fix: create the cache dir before symlinking the re-warm project at it

[zkochan]

The re-warm from #63 broke the first run after merge, on pnpm 11:

[ENOTDIR] ENOTDIR: not a directory, mkdir '.../pnpm11/alotta-files/.rewarm/cache'

Why pnpm 11, and why the smoke test missed it

The cache directory is deleted two rows before update, and a manager whose node_modules row is a registry-free no-op — pnpm restores the tree from the lockfile copy inside node_modules/.pnpm — never recreates it. The re-warm's cache symlink then points at nothing, and the warm-up install dies trying to mkdir cache/store through a dangling link.

The pre-merge smoke test ran pnpm 12 and Bun: pnpm 12 reaches the cache through absolute environment paths (PNPM_HOME, PNPM_CONFIG_CACHE_DIR), never touching the symlink, and Bun's re-downloading node_modules row recreates the directory as a side effect. pnpm 11 is the one manager that both goes through the symlink and skips the re-download — exactly the intersection the test didn't cover.

Fix

Create cwd/cache (idempotently) before symlinking at it. Verified by running pnpm 11 through the full real scenario chain against a local pnpr behind the emulated link: it completes, and its update lands at 1.8s against the 7.5s the cold-cache artifact used to charge it — the pnpm 11 column benefits from #63 just as pnpm 12 does.

Also included: when every measuring job fails, the report job's samples directory never comes into existence, and merge-results died with a raw ENOENT scandir instead of its own explanation. A missing directory is the same situation as an empty one and now produces the same "every measuring job must have failed" error.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

@zkochan, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 11 minutes

Limit details: You’ve used all 2 included reviews currently available.

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 @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

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 within each organization.

For paid Pro and Pro+ reviews, CodeRabbit uses a developer's included PR review attempts over the past 7 days to set the current hourly allowance. At typical activity levels, the full plan allowance applies. Higher sustained activity can lower the allowance until earlier attempts leave the 7-day window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: a2719cfc-037e-43e4-92ff-41c906629b4f

📥 Commits

Reviewing files that changed from the base of the PR and between 2d557be and 6b29db7.

📒 Files selected for processing (2)

• benchmarkFixture.js
• mergeResults.js

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[qodo-code-review]

PR Summary by Qodo

Prevent dangling re-warm cache links and clarify missing results

🐞 Bug fix 🕐 10-20 Minutes

AI Description

• Ensure re-warm cache targets exist before symlinking, preventing pnpm 11 warm-up failures.
• Treat missing result samples like empty uploads, returning the intended all-jobs-failed error.

Diagram

graph TD
  subgraph Benchmark
    A["Cache cleanup"] --> B["Create cache"] --> C["Cache symlink"] --> D["Warm install"]
  end
  subgraph Reporting
    E{"Samples exist?"} -->|Yes| F["Merge runs"]
    E -->|No| G["Clear failure"]
  end

Loading

High-Level Assessment

The current approach is appropriate: recursive mkdir is idempotent and avoids a pre-check race before symlink creation, while catching only ENOENT preserves unexpected filesystem failures. Broader abstractions would add complexity without improving these localized paths.

Files changed (2) +18 / -1

Bug fix (2) +18 / -1

benchmarkFixture.jsCreate the cache target before re-warm symlinking +6/-0

Create the cache target before re-warm symlinking

• Recreates the benchmark cache directory idempotently before linking it into the temporary re-warm project. This prevents pnpm 11 from traversing a dangling cache symlink after prior scenarios remove the directory without recreating it.

benchmarkFixture.js

mergeResults.jsHandle missing sample directories as empty results +12/-1

Handle missing sample directories as empty results

• Converts an ENOENT while reading the samples directory into an empty entry list. The existing no-runs validation then emits the intended explanation that all measuring jobs failed, while other filesystem errors still propagate.

mergeResults.js