<title> hack

closes #1874

changedetectionio/html_tools.py

@@ -1,9 +1,12 @@
 
 from bs4 import BeautifulSoup
 from inscriptis import get_text
-from inscriptis.model.config import ParserConfig
 from jsonpath_ng.ext import parse
 from typing import List
+from inscriptis.css_profiles import CSS_PROFILES, HtmlElement
+from inscriptis.html_properties import Display
+from inscriptis.model.config import ParserConfig
+from xml.sax.saxutils import escape as xml_escape
 import json
 import re
 
@@ -68,10 +71,15 @@ def element_removal(selectors: List[str], html_content):
 
 
 # Return str Utf-8 of matched rules
-def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False):
+def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False, is_rss=False):
     from lxml import etree, html
 
-    tree = html.fromstring(bytes(html_content, encoding='utf-8'))
+    parser = None
+    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 = tree.xpath(xpath_filter.strip(), namespaces={'re': 'http://exslt.org/regular-expressions'})
@@ -94,7 +102,6 @@ def xpath_filter(xpath_filter, html_content, append_pretty_line_formatting=False
 
     return html_block
 
-
 # Extract/find element
 def extract_element(find='title', html_content=''):
 
@@ -260,8 +267,15 @@ def strip_ignore_text(content, wordlist, mode="content"):
 
     return "\n".encode('utf8').join(output)
 
+def cdata_in_document_to_text(html_content: str, render_anchor_tag_content=False) -> str:
+    pattern = '<!\[CDATA\[(\s*(?:.(?<!\]\]>)\s*)*)\]\]>'
+    def repl(m):
+        text = m.group(1)
+        return xml_escape(html_to_text(html_content=text)).strip()
 
-def html_to_text(html_content: str, render_anchor_tag_content=False) -> str:
+    return re.sub(pattern, repl, html_content)
+
+def html_to_text(html_content: str, render_anchor_tag_content=False, is_rss=False) -> str:
     """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
@@ -277,16 +291,21 @@ def html_to_text(html_content: str, render_anchor_tag_content=False) -> str:
     #  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
+            annotation_rules={"a": ["hyperlink"]},
+            display_links=True
         )
-
-    # otherwise set config to None
+    # otherwise set config to None/default
     else:
         parser_config = None
 
-    # get text and annotations via inscriptis
+    # 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)
 
     return text_content

changedetectionio/processors/text_json_diff.py

@@ -11,7 +11,7 @@ from changedetectionio import content_fetcher, html_tools
 from changedetectionio.blueprint.price_data_follower import PRICE_DATA_TRACK_ACCEPT, PRICE_DATA_TRACK_REJECT
 from copy import deepcopy
 from . import difference_detection_processor
-from ..html_tools import PERL_STYLE_REGEX
+from ..html_tools import PERL_STYLE_REGEX, cdata_in_document_to_text
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
 
@@ -153,13 +153,22 @@ class perform_site_check(difference_detection_processor):
 
         is_json = 'application/json' in fetcher.get_all_headers().get('content-type', '').lower()
         is_html = not is_json
+        is_rss = False
+
+        ctype_header = fetcher.get_all_headers().get('content-type', '').lower()
+        # Go into RSS preprocess for converting CDATA/comment to usable text
+        if any(substring in ctype_header for substring in ['application/xml', 'application/rss', 'text/xml']):
+            if '<rss' in fetcher.content[:100].lower():
+                fetcher.content = cdata_in_document_to_text(html_content=fetcher.content)
+                is_rss = True
 
         # source: support, basically treat it as plaintext
         if is_source:
             is_html = False
             is_json = False
 
-        if watch.is_pdf or 'application/pdf' in fetcher.get_all_headers().get('content-type', '').lower():
+        inline_pdf = fetcher.get_all_headers().get('content-disposition', '') and '%PDF-1' in fetcher.content[:10]
+        if watch.is_pdf or 'application/pdf' in fetcher.get_all_headers().get('content-type', '').lower() or inline_pdf:
             from shutil import which
             tool = os.getenv("PDF_TO_HTML_TOOL", "pdftohtml")
             if not which(tool):
@@ -242,7 +251,8 @@ class perform_site_check(difference_detection_processor):
                         if filter_rule[0] == '/' or filter_rule.startswith('xpath:'):
                             html_content += html_tools.xpath_filter(xpath_filter=filter_rule.replace('xpath:', ''),
                                                                     html_content=fetcher.content,
-                                                                    append_pretty_line_formatting=not is_source)
+                                                                    append_pretty_line_formatting=not is_source,
+                                                                    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,
@@ -262,8 +272,9 @@ class perform_site_check(difference_detection_processor):
                     do_anchor = self.datastore.data["settings"]["application"].get("render_anchor_tag_content", False)
                     stripped_text_from_html = \
                         html_tools.html_to_text(
-                            html_content,
-                            render_anchor_tag_content=do_anchor
+                            html_content=html_content,
+                            render_anchor_tag_content=do_anchor,
+                            is_rss=is_rss # #1874 activate the <title workaround hack
                         )
 
         # Re #340 - return the content before the 'ignore text' was applied

changedetectionio/static/js/diff-render.js

@@ -76,7 +76,7 @@ $(document).ready(function () {
             });
         }
 
-        next_diff();
+        $('#jump-next-diff').click();
     }
 
     $('.needs-localtime').each(function () {

changedetectionio/static/js/watch-overview.js

@@ -4,6 +4,14 @@ $(function () {
         $(this).closest('.unviewed').removeClass('unviewed');
     });
 
+    $('td[data-timestamp]').each(function () {
+        $(this).prop('title', new Intl.DateTimeFormat(undefined,
+            {
+                dateStyle: 'full',
+                timeStyle: 'long'
+            }).format($(this).data('timestamp') * 1000));
+    })
+
     $("#checkbox-assign-tag").click(function (e) {
         $('#op_extradata').val(prompt("Enter a tag name"));
     });

changedetectionio/templates/watch-overview.html

@@ -154,8 +154,8 @@
                     {% endfor %}
 
                 </td>
-                <td class="last-checked">{{watch|format_last_checked_time|safe}}</td>
-                <td class="last-changed">{% if watch.history_n >=2 and watch.last_changed >0 %}
+                <td class="last-checked" data-timestamp="{{ watch.last_checked }}">{{watch|format_last_checked_time|safe}}</td>
+                <td class="last-changed" data-timestamp="{{ watch.last_changed }}">{% if watch.history_n >=2 and watch.last_changed >0 %}
                     {{watch.last_changed|format_timestamp_timeago}}
                     {% else %}
                     Not yet

changedetectionio/tests/test_rss.py

@@ -2,12 +2,61 @@
 
 import time
 from flask import url_for
-from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks, extract_rss_token_from_UI
+from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks, extract_rss_token_from_UI, \
+    extract_UUID_from_client
 
 
+def set_original_cdata_xml():
+    test_return_data = """<rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:media="http://search.yahoo.com/mrss/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
+    <channel>
+    <title>Gizi</title>
+    <link>https://test.com</link>
+    <atom:link href="https://testsite.com" rel="self" type="application/rss+xml"/>
+    <description>
+    <![CDATA[ The Future Could Be Here ]]>
+    </description>
+    <language>en</language>
+    <item>
+    <title>
+    <![CDATA[ <img src="https://testsite.com/hacked.jpg"> Hackers can access your computer ]]>
+    </title>
+    <link>https://testsite.com/news/12341234234</link>
+    <description>
+    <![CDATA[ <img class="type:primaryImage" src="https://testsite.com/701c981da04869e.jpg"/><p>The days of Terminator and The Matrix could be closer. But be positive.</p><p><a href="https://testsite.com">Read more link...</a></p> ]]>
+    </description>
+    <category>cybernetics</category>
+    <category>rand corporation</category>
+    <pubDate>Tue, 17 Oct 2023 15:10:00 GMT</pubDate>
+    <guid isPermaLink="false">1850933241</guid>
+    <dc:creator>
+    <![CDATA[ Mr Hacker News ]]>
+    </dc:creator>
+    <media:thumbnail url="https://testsite.com/thumbnail-c224e10d81488e818701c981da04869e.jpg"/>
+    </item>
+
+    <item>
+        <title>    Some other title    </title>
+        <link>https://testsite.com/news/12341234236</link>
+        <description>
+        Some other description
+        </description>
+    </item>    
+    </channel>
+    </rss>
+            """
+
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write(test_return_data)
+
+
+def test_setup(client, live_server):
+    live_server_setup(live_server)
+
 def test_rss_and_token(client, live_server):
+    #    live_server_setup(live_server)
+
     set_original_response()
-    live_server_setup(live_server)
+    rss_token = extract_rss_token_from_UI(client)
 
     # Add our URL to the import page
     res = client.post(
@@ -17,11 +66,11 @@ def test_rss_and_token(client, live_server):
     )
 
     assert b"1 Imported" in res.data
-    rss_token = extract_rss_token_from_UI(client)
 
-    time.sleep(2)
+    wait_for_all_checks(client)
+    set_modified_response()
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
-    time.sleep(2)
+    wait_for_all_checks(client)
 
     # Add our URL to the import page
     res = client.get(
@@ -37,3 +86,80 @@ def test_rss_and_token(client, live_server):
     )
     assert b"Access denied, bad token" not in res.data
     assert b"Random content" in res.data
+
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+def test_basic_cdata_rss_markup(client, live_server):
+    #live_server_setup(live_server)
+
+    set_original_cdata_xml()
+
+    test_url = url_for('test_endpoint', content_type="application/xml", _external=True)
+
+    # 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)
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b'CDATA' not in res.data
+    assert b'<![' not in res.data
+    assert b'Hackers can access your computer' in res.data
+    assert b'The days of Terminator' in res.data
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
+
+def test_rss_xpath_filtering(client, live_server):
+    #live_server_setup(live_server)
+
+    set_original_cdata_xml()
+
+    test_url = url_for('test_endpoint', content_type="application/xml", _external=True)
+
+    res = client.post(
+        url_for("form_quick_watch_add"),
+        data={"url": test_url, "tags": '', 'edit_and_watch_submit_button': 'Edit > Watch'},
+        follow_redirects=True
+    )
+    assert b"Watch added in Paused state, saving will unpause" in res.data
+
+    uuid = extract_UUID_from_client(client)
+    res = client.post(
+        url_for("edit_page", uuid=uuid, unpause_on_save=1),
+        data={
+                "include_filters": "//item/title",
+                "fetch_backend": "html_requests",
+                "headers": "",
+                "proxy": "no-proxy",
+                "tags": "",
+                "url": test_url,
+              },
+        follow_redirects=True
+    )
+    assert b"unpaused" in res.data
+
+    wait_for_all_checks(client)
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b'CDATA' not in res.data
+    assert b'<![' not in res.data
+    # #1874  All but the first <title was getting selected
+    # Convert any HTML with just a top level <title> to <h1> to be sure title renders
+
+    assert b'Hackers can access your computer' in res.data # Should ONLY be selected by the xpath
+    assert b'Some other title' in res.data  # Should ONLY be selected by the xpath
+    assert b'The days of Terminator' not in res.data # Should NOT be selected by the xpath
+    assert b'Some other description' not in res.data  # Should NOT be selected by the xpath
+
+    res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)

changedetectionio/tests/test_xpath_selector.py

@@ -2,7 +2,7 @@
 
 import time
 from flask import url_for
-from . util import live_server_setup
+from .util import live_server_setup, wait_for_all_checks
 
 from ..html_tools import *
 
@@ -86,14 +86,14 @@ def test_check_xpath_filter_utf8(client, live_server):
         follow_redirects=True
     )
     assert b"1 Imported" in res.data
-    time.sleep(1)
+    wait_for_all_checks(client)
     res = client.post(
         url_for("edit_page", uuid="first"),
         data={"include_filters": filter, "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
         follow_redirects=True
     )
     assert b"Updated watch." in res.data
-    time.sleep(3)
+    wait_for_all_checks(client)
     res = client.get(url_for("index"))
     assert b'Unicode strings with encoding declaration are not supported.' not in res.data
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
@@ -140,14 +140,14 @@ def test_check_xpath_text_function_utf8(client, live_server):
         follow_redirects=True
     )
     assert b"1 Imported" in res.data
-    time.sleep(1)
+    wait_for_all_checks(client)
     res = client.post(
         url_for("edit_page", uuid="first"),
         data={"include_filters": filter, "url": test_url, "tags": "", "headers": "", 'fetch_backend': "html_requests"},
         follow_redirects=True
     )
     assert b"Updated watch." in res.data
-    time.sleep(3)
+    wait_for_all_checks(client)
     res = client.get(url_for("index"))
     assert b'Unicode strings with encoding declaration are not supported.' not in res.data
 
@@ -164,7 +164,6 @@ def test_check_xpath_text_function_utf8(client, live_server):
     assert b'Deleted' in res.data
 
 def test_check_markup_xpath_filter_restriction(client, live_server):
-    sleep_time_for_fetch_thread = 3
 
     xpath_filter = "//*[contains(@class, 'sametext')]"
 
@@ -183,7 +182,7 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
     assert b"1 Imported" in res.data
 
     # Give the thread time to pick it up
-    time.sleep(sleep_time_for_fetch_thread)
+    wait_for_all_checks(client)
 
     # Goto the edit page, add our ignore text
     # Add our URL to the import page
@@ -195,7 +194,7 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
     assert b"Updated watch." in res.data
 
     # Give the thread time to pick it up
-    time.sleep(sleep_time_for_fetch_thread)
+    wait_for_all_checks(client)
 
     # view it/reset state back to viewed
     client.get(url_for("diff_history_page", uuid="first"), follow_redirects=True)
@@ -206,7 +205,7 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
     # Trigger a check
     client.get(url_for("form_watch_checknow"), follow_redirects=True)
     # Give the thread time to pick it up
-    time.sleep(sleep_time_for_fetch_thread)
+    wait_for_all_checks(client)
 
     res = client.get(url_for("index"))
     assert b'unviewed' not in res.data
@@ -216,9 +215,6 @@ def test_check_markup_xpath_filter_restriction(client, live_server):
 
 def test_xpath_validation(client, live_server):
 
-    # 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(
@@ -227,7 +223,7 @@ def test_xpath_validation(client, live_server):
         follow_redirects=True
     )
     assert b"1 Imported" in res.data
-    time.sleep(2)
+    wait_for_all_checks(client)
 
     res = client.post(
         url_for("edit_page", uuid="first"),
@@ -244,11 +240,8 @@ def test_check_with_prefix_include_filters(client, live_server):
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
 
-    # Give the endpoint time to spin up
-    time.sleep(1)
-
     set_original_response()
-
+    wait_for_all_checks(client)
     # Add our URL to the import page
     test_url = url_for('test_endpoint', _external=True)
     res = client.post(
@@ -257,7 +250,7 @@ def test_check_with_prefix_include_filters(client, live_server):
         follow_redirects=True
     )
     assert b"1 Imported" in res.data
-    time.sleep(3)
+    wait_for_all_checks(client)
 
     res = client.post(
         url_for("edit_page", uuid="first"),
@@ -266,7 +259,7 @@ def test_check_with_prefix_include_filters(client, live_server):
     )
 
     assert b"Updated watch." in res.data
-    time.sleep(3)
+    wait_for_all_checks(client)
 
     res = client.get(
         url_for("preview_page", uuid="first"),
@@ -277,3 +270,46 @@ def test_check_with_prefix_include_filters(client, live_server):
     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)
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write("""<html>
+       <body>
+     Some initial text<br>
+     <p>Which is across multiple lines</p>
+     <br>
+     So let's see what happens.  <br>
+     <div class="sametext">Some text thats the same</div>
+     <div class="changetext">Some text that will change</div>
+     <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" />
+     </body>
+     </html>
+    """)
+    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)
+
+
+    for r in ['//div', '//a', 'xpath://div', 'xpath://a']:
+        res = client.post(
+            url_for("edit_page", uuid="first"),
+            data={"include_filters": r,
+                  "url": test_url,
+                  "tags": "",
+                  "headers": "",
+                  'fetch_backend': "html_requests"},
+            follow_redirects=True
+        )
+        wait_for_all_checks(client)
+        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"

requirements.txt

@@ -33,7 +33,7 @@ dnspython<2.3.0
 # jq not available on Windows so must be installed manually
 
 # Notification library
-apprise~=1.5.0
+apprise~=1.6.0
 
 # apprise mqtt https://github.com/dgtlmoon/changedetection.io/issues/315
 paho-mqtt