fix comment

.github/dependabot.yml

@@ -4,11 +4,13 @@ updates:
     directory: /
     schedule:
       interval: "weekly"
-    "caronc/apprise":
-      versioning-strategy: "increase"
-      schedule:
-        interval: "daily"
     groups:
       all:
         patterns:
         - "*"
+  - package-ecosystem: pip
+    directory: /
+    schedule:
+      interval: "daily"
+    allow:
+      - dependency-name: "apprise"

.github/workflows/containers.yml

@@ -95,7 +95,7 @@ jobs:
           push: true
           tags: |
             ${{ secrets.DOCKER_HUB_USERNAME }}/changedetection.io:dev,ghcr.io/${{ github.repository }}:dev
-          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8,linux/arm64/v8
+          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8
           cache-from: type=gha
           cache-to: type=gha,mode=max
 
@@ -133,7 +133,7 @@ jobs:
           file: ./Dockerfile
           push: true
           tags: ${{ steps.meta.outputs.tags }}
-          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8,linux/arm64/v8
+          platforms: linux/amd64,linux/arm64,linux/arm/v7,linux/arm/v8
           cache-from: type=gha
           cache-to: type=gha,mode=max
 # Looks like this was disabled

.github/workflows/test-container-build.yml

@@ -38,8 +38,6 @@ jobs:
             dockerfile: ./Dockerfile
           - platform: linux/arm/v8
             dockerfile: ./Dockerfile
-          - platform: linux/arm64/v8
-            dockerfile: ./Dockerfile
           # Alpine Dockerfile platforms (musl via alpine check)
           - platform: linux/amd64
             dockerfile: ./.github/test/Dockerfile-alpine
@@ -76,5 +74,5 @@ jobs:
             file: ${{ matrix.dockerfile }}
             platforms: ${{ matrix.platform }}
             cache-from: type=gha
-            cache-to: type=gha,mode=max
+            cache-to: type=gha,mode=min
 

changedetectionio/__init__.py

@@ -2,7 +2,7 @@
 
 # Read more https://github.com/dgtlmoon/changedetection.io/wiki
 
-__version__ = '0.50.14'
+__version__ = '0.50.15'
 
 from changedetectionio.strtobool import strtobool
 from json.decoder import JSONDecodeError

changedetectionio/blueprint/settings/templates/settings.html

@@ -191,6 +191,12 @@ nav
                         </ul>
                      </span>
                     </fieldset>
+                    <fieldset class="pure-group">
+                        {{ render_checkbox_field(form.application.form.strip_ignored_lines) }}
+                        <span class="pure-form-message-inline">Remove any text that appears in the "Ignore text" from the output (otherwise its just ignored for change-detection)<br>
+                        <i>Note:</i> Changing this will change the status of your existing watches, possibly trigger alerts etc.
+                        </span>
+                    </fieldset>
            </div>
 
             <div class="tab-pane-inner" id="api">

changedetectionio/forms.py

@@ -759,6 +759,7 @@ class processor_text_json_diff_form(commonSettingsForm):
     check_unique_lines = BooleanField('Only trigger when unique lines appear in all history', default=False)
     remove_duplicate_lines = BooleanField('Remove duplicate lines of text', default=False)
     sort_text_alphabetically =  BooleanField('Sort text alphabetically', default=False)
+    strip_ignored_lines = TernaryNoneBooleanField('Strip ignored lines', default=None)
     trim_text_whitespace = BooleanField('Trim whitespace before and after text', default=False)
 
     filter_text_added = BooleanField('Added lines', default=True)
@@ -936,6 +937,7 @@ class globalSettingsApplicationForm(commonSettingsForm):
     removepassword_button = SubmitField('Remove password', render_kw={"class": "pure-button pure-button-primary"})
     render_anchor_tag_content = BooleanField('Render anchor tag content', default=False)
     shared_diff_access = BooleanField('Allow anonymous access to watch history page when password is enabled', default=False, validators=[validators.Optional()])
+    strip_ignored_lines = BooleanField('Strip ignored lines')
     rss_hide_muted_watches = BooleanField('Hide muted watches from RSS feed', default=True,
                                       validators=[validators.Optional()])
     filter_failure_notification_threshold_attempts = IntegerField('Number of times the filter can be missing before sending a notification',

changedetectionio/model/App.py

@@ -57,6 +57,7 @@ class model(dict):
                     'rss_hide_muted_watches': True,
                     'schema_version' : 0,
                     'shared_diff_access': False,
+                    'strip_ignored_lines': False,
                     'tags': {}, #@todo use Tag.model initialisers
                     'timezone': None, # Default IANA timezone name
                     'webdriver_delay': None , # Extra delay in seconds before extracting text

changedetectionio/model/__init__.py

@@ -58,6 +58,7 @@ class watch_base(dict):
             'proxy': None,  # Preferred proxy connection
             'remote_server_reply': None,  # From 'server' reply header
             'sort_text_alphabetically': False,
+            'strip_ignored_lines': None,
             'subtractive_selectors': [],
             'tag': '',  # Old system of text name for a tag, to be removed
             'tags': [],  # list of UUIDs to App.Tags

changedetectionio/processors/text_json_diff/processor.py

@@ -153,12 +153,26 @@ class perform_site_check(difference_detection_processor):
             # CSS Filter, extract the HTML that matches and feed that into the existing inscriptis::get_text
             self.fetcher.content = html_tools.workarounds_for_obfuscations(self.fetcher.content)
             html_content = self.fetcher.content
+            content_type = self.fetcher.get_all_headers().get('content-type', '').lower()
+            is_attachment = 'attachment' in self.fetcher.get_all_headers().get('content-disposition', '').lower() or 'octet-stream' in content_type
 
-            # If not JSON,  and if it's not text/plain..
+            # Try to detect better mime types if its a download or not announced as HTML
-            if 'text/plain' in self.fetcher.get_all_headers().get('content-type', '').lower():
+            if is_attachment:
+                logger.debug(f"Got a reply that may be a download or possibly a text attachment, checking..")
+                try:
+                    import magic
+                    mime = magic.from_buffer(html_content, mime=True)
+                    logger.debug(f"Guessing mime type, original content_type '{content_type}', mime type detected '{mime}'")
+                    if mime and "/" in mime: # looks valid and is a valid mime type
+                        content_type = mime
+                except Exception as e:
+                    logger.error(f"Error getting a more precise mime type from 'magic' library ({str(e)}")
+
+            if 'text/' in content_type and not 'html' in content_type:
                 # Don't run get_text or xpath/css filters on plaintext
                 stripped_text_from_html = html_content
             else:
+                # If not JSON, and if it's not text/plain..
                 # Does it have some ld+json price data? used for easier monitoring
                 update_obj['has_ldjson_price_data'] = html_tools.has_ldjson_product_info(self.fetcher.content)
 
@@ -301,6 +315,11 @@ class perform_site_check(difference_detection_processor):
         text_for_checksuming = stripped_text_from_html
         if text_to_ignore:
             text_for_checksuming = html_tools.strip_ignore_text(stripped_text_from_html, text_to_ignore)
+            # Some people prefer to also completely remove it
+            strip_ignored_lines = watch.get('strip_ignored_lines') if watch.get('strip_ignored_lines') is not None else self.datastore.data['settings']['application'].get('strip_ignored_lines')
+            if strip_ignored_lines:
+                # @todo add test in the 'preview' mode, check the widget works? compare to datastruct
+                stripped_text_from_html = text_for_checksuming
 
         # Re #133 - if we should strip whitespaces from triggering the change detected comparison
         if text_for_checksuming and self.datastore.data['settings']['application'].get('ignore_whitespace', False):

changedetectionio/run_socks_proxy_tests.sh

@@ -9,7 +9,7 @@ set -x
 # SOCKS5 related - start simple Socks5 proxy server
 # SOCKSTEST=xyz should show in the logs of this service to confirm it fetched
 docker run --network changedet-network -d --hostname socks5proxy --rm  --name socks5proxy -p 1080:1080 -e PROXY_USER=proxy_user123 -e PROXY_PASSWORD=proxy_pass123 serjs/go-socks5-proxy
-docker run --network changedet-network -d --hostname socks5proxy-noauth --rm  -p 1081:1080 --name socks5proxy-noauth  serjs/go-socks5-proxy
+docker run --network changedet-network -d --hostname socks5proxy-noauth --rm -p 1081:1080 --name socks5proxy-noauth -e REQUIRE_AUTH=false serjs/go-socks5-proxy
 
 echo "---------------------------------- SOCKS5 -------------------"
 # SOCKS5 related - test from proxies.json

changedetectionio/static/styles/scss/parts/_widgets.scss

@@ -34,7 +34,6 @@
       transition: all 0.2s ease;
       cursor: pointer;
       display: block;
-      min-width: 60px;
       text-align: center;
     }
 

changedetectionio/templates/edit/text-options.html

@@ -26,7 +26,10 @@
                             <li>Changing this will affect the comparison checksum which may trigger an alert</li>
                         </ul>
                 </span>
-
+                <br><br>
+                    <div class="pure-control-group">
+                      {{ render_ternary_field(form.strip_ignored_lines) }}
+                    </div>
                 </fieldset>
 
                 <fieldset>

changedetectionio/tests/test_backend.py

@@ -165,3 +165,130 @@ def test_check_basic_change_detection_functionality(client, live_server, measure
     # Cleanup everything
     res = client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
+
+def test_non_text_mime_or_downloads(client, live_server, measure_memory_usage):
+    """
+
+    https://github.com/dgtlmoon/changedetection.io/issues/3434
+    I noticed that a watched website can be monitored fine as long as the server sends content-type: text/plain; charset=utf-8,
+    but once the server sends content-type: application/octet-stream (which is usually done to force the browser to show the Download dialog),
+    changedetection somehow ignores all line breaks and treats the document file as if everything is on one line.
+
+    :param client:
+    :param live_server:
+    :param measure_memory_usage:
+    :return:
+    """
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write("""some random text that should be split by line
+and not parsed with html_to_text
+this way we know that it correctly parsed as plain text
+\r\n
+ok\r\n
+got it\r\n
+""")
+
+    test_url = url_for('test_endpoint', content_type="application/octet-stream", _external=True)
+
+    # Add our URL to the import page
+    res = client.post(
+        url_for("imports.import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+
+    assert b"1 Imported" in res.data
+
+    wait_for_all_checks(client)
+
+    ### check the front end
+    res = client.get(
+        url_for("ui.ui_views.preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b"some random text that should be split by line\n" in res.data
+    ####
+
+    # Check the snapshot by API that it has linefeeds too
+    watch_uuid = next(iter(live_server.app.config['DATASTORE'].data['watching']))
+    api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
+    res = client.get(
+        url_for("watchhistory", uuid=watch_uuid),
+        headers={'x-api-key': api_key},
+    )
+
+    # Fetch a snapshot by timestamp, check the right one was found
+    res = client.get(
+        url_for("watchsinglehistory", uuid=watch_uuid, timestamp=list(res.json.keys())[-1]),
+        headers={'x-api-key': api_key},
+    )
+    assert b"some random text that should be split by line\n" in res.data
+
+
+    res = client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
+
+
+def test_standard_text_plain(client, live_server, measure_memory_usage):
+    """
+
+    https://github.com/dgtlmoon/changedetection.io/issues/3434
+    I noticed that a watched website can be monitored fine as long as the server sends content-type: text/plain; charset=utf-8,
+    but once the server sends content-type: application/octet-stream (which is usually done to force the browser to show the Download dialog),
+    changedetection somehow ignores all line breaks and treats the document file as if everything is on one line.
+
+    The real bug here can be that it will try to process plain-text as HTML, losing <etc>
+
+    :param client:
+    :param live_server:
+    :param measure_memory_usage:
+    :return:
+    """
+    with open("test-datastore/endpoint-content.txt", "w") as f:
+        f.write("""some random text that should be split by line
+and not parsed with html_to_text
+<title>Even this title should stay because we are just plain text</title>
+this way we know that it correctly parsed as plain text
+\r\n
+ok\r\n
+got it\r\n
+""")
+
+    test_url = url_for('test_endpoint', content_type="text/plain", _external=True)
+
+    # Add our URL to the import page
+    res = client.post(
+        url_for("imports.import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+
+    assert b"1 Imported" in res.data
+
+    wait_for_all_checks(client)
+
+    ### check the front end
+    res = client.get(
+        url_for("ui.ui_views.preview_page", uuid="first"),
+        follow_redirects=True
+    )
+    assert b"some random text that should be split by line\n" in res.data
+    ####
+
+    # Check the snapshot by API that it has linefeeds too
+    watch_uuid = next(iter(live_server.app.config['DATASTORE'].data['watching']))
+    api_key = live_server.app.config['DATASTORE'].data['settings']['application'].get('api_access_token')
+    res = client.get(
+        url_for("watchhistory", uuid=watch_uuid),
+        headers={'x-api-key': api_key},
+    )
+
+    # Fetch a snapshot by timestamp, check the right one was found
+    res = client.get(
+        url_for("watchsinglehistory", uuid=watch_uuid, timestamp=list(res.json.keys())[-1]),
+        headers={'x-api-key': api_key},
+    )
+    assert b"some random text that should be split by line\n" in res.data
+    assert b"<title>Even this title should stay because we are just plain text</title>" in res.data
+
+    res = client.get(url_for("ui.form_delete", uuid="all"), follow_redirects=True)
+

changedetectionio/tests/test_ignore.py

@@ -58,3 +58,39 @@ def test_ignore(client, live_server, measure_memory_usage):
     # Should be in base.html
     assert b'csrftoken' in res.data
 
+
+def test_strip_ignore_lines(client, live_server, measure_memory_usage):
+   #  live_server_setup(live_server) # Setup on conftest per function
+    set_original_ignore_response()
+
+
+    # Goto the settings page, add our ignore text
+    res = client.post(
+        url_for("settings.settings_page"),
+        data={
+            "requests-time_between_check-minutes": 180,
+            "application-ignore_whitespace": "y",
+            "application-strip_ignored_lines": "y",
+            "application-global_ignore_text": "Which is across multiple",
+            'application-fetch_backend': "html_requests"
+        },
+        follow_redirects=True
+    )
+    assert b"Settings updated." in res.data
+
+    test_url = url_for('test_endpoint', _external=True)
+    res = client.post(
+        url_for("imports.import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+    assert b"1 Imported" in res.data
+
+    # Give the thread time to pick it up
+    wait_for_all_checks(client)
+    uuid = next(iter(live_server.app.config['DATASTORE'].data['watching']))
+
+    # It should not be in the preview anymore
+    res = client.get(url_for("ui.ui_views.preview_page", uuid=uuid))
+    assert b'<div class="ignored">' not in res.data
+    assert b'Which is across multiple' not in res.data

requirements.txt

@@ -39,7 +39,7 @@ jsonpath-ng~=1.5.3
 # jq not available on Windows so must be installed manually
 
 # Notification library
-apprise==1.9.3
+apprise==1.9.5
 
 # - Needed for apprise/spush, and maybe others? hopefully doesnt trigger a rust compile.
 # - Requires extra wheel for rPi, adds build time for arm/v8 which is not in piwheels
@@ -51,8 +51,8 @@ cryptography==44.0.1
 # use any version other than 2.0.x due to https://github.com/eclipse/paho.mqtt.python/issues/814
 paho-mqtt!=2.0.*
 
-# Used for CSS filtering
+# Used for CSS filtering, JSON extraction from HTML
-beautifulsoup4>=4.0.0
+beautifulsoup4>=4.0.0,<=4.13.5
 
 # 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