d3ro-voice/apps/desktop/tests
Yun Chan bcd0cf4f23 fix(meeting): keep learned text from injecting prompt lines
Learned text becomes part of a prompt, so a dictionary word or phrase
containing a newline could add its own instruction line. Every value that
reaches the recognition hint or the polish block is now flattened to one
line and length-capped, and a test proves an injected line stays plain text.

Also cover the invariants that keep the correction safe (random inputs,
Hangul untouched, idempotent, no token growth, large transcript) and the
file-transcription wiring that attaches the learned hint to every chunk and
corrects each chunk and segment before it is stored.
2026-11-06 23:00:29 +09:00
..
e2e chore: land in-progress workspace changes before 1.16.2 2026-11-05 21:58:44 +09:00
eval feat(desktop): built-in Han/Eng layout auto-switching, terminal/Vim support, and Unicode input 2026-10-27 10:48:07 +09:00
helpers feat(desktop): send meeting transcript segments and preset prompt edits to the phone 2026-09-27 16:24:25 +09:00
integration feat: 회의 파일 가져오기와 입력 엔진 배포 보완 2026-11-01 19:39:06 +09:00
main fix(meeting): keep learned text from injecting prompt lines 2026-11-06 23:00:29 +09:00
red feat(meeting): ground AI transcript correction in learned vocabulary 2026-11-06 17:22:54 +09:00
renderer feat(desktop): built-in Han/Eng layout auto-switching, terminal/Vim support, and Unicode input 2026-10-27 10:48:07 +09:00
shared feat(shared): gate paid features from one entitlement source 2026-09-16 23:24:18 +09:00
unit chore: land in-progress workspace changes before 1.16.2 2026-11-05 21:58:44 +09:00
setup.ts fix(keybinding): release held shortcuts and modifiers when the screen locks or the PC sleeps 2026-10-21 00:28:40 +09:00