag and not limit_tag in watch['tags']:
+            if active_tag_uuid and not active_tag_uuid in watch['tags']:
                     continue
             if watch.get('last_error'):
                 errored_count += 1
@@ -455,11 +463,12 @@ def changedetection_app(config=None, datastore_o=None):
                                 total=total_count,
                                 per_page=datastore.data['settings']['application'].get('pager_size', 50), css_framework="semantic")
 
-
+        sorted_tags = sorted(datastore.data['settings']['application'].get('tags').items(), key=lambda x: x[1]['title'])
         output = render_template(
             "watch-overview.html",
                                  # Don't link to hosting when we're on the hosting environment
-                                 active_tag=limit_tag,
+                                 active_tag=active_tag,
+                                 active_tag_uuid=active_tag_uuid,
                                  app_rss_token=datastore.data['settings']['application']['rss_access_token'],
                                  datastore=datastore,
                                  errored_count=errored_count,
@@ -474,7 +483,7 @@ def changedetection_app(config=None, datastore_o=None):
                                  sort_attribute=request.args.get('sort') if request.args.get('sort') else request.cookies.get('sort'),
                                  sort_order=request.args.get('order') if request.args.get('order') else request.cookies.get('order'),
                                  system_default_fetcher=datastore.data['settings']['application'].get('fetch_backend'),
-                                 tags=datastore.data['settings']['application'].get('tags'),
+                                 tags=sorted_tags,
                                  watches=sorted_watches
                                  )
 
@@ -1422,6 +1431,13 @@ def changedetection_app(config=None, datastore_o=None):
                     update_q.put(queuedWatchMetaData.PrioritizedItem(priority=1, item={'uuid': uuid, 'skip_when_checksum_same': False}))
             flash("{} watches queued for rechecking".format(len(uuids)))
 
+        elif (op == 'clear-errors'):
+            for uuid in uuids:
+                uuid = uuid.strip()
+                if datastore.data['watching'].get(uuid):
+                    datastore.data['watching'][uuid]["last_error"] = False
+            flash(f"{len(uuids)} watches errors cleared")
+
         elif (op == 'clear-history'):
             for uuid in uuids:
                 uuid = uuid.strip()

changedetectionio/processors/__init__.py

@@ -75,12 +75,12 @@ class difference_detection_processor():
 
         proxy_url = None
         if preferred_proxy_id:
-            # Custom browser endpoints should not have a proxy added
-            if not preferred_proxy_id.startswith('ui-'):
+            # Custom browser endpoints should NOT have a proxy added
+            if not prefer_fetch_backend.startswith('extra_browser_'):
                 proxy_url = self.datastore.proxy_list.get(preferred_proxy_id).get('url')
                 logger.debug(f"Selected proxy key '{preferred_proxy_id}' as proxy URL '{proxy_url}' for {url}")
             else:
-                logger.debug(f"Skipping adding proxy data when custom Browser endpoint is specified.")
+                logger.debug(f"Skipping adding proxy data when custom Browser endpoint is specified. ")
 
         # Now call the fetcher (playwright/requests/etc) with arguments that only a fetcher would need.
         # When browser_connection_url is None, it method should default to working out whats the best defaults (os env vars etc)

changedetectionio/processors/text_json_diff.py

@@ -119,7 +119,7 @@ class perform_site_check(difference_detection_processor):
         include_filters_from_tags = self.datastore.get_tag_overrides_for_watch(uuid=uuid, attr='include_filters')
 
         # 1845 - remove duplicated filters in both group and watch include filter
-        include_filters_rule = list({*watch.get('include_filters', []), *include_filters_from_tags})
+        include_filters_rule = list(dict.fromkeys(watch.get('include_filters', []) + include_filters_from_tags))
 
         subtractive_selectors = [*self.datastore.get_tag_overrides_for_watch(uuid=uuid, attr='subtractive_selectors'),
                                  *watch.get("subtractive_selectors", []),

changedetectionio/static/js/browser-steps.js

@@ -160,6 +160,12 @@ $(document).ready(function () {
                     e.offsetX > item.left * y_scale && e.offsetX < item.left * y_scale + item.width * y_scale
 
                 ) {
+                    // Ignore really large ones, because we are scraping 'div' also from xpath_element_scraper but
+                    // that div or whatever could be some wrapper and would generally make you select the whole page
+                    if (item.width > 800 && item.height > 400) {
+                        return
+                    }
+
                     // There could be many elements here, record them all and then we'll find out which is the most 'useful'
                     // (input, textarea, button, A etc)
                     if (item.width < xpath_data['browser_width']) {

changedetectionio/static/styles/scss/styles.scss

@@ -1096,3 +1096,16 @@ ul {
   white-space: nowrap;
 }
 
+#chrome-extension-link {
+  img {
+    height: 21px;
+    padding: 2px;
+    vertical-align: middle;
+  }
+
+  padding: 9px;
+  border: 1px solid var(--color-grey-800);
+  border-radius: 10px;
+  vertical-align: middle;
+}
+

changedetectionio/static/styles/styles.css

@@ -1180,3 +1180,13 @@ ul {
   .restock-label.not-in-stock {
     background-color: var(--color-background-button-cancel);
     color: #777; }
+
+#chrome-extension-link {
+  padding: 9px;
+  border: 1px solid var(--color-grey-800);
+  border-radius: 10px;
+  vertical-align: middle; }
+  #chrome-extension-link img {
+    height: 21px;
+    padding: 2px;
+    vertical-align: middle; }

changedetectionio/templates/base.html

@@ -147,7 +147,19 @@
     <section class="content">
         <div id="overlay">
             <div class="content">
-                <strong>changedetection.io needs your support!</strong><br>
+                <h4>Try our Chrome extension</h4>
+                <p>
+                    <a id="chrome-extension-link"
+                       title="Try our new Chrome Extension!"
+                       href="https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop">
+                        <img src="{{url_for('static_content', group='images', filename='Google-Chrome-icon.png')}}">
+                        Chrome Webstore
+                    </a>
+                </p>
+
+                Easily add the current web-page from your browser directly into your changedetection.io tool, more great features coming soon!
+
+                <h4>Changedetection.io needs your support!</h4>
                 <p>
                     You can help us by supporting changedetection.io on these platforms;
                 </p>

changedetectionio/templates/import.html

@@ -107,7 +107,7 @@
                                     <option value="" style="color: #aaa"> -- none --</option>
                                     <option value="url">URL</option>
                                     <option value="title">Title</option>
-                                    <option value="include_filter">CSS/xPath filter</option>
+                                    <option value="include_filters">CSS/xPath filter</option>
                                     <option value="tag">Group / Tag name(s)</option>
                                     <option value="interval_minutes">Recheck time (minutes)</option>
                                 </select></td>

changedetectionio/templates/settings.html

@@ -168,12 +168,12 @@ nav
            </div>
 
             <div class="tab-pane-inner" id="api">
-
+                <h4>API Access</h4>
                 <p>Drive your changedetection.io via API, More about <a href="https://github.com/dgtlmoon/changedetection.io/wiki/API-Reference">API access here</a></p>
 
                 <div class="pure-control-group">
                     {{ render_checkbox_field(form.application.form.api_access_token_enabled) }}
-                    <div class="pure-form-message-inline">Restrict API access limit by using <code>x-api-key</code> header</div><br>
+                    <div class="pure-form-message-inline">Restrict API access limit by using <code>x-api-key</code> header - required for the Chrome Extension to work</div><br>
                     <div class="pure-form-message-inline"><br>API Key <span id="api-key">{{api_key}}</span>
                         <span style="display:none;" id="api-key-copy" >copy</span>
                     </div>
@@ -181,6 +181,20 @@ nav
                 <div class="pure-control-group">
                     <a href="{{url_for('settings_reset_api_key')}}" class="pure-button button-small button-cancel">Regenerate API key</a>
                 </div>
+                <div class="pure-control-group">
+                    <h4>Chrome Extension</h4>
+                    <p>Easily add any web-page to your changedetection.io installation from within Chrome.</p>
+                    <strong>Step 1</strong> Install the extension, <strong>Step 2</strong> Navigate to this page,
+                    <strong>Step 3</strong> Open the extension from the toolbar and click "<i>Sync API Access</i>"
+                    <p>
+                        <a id="chrome-extension-link"
+                           title="Try our new Chrome Extension!"
+                           href="https://chromewebstore.google.com/detail/changedetectionio-website/kefcfmgmlhmankjmnbijimhofdjekbop">
+                            <img src="{{ url_for('static_content', group='images', filename='Google-Chrome-icon.png') }}">
+                            Chrome Webstore
+                        </a>
+                    </p>
+                </div>
             </div>
             <div class="tab-pane-inner" id="proxies">
                 <div id="recommended-proxy">

changedetectionio/templates/watch-overview.html

@@ -1,6 +1,6 @@
 {% extends 'base.html' %}
 {% block content %}
-{% from '_helpers.jinja' import render_simple_field, render_field, render_nolabel_field %}
+{% from '_helpers.jinja' import render_simple_field, render_field, render_nolabel_field, sort_by_title %}
 <script src="{{url_for('static_content', group='js', filename='jquery-3.6.0.min.js')}}"></script>
 <script src="{{url_for('static_content', group='js', filename='watch-overview.js')}}" defer></script>
 
@@ -13,7 +13,7 @@
             <div id="watch-add-wrapper-zone">
 
                     {{ render_nolabel_field(form.url, placeholder="https://...", required=true) }}
-                    {{ render_nolabel_field(form.tags, value=tags[active_tag].title if active_tag else '', placeholder="watch label / tag") }}
+                    {{ render_nolabel_field(form.tags, value=active_tag.title if active_tag else '', placeholder="watch label / tag") }}
                     {{ render_nolabel_field(form.watch_submit_button, title="Watch this URL!" ) }}
                     {{ render_nolabel_field(form.edit_and_watch_submit_button, title="Edit first then Watch") }}
             </div>
@@ -37,6 +37,7 @@
         <button class="pure-button button-secondary button-xsmall" name="op" value="assign-tag" id="checkbox-assign-tag">Tag</button>
         <button class="pure-button button-secondary button-xsmall" name="op" value="mark-viewed">Mark viewed</button>
         <button class="pure-button button-secondary button-xsmall" name="op" value="notification-default">Use default notification</button>
+        <button class="pure-button button-secondary button-xsmall" name="op" value="clear-errors">Clear errors</button>
         <button class="pure-button button-secondary button-xsmall" style="background: #dd4242;" name="op" value="clear-history">Clear/reset history</button>
         <button class="pure-button button-secondary button-xsmall" style="background: #dd4242;" name="op" value="delete">Delete</button>
     </div>
@@ -46,11 +47,13 @@
     {% if search_q %}<div id="search-result-info">Searching "<strong><i>{{search_q}}</i></strong>"</div>{% endif %}
     <div>
         <a href="{{url_for('index')}}" class="pure-button button-tag {{'active' if not active_tag }}">All</a>
-        {% for uuid, tag in tags.items() %}
-            {% if tag != "" %}
-                <a href="{{url_for('index', tag=uuid) }}" class="pure-button button-tag {{'active' if active_tag == uuid }}">{{ tag.title }}</a>
-            {% endif %}
-        {% endfor %}
+
+    <!-- tag list -->
+    {% for uuid, tag in tags %}
+        {% if tag != "" %}
+            <a href="{{url_for('index', tag=uuid) }}" class="pure-button button-tag {{'active' if active_tag_uuid == uuid }}">{{ tag.title }}</a>
+        {% endif %}
+    {% endfor %}
     </div>
 
     {% set sort_order = sort_order or 'asc' %}
@@ -197,8 +200,8 @@
             </li>
             {% endif %}
             <li>
-               <a href="{{ url_for('form_watch_checknow', tag=active_tag, with_errors=request.args.get('with_errors',0)) }}" class="pure-button button-tag ">Recheck
-                all {% if active_tag%} in "{{tags[active_tag].title}}"{%endif%}</a>
+               <a href="{{ url_for('form_watch_checknow', tag=active_tag_uuid, with_errors=request.args.get('with_errors',0)) }}" class="pure-button button-tag ">Recheck
+                all {% if active_tag_uuid %} in "{{active_tag.title}}"{%endif%}</a>
             </li>
             <li>
                 <a href="{{ url_for('rss', tag=active_tag , token=app_rss_token)}}"><img alt="RSS Feed" id="feed-icon" src="{{url_for('static_content', group='images', filename='Generic_Feed-icon.svg')}}" height="15"></a>

changedetectionio/tests/fetchers/test_custom_js_before_content.py

@@ -0,0 +1,56 @@
+import os
+from flask import url_for
+from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_client
+
+
+def test_execute_custom_js(client, live_server):
+
+    live_server_setup(live_server)
+    assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
+
+    test_url = url_for('test_interactive_html_endpoint', _external=True)
+    test_url = test_url.replace('localhost.localdomain', 'cdio')
+    test_url = test_url.replace('localhost', 'cdio')
+
+    res = client.post(
+        url_for("form_quick_watch_add"),
+        data={"url": test_url, "tags": '', 'edit_and_watch_submit_button': 'Edit > Watch'},
+        follow_redirects=True
+    )
+
+    assert b"Watch added in Paused state, saving will unpause" in res.data
+
+    res = client.post(
+        url_for("edit_page", uuid="first", unpause_on_save=1),
+        data={
+            "url": test_url,
+            "tags": "",
+            'fetch_backend': "html_webdriver",
+            'webdriver_js_execute_code': 'document.querySelector("button[name=test-button]").click();',
+            'headers': "testheader: yes\buser-agent: MyCustomAgent",
+        },
+        follow_redirects=True
+    )
+    assert b"unpaused" in res.data
+    wait_for_all_checks(client)
+
+    uuid = extract_UUID_from_client(client)
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid].history_n >= 1, "Watch history had atleast 1 (everything fetched OK)"
+
+    assert b"This text should be removed" not in res.data
+
+    # Check HTML conversion detected and workd
+    res = client.get(
+        url_for("preview_page", uuid=uuid),
+        follow_redirects=True
+    )
+    assert b"This text should be removed" not in res.data
+    assert b"I smell JavaScript because the button was pressed" in res.data
+
+    assert b"testheader: yes" in res.data
+    assert b"user-agent: mycustomagent" in res.data
+
+    client.get(
+        url_for("form_delete", uuid="all"),
+        follow_redirects=True
+    )

changedetectionio/tests/restock/test_restock.py

@@ -95,7 +95,7 @@ def test_restock_detection(client, live_server):
 
     # We should have a notification
     time.sleep(2)
-    assert os.path.isfile("test-datastore/notification.txt")
+    assert os.path.isfile("test-datastore/notification.txt"), "Notification received"
     os.unlink("test-datastore/notification.txt")
 
     # Default behaviour is to only fire notification when it goes OUT OF STOCK -> IN STOCK
@@ -103,4 +103,9 @@ def test_restock_detection(client, live_server):
     set_original_response()
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     wait_for_all_checks(client)
-    assert not os.path.isfile("test-datastore/notification.txt")
+    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"
+    res = client.get(url_for("index"))
+    assert b'not-in-stock' in res.data, "Correctly showing NOT IN STOCK in the list after it changed from IN STOCK"
+

changedetectionio/tests/test_group.py

@@ -321,3 +321,154 @@ def test_clone_tag_on_quickwatchform_add(client, live_server):
 
     res = client.get(url_for("tags.delete_all"), follow_redirects=True)
     assert b'All tags deleted' in res.data
+
+def test_order_of_filters_tag_filter_and_watch_filter(client, live_server):
+
+    # Add a tag with some config, import a tag and it should roughly work
+    res = client.post(
+        url_for("tags.form_tag_add"),
+        data={"name": "test-tag-keep-order"},
+        follow_redirects=True
+    )
+    assert b"Tag added" in res.data
+    assert b"test-tag-keep-order" in res.data
+    tag_filters = [
+            '#only-this', # duplicated filters
+            '#only-this',
+            '#only-this',
+            '#only-this',
+            ]
+
+    res = client.post(
+        url_for("tags.form_tag_edit_submit", uuid="first"),
+        data={"name": "test-tag-keep-order",
+              "include_filters": '\n'.join(tag_filters) },
+        follow_redirects=True
+    )
+    assert b"Updated" in res.data
+    tag_uuid = get_UUID_for_tag_name(client, name="test-tag-keep-order")
+    res = client.get(
+        url_for("tags.form_tag_edit", uuid="first")
+    )
+    assert b"#only-this" in res.data
+
+
+    d = """<html>
+       <body>
+     Some initial text<br>
+     <p id="only-this">And 1 this</p>
+     <br>
+     <p id="not-this">And 2 this</p>
+     <p id="">And 3 this</p><!--/html/body/p[3]/-->
+     <p id="">And 4 this</p><!--/html/body/p[4]/-->
+     <p id="">And 5 this</p><!--/html/body/p[5]/-->
+     <p id="">And 6 this</p><!--/html/body/p[6]/-->
+     <p id="">And 7 this</p><!--/html/body/p[7]/-->
+     <p id="">And 8 this</p><!--/html/body/p[8]/-->
+     <p id="">And 9 this</p><!--/html/body/p[9]/-->
+     <p id="">And 10 this</p><!--/html/body/p[10]/-->
+     <p id="">And 11 this</p><!--/html/body/p[11]/-->
+     <p id="">And 12 this</p><!--/html/body/p[12]/-->
+     <p id="">And 13 this</p><!--/html/body/p[13]/-->
+     <p id="">And 14 this</p><!--/html/body/p[14]/-->
+     <p id="not-this">And 15 this</p><!--/html/body/p[15]/-->
+     </body>
+     </html>
+    """
+
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write(d)
+
+    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
+    wait_for_all_checks(client)
+
+    filters = [
+            '/html/body/p[3]',
+            '/html/body/p[4]',
+            '/html/body/p[5]',
+            '/html/body/p[6]',
+            '/html/body/p[7]',
+            '/html/body/p[8]',
+            '/html/body/p[9]',
+            '/html/body/p[10]',
+            '/html/body/p[11]',
+            '/html/body/p[12]',
+            '/html/body/p[13]', # duplicated tags
+            '/html/body/p[13]',
+            '/html/body/p[13]',
+            '/html/body/p[13]',
+            '/html/body/p[13]',
+            '/html/body/p[14]',
+            ]
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": '\n'.join(filters),
+            "url": test_url,
+            "tags": "test-tag-keep-order",
+            "headers": "",
+            'fetch_backend': "html_requests"},
+        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"And 1 this" in res.data  # test-tag-keep-order
+
+    a_tag_filter_check = b'And 1 this' #'#only-this' of tag_filters
+    # check there is no duplication of tag_filters
+    assert res.data.count(a_tag_filter_check) == 1, f"duplicated filters didn't removed {res.data.count(a_tag_filter_check)} of {a_tag_filter_check} in {res.data=}"
+
+    a_filter_check = b"And 13 this" # '/html/body/p[13]'
+    # check there is no duplication of filters
+    assert res.data.count(a_filter_check) == 1, f"duplicated filters didn't removed. {res.data.count(a_filter_check)} of {a_filter_check} in {res.data=}"
+
+    a_filter_check_not_include = b"And 2 this" # '/html/body/p[2]'
+    assert a_filter_check_not_include not in res.data
+
+    checklist = [
+            b"And 3 this",
+            b"And 4 this",
+            b"And 5 this",
+            b"And 6 this",
+            b"And 7 this",
+            b"And 8 this",
+            b"And 9 this",
+            b"And 10 this",
+            b"And 11 this",
+            b"And 12 this",
+            b"And 13 this",
+            b"And 14 this",
+            b"And 1 this", # result of filter from tag.
+            ]
+    # check whether everything a user requested is there
+    for test in checklist:
+        assert test in res.data
+
+    # check whether everything a user requested is in order of filters.
+    n = 0
+    for test in checklist:
+        t_index = res.data[n:].find(test)
+        # if the text is not searched, return -1.
+        assert t_index >= 0, f"""failed because {test=} not in {res.data[n:]=}
+#####################
+Looks like some feature changed the order of result of filters.
+#####################
+the {test} appeared before. {test in res.data[:n]=}
+{res.data[:n]=}
+        """
+        n += t_index + len(test)
+
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data

changedetectionio/tests/test_pdf.py

@@ -29,7 +29,8 @@ def test_fetch_pdf(client, live_server):
         follow_redirects=True
     )
 
-    assert b'PDF-1.5' not in res.data
+    # PDF header should not be there (it was converted to text)
+    assert b'PDF' not in res.data[:10]
     assert b'hello world' in res.data
 
     # So we know if the file changes in other ways

changedetectionio/tests/util.py

@@ -242,5 +242,28 @@ def live_server_setup(live_server):
             resp.headers['Content-Type'] = 'application/pdf'
             return resp
 
+    @live_server.app.route('/test-interactive-html-endpoint')
+    def test_interactive_html_endpoint():
+        header_text=""
+        for k,v in request.headers.items():
+            header_text += f"{k}: {v}<br>"
+
+        resp = make_response(f"""
+        <html>
+          <body>
+          Primitive JS check for <pre>changedetectionio/tests/visualselector/test_fetch_data.py</pre>
+            <p id="remove">This text should be removed</p>
+              <form onsubmit="event.preventDefault();">
+            <!-- obfuscated text so that we dont accidentally get a false positive due to conversion of the source :) --->
+                <button name="test-button" onclick="getElementById('remove').remove();getElementById('some-content').innerHTML = atob('SSBzbWVsbCBKYXZhU2NyaXB0IGJlY2F1c2UgdGhlIGJ1dHRvbiB3YXMgcHJlc3NlZCE=')">Click here</button>
+                <div id=some-content></div>
+                <pre>
+                {header_text.lower()}
+                </pre>
+              </body>
+         </html>""", 200)
+        resp.headers['Content-Type'] = 'text/html'
+        return resp
+
     live_server.start()
 

changedetectionio/tests/visualselector/test_fetch_data.py

@@ -7,15 +7,19 @@ from ..util import live_server_setup, wait_for_all_checks, extract_UUID_from_cli
 def test_setup(client, live_server):
     live_server_setup(live_server)
 
+
 # Add a site in paused mode, add an invalid filter, we should still have visual selector data ready
 def test_visual_selector_content_ready(client, live_server):
+
     import os
     import json
 
     assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
 
     # Add our URL to the import page, because the docker container (playwright/selenium) wont be able to connect to our usual test url
-    test_url = "https://changedetection.io/ci-test/test-runjs.html"
+    test_url = url_for('test_interactive_html_endpoint', _external=True)
+    test_url = test_url.replace('localhost.localdomain', 'cdio')
+    test_url = test_url.replace('localhost', 'cdio')
 
     res = client.post(
         url_for("form_quick_watch_add"),
@@ -23,28 +27,31 @@ def test_visual_selector_content_ready(client, live_server):
         follow_redirects=True
     )
     assert b"Watch added in Paused state, saving will unpause" in res.data
-
+    uuid = extract_UUID_from_client(client)
     res = client.post(
-        url_for("edit_page", uuid="first", unpause_on_save=1),
+        url_for("edit_page", uuid=uuid, unpause_on_save=1),
         data={
-              "url": test_url,
-              "tags": "",
-              "headers": "",
-              'fetch_backend': "html_webdriver",
-              'webdriver_js_execute_code': 'document.querySelector("button[name=test-button]").click();'
+            "url": test_url,
+            "tags": "",
+            # For now, cookies doesnt work in headers because it must be a full cookiejar object
+            'headers': "testheader: yes\buser-agent: MyCustomAgent",
+            'fetch_backend': "html_webdriver",
         },
         follow_redirects=True
     )
     assert b"unpaused" in res.data
     wait_for_all_checks(client)
-    uuid = extract_UUID_from_client(client)
 
-    # Check the JS execute code before extract worked
+
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid].history_n >= 1, "Watch history had atleast 1 (everything fetched OK)"
+
     res = client.get(
-        url_for("preview_page", uuid="first"),
+        url_for("preview_page", uuid=uuid),
         follow_redirects=True
     )
-    assert b'I smell JavaScript' in res.data
+    assert b"testheader: yes" in res.data
+    assert b"user-agent: mycustomagent" in res.data
+
 
     assert os.path.isfile(os.path.join('test-datastore', uuid, 'last-screenshot.png')), "last-screenshot.png should exist"
     assert os.path.isfile(os.path.join('test-datastore', uuid, 'elements.json')), "xpath elements.json data should exist"
@@ -74,30 +81,33 @@ def test_visual_selector_content_ready(client, live_server):
 
 def test_basic_browserstep(client, live_server):
 
-    assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
     #live_server_setup(live_server)
+    assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
 
-    # Add our URL to the import page, because the docker container (playwright/selenium) wont be able to connect to our usual test url
-    test_url = "https://changedetection.io/ci-test/test-runjs.html"
+    test_url = url_for('test_interactive_html_endpoint', _external=True)
+    test_url = test_url.replace('localhost.localdomain', 'cdio')
+    test_url = test_url.replace('localhost', 'cdio')
 
     res = client.post(
         url_for("form_quick_watch_add"),
         data={"url": test_url, "tags": '', 'edit_and_watch_submit_button': 'Edit > Watch'},
         follow_redirects=True
     )
+
     assert b"Watch added in Paused state, saving will unpause" in res.data
 
     res = client.post(
         url_for("edit_page", uuid="first", unpause_on_save=1),
         data={
-              "url": test_url,
-              "tags": "",
-              "headers": "",
-              'fetch_backend': "html_webdriver",
-              'browser_steps-0-operation': 'Goto site',
-              'browser_steps-1-operation': 'Click element',
-              'browser_steps-1-selector': 'button[name=test-button]',
-              'browser_steps-1-optional_value': ''
+            "url": test_url,
+            "tags": "",
+            'fetch_backend': "html_webdriver",
+            'browser_steps-0-operation': 'Goto site',
+            'browser_steps-1-operation': 'Click element',
+            'browser_steps-1-selector': 'button[name=test-button]',
+            'browser_steps-1-optional_value': '',
+            # For now, cookies doesnt work in headers because it must be a full cookiejar object
+            'headers': "testheader: yes\buser-agent: MyCustomAgent",
         },
         follow_redirects=True
     )
@@ -105,6 +115,9 @@ def test_basic_browserstep(client, live_server):
     wait_for_all_checks(client)
 
     uuid = extract_UUID_from_client(client)
+    assert live_server.app.config['DATASTORE'].data['watching'][uuid].history_n >= 1, "Watch history had atleast 1 (everything fetched OK)"
+
+    assert b"This text should be removed" not in res.data
 
     # Check HTML conversion detected and workd
     res = client.get(
@@ -114,13 +127,19 @@ def test_basic_browserstep(client, live_server):
     assert b"This text should be removed" not in res.data
     assert b"I smell JavaScript because the button was pressed" in res.data
 
+    assert b"testheader: yes" in res.data
+    assert b"user-agent: mycustomagent" in res.data
+
+    four_o_four_url =  url_for('test_endpoint', status_code=404, _external=True)
+    four_o_four_url = four_o_four_url.replace('localhost.localdomain', 'cdio')
+    four_o_four_url = four_o_four_url.replace('localhost', 'cdio')
+
     # now test for 404 errors
     res = client.post(
         url_for("edit_page", uuid=uuid, unpause_on_save=1),
         data={
-              "url": "https://changedetection.io/404",
+              "url": four_o_four_url,
               "tags": "",
-              "headers": "",
               'fetch_backend': "html_webdriver",
               'browser_steps-0-operation': 'Goto site',
               'browser_steps-1-operation': 'Click element',

requirements.txt

@@ -1,7 +1,7 @@
 # Used by Pyppeteer
 pyee
 
-eventlet>=0.33.3 # related to dnspython fixes
+eventlet==0.33.3 # related to dnspython fixes
 feedgen~=0.9
 flask-compress
 # 0.6.3 included compatibility fix for werkzeug 3.x (2.x had deprecation of url handlers)
@@ -9,6 +9,7 @@ flask-login>=0.6.3
 flask-paginate
 flask_expects_json~=1.7
 flask_restful
+flask_cors # For the Chrome extension to operate
 flask_wtf~=1.2
 flask~=2.3
 inscriptis~=2.2
@@ -22,21 +23,25 @@ validators~=0.21
 brotli~=1.0
 requests[socks]
 
-urllib3>1.26
+urllib3==1.26.18
 chardet>2.3.0
 
 wtforms~=3.0
 jsonpath-ng~=1.5.3
 
-dnspython~=2.4 # related to eventlet fixes
+# Pinned: module 'eventlet.green.select' has no attribute 'epoll'
+# https://github.com/eventlet/eventlet/issues/805#issuecomment-1640463482
+dnspython==2.3.0 # related to eventlet fixes
 
 # jq not available on Windows so must be installed manually
 
 # Notification library
-apprise~=1.7.1
+apprise~=1.7.4
 
 # apprise mqtt https://github.com/dgtlmoon/changedetection.io/issues/315
-paho-mqtt
+# and 2.0.0 https://github.com/dgtlmoon/changedetection.io/issues/2241 not yet compatible
+# use v1.x due to https://github.com/eclipse/paho.mqtt.python/issues/814
+paho-mqtt < 2.0.0
 
 # This mainly affects some ARM builds, which unlike the other builds ignores "ARG CRYPTOGRAPHY_DONT_BUILD_RUST=1"
 # so without this pinning, the newer versions on ARM will forcefully try to build rust, which results in "rust compiler not found"
@@ -70,7 +75,7 @@ pillow
 # playwright is installed at Dockerfile build time because it's not available on all platforms
 
 # experimental release
-pyppeteer-ng==2.0.0rc2
+pyppeteer-ng==2.0.0rc5
 
 # Include pytest, so if theres a support issue we can ask them to run these tests on their setup
 pytest ~=7.2