Parallel checksum calculation for file invalidation

[nikblanchet]

Summary

File invalidation checksums are currently computed sequentially in FileTracker. Parallelize checksum computation for significant performance improvement.

Current Behavior

• FileTracker.createSnapshot() computes checksums sequentially
• 100-500 files: ~200ms
• 1000+ files: ~400ms

Proposed Enhancement

Use Promise.all() / asyncio.gather() to compute checksums in parallel.

Expected Impact

• 4x speedup: 200ms → 50ms (100-500 files)
• 4x speedup: 400ms → 100ms (1000+ files)

Implementation

TypeScript (cli/src/utils/file-tracker.ts:34):

const snapshotPromises = filepaths.map(async (filepath) => {
  // Compute checksum...
});
const results = await Promise.all(snapshotPromises);

Python (analyzer/src/utils/file_tracker.py:52):

async def create_snapshot_async(filepaths: list[str]) -> dict[str, FileSnapshot]:
    tasks = [compute_checksum(fp) for fp in filepaths]
    results = await asyncio.gather(*tasks)

Effort

~1 hour

Related

• Code review PR Session 2: Audit State Schema & Persistence (Issue #216) #365