Skip to content

Conversation

@ComputelessComputer
Copy link
Collaborator

No description provided.

@ComputelessComputer ComputelessComputer merged commit 8a1ea10 into main Dec 11, 2024
@ComputelessComputer ComputelessComputer deleted the feat/editor branch December 11, 2024 22:26
yujonglee pushed a commit that referenced this pull request Feb 20, 2025
devin-ai-integration bot added a commit that referenced this pull request Nov 23, 2025
- Issue #1: Fix partial hints index mismatch after filtering
  - Added index mapping to track position changes when filtering partial words
  - Changed filtering condition from > to >= to keep consecutive words
  - Added comprehensive tests for index adjustment scenarios

- Issue #3: Add memory leak prevention safeguards
  - Implemented MAX_PARTIAL_WORDS (1000) and MAX_PARTIAL_HINTS (1000) limits
  - Added automatic cleanup when limits are exceeded
  - Adjusted hint indices when trimming old partial words

- Issue #4: Add input validation for transcript handling
  - Validate channelIndex bounds (0-255)
  - Verify words are properly ordered by timestamp
  - Handle undefined channelIndex and missing alternatives gracefully

- Issue #5: Improve string matching and add logging
  - Added console.warn when words not found in transcript
  - Log includes word, transcript, position, and wordIndex for debugging
  - Added tests for edge cases (empty transcript, empty words array)

- Issue #8: Make gap threshold configurable
  - Added optional maxGapMs parameter to buildSegments
  - Defaults to 2000ms when not specified
  - Allows per-call customization of segment gap threshold

All changes follow TDD approach with tests written first, then implementations.
All unit tests passing (12 transcript tests, 8 utils tests, 6+ segment tests).

Co-Authored-By: yujonglee <yujonglee.dev@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants