v0.1.0 — first release

First public release of ytmla.

Tooling for putting multi-language audio tracks on YouTube videos — the part the Data API covers, plus the part it does not.

The API half is proven (used end to end against a live channel: 3 languages x 6 episodes, long-form and Shorts):

• auth — one-time OAuth consent
• upload — upload a video and verify it landed
• localize — apply localized titles and descriptions
• dub — fit an audio track to a video's exact length
• check — preflight whether YouTube will accept a track

The Studio half is not proven. Multi-language audio has no Data API endpoint — verified against the v3 discovery document; there is no audio/localization write part on videos.insert/videos.update. So the track has to be attached through the Studio UI, which studio drives with Playwright. Studio's DOM changes without notice, so treat the selector map as a starting point, and run --dry-run --headed first.

Install: pip install ytmla · Studio extra: pip install "ytmla[studio]"