fix(subs): stop segmented WebVTT merge from stripping formatting

Merge cues at the string level instead of round-tripping through
pycaption, which dropped inline tags and cue settings. STYLE and
REGION blocks are kept, duplicate boundary cues splice correctly,
and a bad timing line now skips the cue instead of crashing.

Also rank pysubs2 above SubtitleEdit for ASS to TTML/VTT since
SubtitleEdit flattens inline styling there. Documented the measured
backend fidelity table and added regression tests for the merge.
This commit is contained in:
imSp4rky
2026-07-12 09:39:39 -06:00
parent 6ef24661c8
commit d7a2bb6b8e
5 changed files with 228 additions and 208 deletions
+12
View File
@@ -87,6 +87,18 @@ def test_subtitleedit_ranks_first_when_available(monkeypatch):
assert chain[0] == "subtitleedit"
def test_styled_ass_to_ttml_vtt_prefers_pysubs2_over_subtitleedit(monkeypatch):
# Measured: SE flattens ASS inline styling to TTML/VTT; pysubs2 keeps i/b/u.
monkeypatch.setattr(binaries, "SubtitleEdit", "/usr/bin/seconv")
for target in (Codec.TimedTextMarkupLang, Codec.WebVTT):
chain = [b.name for b in sc.resolve_backends(Codec.SubStationAlphav4, target)]
assert chain[0] == "pysubs2", f"expected pysubs2 first for ASS->{target.name}, got {chain}"
assert "subtitleedit" in chain # still a fallback
# ASS->SRT keeps SE first (only backend that keeps colour there).
chain = [b.name for b in sc.resolve_backends(Codec.SubStationAlphav4, Codec.SubRip)]
assert chain[0] == "subtitleedit"
def test_pin_then_fallback_orders_pin_first():
chain = [b.name for b in sc.resolve_backends(Codec.WebVTT, Codec.SubRip, pin="pysubs2")]
assert chain[0] == "pysubs2"