cacheable - fix: count falsy values in delete stats

[jaredwray]

Please check if the PR fulfills these requirements

• Followed the Contributing guidelines and Code of Conduct
• Tests for the changes have been added (for bug fixes/features) with 100% code coverage.

What kind of change does this PR introduce? (Bug fix, feature, docs update, ...)

Bug fix. This is a follow-up to #1742. The review of #1742 found this bug.

Problem

When stats are on, Cacheable.delete(key) read the value with get(). It changed the stats only when the value was truthy. Thus a cached 0, "", false, null, or undefined did not decrease count, ksize, or vsize, and did not increase deletes. take(key) calls delete, so it had the same bug.

Example: set("hits", 0), then delete("hits"). After this, stats.count stays at 1.

deleteMany does not have this bug after #1742. So delete and deleteMany gave different stats for the same keys.

Change

packages/cacheable/src/index.ts, deleteInternal:

• It reads the raw entry with this._primary.get(key, { raw: true }). It changes the stats for each entry that is in the cache. It skips a missing key.
• It sizes entry.value, which is the value that set sized.
• It does not use getRaw(). Keyv get() catches a store error, but getRaw() does not. With getRaw(), a failed primary read makes delete() reject before the delete starts.
• I removed the /* v8 ignore next */ comment. The tests now reach both branches.

Verification

• pnpm test passes in packages/cacheable, packages/cache-manager, and packages/flat-cache, with a local Redis. I did not run packages/net, because its tests need the mockhttp container.
• pnpm build passes in packages/cacheable.
• Coverage of cacheable/src/index.ts is the same as on main. The only branch that is not covered is line 422 (the sync setter). This PR does not change it.
• A Fable code review found no defects in the fix. It also ran two mutation checks on the extended test (results below).

Test audit:

• Extended cacheable stats enabled > should get stats on delete. This test owns the stats contract of delete, so I extended it and did not add a near-duplicate.
  • Contract: delete changes the stats for each key that is in the cache, for all values. A missing key does not change the stats.
  • The test deletes keys with the values "value", 0, "", false, null, and undefined, and one missing key. It compares ksize and vsize with a cache that holds only the remaining key.
  • It fails on main (deletes is 1, expected 6). It also fails with get() and an !== undefined check (deletes is 5, expected 6), because of the cached undefined value.
• Added cacheable stats enabled > should delete with stats when the primary store read fails.
  • Contract: a store error on the stats read does not stop delete. This was the behavior on main.
  • It fails if a change uses getRaw() for the stats read ("promise rejected ... instead of resolving"). The first version of this fix used getRaw(), so this regression is credible.
  • No other test covers a failed primary read in delete. The test mocks only the store get, which is the real boundary for this error.

---

Generated by Claude Code

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-18T01:05:38.100843Z	e293224	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (dd4562d) to head (e293224).

Additional details and impacted files

@@            Coverage Diff            @@
##              main     #1743   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           29        29           
  Lines         3615      3620    +5     
  Branches       810       811    +1     
=========================================
+ Hits          3615      3620    +5     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.