sudacode/SubMiner

Fork 0

mirror of https://github.com/ksyasuda/SubMiner.git synced 2026-08-04 19:21:33 -07:00

Files

T

History

sudacode b0a2ce6e8a

perf(tokenizer): skip character-name lookups where no name can start

The greedy name pre-pass asked the Yomitan backend at every Japanese
position, because a character name can begin mid-token. With the character
dictionary enabled that roughly doubled the round trips per line (measured
10 -> 21 on a 23-char line).

SubMiner generates the character dictionary, so the cached snapshots already
list every form a character entry can be matched by (term and reading). Those
forms are installed into the scan runtime once per media and the pre-pass now
probes only positions where one of them starts, compared after kana
normalization so a katakana name still matches a hiragana reading form. The
overhead drops to zero (21 -> 10, the same as with the dictionary disabled).

Fail-safe: with no candidate list (no media id, no cached snapshot, failed
install) the pre-pass keeps its exhaustive behavior, so stale character data
costs speed rather than a missing name. Halfwidth katakana positions bypass
the filter since kana normalization does not fold them.

The candidate lookup is consulted per subtitle line, so it caches its snapshot
directory signature for 5s; dictionary writes still call invalidate().

2026-08-04 01:56:44 -07:00

background-config-log-level.md

fix(logging): surface subtitle processing debug/warn logs (#182 )

2026-08-03 20:44:39 -07:00

README.md

docs(changelog): require reconciled fragments, not just new ones (#113 )

2026-06-06 00:55:34 -07:00

security-audit-high-advisories.md

fix(streaming): keep subtitle tokenization prefetch warm for full episodes (#183 )

2026-08-03 21:22:18 -07:00

stream-subtitle-tokenization-cache.md

fix(streaming): keep subtitle tokenization prefetch warm for full episodes (#183 )

2026-08-03 21:22:18 -07:00

subsync-reference-target-picker.md

feat(subsync): add reference and target subtitle track picker (#181 )

2026-08-03 01:00:14 -07:00

subtitle-plain-line-on-cache-miss.md

fix(overlay): show plain subtitle line immediately on tokenization cache miss (#184 )

2026-08-04 01:55:52 -07:00

subtitle-tokenization-performance.md

perf(tokenizer): skip character-name lookups where no name can start

2026-08-04 01:56:44 -07:00

README.md

Changelog Fragments

Add one .md file per user-visible PR in this directory.

Use this format:

type: added
area: overlay

- Added keyboard navigation for Yomitan popups.
- Added auto-pause toggle when opening the popup.

For breaking changes, add breaking: true:

type: changed
area: config
breaking: true

- Renamed `foo.bar` to `foo.baz`.

Rules:

type required: added, changed, fixed, docs, or internal
area required: short product area like overlay, launcher, release
breaking optional: set to true to flag as a breaking change
each non-empty body line becomes a bullet
README.md is ignored by the generator
if a PR should not produce release notes, apply the skip-changelog label instead of adding a fragment

PR branch workflow:

Before adding a fragment or bullet, inspect the changes/*.md files already changed in the PR
If the new work fixes, modifies, renames, or supersedes behavior introduced or referenced by that fragment, edit or remove the stale bullet instead of adding follow-up churn
Add a new bullet only when it describes a truly separate user-visible outcome
Multiple fragment files are allowed when one PR has genuinely separate release-note outcomes, but keep them minimized and current

How fragments turn into a release:

At release time, bun run changelog:build (and bun run changelog:prerelease-notes) pipes every pending fragment through claude -p to merge related items, drop noise, and rewrite into a clean user-facing release body. Write fragments as raw, informative notes — don't worry about polished prose, deduping across PRs, or line-by-line phrasing. The polish step handles all of that.
The polish step treats pending fragments as the final release outcome, not prerelease history. If a feature is added and then renamed or fixed before the stable cut, ship the final feature bullet instead of separate prerelease-only breaking/fix entries.
GitHub release notes and prerelease notes use short top-level items with nested bullets for the change, user benefit, and any useful action note. The stable CHANGELOG.md can stay in compact single-line bullets.
internal fragments stay in CHANGELOG.md (inside a collapsed <details> block) but are dropped from the GitHub release notes entirely.
The polished CHANGELOG.md and release/release-notes.md are committed and reviewed before tagging — edit the Markdown by hand if Claude misses something.

Prerelease notes:

prerelease tags like v0.11.3-beta.1 and v0.11.3-rc.1 reuse the current pending fragments to generate release/prerelease-notes.md
existing prerelease notes are a reviewed baseline; later prerelease runs should replace stale beta/RC wording with the current outcome instead of appending fix churn
prerelease note generation does not consume fragments and does not update CHANGELOG.md or docs-site/changelog.md
the final stable release is the point where bun run changelog:build consumes fragments into the stable changelog and release notes