[ryansolid]

fix(signals): a store key's delete decrements the key count

What

deleteProperty in packages/signals/src/store/store.ts now decrements the store's key count (kc) when an own key of the draft leaves: a plain clone's key, or an overlay's own add. An overlay's delete of a committed key is unchanged: recorded in del and counted at the rebuild, as before.

Why

kc gates the overlay draft (an owned plain object above 32 keys drafts as an overlay over its committed backing) and caps the rebuild. The set trap incremented it for every new key; nothing decremented it on a delete outside the overlay rebuild paths. So the natural keyed-selection shape — a store holding selected[id], written as "delete the previous key, set the new one" — drifted the count up by one per write. After 32 selections a three-key object crossed the overlay threshold and, since any delete makes an overlay rebuild its backing (#3689), every commit after paid a full clone of the object. The user's code was right; the store miscounted it. Any keyed store that adds and deletes keys over time (a selection map, a cache, a set of open rows) hit the same drift.

Found on the Octane js-framework fixture while asking why Solid's selection cost three times Vue Vapor's compiled selector: the profile showed materializePB (the overlay rebuild) on a store that should never have been an overlay.

Measured

Call profiler on the unminified fixture, Solid on next, the fixture's original delete form, µs per select click:

	wall	store write path
before	5.47	materializePB 1.37, ensurePB 0.32, storeSetter 1.69 incl
after	3.38	storeSetter 0.89 incl, ensurePB 0.22, settle 0.18
Vue Vapor (compiled selector)	1.65	—

Octane harness, all targets rerun: select 7.2 → 6.2 µs, select_lots 4.4 → 3.7 µs (Octane 5.6 / 3.7, Vapor 2.8 / 1.8); the suite's Solid ÷ Octane geomean 1.19 → 1.09. The other ops are within run-to-run noise. The remaining gap to Vapor is the transaction itself against a map lookup, which is the 2.0 selection-through-store design, not this bug.

Tests

tests/store/delete-key-count.test.ts: 200 delete-then-set selections stay at two keys and never draft as an overlay; delete and re-add in one draft; delete of a never-committed key; an overlay's own add deleted again, and an overlay's committed key deleted (counted at the rebuild). The #3689 overlay rebuild tests pass unchanged. Signals (5343), solid (869), web default (1379), server (1719) and hydrate (509) suites pass.

🤖 Generated with Claude Code

[changeset-bot]

🦋 Changeset detected

Latest commit: bcd6e02

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 14 packages

Name	Type
@solidjs/signals	Patch
test-integration	Patch
@solidjs/web	Patch
@solidjs/babel-plugin	Patch
@solidjs/compiler	Patch
@solidjs/diagnostics	Patch
@solidjs/element	Patch
@solidjs/h	Patch
@solidjs/html	Patch
solid-js	Patch
@solidjs/universal	Patch
board-example	Patch
server-dbmon-example	Patch
todos-server-example	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[github-actions]

Size (brotli, eager entry chunk)

scenario	head	vs base	minified vs base	minified vs recorded	cap		lazy chunks (not counted)
signals: core floor (createSignal/Memo/Effect/Root/flush)	7.25 KB	0 B	0 B	+4 B	7.25 KB	⚠️ over by 3 B, 16 B minified headroom
signals: + createStore	14.79 KB	+27 B (+0.2%)	+33 B	+4 B	14.82 KB	✅
signals: + isPending/latest	9.96 KB	0 B	0 B	+55 B	9.96 KB	✅
signals: + until/refresh (truth readers, no optimism)	7.85 KB	0 B	0 B	+4 B	7.86 KB	✅
app: render + one signal (the simple-app floor)	9.55 KB	0 B	0 B	+4 B	9.56 KB	✅
app: hydrating (no stores) with Show/For/Loading/Errored/lazy	18.20 KB	0 B	0 B	+4 B	18.19 KB	⚠️ over by 10 B, 16 B minified headroom	lazy-page.js 0.04 KB
app: hydrating + every store primitive family	30.12 KB	−9 B (−0.0%)	+33 B	0 B	30.14 KB	✅	lazy-page.js 0.04 KB
app: CSR with Show/For/Loading/Errored/lazy	13.17 KB	0 B	0 B	+4 B	13.14 KB	⚠️ over by 27 B, 16 B minified headroom	lazy-page.js 0.04 KB
app: CSR, observe tier (same app on the observe artifacts)	14.88 KB	0 B	0 B	+4 B	14.91 KB	✅	lazy-page.js 0.04 KB
app: CSR, observe tier + attribution engine enabled	29.48 KB	0 B	0 B	+4 B	29.48 KB	✅	lazy-page.js 0.04 KB
app: compiled floor (one template, one text hole, one delegated click)	9.76 KB	0 B	0 B	+6 B	9.77 KB	✅
app: compiled CSR (JSX todo app: spread/merge/omit, events, class/style, keyed For, Show, Loading + lazy, store)	25.95 KB	+1 B (+0.0%)	+33 B	+4 B	26.03 KB	✅	stats.js 0.18 KB
app: compiled hydrating (the same JSX todo app through hydrate(), compiled hydratable)	31.96 KB	−13 B (−0.0%)	+33 B	+4 B	31.98 KB	✅	stats.js 0.20 KB
frames: eager client consumer (frames client + transport, lazy codec)	12.23 KB	0 B	0 B	0 B	12.25 KB	✅
page: base server components (hydrating + dynamic + frames + sf reference)	33.40 KB	0 B	0 B	+4 B	33.46 KB	✅	assets.js 0.78 KB, decode.js 6.25 KB, lazy-page.js 0.04 KB, regions.js 1.07 KB, trace.js 8.99 KB, wire.js 1.33 KB
page: live server components (base + live/GET + action + isPending/latest)	37.63 KB	0 B	0 B	0 B	37.65 KB	✅	assets.js 0.78 KB, decode.js 6.25 KB, lazy-page.js 0.04 KB, regions.js 1.07 KB, trace.js 8.99 KB, wire.js 1.33 KB
page: compiled base server components (the base page as JSX: templates with class/style/attributes/events, For/Show; no spread)	35.35 KB	0 B	0 B	+4 B	35.42 KB	✅	assets.js 0.78 KB, decode.js 6.25 KB, regions.js 1.06 KB, sc-comments.js 0.20 KB, trace.js 9.01 KB, wire.js 1.33 KB
page: compiled live server components (the compiled base page + live/GET + action + isPending/latest)	41.43 KB	0 B	0 B	+54 B	41.48 KB	✅	eager (counted): web.js 21.55 KB; assets.js 0.78 KB, decode.js 6.25 KB, regions.js 1.07 KB, sc-comments.js 0.19 KB, trace.js 9.03 KB, wire.js 1.33 KB
page: base + router (base page + @solidjs/router: createRouter, two routes, preload, useNavigate)	40.87 KB	0 B	0 B	+7 B	40.90 KB	✅	assets.js 0.78 KB, decode.js 6.25 KB, lazy-page.js 0.04 KB, regions.js 1.07 KB, server.js 1.02 KB, serverForms.js 3.60 KB, trace.js 9.05 KB, wire.js 1.33 KB
page: live + router (live page + @solidjs/router: createRouter, two routes, preload, useNavigate)	47.09 KB	0 B	0 B	+1 B	47.05 KB	⚠️ over by 37 B, 19 B minified headroom	eager (counted): client.js 26.75 KB; assets.js 0.78 KB, decode.js 6.25 KB, lazy-page.js 0.04 KB, regions.js 1.07 KB, server.js 1.02 KB, serverForms.js 3.30 KB, trace.js 9.05 KB, wire.js 1.34 KB
server: floor (getRequestEvent + isServer)	1.33 KB	0 B	0 B	0 B	1.34 KB	✅
server: renderToString (the server-render floor)	20.62 KB	0 B	0 B	−23 B	20.59 KB	⚠️ over by 34 B, 43 B minified headroom

⚠️ Over the brotli cap within the minified allowance (passes)

• signals: core floor (createSignal/Memo/Effect/Root/flush): over brotli cap by 3 B; minified 19,777 B vs 19,773 B recorded with the cap (+4 B) — 16 B of the 20 B minified allowance left; +0 B minified over this PR's base
• app: hydrating (no stores) with Show/For/Loading/Errored/lazy: over brotli cap by 10 B; minified 53,240 B vs 53,236 B recorded with the cap (+4 B) — 16 B of the 20 B minified allowance left; +0 B minified over this PR's base
• app: CSR with Show/For/Loading/Errored/lazy: over brotli cap by 27 B; minified 36,904 B vs 36,900 B recorded with the cap (+4 B) — 16 B of the 20 B minified allowance left; +0 B minified over this PR's base
• page: live + router (live page + @solidjs/router: createRouter, two routes, preload, useNavigate): over brotli cap by 37 B; minified 141,256 B vs 141,255 B recorded with the cap (+1 B) — 19 B of the 20 B minified allowance left; +0 B minified over this PR's base
• server: renderToString (the server-render floor): over brotli cap by 34 B; minified 72,131 B vs 72,154 B recorded with the cap (−23 B) — 43 B of the 20 B minified allowance left; +0 B minified over this PR's base

_{Bundled with Rolldown (what Vite ships), brotli q11, decimal KB. A scenario fails only when it is over its brotli cap and its minified size is more than 20 B over the minified recorded with the cap; over the cap within that allowance is brotli layout noise and passes with a warning. Caps and their recorded minified in scripts/size/scenarios.js; the floor and page caps in floor-caps.json are frozen (lower only, or Size-Exception: in the PR body). npm run ratchet lowers caps per RC; it never raises one (scripts/size/README.md).}

[coveralls]

Coverage Report for CI Build 39104313478

Coverage remained the same at 77.564%

Details

• Coverage remained the same as the base build.
• Patch coverage: No coverable lines changed in this PR.
• No coverage regressions found.

Uncovered Changes

No uncovered changes found.

Coverage Regressions

No coverage regressions found.

---

Coverage Stats

Relevant Lines:	1339
Covered Lines:	1097
Line Coverage:	81.93%
Relevant Branches:	1050
Covered Branches:	756
Branch Coverage:	72.0%
Branches in Coverage %:	Yes
Coverage Strength:	27.69 hits per line

---

💛 - Coveralls

[codspeed]

Merging this PR will not alter performance

⚠️ 2 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• readAllowed
• propagation:diamond

✅ 196 untouched benchmarks
⏩ 5 skipped benchmarks¹

---

_{Comparing fix/store-delete-key-count (bcd6e02) with next (7093b07)}

Footnotes

1. 5 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩