Attempt to handle #text

.github/workflows/test-only.yml

@@ -30,7 +30,10 @@ jobs:
 
           # Selenium+browserless
           docker run --network changedet-network -d --hostname selenium  -p 4444:4444 --rm --shm-size="2g"  selenium/standalone-chrome:4
-          docker run --network changedet-network -d --hostname browserless -e "FUNCTION_BUILT_INS=[\"fs\",\"crypto\"]" -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm  -p 3000:3000  --shm-size="2g"  browserless/chrome:1.60-chrome-stable
+          docker run --network changedet-network -d --name browserless --hostname browserless -e "FUNCTION_BUILT_INS=[\"fs\",\"crypto\"]" -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm  -p 3000:3000  --shm-size="2g"  browserless/chrome:1.60-chrome-stable
+          
+          # For accessing custom browser tests
+          docker run --network changedet-network -d --name browserless-custom-url --hostname browserless-custom-url -e "FUNCTION_BUILT_INS=[\"fs\",\"crypto\"]" -e "DEFAULT_LAUNCH_ARGS=[\"--window-size=1920,1080\"]" --rm --shm-size="2g"  browserless/chrome:1.60-chrome-stable
 
       - name: Build changedetection.io container for testing
         run: |         
@@ -86,6 +89,12 @@ jobs:
           # And again with PLAYWRIGHT_DRIVER_URL=..
           cd ..
 
+      - name: Test custom browser URL
+        run: |
+          cd changedetectionio
+          ./run_custom_browser_url_tests.sh
+          cd ..
+
       - name: Test changedetection.io container starts+runs basically without error
         run: |
           docker run -p 5556:5000 -d test-changedetectionio

README.md

@@ -268,3 +268,7 @@ I offer commercial support, this software is depended on by network security, ae
 [license-shield]: https://img.shields.io/github/license/dgtlmoon/changedetection.io.svg?style=for-the-badge
 [release-link]: https://github.com/dgtlmoon/changedetection.io/releases
 [docker-link]: https://hub.docker.com/r/dgtlmoon/changedetection.io
+
+## Third-party licenses
+
+changedetectionio.html_tools.elementpath_tostring: Copyright (c), 2018-2021, SISSA (Scuola Internazionale Superiore di Studi Avanzati), Licensed under [MIT license](https://github.com/sissaschool/elementpath/blob/master/LICENSE)

changedetectionio/__init__.py

@@ -614,6 +614,8 @@ def changedetection_app(config=None, datastore_o=None):
         # For the form widget tag uuid lookup
         form.tags.datastore = datastore # in _value
 
+        for p in datastore.extra_browsers:
+            form.fetch_backend.choices.append(p)
 
         form.fetch_backend.choices.append(("system", 'System settings default'))
 
@@ -714,7 +716,7 @@ def changedetection_app(config=None, datastore_o=None):
             system_uses_webdriver = datastore.data['settings']['application']['fetch_backend'] == 'html_webdriver'
 
             is_html_webdriver = False
-            if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver':
+            if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
                 is_html_webdriver = True
 
             # Only works reliably with Playwright
@@ -819,6 +821,16 @@ def changedetection_app(config=None, datastore_o=None):
 
         return output
 
+    @app.route("/settings/reset-api-key", methods=['GET'])
+    @login_optionally_required
+    def settings_reset_api_key():
+        import secrets
+        secret = secrets.token_hex(16)
+        datastore.data['settings']['application']['api_access_token'] = secret
+        datastore.needs_write_urgent = True
+        flash("API Key was regenerated.")
+        return redirect(url_for('settings_page')+'#api')
+
     @app.route("/import", methods=['GET', "POST"])
     @login_optionally_required
     def import_page():
@@ -977,7 +989,7 @@ def changedetection_app(config=None, datastore_o=None):
         system_uses_webdriver = datastore.data['settings']['application']['fetch_backend'] == 'html_webdriver'
 
         is_html_webdriver = False
-        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver':
+        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
             is_html_webdriver = True
 
         password_enabled_and_share_is_off = False
@@ -1031,7 +1043,7 @@ def changedetection_app(config=None, datastore_o=None):
 
 
         is_html_webdriver = False
-        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver':
+        if (watch.get('fetch_backend') == 'system' and system_uses_webdriver) or watch.get('fetch_backend') == 'html_webdriver' or watch.get('fetch_backend', '').startswith('extra_browser_'):
             is_html_webdriver = True
 
         # Never requested successfully, but we detected a fetch error

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

@@ -69,11 +69,12 @@ xpath://body/div/span[contains(@class, 'example-class')]",
                                 {% endif %}
                             </ul>
                         </li>
-                        <li>XPath - Limit text to this XPath rule, simply start with a forward-slash,
+                        <li>XPath - Limit text to this XPath rule, simply start with a forward-slash. To specify XPath to be used explicitly or the XPath rule starts with an XPath function: Prefix with <code>xpath:</code>
                             <ul>
-                                <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath://*[contains(@class, 'sametext')]</code>, <a
+                                <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath:count(//*[contains(@class, 'sametext')])</code>, <a
                                 href="http://xpather.com/" target="new">test your XPath here</a></li>
                                 <li>Example: Get all titles from an RSS feed <code>//title/text()</code></li>
+                                <li>To use XPath1.0: Prefix with <code>xpath1:</code></li>
                             </ul>
                             </li>
                     </ul>

changedetectionio/content_fetcher.py

@@ -96,6 +96,7 @@ class Fetcher():
     content = None
     error = None
     fetcher_description = "No description"
+    browser_connection_url = None
     headers = {}
     status_code = None
     webdriver_js_execute_code = None
@@ -251,14 +252,16 @@ class base_html_playwright(Fetcher):
 
     proxy = None
 
-    def __init__(self, proxy_override=None):
+    def __init__(self, proxy_override=None, browser_connection_url=None):
         super().__init__()
-        # .strip('"') is going to save someone a lot of time when they accidently wrap the env value
+
         self.browser_type = os.getenv("PLAYWRIGHT_BROWSER_TYPE", 'chromium').strip('"')
-        self.command_executor = os.getenv(
+
-            "PLAYWRIGHT_DRIVER_URL",
+        # .strip('"') is going to save someone a lot of time when they accidently wrap the env value
-            'ws://playwright-chrome:3000'
+        if not browser_connection_url:
-        ).strip('"')
+            self.browser_connection_url = os.getenv("PLAYWRIGHT_DRIVER_URL", 'ws://playwright-chrome:3000').strip('"')
+        else:
+            self.browser_connection_url = browser_connection_url
 
         # If any proxy settings are enabled, then we should setup the proxy object
         proxy_args = {}
@@ -444,7 +447,7 @@ class base_html_playwright(Fetcher):
             # Seemed to cause a connection Exception even tho I can see it connect
             # self.browser = browser_type.connect(self.command_executor, timeout=timeout*1000)
             # 60,000 connection timeout only
-            browser = browser_type.connect_over_cdp(self.command_executor, timeout=60000)
+            browser = browser_type.connect_over_cdp(self.browser_connection_url, timeout=60000)
 
             # SOCKS5 with authentication is not supported (yet)
             # https://github.com/microsoft/playwright/issues/10567
@@ -504,7 +507,11 @@ class base_html_playwright(Fetcher):
             self.status_code = response.status
 
             if self.status_code != 200 and not ignore_status_codes:
-                raise Non200ErrorCodeReceived(url=url, status_code=self.status_code)
+
+                screenshot=self.page.screenshot(type='jpeg', full_page=True,
+                                     quality=int(os.getenv("PLAYWRIGHT_SCREENSHOT_QUALITY", 72)))
+
+                raise Non200ErrorCodeReceived(url=url, status_code=self.status_code, screenshot=screenshot)
 
             if len(self.page.content().strip()) == 0:
                 context.close()
@@ -555,8 +562,6 @@ class base_html_webdriver(Fetcher):
     else:
         fetcher_description = "WebDriver Chrome/Javascript"
 
-    command_executor = ''
-
     # 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',
@@ -564,12 +569,15 @@ class base_html_webdriver(Fetcher):
                                         'socksProxy', 'socksVersion', 'socksUsername', 'socksPassword']
     proxy = None
 
-    def __init__(self, proxy_override=None):
+    def __init__(self, proxy_override=None, browser_connection_url=None):
         super().__init__()
         from selenium.webdriver.common.proxy import Proxy as SeleniumProxy
 
         # .strip('"') is going to save someone a lot of time when they accidently wrap the env value
-        self.command_executor = os.getenv("WEBDRIVER_URL", 'http://browser-chrome:4444/wd/hub').strip('"')
+        if not browser_connection_url:
+            self.browser_connection_url = os.getenv("WEBDRIVER_URL", 'http://browser-chrome:4444/wd/hub').strip('"')
+        else:
+            self.browser_connection_url = browser_connection_url
 
         # If any proxy settings are enabled, then we should setup the proxy object
         proxy_args = {}
@@ -611,7 +619,7 @@ class base_html_webdriver(Fetcher):
             options.proxy = self.proxy
 
         self.driver = webdriver.Remote(
-            command_executor=self.command_executor,
+            command_executor=self.browser_connection_url,
             options=options)
 
         try:
@@ -666,9 +674,10 @@ class base_html_webdriver(Fetcher):
 class html_requests(Fetcher):
     fetcher_description = "Basic fast Plaintext/HTTP Client"
 
-    def __init__(self, proxy_override=None):
+    def __init__(self, proxy_override=None, browser_connection_url=None):
         super().__init__()
         self.proxy_override = proxy_override
+        # browser_connection_url is none because its always 'launched locally'
 
     def run(self,
             url,

changedetectionio/forms.py

@@ -168,7 +168,9 @@ class ValidateContentFetcherIsReady(object):
     def __call__(self, form, field):
         import urllib3.exceptions
         from changedetectionio import content_fetcher
+        return
 
+# AttributeError: module 'changedetectionio.content_fetcher' has no attribute 'extra_browser_unlocked<>ASDF213r123r'
         # Better would be a radiohandler that keeps a reference to each class
         if field.data is not None and field.data != 'system':
             klass = getattr(content_fetcher, field.data)
@@ -326,11 +328,30 @@ class ValidateCSSJSONXPATHInput(object):
                 return
 
             # Does it look like XPath?
-            if line.strip()[0] == '/':
+            if line.strip()[0] == '/' or line.strip().startswith('xpath:'):
+                if not self.allow_xpath:
+                    raise ValidationError("XPath not permitted in this field!")
+                from lxml import etree, html
+                import elementpath
+                # xpath 2.0-3.1
+                from elementpath.xpath3 import XPath3Parser
+                tree = html.fromstring("<html></html>")
+                line = line.replace('xpath:', '')
+
+                try:
+                    elementpath.select(tree, line.strip(), parser=XPath3Parser)
+                except elementpath.ElementPathError as e:
+                    message = field.gettext('\'%s\' is not a valid XPath expression. (%s)')
+                    raise ValidationError(message % (line, str(e)))
+                except:
+                    raise ValidationError("A system-error occurred when validating your XPath expression")
+
+            if line.strip().startswith('xpath1:'):
                 if not self.allow_xpath:
                     raise ValidationError("XPath not permitted in this field!")
                 from lxml import etree, html
                 tree = html.fromstring("<html></html>")
+                line = re.sub(r'^xpath1:', '', line)
 
                 try:
                     tree.xpath(line.strip())
@@ -496,6 +517,12 @@ class SingleExtraProxy(Form):
     proxy_url = StringField('Proxy URL', [validators.Optional()], render_kw={"placeholder": "socks5:// or regular proxy http://user:pass@...:3128", "size":50})
     # @todo do the validation here instead
 
+class SingleExtraBrowser(Form):
+    browser_name = StringField('Name', [validators.Optional()], render_kw={"placeholder": "Name"})
+    browser_connection_url = StringField('Browser connection URL', [validators.Optional()], render_kw={"placeholder": "wss://brightdata... wss://oxylabs etc", "size":50})
+    # @todo do the validation here instead
+
+
 # datastore.data['settings']['requests']..
 class globalSettingsRequestForm(Form):
     time_between_check = FormField(TimeBetweenCheckForm)
@@ -504,6 +531,7 @@ class globalSettingsRequestForm(Form):
                                   render_kw={"style": "width: 5em;"},
                                   validators=[validators.NumberRange(min=0, message="Should contain zero or more seconds")])
     extra_proxies = FieldList(FormField(SingleExtraProxy), min_entries=5)
+    extra_browsers = FieldList(FormField(SingleExtraBrowser), min_entries=5)
 
     def validate_extra_proxies(self, extra_validators=None):
         for e in self.data['extra_proxies']:

changedetectionio/html_tools.py

@@ -69,10 +69,89 @@ def element_removal(selectors: List[str], html_content):
     selector = ",".join(selectors)
     return subtractive_css_selector(selector, html_content)
 
+def elementpath_tostring(obj):
+    """
+    change elementpath.select results to string type
+    # The MIT License (MIT), Copyright (c), 2018-2021, SISSA (Scuola Internazionale Superiore di Studi Avanzati)
+    # https://github.com/sissaschool/elementpath/blob/dfcc2fd3d6011b16e02bf30459a7924f547b47d0/elementpath/xpath_tokens.py#L1038
+    """
+
+    import elementpath
+    from decimal import Decimal
+    import math
+
+    if obj is None:
+        return ''
+    # https://elementpath.readthedocs.io/en/latest/xpath_api.html#elementpath.select
+    elif isinstance(obj, elementpath.XPathNode):
+        return obj.string_value
+    elif isinstance(obj, bool):
+        return 'true' if obj else 'false'
+    elif isinstance(obj, Decimal):
+        value = format(obj, 'f')
+        if '.' in value:
+            return value.rstrip('0').rstrip('.')
+        return value
+
+    elif isinstance(obj, float):
+        if math.isnan(obj):
+            return 'NaN'
+        elif math.isinf(obj):
+            return str(obj).upper()
+
+        value = str(obj)
+        if '.' in value:
+            value = value.rstrip('0').rstrip('.')
+        if '+' in value:
+            value = value.replace('+', '')
+        if 'e' in value:
+            return value.upper()
+        return value
+
+    return str(obj)
 
 # Return str Utf-8 of matched rules
 def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_rss=False):
     from lxml import etree, html
+    import elementpath
+    # xpath 2.0-3.1
+    from elementpath.xpath3 import XPath3Parser
+
+    parser = etree.HTMLParser()
+    if is_rss:
+        # So that we can keep CDATA for cdata_in_document_to_text() to process
+        parser = etree.XMLParser(strip_cdata=False)
+
+    tree = html.fromstring(bytes(html_content, encoding='utf-8'), parser=parser)
+    html_block = ""
+
+    r = elementpath.select(tree, xpath_filter.strip(), namespaces={'re': 'http://exslt.org/regular-expressions'}, parser=XPath3Parser)
+    #@note: //title/text() wont work where <title>CDATA..
+
+    if type(r) != list:
+        r = [r]
+
+    for element in r:
+        # When there's more than 1 match, then add the suffix to separate each line
+        # And where the matched result doesn't include something that will cause Inscriptis to add a newline
+        # (This way each 'match' reliably has a new-line in the diff)
+        # Divs are converted to 4 whitespaces by inscriptis
+        if append_pretty_line_formatting and len(html_block) and (not hasattr( element, 'tag' ) or not element.tag in (['br', 'hr', 'div', 'p'])):
+            html_block += TEXT_FILTER_LIST_LINE_SUFFIX
+
+        if type(element) == str:
+            html_block += element
+        elif issubclass(type(element), etree._Element) or issubclass(type(element), etree._ElementTree):
+            html_block += etree.tostring(element, pretty_print=True).decode('utf-8')
+        else:
+            html_block += elementpath_tostring(element)
+
+    return html_block
+
+# Return str Utf-8 of matched rules
+# 'xpath1:'
+def xpath1_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_rss=False):
+    from lxml import etree, html
 
     parser = None
     if is_rss:

changedetectionio/model/App.py

@@ -16,6 +16,7 @@ class model(dict):
                 },
                 'requests': {
                     'extra_proxies': [], # Configurable extra proxies via the UI
+                    'extra_browsers': [],  # Configurable extra proxies via the UI
                     'jitter_seconds': 0,
                     'proxy': None, # Preferred proxy connection
                     'time_between_check': {'weeks': None, 'days': None, 'hours': 3, 'minutes': None, 'seconds': None},

changedetectionio/processors/__init__.py

@@ -8,11 +8,12 @@ from distutils.util import strtobool
 
 class difference_detection_processor():
 
+    browser_steps = None
     datastore = None
     fetcher = None
     screenshot = None
+    watch = None
     xpath_data = None
-    browser_steps = None
 
     def __init__(self, *args, datastore, watch_uuid, **kwargs):
         super().__init__(*args, **kwargs)
@@ -40,6 +41,18 @@ class difference_detection_processor():
         if not prefer_fetch_backend or prefer_fetch_backend == 'system':
             prefer_fetch_backend = self.datastore.data['settings']['application'].get('fetch_backend')
 
+        # In the case that the preferred fetcher was a browser config with custom connection URL..
+        # @todo - on save watch, if its extra_browser_ then it should be obvious it will use playwright (like if its requests now..)
+        browser_connection_url = None
+        if prefer_fetch_backend.startswith('extra_browser_'):
+            (t, key) = prefer_fetch_backend.split('extra_browser_')
+            connection = list(
+                filter(lambda s: (s['browser_name'] == key), self.datastore.data['settings']['requests'].get('extra_browsers', [])))
+            if connection:
+                prefer_fetch_backend = 'base_html_playwright'
+                browser_connection_url = connection[0].get('browser_connection_url')
+
+
         # Grab the right kind of 'fetcher', (playwright, requests, etc)
         if hasattr(content_fetcher, prefer_fetch_backend):
             fetcher_obj = getattr(content_fetcher, prefer_fetch_backend)
@@ -54,8 +67,9 @@ class difference_detection_processor():
             print(f"Using proxy Key: {preferred_proxy_id} as Proxy URL {proxy_url}")
 
         # 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)
         self.fetcher = fetcher_obj(proxy_override=proxy_url,
-                                   #browser_url_extra/configurable browser url=...
+                                   browser_connection_url=browser_connection_url
                                    )
 
         if self.watch.has_browser_steps:

changedetectionio/processors/text_json_diff.py

@@ -173,6 +173,11 @@ class perform_site_check(difference_detection_processor):
                                                                     html_content=self.fetcher.content,
                                                                     append_pretty_line_formatting=not watch.is_source_type_url,
                                                                     is_rss=is_rss)
+                        elif filter_rule.startswith('xpath1:'):
+                            html_content += html_tools.xpath1_filter(xpath_filter=filter_rule.replace('xpath1:', ''),
+                                                                    html_content=self.fetcher.content,
+                                                                    append_pretty_line_formatting=not watch.is_source_type_url,
+                                                                    is_rss=is_rss)
                         else:
                             # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
                             html_content += html_tools.include_filters(include_filters=filter_rule,

changedetectionio/res/xpath_element_scraper.js

@@ -170,9 +170,12 @@ if (include_filters.length) {
 
         try {
             // is it xpath?
-            if (f.startsWith('/') || f.startsWith('xpath:')) {
+            if (f.startsWith('/') || f.startsWith('xpath')) {
-                q = document.evaluate(f.replace('xpath:', ''), document, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null).singleNodeValue;
+                var qry_f = f.replace(/xpath(:|\d:)/, '')
+                console.log("[xpath] Scanning for included filter " + qry_f)
+                q = document.evaluate(qry_f, document, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null).singleNodeValue;
             } else {
+                console.log("[css] Scanning for included filter " + f)
                 q = document.querySelector(f);
             }
         } catch (e) {
@@ -182,8 +185,18 @@ if (include_filters.length) {
         }
 
         if (q) {
+            // Try to resolve //something/text() back to its /something so we can atleast get the bounding box
+            try {
+                if (typeof q.nodeName == 'string' && q.nodeName === '#text') {
+                    q = q.parentElement
+                }
+            } catch (e) {
+                console.log(e)
+                console.log("xpath_element_scraper: #text resolver")
+            }
+
             // #1231 - IN the case XPath attribute filter is applied, we will have to traverse up and find the element.
-            if (q.hasOwnProperty('getBoundingClientRect')) {
+            if (typeof q.getBoundingClientRect == 'function') {
                 bbox = q.getBoundingClientRect();
                 console.log("xpath_element_scraper: Got filter element, scroll from top was " + scroll_y)
             } else {
@@ -192,7 +205,8 @@ if (include_filters.length) {
                     bbox = q.ownerElement.getBoundingClientRect();
                     console.log("xpath_element_scraper: Got filter by ownerElement element, scroll from top was " + scroll_y)
                 } catch (e) {
-                    console.log("xpath_element_scraper: error looking up ownerElement")
+                    console.log(e)
+                    console.log("xpath_element_scraper: error looking up q.ownerElement")
                 }
             }
         }

changedetectionio/run_custom_browser_url_tests.sh

@@ -0,0 +1,44 @@
+#!/bin/bash
+
+# run some tests and look if the 'custom-browser-search-string=1' connect string appeared in the correct containers
+
+# enable debug
+set -x
+
+# A extra browser is configured, but we never chose to use it, so it should NOT show in the logs
+docker run --rm -e "PLAYWRIGHT_DRIVER_URL=ws://browserless:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/custom_browser_url/test_custom_browser_url.py::test_request_not_via_custom_browser_url'
+docker logs browserless-custom-url &>log.txt
+grep 'custom-browser-search-string=1' log.txt
+if [ $? -ne 1 ]
+then
+  echo "Saw a request in 'browserless-custom-url' container with 'custom-browser-search-string=1' when I should not"
+  exit 1
+fi
+
+docker logs browserless &>log.txt
+grep 'custom-browser-search-string=1' log.txt
+if [ $? -ne 1 ]
+then
+  echo "Saw a request in 'browser' container with 'custom-browser-search-string=1' when I should not"
+  exit 1
+fi
+
+# Special connect string should appear in the custom-url container, but not in the 'default' one
+docker run --rm -e "PLAYWRIGHT_DRIVER_URL=ws://browserless:3000" --network changedet-network test-changedetectionio  bash -c 'cd changedetectionio;pytest tests/custom_browser_url/test_custom_browser_url.py::test_request_via_custom_browser_url'
+docker logs browserless-custom-url &>log.txt
+grep 'custom-browser-search-string=1' log.txt
+if [ $? -ne 0 ]
+then
+  echo "Did not see request in 'browserless-custom-url' container with 'custom-browser-search-string=1' when I should"
+  exit 1
+fi
+
+docker logs browserless &>log.txt
+grep 'custom-browser-search-string=1' log.txt
+if [ $? -ne 1 ]
+then
+  echo "Saw a request in 'browser' container with 'custom-browser-search-string=1' when I should not"
+  exit 1
+fi
+
+

changedetectionio/static/js/visual-selector.js

@@ -149,7 +149,7 @@ $(document).ready(function () {
             // @todo In the future paint all that match
             for (const c of current_default_xpath) {
                 for (var i = selector_data['size_pos'].length; i !== 0; i--) {
-                    if (selector_data['size_pos'][i - 1].xpath === c) {
+                    if (selector_data['size_pos'][i - 1].xpath.trim() === c.trim()) {
                         console.log("highlighting " + c);
                         current_selected_i = i - 1;
                         highlight_current_selected_i();

changedetectionio/static/styles/scss/parts/_extra_browsers.scss

@@ -0,0 +1,24 @@
+ul#requests-extra_browsers {
+  list-style: none;
+  /* tidy up the table to look more "inline" */
+  li {
+    > label {
+      display: none;
+    }
+
+  }
+
+  /* each proxy entry is a `table` */
+  table {
+    tr {
+      display: inline;
+    }
+  }
+}
+
+#extra-browsers-setting {
+  border: 1px solid var(--color-grey-800);
+  border-radius: 4px;
+  margin: 1em;
+   padding: 1em;
+}

changedetectionio/static/styles/scss/parts/_extra_proxies.scss

@@ -60,3 +60,10 @@ body.proxy-check-active {
 
   padding-bottom: 1em;
 }
+
+#extra-proxies-setting {
+  border: 1px solid var(--color-grey-800);
+  border-radius: 4px;
+    margin: 1em;
+   padding: 1em;
+}

changedetectionio/static/styles/scss/styles.scss

@@ -5,6 +5,7 @@
 @import "parts/_arrows";
 @import "parts/_browser-steps";
 @import "parts/_extra_proxies";
+@import "parts/_extra_browsers";
 @import "parts/_pagination";
 @import "parts/_spinners";
 @import "parts/_variables";

changedetectionio/static/styles/styles.css

@@ -128,6 +128,27 @@ body.proxy-check-active #request .proxy-timing {
     border-radius: 4px;
     padding: 1em; }
 
+#extra-proxies-setting {
+  border: 1px solid var(--color-grey-800);
+  border-radius: 4px;
+  margin: 1em;
+  padding: 1em; }
+
+ul#requests-extra_browsers {
+  list-style: none;
+  /* tidy up the table to look more "inline" */
+  /* each proxy entry is a `table` */ }
+  ul#requests-extra_browsers li > label {
+    display: none; }
+  ul#requests-extra_browsers table tr {
+    display: inline; }
+
+#extra-browsers-setting {
+  border: 1px solid var(--color-grey-800);
+  border-radius: 4px;
+  margin: 1em;
+  padding: 1em; }
+
 .pagination-page-info {
   color: #fff;
   font-size: 0.85rem;

changedetectionio/store.py

@@ -633,6 +633,18 @@ class ChangeDetectionStore:
 
         return {}
 
+    @property
+    def extra_browsers(self):
+        res = []
+        p = list(filter(
+            lambda s: (s.get('browser_name') and s.get('browser_connection_url')),
+            self.__data['settings']['requests'].get('extra_browsers', [])))
+        if p:
+            for i in p:
+                res.append(("extra_browser_"+i['browser_name'], i['browser_name']))
+
+        return res
+
     def tag_exists_by_name(self, tag_name):
         return any(v.get('title', '').lower() == tag_name.lower() for k, v in self.__data['settings']['application']['tags'].items())
 
@@ -835,4 +847,14 @@ class ChangeDetectionStore:
             if not watch.get('date_created'):
                 self.data['watching'][uuid]['date_created'] = i
             i+=1
-        return
+        return
+
+    # #1774 - protect xpath1 against migration
+    def update_14(self):
+        for awatch in self.__data["watching"]:
+            if self.__data["watching"][awatch]['include_filters']:
+                for num, selector in enumerate(self.__data["watching"][awatch]['include_filters']):
+                    if selector.startswith('/'):
+                        self.__data["watching"][awatch]['include_filters'][num] = 'xpath1:' + selector
+                    if selector.startswith('xpath:'):
+                        self.__data["watching"][awatch]['include_filters'][num] = selector.replace('xpath:', 'xpath1:', 1)

changedetectionio/templates/edit.html

@@ -290,11 +290,12 @@ xpath://body/div/span[contains(@class, 'example-class')]",
                                 {% endif %}
                             </ul>
                         </li>
-                        <li>XPath - Limit text to this XPath rule, simply start with a forward-slash,
+                        <li>XPath - Limit text to this XPath rule, simply start with a forward-slash. To specify XPath to be used explicitly or the XPath rule starts with an XPath function: Prefix with <code>xpath:</code>
                             <ul>
-                                <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath://*[contains(@class, 'sametext')]</code>, <a
+                                <li>Example:  <code>//*[contains(@class, 'sametext')]</code> or <code>xpath:count(//*[contains(@class, 'sametext')])</code>, <a
                                 href="http://xpather.com/" target="new">test your XPath here</a></li>
                                 <li>Example: Get all titles from an RSS feed <code>//title/text()</code></li>
+                                <li>To use XPath1.0: Prefix with <code>xpath1:</code></li>
                             </ul>
                             </li>
                     </ul>

changedetectionio/templates/settings.html

@@ -178,6 +178,9 @@ nav
                         <span style="display:none;" id="api-key-copy" >copy</span>
                     </div>
                 </div>
+                <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>
             <div class="tab-pane-inner" id="proxies">
                 <div id="recommended-proxy">
@@ -227,11 +230,15 @@ nav
                 </p>
                <p><strong>Tip</strong>: "Residential" and "Mobile" proxy type can be more successfull than "Data Center" for blocked websites.
 
-                <div class="pure-control-group">
+                <div class="pure-control-group" id="extra-proxies-setting">
                 {{ render_field(form.requests.form.extra_proxies) }}
                 <span class="pure-form-message-inline">"Name" will be used for selecting the proxy in the Watch Edit settings</span><br>
                 <span class="pure-form-message-inline">SOCKS5 proxies with authentication are only supported with 'plain requests' fetcher, for other fetchers you should whitelist the IP access instead</span>
                 </div>
+                <div class="pure-control-group" id="extra-browsers-setting">
+                    <span class="pure-form-message-inline"><i>Extra Browsers</i> allow changedetection.io to communicate with a different web-browser.</span><br>
+                  {{ render_field(form.requests.form.extra_browsers) }}
+                </div>
             </div>
             <div id="actions">
                 <div class="pure-control-group">

changedetectionio/templates/watch-overview.html

@@ -104,8 +104,9 @@
 
                     {% if watch.get_fetch_backend == "html_webdriver"
                          or (  watch.get_fetch_backend == "system" and system_default_fetcher == 'html_webdriver'  )
+                         or "extra_browser_" in watch.get_fetch_backend
                     %}
-                    <img class="status-icon" src="{{url_for('static_content', group='images', filename='Google-Chrome-icon.png')}}" title="Using a chrome browser" >
+                    <img class="status-icon" src="{{url_for('static_content', group='images', filename='Google-Chrome-icon.png')}}" title="Using a Chrome browser" >
                     {% endif %}
 
                     {%if watch.is_pdf  %}<img class="status-icon" src="{{url_for('static_content', group='images', filename='pdf-icon.svg')}}" title="Converting PDF to text" >{% endif %}

changedetectionio/tests/custom_browser_url/__init__.py

@@ -0,0 +1 @@
+# placeholder

changedetectionio/tests/custom_browser_url/test_custom_browser_url.py

@@ -0,0 +1,89 @@
+# !/usr/bin/python3
+import os
+
+from flask import url_for
+from ..util import live_server_setup, wait_for_all_checks
+
+def do_test(client, live_server, make_test_use_extra_browser=False):
+
+    # Grep for this string in the logs?
+    test_url = f"https://changedetection.io/ci-test.html"
+    custom_browser_name = 'custom browser URL'
+
+    # needs to be set and something like 'ws://127.0.0.1:3000?stealth=1&--disable-web-security=true'
+    assert os.getenv('PLAYWRIGHT_DRIVER_URL'), "Needs PLAYWRIGHT_DRIVER_URL set for this test"
+
+    #####################
+    res = client.post(
+        url_for("settings_page"),
+        data={"application-empty_pages_are_a_change": "",
+              "requests-time_between_check-minutes": 180,
+              'application-fetch_backend': "html_webdriver",
+              # browserless-custom-url is setup in  .github/workflows/test-only.yml
+              # the test script run_custom_browser_url_test.sh will look for 'custom-browser-search-string' in the container logs
+              'requests-extra_browsers-0-browser_connection_url': 'ws://browserless-custom-url:3000?stealth=1&--disable-web-security=true&custom-browser-search-string=1',
+              'requests-extra_browsers-0-browser_name': custom_browser_name
+              },
+        follow_redirects=True
+    )
+
+    assert b"Settings updated." in res.data
+
+    # Add our URL to the import page
+    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)
+
+    if make_test_use_extra_browser:
+
+        # So the name should appear in the edit page under "Request" > "Fetch Method"
+        res = client.get(
+            url_for("edit_page", uuid="first"),
+            follow_redirects=True
+        )
+        assert b'custom browser URL' in res.data
+
+        res = client.post(
+            url_for("edit_page", uuid="first"),
+            data={
+                  "url": test_url,
+                  "tags": "",
+                  "headers": "",
+                  'fetch_backend': f"extra_browser_{custom_browser_name}",
+                  'webdriver_js_execute_code': ''
+            },
+            follow_redirects=True
+        )
+
+        assert b"Updated watch." in res.data
+        wait_for_all_checks(client)
+
+    # Force recheck
+    res = client.get(url_for("form_watch_checknow"), follow_redirects=True)
+    assert b'1 watches queued for rechecking.' in res.data
+
+    wait_for_all_checks(client)
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b'cool it works' in res.data
+
+
+# Requires playwright to be installed
+def test_request_via_custom_browser_url(client, live_server):
+    live_server_setup(live_server)
+    # We do this so we can grep the logs of the custom container and see if the request actually went through that container
+    do_test(client, live_server, make_test_use_extra_browser=True)
+
+
+def test_request_not_via_custom_browser_url(client, live_server):
+    live_server_setup(live_server)
+    # We do this so we can grep the logs of the custom container and see if the request actually went through that container
+    do_test(client, live_server, make_test_use_extra_browser=False)

changedetectionio/tests/test_extract_regex.py

@@ -227,9 +227,6 @@ def test_regex_error_handling(client, live_server):
         follow_redirects=True
     )
 
-    with open('/tmp/fuck.html', 'wb') as f:
-        f.write(res.data)
-
     assert b'is not a valid regular expression.' in res.data
 
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)

changedetectionio/tests/test_xpath_selector.py

@@ -6,9 +6,11 @@ from .util import live_server_setup, wait_for_all_checks
 
 from ..html_tools import *
 
+
 def test_setup(live_server):
     live_server_setup(live_server)
 
+
 def set_original_response():
     test_return_data = """<html>
        <body>
@@ -26,6 +28,7 @@ def set_original_response():
         f.write(test_return_data)
     return None
 
+
 def set_modified_response():
     test_return_data = """<html>
        <body>
@@ -44,11 +47,12 @@ def set_modified_response():
 
     return None
 
+
 # Handle utf-8 charset replies https://github.com/dgtlmoon/changedetection.io/pull/613
 def test_check_xpath_filter_utf8(client, live_server):
-    filter='//item/*[self::description]'
+    filter = '//item/*[self::description]'
 
-    d='''<?xml version="1.0" encoding="UTF-8"?>
+    d = '''<?xml version="1.0" encoding="UTF-8"?>
 <rss xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
 	<channel>
 		<title>rpilocator.com</title>
@@ -102,9 +106,9 @@ def test_check_xpath_filter_utf8(client, live_server):
 
 # Handle utf-8 charset replies https://github.com/dgtlmoon/changedetection.io/pull/613
 def test_check_xpath_text_function_utf8(client, live_server):
-    filter='//item/title/text()'
+    filter = '//item/title/text()'
 
-    d='''<?xml version="1.0" encoding="UTF-8"?>
+    d = '''<?xml version="1.0" encoding="UTF-8"?>
 <rss xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
 	<channel>
 		<title>rpilocator.com</title>
@@ -163,15 +167,12 @@ def test_check_xpath_text_function_utf8(client, live_server):
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
 
-def test_check_markup_xpath_filter_restriction(client, live_server):
 
+def test_check_markup_xpath_filter_restriction(client, live_server):
     xpath_filter = "//*[contains(@class, 'sametext')]"
 
     set_original_response()
 
-    # Give the endpoint time to spin up
-    time.sleep(1)
-
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
     res = client.post(
@@ -214,7 +215,6 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
 
 
 def test_xpath_validation(client, live_server):
-
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
     res = client.post(
@@ -235,6 +235,48 @@ def test_xpath_validation(client, live_server):
     assert b'Deleted' in res.data
 
 
+def test_xpath23_prefix_validation(client, live_server):
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": "xpath:/something horrible", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+    assert b"is not a valid XPath expression" in res.data
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
+
+
+def test_xpath1_validation(client, live_server):
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": "xpath1:/something horrible", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
+        follow_redirects=True
+    )
+    assert b"is not a valid XPath expression" in res.data
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
+
+
 # actually only really used by the distll.io importer, but could be handy too
 def test_check_with_prefix_include_filters(client, live_server):
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
@@ -254,7 +296,8 @@ def test_check_with_prefix_include_filters(client, live_server):
 
     res = client.post(
         url_for("edit_page", uuid="first"),
-        data={"include_filters":  "xpath://*[contains(@class, 'sametext')]", "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
+        data={"include_filters": "xpath://*[contains(@class, 'sametext')]", "url": test_url, "tags": "", "headers": "",
+              'fetch_backend': "html_requests"},
         follow_redirects=True
     )
 
@@ -266,13 +309,15 @@ def test_check_with_prefix_include_filters(client, live_server):
         follow_redirects=True
     )
 
-    assert b"Some text thats the same" in res.data #in selector
+    assert b"Some text thats the same" in res.data  # in selector
-    assert b"Some text that will change" not in res.data #not in selector
+    assert b"Some text that will change" not in res.data  # not in selector
 
     client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+
 def test_various_rules(client, live_server):
     # Just check these don't error
-    #live_server_setup(live_server)
+    # live_server_setup(live_server)
     with open("test-datastore/endpoint-content.txt", "w") as f:
         f.write("""<html>
        <body>
@@ -285,10 +330,11 @@ def test_various_rules(client, live_server):
      <a href=''>some linky </a>
      <a href=''>another some linky </a>
      <!-- related to https://github.com/dgtlmoon/changedetection.io/pull/1774 -->
-     <input   type="email"   id="email" />
+     <input   type="email"   id="email" />     
      </body>
      </html>
     """)
+
     test_url = url_for('test_endpoint', _external=True)
     res = client.post(
         url_for("import_page"),
@@ -298,7 +344,6 @@ def test_various_rules(client, live_server):
     assert b"1 Imported" in res.data
     wait_for_all_checks(client)
 
-
     for r in ['//div', '//a', 'xpath://div', 'xpath://a']:
         res = client.post(
             url_for("edit_page", uuid="first"),
@@ -313,3 +358,153 @@ def test_various_rules(client, live_server):
         assert b"Updated watch." in res.data
         res = client.get(url_for("index"))
         assert b'fetch-error' not in res.data, f"Should not see errors after '{r} filter"
+
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+    assert b'Deleted' in res.data
+
+
+def test_xpath_20(client, live_server):
+    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)
+
+    set_original_response()
+
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": "//*[contains(@class, 'sametext')]|//*[contains(@class, 'changetext')]",
+              "url": test_url,
+              "tags": "",
+              "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"Some text thats the same" in res.data  # in selector
+    assert b"Some text that will change" in res.data  # in selector
+
+    client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+
+def test_xpath_20_function_count(client, live_server):
+    set_original_response()
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": "xpath:count(//div) * 123456789987654321",
+              "url": test_url,
+              "tags": "",
+              "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"246913579975308642" in res.data  # in selector
+
+    client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+
+def test_xpath_20_function_count2(client, live_server):
+    set_original_response()
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={"include_filters": "/html/body/count(div) * 123456789987654321",
+              "url": test_url,
+              "tags": "",
+              "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"246913579975308642" in res.data  # in selector
+
+    client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+
+def test_xpath_20_function_string_join_matches(client, live_server):
+    set_original_response()
+
+    # Add our URL to the import page
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+    wait_for_all_checks(client)
+
+    res = client.post(
+        url_for("edit_page", uuid="first"),
+        data={
+            "include_filters": "xpath:string-join(//*[contains(@class, 'sametext')]|//*[matches(@class, 'changetext')], 'specialconjunction')",
+            "url": test_url,
+            "tags": "",
+            "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"Some text thats the samespecialconjunctionSome text that will change" in res.data  # in selector
+
+    client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+

changedetectionio/tests/test_xpath_selector_unit.py

@@ -0,0 +1,203 @@
+import sys
+import os
+import pytest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import html_tools
+
+# test generation guide.
+# 1. Do not include encoding in the xml declaration if the test object is a str type.
+# 2. Always paraphrase test.
+
+hotels = """
+<hotel>
+  <branch location="California">
+    <staff>
+      <given_name>Christopher</given_name>
+      <surname>Anderson</surname>
+      <age>25</age>
+    </staff>
+    <staff>
+      <given_name>Christopher</given_name>
+      <surname>Carter</surname>
+      <age>30</age>
+    </staff>
+  </branch>
+  <branch location="Las Vegas">
+    <staff>
+      <given_name>Lisa</given_name>
+      <surname>Walker</surname>
+      <age>60</age>
+    </staff>
+    <staff>
+      <given_name>Jessica</given_name>
+      <surname>Walker</surname>
+      <age>32</age>
+    </staff>
+    <staff>
+      <given_name>Jennifer</given_name>
+      <surname>Roberts</surname>
+      <age>50</age>
+    </staff>
+  </branch>
+</hotel>"""
+
+@pytest.mark.parametrize("html_content", [hotels])
+@pytest.mark.parametrize("xpath, answer", [('(//staff/given_name, //staff/age)', '25'),
+                          ("xs:date('2023-10-10')", '2023-10-10'),
+                          ("if (/hotel/branch[@location = 'California']/staff[1]/age = 25) then 'is 25' else 'is not 25'", 'is 25'),
+                          ("if (//hotel/branch[@location = 'California']/staff[1]/age = 25) then 'is 25' else 'is not 25'", 'is 25'),
+                          ("if (count(/hotel/branch/staff) = 5) then true() else false()", 'true'),
+                          ("if (count(//hotel/branch/staff) = 5) then true() else false()", 'true'),
+                          ("for $i in /hotel/branch/staff return if ($i/age >= 40) then upper-case($i/surname) else lower-case($i/surname)", 'anderson'),
+                          ("given_name  =  'Christopher' and age  =  40", 'false'),
+                          ("//given_name  =  'Christopher' and //age  =  40", 'false'),
+                          #("(staff/given_name, staff/age)", 'Lisa'),
+                          ("(//staff/given_name, //staff/age)", 'Lisa'),
+                          #("hotel/branch[@location = 'California']/staff/age union hotel/branch[@location = 'Las Vegas']/staff/age", ''),
+                          ("(//hotel/branch[@location = 'California']/staff/age union //hotel/branch[@location = 'Las Vegas']/staff/age)", '60'),
+                          ("(200 to 210)", "205"),
+                          ("(//hotel/branch[@location = 'California']/staff/age union //hotel/branch[@location = 'Las Vegas']/staff/age)", "50"),
+                          ("(1, 9, 9, 5)", "5"),
+                          ("(3, (), (14, 15), 92, 653)", "653"),
+                          ("for $i in /hotel/branch/staff return $i/given_name", "Christopher"),
+                          ("for $i in //hotel/branch/staff return $i/given_name", "Christopher"),
+                          ("distinct-values(for $i in /hotel/branch/staff return $i/given_name)", "Jessica"),
+                          ("distinct-values(for $i in //hotel/branch/staff return $i/given_name)", "Jessica"),
+                          ("for $i in (7 to  15) return $i*10", "130"),
+                          ("some $i in /hotel/branch/staff satisfies $i/age < 20", "false"),
+                          ("some $i in //hotel/branch/staff satisfies $i/age < 20", "false"),
+                          ("every $i in /hotel/branch/staff satisfies $i/age > 20", "true"),
+                          ("every $i in //hotel/branch/staff satisfies $i/age > 20 ", "true"),
+                          ("let $x := branch[@location = 'California'], $y := branch[@location = 'Las Vegas'] return (avg($x/staff/age), avg($y/staff/age))", "27.5"),
+                          ("let $x := //branch[@location = 'California'], $y := //branch[@location = 'Las Vegas'] return (avg($x/staff/age), avg($y/staff/age))", "27.5"),
+                          ("let $nu := 1, $de := 1000 return  'probability = ' || $nu div $de * 100 || '%'", "0.1%"),
+                          ("let $nu := 2, $probability := function ($argument) { 'probability = ' ||  $nu div $argument  * 100 || '%'}, $de := 5 return $probability($de)", "40%"),
+                          ("'XPATH2.0-3.1 dissemination' instance of xs:string ", "true"),
+                          ("'new stackoverflow question incoming' instance of xs:integer ", "false"),
+                          ("'50000' cast as xs:integer", "50000"),
+                          ("//branch[@location = 'California']/staff[1]/surname eq 'Anderson'", "true"),
+                          ("fn:false()", "false")])
+def test_hotels(html_content, xpath, answer):
+    html_content = html_tools.xpath_filter(xpath, html_content, append_pretty_line_formatting=True)
+    assert type(html_content) == str
+    assert answer in html_content
+
+
+
+branches_to_visit = """<?xml version="1.0" ?>
+  <branches_to_visit>
+     <manager name="Godot" room_no="501">
+         <branch>Area 51</branch>
+         <branch>A place with no name</branch>
+         <branch>Stalsk12</branch>
+     </manager>
+      <manager name="Freya" room_no="305">
+         <branch>Stalsk12</branch>
+         <branch>Barcelona</branch>
+         <branch>Paris</branch>
+     </manager>
+ </branches_to_visit>"""
+@pytest.mark.parametrize("html_content", [branches_to_visit])
+@pytest.mark.parametrize("xpath, answer", [
+    ("manager[@name = 'Godot']/branch union manager[@name = 'Freya']/branch", "Area 51"),
+    ("//manager[@name = 'Godot']/branch union //manager[@name = 'Freya']/branch", "Stalsk12"),
+    ("manager[@name = 'Godot']/branch | manager[@name = 'Freya']/branch", "Stalsk12"),
+    ("//manager[@name = 'Godot']/branch | //manager[@name = 'Freya']/branch", "Stalsk12"),
+    ("manager/branch intersect manager[@name = 'Godot']/branch", "A place with no name"),
+    ("//manager/branch intersect //manager[@name = 'Godot']/branch", "A place with no name"),
+    ("manager[@name = 'Godot']/branch intersect manager[@name = 'Freya']/branch", ""),
+    ("manager/branch except manager[@name = 'Godot']/branch", "Barcelona"),
+    ("manager[@name = 'Godot']/branch[1]  eq 'Area 51'", "true"),
+    ("//manager[@name = 'Godot']/branch[1]  eq 'Area 51'", "true"),
+    ("manager[@name = 'Godot']/branch[1]  eq 'Seoul'", "false"),
+    ("//manager[@name = 'Godot']/branch[1]  eq 'Seoul'", "false"),
+    ("manager[@name = 'Godot']/branch[2] eq manager[@name = 'Freya']/branch[2]", "false"),
+    ("//manager[@name = 'Godot']/branch[2] eq //manager[@name = 'Freya']/branch[2]", "false"),
+    ("manager[1]/@room_no lt manager[2]/@room_no", "false"),
+    ("//manager[1]/@room_no lt //manager[2]/@room_no", "false"),
+    ("manager[1]/@room_no gt manager[2]/@room_no", "true"),
+    ("//manager[1]/@room_no gt //manager[2]/@room_no", "true"),
+    ("manager[@name = 'Godot']/branch[1]  = 'Area 51'", "true"),
+    ("//manager[@name = 'Godot']/branch[1]  = 'Area 51'", "true"),
+    ("manager[@name = 'Godot']/branch[1]  = 'Seoul'", "false"),
+    ("//manager[@name = 'Godot']/branch[1]  = 'Seoul'", "false"),
+    ("manager[@name = 'Godot']/branch  = 'Area 51'", "true"),
+    ("//manager[@name = 'Godot']/branch  = 'Area 51'", "true"),
+    ("manager[@name = 'Godot']/branch  = 'Barcelona'", "false"),
+    ("//manager[@name = 'Godot']/branch  = 'Barcelona'", "false"),
+    ("manager[1]/@room_no > manager[2]/@room_no", "true"),
+    ("//manager[1]/@room_no > //manager[2]/@room_no", "true"),
+    ("manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is manager[1]/branch[1]", "false"),
+    ("//manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is //manager[1]/branch[1]", "false"),
+    ("manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is manager[1]/branch[3]", "true"),
+    ("//manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is //manager[1]/branch[3]", "true"),
+    ("manager[@name = 'Godot']/branch[ . = 'Stalsk12'] <<  manager[1]/branch[1]", "false"),
+    ("//manager[@name = 'Godot']/branch[ . = 'Stalsk12'] <<  //manager[1]/branch[1]", "false"),
+    ("manager[@name = 'Godot']/branch[ . = 'Stalsk12']  >>  manager[1]/branch[1]", "true"),
+    ("//manager[@name = 'Godot']/branch[ . = 'Stalsk12'] >>  //manager[1]/branch[1]", "true"),
+    ("manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is manager[@name = 'Freya']/branch[ . = 'Stalsk12']", "false"),
+    ("//manager[@name = 'Godot']/branch[ . = 'Stalsk12'] is //manager[@name = 'Freya']/branch[ . = 'Stalsk12']", "false"),
+    ("manager[1]/@name || manager[2]/@name", "GodotFreya"),
+    ("//manager[1]/@name || //manager[2]/@name", "GodotFreya"),
+                          ])
+def test_branches_to_visit(html_content, xpath, answer):
+    html_content = html_tools.xpath_filter(xpath, html_content, append_pretty_line_formatting=True)
+    assert type(html_content) == str
+    assert answer in html_content
+
+trips = """
+<trips>
+   <trip reservation_number="10">
+       <depart>2023-10-06</depart>
+       <arrive>2023-10-10</arrive>
+       <traveler name="Christopher Anderson">
+           <duration>4</duration>
+           <price>2000.00</price>
+       </traveler>
+   </trip>
+   <trip reservation_number="12">
+       <depart>2023-10-06</depart>
+       <arrive>2023-10-12</arrive>
+       <traveler name="Frank Carter">
+           <duration>6</duration>
+           <price>3500.34</price>
+       </traveler>
+   </trip>
+</trips>"""
+@pytest.mark.parametrize("html_content", [trips])
+@pytest.mark.parametrize("xpath, answer", [
+    ("1 + 9 * 9 + 5 div 5", "83"),
+    ("(1 + 9 * 9 + 5) div 6", "14.5"),
+    ("23 idiv 3", "7"),
+    ("23 div 3", "7.66666666"),
+    ("for $i in ./trip return $i/traveler/duration * $i/traveler/price", "21002.04"),
+    ("for $i in ./trip return $i/traveler/duration ", "4"),
+    ("for $i in .//trip return $i/traveler/duration * $i/traveler/price", "21002.04"),
+    ("sum(for $i in ./trip return $i/traveler/duration * $i/traveler/price)", "29002.04"),
+    ("sum(for $i in .//trip return $i/traveler/duration * $i/traveler/price)", "29002.04"),
+    #("trip[1]/depart - trip[1]/arrive", "fail_to_get_answer"),
+    #("//trip[1]/depart - //trip[1]/arrive", "fail_to_get_answer"),
+    #("trip[1]/depart + trip[1]/arrive", "fail_to_get_answer"),
+    #("xs:date(trip[1]/depart) + xs:date(trip[1]/arrive)", "fail_to_get_answer"),
+    ("(//trip[1]/arrive cast as xs:date) - (//trip[1]/depart cast as xs:date)", "P4D"),
+    ("(//trip[1]/depart cast as xs:date) - (//trip[1]/arrive cast as xs:date)", "-P4D"),
+    ("(//trip[1]/depart cast as xs:date) + xs:dayTimeDuration('P3D')", "2023-10-09"),
+    ("(//trip[1]/depart cast as xs:date) - xs:dayTimeDuration('P3D')", "2023-10-03"),
+    ("(456, 623) instance of xs:integer", "false"),
+    ("(456, 623) instance of xs:integer*", "true"),
+    ("/trips/trip instance of element()", "false"),
+    ("/trips/trip instance of element()*", "true"),
+    ("/trips/trip[1]/arrive instance of xs:date", "false"),
+    ("date(/trips/trip[1]/arrive) instance of xs:date", "true"),
+    ("'8' cast as xs:integer", "8"),
+    ("'11.1E3' cast as xs:double", "11100"),
+    ("6.5 cast as xs:integer", "6"),
+    #("/trips/trip[1]/arrive cast as xs:dateTime", "fail_to_get_answer"),
+    ("/trips/trip[1]/arrive cast as xs:date", "2023-10-10"),
+    ("('2023-10-12') cast as xs:date", "2023-10-12"),
+    ("for $i in //trip return concat($i/depart, '  ', $i/arrive)", "2023-10-06  2023-10-10"),
+                          ])
+def test_trips(html_content, xpath, answer):
+    html_content = html_tools.xpath_filter(xpath, html_content, append_pretty_line_formatting=True)
+    assert type(html_content) == str
+    assert answer in html_content

requirements.txt

@@ -46,6 +46,9 @@ beautifulsoup4
 # XPath filtering, lxml is required by bs4 anyway, but put it here to be safe.
 lxml
 
+# XPath 2.0-3.1 support
+elementpath
+
 selenium~=4.14.0
 
 werkzeug~=3.0