diff options
| author | Christophe Besson <cbesson@gmail.com> | 2026-08-30 16:31:34 +0200 |
|---|---|---|
| committer | Christophe Besson <cbesson@gmail.com> | 2026-08-30 16:31:34 +0200 |
| commit | 854b76338072979526ce36fbc176c82f6fc5d0bd (patch) | |
| tree | 6db723f22ba49b892f07942ea93d04076d0f72fe /packages/meshbay-node/tests/test_index_progress.py | |
| parent | 4b64ba2513d3985c843167153311436a80474149 (diff) | |
| download | meshbay-854b76338072979526ce36fbc176c82f6fc5d0bd.tar.gz | |
fix(node): stop a numbered saga all matching its first film
Every "<Saga> Episode <N> - <subtitle>" file in a numbered franchise
resolved to the series' first entry (a real, older film). `sequel_variants`
stripped "Episode <N>" and offered the bare "<Saga>" as a candidate query;
that matches the first film's original_title at ratio 1.0 and beat PASS 1's
correct-but-lower hit. A franchise's bare name is very often a real,
different film.
When a Part/Episode/Chapitre/… keyword carried the index, sequel_variants
no longer emits the bare base — only "<base> <digit>" and "<base> <roman>".
Without a keyword ("<Franchise> 3") the bare base is still offered, so that
fix is untouched. Verified live against TMDB: the franchise's episodes each
resolve to their own entry; the earlier numbered-sequel, two-part-film and
franchise-subtitle regressions all hold.
docs/mediacenter.md §10.3.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018BMLQjqFGCize2KtNBT79v
Diffstat (limited to 'packages/meshbay-node/tests/test_index_progress.py')
0 files changed, 0 insertions, 0 deletions