docs(topology): retain eslint with benchmark evidence (sc-1678) - #398
docs(topology): retain eslint with benchmark evidence (sc-1678)#398norvalbv wants to merge 1 commit into
Conversation
## Summary - benchmark a dependency-free filesystem walker against Devkit's incumbent ESLint topology gate on disposable fixtures - preserve the full coverage accounting: generic placement/ignore behavior, import walls, illegal empty directories, and Electron's six hand-written folder trees - retain ESLint because the faster candidate does not yet preserve the incumbent policy surface - record the prior-art and pre-implementation critique alongside raw 20-sample evidence ## Decision The direct walker is materially cheaper, but this PR deliberately makes no runtime change. On Node 24.19.0 it reduced median CPU by 75.5% on the clean lane and 76.0% on the placement lane, with median tree RSS down 31.8% and 31.7%. It still misses three required responsibilities: 1. Electron renderer/main/shared/preload/socket/vercel folder trees are hand-written in the ESLint preset and are not represented in guard config. 2. Electron import walls require parsed/resolved imports. 3. Illegal empty directories are observable by the current filesystem-aware ESLint plugin but not by the candidate's file-only walk. The ignored-directory fixture now uses an invalid selected `.ts` file, so the zero-diagnostic result proves actual ignore semantics rather than extension non-selection. ## Benchmark | Lane | Current ESLint median CPU | Direct walker median CPU | CPU delta | RSS delta | | --- | ---: | ---: | ---: | ---: | | Clean | 490 ms | 120 ms | -75.5% | -31.8% | | Placement | 500 ms | 120 ms | -76.0% | -31.7% | Method: Node 24.19.0, 3 discarded warm-ups, 20 measured samples per lane, fresh disposable fixture roots, wall/CPU/process-tree RSS captured in raw `results.json`. ## Validation - benchmark result assertions - all benchmark scripts pass `node --check` - ignored `.ts` directory parity verified - decision-record checker - `bun run lint` - `bun run typecheck` - `bun run lint:structure` - `git diff --check` - GitNexus staged change analysis: LOW, documentation/evidence only Shortcut: sc-1678
|
Warning Review limit reached
Next review available in: 50 minutes You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (8)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
1 participant
Summary
Decision
The direct walker is materially cheaper, but this PR deliberately makes no runtime change. On Node 24.19.0 it reduced median CPU by 75.5% on the clean lane and 76.0% on the placement lane, with median tree RSS down 31.8% and 31.7%. It still misses three required responsibilities:
The ignored-directory fixture now uses an invalid selected
.tsfile, so the zero-diagnostic result proves actual ignore semantics rather than extension non-selection.Benchmark
Method: Node 24.19.0, 3 discarded warm-ups, 20 measured samples per lane, fresh disposable fixture roots, wall/CPU/process-tree RSS captured in raw
results.json.Validation
node --check.tsdirectory parity verifiedbun run lintbun run typecheckbun run lint:structuregit diff --checkShortcut: sc-1678