lee/changedetection.io
1
0
Fork 0
mirror of https://github.com/dgtlmoon/changedetection.io.git synced 2026-05-06 01:31:06 +00:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

Ability to apply filters (first, last etc)

Browse Source
This commit is contained in:
dgtlmoon
2025-10-10 17:26:20 +02:00
parent f02fb7406d
commit 709dadc492
3 changed files with 48 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
changedetectionio/processors/text_json_diff/processor.py
+5
View File
@@ -397,6 +397,11 @@ class perform_site_check(difference_detection_processor):
# RSS preprocessing
if stream_content_type.is_rss:
content = content_processor.preprocess_rss(content)
if self.datastore.data["settings"]["application"].get("rss_reader_mode"):
# Now just becomes regular HTML that can have xpath/CSS applied (first of the set etc)
stream_content_type.is_rss = False
stream_content_type.is_html = True
self.fetcher.content = content
# PDF preprocessing
if watch.is_pdf or stream_content_type.is_pdf:
changedetectionio/rss_tools.py
+16 -2
View File
@@ -109,8 +109,22 @@ def format_rss_items(rss_content: str, render_anchor_tag_content=False) -> str:
if item_parts:
formatted_items.append('\n'.join(item_parts))
# Join all items with <br><br><hr>
return '<html><body>'+'<br><hr><br>'.join(formatted_items)
# Wrap each item in a div with classes (first, last, item-N)
items_html = []
total_items = len(formatted_items)
for idx, item in enumerate(formatted_items):
classes = ['rss-item']
if idx == 0:
classes.append('first')
if idx == total_items - 1:
classes.append('last')
classes.append(f'item-{idx + 1}')
class_str = ' '.join(classes)
items_html.append(f'<div class="{class_str}">{item}</div>')
# Join items with two <br> tags
return f'<html><body>{"\n<br><br>".join(items_html)}</body></html>'
except Exception as e:
logger.warning(f"Error formatting RSS items: {str(e)}")
changedetectionio/tests/test_rss_reader_mode.py
+27
View File
@@ -25,6 +25,8 @@ def set_original_cdata_xml():
<pubDate>Thu, 07 Aug 2025 00:00:00 GMT</pubDate>
<description><p>Wet noodles escape<br><p>they also found themselves outside</p> </description>
</item>
<item>
<title>TS-2025-004</title>
<link>https://wetscale.com/security-bulletins/#ts-2025-004</link>
@@ -69,3 +71,28 @@ def test_rss_reader_mode(client, live_server, measure_memory_usage):
assert 'PubDate: Thu, 07 Aug 2025 00:00:00 GMT' in snapshot_contents
delete_all_watches(client)
def test_rss_reader_mode_with_css_filters(client, live_server, measure_memory_usage):
set_original_cdata_xml()
# Rarely do endpoints give the right header, usually just text/xml, so we check also for <rss
# This also triggers the automatic CDATA text parser so the RSS goes back a nice content list
test_url = url_for('test_endpoint', content_type="text/xml; charset=UTF-8", _external=True)
live_server.app.config['DATASTORE'].data['settings']['application']['rss_reader_mode'] = True
# Add our URL to the import page
uuid = client.application.config.get('DATASTORE').add_watch(url=test_url, extras={'include_filters': [".last"]})
client.get(url_for("ui.form_watch_checknow"), follow_redirects=True)
wait_for_all_checks(client)
watch = live_server.app.config['DATASTORE'].data['watching'][uuid]
dates = list(watch.history.keys())
snapshot_contents = watch.get_history_snapshot(dates[0])
assert 'Wet noodles escape' not in snapshot_contents
assert '<br>' not in snapshot_contents
assert '&lt;' not in snapshot_contents
assert 'The days of Terminator and The Matrix' in snapshot_contents
delete_all_watches(client)