v1.28.2

Extractors

Additions

• [cyberdrop] add extractor for media URLs (#2496)
• [itaku] add search extractor (#6613)
• [lofter] add initial support (#650, #2294, #4095, #4728, #5656, #6607)
• [yiffverse] add support (#6611)

Fixes

• [facebook] decode Unicode surrogate pairs in metadata values (#6599)
• [zerochan] parse API responses manually when receiving invalid JSON (#6632)
• [zerochan] fix source metadata extraction when not logged in

Improvements

• [bilibili] extract files from module_top entries (#6687)
• [bilibili] support /upload/opus URLs (#6687)
• [bluesky] default to posts timeline when reposts or quoted is enabled (#6583)
• [common] simplify HTTP error messages
• [common] detect DDoS-Guard challenge pages
• [deviantart] improve tiptap markup to HTML conversion (#6686)
  • fix KeyError: 'attrs' for links without href
  • support heading content blocks
  • support strike text markers
• [instagram] extract date metadata for stories (#6677)
• [kemonoparty:favorite] support new URL format (#6676)
• [saint] support saint2.cr URLs (#6692)
• [tapas] improve extractor hierarchy (#6680)

Options

• [cohost] add avatar and background options (#6656)

Miscellaneous

• support * wildcards for parent>child categories, for example reddit>* (#6673)
• use latest Firefox UA as default user-agent
• use random unused port for "user-agent": "browser" requests