0.49.17

changedetectionio/__init__.py

@@ -2,7 +2,7 @@
 
 # Read more https://github.com/dgtlmoon/changedetection.io/wiki
 
-__version__ = '0.49.15'
+__version__ = '0.49.17'
 
 from changedetectionio.strtobool import strtobool
 from json.decoder import JSONDecodeError

changedetectionio/blueprint/browser_steps/__init__.py

@@ -168,9 +168,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
             step_optional_value = request.form.get('optional_value')
             is_last_step = strtobool(request.form.get('is_last_step'))
 
-            # @todo try.. accept.. nice errors not popups..
             try:
-
                 browsersteps_sessions[browsersteps_session_id]['browserstepper'].call_action(action_name=step_operation,
                                          selector=step_selector,
                                          optional_value=step_optional_value)

changedetectionio/blueprint/browser_steps/browser_steps.py

@@ -62,23 +62,6 @@ class steppable_browser_interface():
     def __init__(self, start_url):
         self.start_url = start_url
 
-    def safe_page_operation(self, operation_fn, default_return=None):
-        """Safely execute a page operation with error handling"""
-        if self.page is None:
-            logger.warning("Attempted operation on None page object")
-            return default_return
-            
-        try:
-            return operation_fn()
-        except Exception as e:
-            logger.debug(f"Page operation failed: {str(e)}")
-            # Try to reclaim memory if possible
-            try:
-                self.page.request_gc()
-            except:
-                pass
-            return default_return
-
     # Convert and perform "Click Button" for example
     def call_action(self, action_name, selector=None, optional_value=None):
         if self.page is None:
@@ -109,20 +92,11 @@ class steppable_browser_interface():
         if optional_value and ('{%' in optional_value or '{{' in optional_value):
             optional_value = jinja_render(template_str=optional_value)
 
-        try:
+
         action_handler(selector, optional_value)
         # Safely wait for timeout
-            def wait_timeout():
         self.page.wait_for_timeout(1.5 * 1000)
-            self.safe_page_operation(wait_timeout)
         logger.debug(f"Call action done in {time.time()-now:.2f}s")
-        except Exception as e:
-            logger.error(f"Error executing action '{call_action_name}': {str(e)}")
-            # Request garbage collection to free up resources after error
-            try:
-                self.page.request_gc()
-            except:
-                pass
 
     def action_goto_url(self, selector=None, value=None):
         if not value:
@@ -130,11 +104,7 @@ class steppable_browser_interface():
             return None
             
         now = time.time()
-        
+        response = self.page.goto(value, timeout=0, wait_until='load')
-        def goto_operation():
-            return self.page.goto(value, timeout=0, wait_until='load')
-            
-        response = self.safe_page_operation(goto_operation)
         logger.debug(f"Time to goto URL {time.time()-now:.2f}s")
         return response
 
@@ -147,61 +117,47 @@ class steppable_browser_interface():
         if not value or not len(value.strip()):
             return
             
-        def click_operation():
         elem = self.page.get_by_text(value)
         if elem.count():
             elem.first.click(delay=randint(200, 500), timeout=self.action_timeout)
 
-        self.safe_page_operation(click_operation)
 
     def action_click_element_containing_text_if_exists(self, selector=None, value=''):
         logger.debug("Clicking element containing text if exists")
         if not value or not len(value.strip()):
             return
             
-        def click_if_exists_operation():
         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=self.action_timeout)
                 
-        self.safe_page_operation(click_if_exists_operation)
 
     def action_enter_text_in_field(self, selector, value):
         if not selector or not len(selector.strip()):
             return
 
-        def fill_operation():
         self.page.fill(selector, value, timeout=self.action_timeout)
 
-        self.safe_page_operation(fill_operation)
-
     def action_execute_js(self, selector, value):
         if not value:
             return None
             
-        def evaluate_operation():
         return self.page.evaluate(value)
 
-        return self.safe_page_operation(evaluate_operation)
-
     def action_click_element(self, selector, value):
         logger.debug("Clicking element")
         if not selector or not len(selector.strip()):
             return
 
-        def click_operation():
         self.page.click(selector=selector, timeout=self.action_timeout + 20 * 1000, delay=randint(200, 500))
 
-        self.safe_page_operation(click_operation)
-
     def action_click_element_if_exists(self, selector, value):
         import playwright._impl._errors as _api_types
         logger.debug("Clicking element if exists")
         if not selector or not len(selector.strip()):
             return
             
-        def click_if_exists_operation():
         try:
             self.page.click(selector, timeout=self.action_timeout, delay=randint(200, 500))
         except _api_types.TimeoutError:
@@ -210,7 +166,6 @@ class steppable_browser_interface():
             # Element was there, but page redrew and now its long long gone
             return
                 
-        self.safe_page_operation(click_if_exists_operation)
 
     def action_click_x_y(self, selector, value):
         if not value or not re.match(r'^\s?\d+\s?,\s?\d+\s?$', value):
@@ -222,10 +177,8 @@ class steppable_browser_interface():
             x = int(float(x.strip()))
             y = int(float(y.strip()))
             
-            def click_xy_operation():
             self.page.mouse.click(x=x, y=y, delay=randint(200, 500))
                 
-            self.safe_page_operation(click_xy_operation)
         except Exception as e:
             logger.error(f"Error parsing x,y coordinates: {str(e)}")
 
@@ -233,27 +186,17 @@ class steppable_browser_interface():
         if not selector or not len(selector.strip()):
             return
 
-        def select_operation():
         self.page.select_option(selector, label=value, timeout=self.action_timeout)
 
-        self.safe_page_operation(select_operation)
-
     def action_scroll_down(self, selector, value):
-        def scroll_operation():
         # Some sites this doesnt work on for some reason
         self.page.mouse.wheel(0, 600)
         self.page.wait_for_timeout(1000)
 
-        self.safe_page_operation(scroll_operation)
-
     def action_wait_for_seconds(self, selector, value):
         try:
             seconds = float(value.strip()) if value else 1.0
-            
-            def wait_operation():
             self.page.wait_for_timeout(seconds * 1000)
-                
-            self.safe_page_operation(wait_operation)
         except (ValueError, TypeError) as e:
             logger.error(f"Invalid value for wait_for_seconds: {str(e)}")
 
@@ -263,14 +206,11 @@ class steppable_browser_interface():
             
         import json
         v = json.dumps(value)
-        
-        def wait_for_text_operation():
         self.page.wait_for_function(
             f'document.querySelector("body").innerText.includes({v});',
             timeout=30000
         )
             
-        self.safe_page_operation(wait_for_text_operation)
 
     def action_wait_for_text_in_element(self, selector, value):
         if not selector or not value:
@@ -280,68 +220,48 @@ class steppable_browser_interface():
         s = json.dumps(selector)
         v = json.dumps(value)
         
-        def wait_for_text_in_element_operation():
         self.page.wait_for_function(
             f'document.querySelector({s}).innerText.includes({v});',
             timeout=30000
         )
 
-        self.safe_page_operation(wait_for_text_in_element_operation)
-
     # @todo - in the future make some popout interface to capture what needs to be set
     # https://playwright.dev/python/docs/api/class-keyboard
     def action_press_enter(self, selector, value):
-        def press_operation():
         self.page.keyboard.press("Enter", delay=randint(200, 500))
             
-        self.safe_page_operation(press_operation)
 
     def action_press_page_up(self, selector, value):
-        def press_operation():
         self.page.keyboard.press("PageUp", delay=randint(200, 500))
 
-        self.safe_page_operation(press_operation)
-
     def action_press_page_down(self, selector, value):
-        def press_operation():
         self.page.keyboard.press("PageDown", delay=randint(200, 500))
 
-        self.safe_page_operation(press_operation)
-
     def action_check_checkbox(self, selector, value):
         if not selector:
             return
 
-        def check_operation():
         self.page.locator(selector).check(timeout=self.action_timeout)
 
-        self.safe_page_operation(check_operation)
-
     def action_uncheck_checkbox(self, selector, value):
         if not selector:
             return
             
-        def uncheck_operation():
         self.page.locator(selector).uncheck(timeout=self.action_timeout)
             
-        self.safe_page_operation(uncheck_operation)
 
     def action_remove_elements(self, selector, value):
         """Removes all elements matching the given selector from the DOM."""
         if not selector:
             return
             
-        def remove_operation():
         self.page.locator(selector).evaluate_all("els => els.forEach(el => el.remove())")
 
-        self.safe_page_operation(remove_operation)
-
     def action_make_all_child_elements_visible(self, selector, value):
         """Recursively makes all child elements inside the given selector fully visible."""
         if not selector:
             return
             
-        def make_visible_operation():
         self.page.locator(selector).locator("*").evaluate_all("""
             els => els.forEach(el => {
                 el.style.display = 'block';   // Forces it to be displayed
@@ -355,8 +275,6 @@ class steppable_browser_interface():
             })
         """)
 
-        self.safe_page_operation(make_visible_operation)
-
 # Responsible for maintaining a live 'context' with the chrome CDP
 # @todo - how long do contexts live for anyway?
 class browsersteps_live_ui(steppable_browser_interface):

changedetectionio/content_fetchers/playwright.py

@@ -194,7 +194,6 @@ class fetcher(Fetcher):
             browsersteps_interface.page = self.page
 
             response = browsersteps_interface.action_goto_url(value=url)
-            self.headers = response.all_headers()
 
             if response is None:
                 context.close()
@@ -202,6 +201,8 @@ class fetcher(Fetcher):
                 logger.debug("Content Fetcher > Response object from the browser communication was none")
                 raise EmptyReply(url=url, status_code=None)
 
+            self.headers = response.all_headers()
+
             try:
                 if self.webdriver_js_execute_code is not None and len(self.webdriver_js_execute_code):
                     browsersteps_interface.action_execute_js(value=self.webdriver_js_execute_code, selector=None)

changedetectionio/content_fetchers/requests.py

@@ -28,6 +28,7 @@ class fetcher(Fetcher):
 
         import chardet
         import requests
+        from requests.exceptions import ProxyError, ConnectionError, RequestException
 
         if self.browser_steps_get_valid_steps():
             raise BrowserStepsInUnsupportedFetcher(url=url)
@@ -52,7 +53,7 @@ class fetcher(Fetcher):
         if strtobool(os.getenv('ALLOW_FILE_URI', 'false')) and url.startswith('file://'):
             from requests_file import FileAdapter
             session.mount('file://', FileAdapter())
-
+        try:
             r = session.request(method=request_method,
                                 data=request_body.encode('utf-8') if type(request_body) is str else request_body,
                                 url=url,
@@ -60,6 +61,11 @@ class fetcher(Fetcher):
                                 timeout=timeout,
                                 proxies=proxies,
                                 verify=False)
+        except Exception as e:
+            msg = str(e)
+            if proxies and 'SOCKSHTTPSConnectionPool' in msg:
+                msg = f"Proxy connection failed? {msg}"
+            raise Exception(msg) from e
 
         # If the response did not tell us what encoding format to expect, Then use chardet to override what `requests` thinks.
         # For example - some sites don't tell us it's utf-8, but return utf-8 content

changedetectionio/content_fetchers/webdriver_selenium.py

@@ -10,16 +10,13 @@ class fetcher(Fetcher):
     else:
         fetcher_description = "WebDriver Chrome/Javascript"
 
-    # Configs for Proxy setup
-    # In the ENV vars, is prefixed with "webdriver_", so it is for example "webdriver_sslProxy"
-    selenium_proxy_settings_mappings = ['proxyType', 'ftpProxy', 'httpProxy', 'noProxy',
-                                        'proxyAutoconfigUrl', 'sslProxy', 'autodetect',
-                                        'socksProxy', 'socksVersion', 'socksUsername', 'socksPassword']
     proxy = None
+    proxy_url = None
 
     def __init__(self, proxy_override=None, custom_browser_connection_url=None):
         super().__init__()
-        from selenium.webdriver.common.proxy import Proxy as SeleniumProxy
+        from urllib.parse import urlparse
+        from selenium.webdriver.common.proxy import Proxy
 
         # .strip('"') is going to save someone a lot of time when they accidently wrap the env value
         if not custom_browser_connection_url:
@@ -28,25 +25,27 @@ class fetcher(Fetcher):
             self.browser_connection_is_custom = True
             self.browser_connection_url = custom_browser_connection_url
 
-        # If any proxy settings are enabled, then we should setup the proxy object
-        proxy_args = {}
-        for k in self.selenium_proxy_settings_mappings:
-            v = os.getenv('webdriver_' + k, False)
-            if v:
-                proxy_args[k] = v.strip('"')
 
-        # Map back standard HTTP_ and HTTPS_PROXY to webDriver httpProxy/sslProxy
+        ##### PROXY SETUP #####
-        if not proxy_args.get('webdriver_httpProxy') and self.system_http_proxy:
-            proxy_args['httpProxy'] = self.system_http_proxy
-        if not proxy_args.get('webdriver_sslProxy') and self.system_https_proxy:
-            proxy_args['httpsProxy'] = self.system_https_proxy
 
-        # Allows override the proxy on a per-request basis
+        proxy_sources = [
-        if proxy_override is not None:
+            self.system_http_proxy,
-            proxy_args['httpProxy'] = proxy_override
+            self.system_https_proxy,
+            os.getenv('webdriver_proxySocks'),
+            os.getenv('webdriver_socksProxy'),
+            os.getenv('webdriver_proxyHttp'),
+            os.getenv('webdriver_httpProxy'),
+            os.getenv('webdriver_proxyHttps'),
+            os.getenv('webdriver_httpsProxy'),
+            os.getenv('webdriver_sslProxy'),
+            proxy_override, # last one should override
+        ]
+        # The built in selenium proxy handling is super unreliable!!! so we just grab which ever proxy setting we can find and throw it in --proxy-server=
+        for k in filter(None, proxy_sources):
+            if not k:
+                continue
+            self.proxy_url = k.strip()
 
-        if proxy_args:
-            self.proxy = SeleniumProxy(raw=proxy_args)
 
     def run(self,
             url,
@@ -59,9 +58,7 @@ class fetcher(Fetcher):
             is_binary=False,
             empty_pages_are_a_change=False):
 
-        from selenium import webdriver
         from selenium.webdriver.chrome.options import Options as ChromeOptions
-        from selenium.common.exceptions import WebDriverException
         # request_body, request_method unused for now, until some magic in the future happens.
 
         options = ChromeOptions()
@@ -76,30 +73,48 @@ class fetcher(Fetcher):
         for opt in CHROME_OPTIONS:
             options.add_argument(opt)
 
-        if self.proxy:
+        # 1. proxy_config /Proxy(proxy_config) selenium object is REALLY unreliable
-            options.proxy = self.proxy
+        # 2. selenium-wire cant be used because the websocket version conflicts with pypeteer-ng
+        # 3. selenium only allows ONE runner at a time by default!
+        # 4. driver must use quit() or it will continue to block/hold the selenium process!!
 
-        self.driver = webdriver.Remote(
+        if self.proxy_url:
-            command_executor=self.browser_connection_url,
+            options.add_argument(f'--proxy-server={self.proxy_url}')
-            options=options)
+
+        from selenium.webdriver.remote.remote_connection import RemoteConnection
+        from selenium.webdriver.remote.webdriver import WebDriver as RemoteWebDriver
+        driver = None
+        try:
+            # Create the RemoteConnection and set timeout (e.g., 30 seconds)
+            remote_connection = RemoteConnection(
+                self.browser_connection_url,
+            )
+            remote_connection.set_timeout(30)  # seconds
+
+            # Now create the driver with the RemoteConnection
+            driver = RemoteWebDriver(
+                command_executor=remote_connection,
+                options=options
+            )
+
+            driver.set_page_load_timeout(int(os.getenv("WEBDRIVER_PAGELOAD_TIMEOUT", 45)))
+        except Exception as e:
+            if driver:
+                driver.quit()
+            raise e
 
         try:
-            self.driver.get(url)
+            driver.get(url)
-        except WebDriverException as e:
-            # Be sure we close the session window
-            self.quit()
-            raise
 
             if not "--window-size" in os.getenv("CHROME_OPTIONS", ""):
-            self.driver.set_window_size(1280, 1024)
+                driver.set_window_size(1280, 1024)
 
-        self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+            driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
 
             if self.webdriver_js_execute_code is not None:
-            self.driver.execute_script(self.webdriver_js_execute_code)
+                driver.execute_script(self.webdriver_js_execute_code)
                 # Selenium doesn't automatically wait for actions as good as Playwright, so wait again
-            self.driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
+                driver.implicitly_wait(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)))
-
 
             # @todo - how to check this? is it possible?
             self.status_code = 200
@@ -108,27 +123,12 @@ class fetcher(Fetcher):
 
             # @todo - dom wait loaded?
             time.sleep(int(os.getenv("WEBDRIVER_DELAY_BEFORE_CONTENT_READY", 5)) + self.render_extract_delay)
-        self.content = self.driver.page_source
+            self.content = driver.page_source
             self.headers = {}
-
+            self.screenshot = driver.get_screenshot_as_png()
-        self.screenshot = self.driver.get_screenshot_as_png()
-
-    # Does the connection to the webdriver work? run a test connection.
-    def is_ready(self):
-        from selenium import webdriver
-        from selenium.webdriver.chrome.options import Options as ChromeOptions
-
-        self.driver = webdriver.Remote(
-            command_executor=self.command_executor,
-            options=ChromeOptions())
-
-        # driver.quit() seems to cause better exceptions
-        self.quit()
-        return True
-
-    def quit(self, watch=None):
-        if self.driver:
-            try:
-                self.driver.quit()
         except Exception as e:
-                logger.debug(f"Content Fetcher > Exception in chrome shutdown/quit {str(e)}")
+            driver.quit()
+            raise e
+
+        driver.quit()
+

changedetectionio/forms.py

@@ -224,27 +224,37 @@ class StringDictKeyValue(StringField):
 
     def _value(self):
         if self.data:
-            output = u''
+            output = ''
-            for k in self.data.keys():
+            for k, v in self.data.items():
-                output += "{}: {}\r\n".format(k, self.data[k])
+                output += f"{k}: {v}\r\n"
-
             return output
         else:
-            return u''
+            return ''
 
-    # incoming
+    # incoming data processing + validation
     def process_formdata(self, valuelist):
+        self.data = {}
+        errors = []
         if valuelist:
-            self.data = {}
+            # Remove empty strings (blank lines)
-            # Remove empty strings
+            cleaned = [line.strip() for line in valuelist[0].split("\n") if line.strip()]
-            cleaned = list(filter(None, valuelist[0].split("\n")))
+            for idx, s in enumerate(cleaned, start=1):
-            for s in cleaned:
+                if ':' not in s:
-                parts = s.strip().split(':', 1)
+                    errors.append(f"Line {idx} is missing a ':' separator.")
-                if len(parts) == 2:
+                    continue
-                    self.data.update({parts[0].strip(): parts[1].strip()})
+                parts = s.split(':', 1)
+                key = parts[0].strip()
+                value = parts[1].strip()
 
-        else:
+                if not key:
-            self.data = {}
+                    errors.append(f"Line {idx} has an empty key.")
+                if not value:
+                    errors.append(f"Line {idx} has an empty value.")
+
+                self.data[key] = value
+
+        if errors:
+            raise ValidationError("Invalid input:\n" + "\n".join(errors))
 
 class ValidateContentFetcherIsReady(object):
     """

changedetectionio/html_tools.py

@@ -309,10 +309,10 @@ def extract_json_as_string(content, json_filter, ensure_is_ldjson_info_type=None
         soup = BeautifulSoup(content, 'html.parser')
 
         if ensure_is_ldjson_info_type:
-            bs_result = soup.findAll('script', {"type": "application/ld+json"})
+            bs_result = soup.find_all('script', {"type": "application/ld+json"})
         else:
-            bs_result = soup.findAll('script')
+            bs_result = soup.find_all('script')
-        bs_result += soup.findAll('body')
+        bs_result += soup.find_all('body')
 
         bs_jsons = []
         for result in bs_result:
@@ -436,55 +436,27 @@ 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_sub_worker(conn, html_content: str, render_anchor_tag_content=False, is_rss=False):
+# NOTE!! ANYTHING LIBXML, HTML5LIB ETC WILL CAUSE SOME SMALL MEMORY LEAK IN THE LOCAL "LIB" IMPLEMENTATION OUTSIDE PYTHON
 
+
+def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False, timeout=10) -> 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
-
-    :param html_content: string with html content
-    :param render_anchor_tag_content: boolean flag indicating whether to extract
-    hyperlinks (the anchor tag content) together with text. This refers to the
-    'href' inside 'a' tags.
-    Anchor tag content is rendered in the following manner:
-    '[ text ](anchor tag content)'
-    :return: extracted text from the HTML
-    """
-    #  if anchor tag content flag is set to True define a config for
-    #  extracting this content
     if render_anchor_tag_content:
         parser_config = ParserConfig(
             annotation_rules={"a": ["hyperlink"]},
             display_links=True
         )
-    # otherwise set config to None/default
     else:
         parser_config = None
 
-    # RSS Mode - Inscriptis will treat `title` as something else.
-    # Make it as a regular block display element (//item/title)
-    # This is a bit of a hack - the real way it to use XSLT to convert it to HTML #1874
     if is_rss:
         html_content = re.sub(r'<title([\s>])', r'<h1\1', html_content)
         html_content = re.sub(r'</title>', r'</h1>', html_content)
 
     text_content = get_text(html_content, config=parser_config)
-    conn.send(text_content)
+    return text_content
-    conn.close()
-
-# NOTE!! ANYTHING LIBXML, HTML5LIB ETC WILL CAUSE SOME SMALL MEMORY LEAK IN THE LOCAL "LIB" IMPLEMENTATION OUTSIDE PYTHON
-def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False):
-    from multiprocessing import Process, Pipe
-
-    parent_conn, child_conn = Pipe()
-    p = Process(target=html_to_text_sub_worker, args=(child_conn, html_content, render_anchor_tag_content, is_rss))
-    p.start()
-    text = parent_conn.recv()
-    p.join()
-    return text
 
 # Does LD+JSON exist with a @type=='product' and a .price set anywhere?
 def has_ldjson_product_info(content):

changedetectionio/run_basic_tests.sh

@@ -38,6 +38,9 @@ pytest tests/test_backend.py
 pytest tests/test_rss.py
 pytest tests/test_unique_lines.py
 
+# Try high concurrency
+FETCH_WORKERS=130 pytest  tests/test_history_consistency.py -v -l
+
 # Check file:// will pickup a file when enabled
 echo "Hello world" > /tmp/test-file.txt
 ALLOW_FILE_URI=yes pytest tests/test_security.py

changedetectionio/run_proxy_tests.sh

@@ -82,3 +82,25 @@ done
 
 
 docker kill squid-one squid-two squid-custom
+
+# Test that the UI is returning the correct error message when a proxy is not available
+
+# Requests
+docker run --network changedet-network \
+  test-changedetectionio \
+  bash -c 'cd changedetectionio && pytest tests/proxy_list/test_proxy_noconnect.py'
+
+# Playwright
+docker run --network changedet-network \
+  test-changedetectionio \
+  bash -c 'cd changedetectionio && PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py'
+
+# Puppeteer fast
+docker run --network changedet-network \
+  test-changedetectionio \
+  bash -c 'cd changedetectionio && FAST_PUPPETEER_CHROME_FETCHER=1 PLAYWRIGHT_DRIVER_URL=ws://sockpuppetbrowser:3000 pytest tests/proxy_list/test_proxy_noconnect.py'
+
+# Selenium
+docker run --network changedet-network \
+  test-changedetectionio \
+  bash -c 'cd changedetectionio && WEBDRIVER_URL=http://selenium:4444/wd/hub pytest tests/proxy_list/test_proxy_noconnect.py'

changedetectionio/tests/proxy_list/test_proxy_noconnect.py

@@ -0,0 +1,68 @@
+#!/usr/bin/env python3
+
+from flask import url_for
+from ..util import live_server_setup, wait_for_all_checks
+import os
+from ... import strtobool
+
+
+# Just to be sure the UI outputs the right error message on proxy connection failed
+# docker run -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome:4
+# PLAYWRIGHT_DRIVER_URL=ws://127.0.0.1:3000 pytest tests/proxy_list/test_proxy_noconnect.py
+# FAST_PUPPETEER_CHROME_FETCHER=True PLAYWRIGHT_DRIVER_URL=ws://127.0.0.1:3000 pytest tests/proxy_list/test_proxy_noconnect.py
+# WEBDRIVER_URL=http://127.0.0.1:4444/wd/hub pytest tests/proxy_list/test_proxy_noconnect.py
+
+def test_proxy_noconnect_custom(client, live_server, measure_memory_usage):
+    live_server_setup(live_server)
+
+    # Goto settings, add our custom one
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={
+            "requests-time_between_check-minutes": 180,
+            "application-ignore_whitespace": "y",
+            "application-fetch_backend": 'html_webdriver' if os.getenv('PLAYWRIGHT_DRIVER_URL') or os.getenv("WEBDRIVER_URL") else 'html_requests',
+            "requests-extra_proxies-0-proxy_name": "custom-test-proxy",
+            # test:awesome is set in tests/proxy_list/squid-passwords.txt
+            "requests-extra_proxies-0-proxy_url": "http://127.0.0.1:3128",
+        },
+        follow_redirects=True
+    )
+
+    assert b"Settings updated." in res.data
+
+    test_url = "https://changedetection.io"
+    res = client.post(
+        url_for("ui.ui_views.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
+
+    options = {
+        "url": test_url,
+        "fetch_backend": "html_webdriver" if os.getenv('PLAYWRIGHT_DRIVER_URL') or os.getenv("WEBDRIVER_URL") else "html_requests",
+        "proxy": "ui-0custom-test-proxy",
+    }
+
+    res = client.post(
+        url_for("ui.ui_edit.edit_page", uuid="first", unpause_on_save=1),
+        data=options,
+        follow_redirects=True
+    )
+    assert b"unpaused" in res.data
+    import time
+    wait_for_all_checks(client)
+
+    # Requests default
+    check_string = b'Cannot connect to proxy'
+
+    if os.getenv('PLAYWRIGHT_DRIVER_URL') or strtobool(os.getenv('FAST_PUPPETEER_CHROME_FETCHER', 'False')) or os.getenv("WEBDRIVER_URL"):
+        check_string = b'ERR_PROXY_CONNECTION_FAILED'
+
+
+    res = client.get(url_for("watchlist.index"))
+    #with open("/tmp/debug.html", 'wb') as f:
+    #    f.write(res.data)
+    assert check_string in res.data

changedetectionio/tests/test_history_consistency.py

@@ -10,8 +10,8 @@ from urllib.parse import urlparse, parse_qs
 
 def test_consistent_history(client, live_server, measure_memory_usage):
     live_server_setup(live_server)
-
+    workers = int(os.getenv("FETCH_WORKERS", 10))
-    r = range(1, 30)
+    r = range(1, 10+workers)
 
     for one in r:
         test_url = url_for('test_endpoint', content_type="text/html", content=str(one), _external=True)
@@ -46,9 +46,10 @@ def test_consistent_history(client, live_server, measure_memory_usage):
 
     # assert the right amount of watches was found in the JSON
     assert len(json_obj['watching']) == len(r), "Correct number of watches was found in the JSON"
-
+    i=0
     # each one should have a history.txt containing just one line
     for w in json_obj['watching'].keys():
+        i+=1
         history_txt_index_file = os.path.join(live_server.app.config['DATASTORE'].datastore_path, w, 'history.txt')
         assert os.path.isfile(history_txt_index_file), f"History.txt should exist where I expect it at {history_txt_index_file}"
 
@@ -58,8 +59,8 @@ def test_consistent_history(client, live_server, measure_memory_usage):
             assert len(tmp_history) == 1, "History.txt should contain 1 line"
 
         # Should be two files,. the history.txt , and the snapshot.txt
-        files_in_watch_dir = os.listdir(os.path.join(live_server.app.config['DATASTORE'].datastore_path,
+        files_in_watch_dir = os.listdir(os.path.join(live_server.app.config['DATASTORE'].datastore_path, w))
-                                                     w))
+
         # Find the snapshot one
         for fname in files_in_watch_dir:
             if fname != 'history.txt' and 'html' not in fname:
@@ -75,7 +76,6 @@ def test_consistent_history(client, live_server, measure_memory_usage):
 
         assert len(files_in_watch_dir) == 3, "Should be just three files in the dir, html.br snapshot, history.txt and the extracted text snapshot"
 
-
     json_db_file = os.path.join(live_server.app.config['DATASTORE'].datastore_path, 'url-watches.json')
     with open(json_db_file, 'r') as f:
         assert '"default"' not in f.read(), "'default' probably shouldnt be here, it came from when the 'default' Watch vars were accidently being saved"

changedetectionio/tests/test_request.py

@@ -424,3 +424,27 @@ def test_headers_textfile_in_request(client, live_server, measure_memory_usage):
     # unlink headers.txt on start/stop
     res = client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
+
+def test_headers_validation(client, live_server):
+    #live_server_setup(live_server)
+
+    test_url = url_for('test_headers', _external=True)
+    res = client.post(
+        url_for("imports.import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+
+    res = client.post(
+        url_for("ui.ui_edit.edit_page", uuid="first"),
+        data={
+            "url": test_url,
+            "fetch_backend": 'html_requests',
+            "headers": "User-AGent agent-from-watch\r\nsadfsadfsadfsdaf\r\n:foobar"},
+        follow_redirects=True
+    )
+
+    assert b"Line 1 is missing a ':' separator." in res.data
+    assert b"Line 3 has an empty key." in res.data
+

changedetectionio/tests/util.py

@@ -126,18 +126,51 @@ def extract_UUID_from_client(client):
     uuid = m.group(1)
     return uuid.strip()
 
-def wait_for_all_checks(client):
+
-    # actually this is not entirely true, it can still be 'processing' but not in the queue
+def wait_for_all_checks(client=None):
-    # Loop waiting until done..
+    """
+    Waits until the queue is empty and remains empty for at least `required_empty_duration` seconds,
+    and also ensures no running threads have `current_uuid` set.
+    Retries for up to `max_attempts` times, sleeping `wait_between_attempts` seconds between checks.
+    """
+    from changedetectionio.flask_app import update_q as global_update_q, running_update_threads
+
+    # Configuration
     attempt = 0
-    # because sub-second rechecks are problematic in testing, use lots of delays
+    i=0
-    time.sleep(1)
+    max_attempts = 60
-    while attempt < 60:
+    wait_between_attempts = 2
-        res = client.get(url_for("watchlist.index"))
+    required_empty_duration = 2
-        if not b'Checking now' in res.data:
+
+    logger = logging.getLogger()
+    time.sleep(1.2)
+
+    empty_since = None
+
+    while attempt < max_attempts:
+        q_length = global_update_q.qsize()
+
+        # Check if any threads are still processing
+        time.sleep(1.2)
+        any_threads_busy = any(t.current_uuid for t in running_update_threads)
+
+
+        if q_length == 0 and not any_threads_busy:
+            if empty_since is None:
+                empty_since = time.time()
+                logger.info(f"Queue empty and no active threads at attempt {attempt}, starting empty timer...")
+            elif time.time() - empty_since >= required_empty_duration:
+                logger.info(f"Queue has been empty and threads idle for {required_empty_duration} seconds. Done waiting.")
                 break
-        logging.getLogger().info("Waiting for watch-list to not say 'Checking now'.. {}".format(attempt))
+            else:
-        time.sleep(1)
+                logger.info(f"Still waiting: queue empty and no active threads, but not yet {required_empty_duration} seconds...")
+        else:
+            if q_length != 0:
+                logger.info(f"Queue not empty (size={q_length}), resetting timer.")
+            if any_threads_busy:
+                busy_threads = [t.name for t in running_update_threads if t.current_uuid]
+                logger.info(f"Threads still busy: {busy_threads}, resetting timer.")
+            empty_since = None
         attempt += 1
 
     time.sleep(1)

requirements.txt

@@ -42,7 +42,7 @@ paho-mqtt!=2.0.*
 cryptography~=42.0.8
 
 # Used for CSS filtering
-beautifulsoup4
+beautifulsoup4>=4.0.0
 
 # XPath filtering, lxml is required by bs4 anyway, but put it here to be safe.
 # #2328 - 5.2.0 and 5.2.1 had extra CPU flag CFLAGS set which was not compatible on older hardware
@@ -70,7 +70,7 @@ jq~=1.3; python_version >= "3.8" and sys_platform == "linux"
 
 # playwright is installed at Dockerfile build time because it's not available on all platforms
 
-pyppeteer-ng==2.0.0rc9
+pyppeteer-ng==2.0.0rc10
 
 pyppeteerstealth>=0.0.4