MatchOptions.trustedIds: a candidate fetched by a site-given ID is at least
multi_evidence even when titles differ; year/type conflicts stay candidates.
Differential group site-ids-* against MP search/result.py _match_source.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
MP MetaInfo + TheMovieDbModule name matching vs parse/recall/resolveIdentity,
TMDB responses replayed offline; includes the category-alias misbinding
regression from the user stack.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Strip [电视剧]/[动漫]/[音乐]… prefixes into parsed.category instead of titleCn,
split RSS-embedded site descriptions into the subtitle, take the first Chinese
name of the subtitle as titleCn and the rest as altTitles, and stop reading
'Travis - 12 Memories' as episode 12. PARSER_VERSION 2.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
- rules/: RuleExpr tree, '>' layers with ! & | and parens, position-aware errors
(unknown atom with suggestions, unsupported music atoms, parens, empty layer)
- builtin keyword table aligned with MP BUILTIN_RULE_SET; field-based judgement
where parsed fields exist, MP regex otherwise; custom atoms (MP 7 fields)
- chain evaluation: hard excludes, ladder pri 100-k, groups AND-ed, pri from last
laddered group, readable per-atom trace
- slot: SlotSpec.rules, rule.hard_exclude / rule.no_layer_matched violations,
pri basis switches to priority when a rule ladder is present
- differential: 66 rule chains x 32 torrents vs MP python path (parse 64/64,
pri 2090/2112, 22 registered mp_bug)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
