feat: expand t.co links; linkify bare URLs in tweet and article text

Browse source

Frontend:
- resolveEntityBounds: try multiple candidate strings in order (u.url
  first, since that's the t.co short URL that appears in full_text)
- normalizeUrlAnn: multi-candidate search; href = expanded > url,
  display = display_url > expanded > url
- linkifyText(): regex linkifier for entity-less bare URLs; trims
  trailing punctuation [.,;:!?)] before linking; used in both
  renderTweetTextJSX and renderInlineJSX including their early-return
  paths (anns.length === 0) that previously bypassed linkification
- renderInlineJSX: fix mention href mention.name → screen_name;
  replace t.co segment text with url.display when entity covers it

Scraper (vendor/twitter/scrape_user_tweet_contents.py):
- extract_tweet_data: when note_tweet text is used, pull urls/mentions/
  hashtags/symbols from note_result.entity_set (correct indices for the
  note text); keep media from legacy.entities (no note media downloads)

...

This commit is contained in:

TheGeneralist 2026-07-12 15:34:39 +02:00

parent 900e33fa60

commit a06c541605

Signed by: thegeneralist01

SSH key fingerprint: SHA256:pp9qddbCNmVNoSjevdvQvM5z0DHN7LTa8qBMbcMq/R4

4 changed files with 113 additions and 50 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

47

crates/archivr-server/static/assets/index-C98W3QeO.js

View file

File diff suppressed because one or more lines are too long