3d5dfd2a88
- Remove legacy compute-worker monolith under `src/compute/`, including PDF and Whisper inference, control-plane, and runtime orchestration - Move PDF and Whisper inference logic to new `src/inference/` module - Move control-plane, orchestrator, and state machine logic to `src/operations/` - Move NATS/JetStream adapters to `src/infrastructure/` - Move job orchestration, progress, and artifact persistence to `src/jobs/` - Update imports throughout tests and main app to reference new module structure - Update Dockerfile and scripts to use new asset paths - Add new entrypoints: `src/api/app.ts` and `src/api/contracts.ts` - Remove obsolete files and update `.gitignore` for new dev artifacts This refactor modularizes the compute-worker, improving maintainability and separation of concerns. No functional changes to inference or orchestration logic. BREAKING CHANGE: compute-worker internal APIs, directory structure, and imports have changed; downstream code must update imports and integration points.
35 lines
1.5 KiB
TypeScript
35 lines
1.5 KiB
TypeScript
import { describe, expect, test } from 'vitest';
|
|
import {
|
|
mapWordsToSentenceOffsets,
|
|
} from '../../../src/inference';
|
|
|
|
describe('whisper alignment mapping', () => {
|
|
test('maps words to sentence offsets with punctuation and repeated spaces', () => {
|
|
const aligned = mapWordsToSentenceOffsets('Hello, world again.', [
|
|
{ word: 'Hello', start: 0, end: 0.25 },
|
|
{ word: 'world', start: 0.25, end: 0.5 },
|
|
{ word: 'again', start: 0.5, end: 1.0 },
|
|
]);
|
|
|
|
expect(aligned.words).toHaveLength(3);
|
|
expect(aligned.words[0].charStart).toBe(0);
|
|
expect(aligned.words[0].charEnd).toBe(5);
|
|
expect(aligned.words[1].charStart).toBeGreaterThan(aligned.words[0].charEnd);
|
|
expect(aligned.words[2].charStart).toBeGreaterThan(aligned.words[1].charEnd);
|
|
expect(aligned.words[2].charEnd).toBeLessThanOrEqual('Hello, world again.'.length);
|
|
});
|
|
|
|
test('joins line-break hyphenation across unicode letters', () => {
|
|
// "Über-\n mensch" must normalize to "Übermensch" so the offsets match the
|
|
// client char map. This only works with the unicode-aware hyphen regex that
|
|
// is kept in lock-step across nlp.ts / alignment-map.ts / highlight-char-map.ts.
|
|
const aligned = mapWordsToSentenceOffsets('Über-\n mensch walks', [
|
|
{ word: 'Übermensch', start: 0, end: 0.5 },
|
|
{ word: 'walks', start: 0.5, end: 1.0 },
|
|
]);
|
|
|
|
expect(aligned.words[0].charStart).toBe(0);
|
|
expect(aligned.words[0].charEnd).toBe('Übermensch'.length);
|
|
expect(aligned.words[1].charStart).toBeGreaterThanOrEqual(aligned.words[0].charEnd);
|
|
});
|
|
});
|