perf(spend): bound initial ledger layout on long ranges

[steipete]

Long Usage & Spend ranges initially lay out every daily ledger row. Show the newest 30 rows with the existing Show all / Show less control, keep full-period totals and chart data, and group provider-icon tooltips per row. Shared daily/hourly chart styling offsets the change: production code is 71 insertions and 72 deletions.

This adopts the ledger portion of #3998. Thanks @Yuxin-Qiao; co-author credit is in the commit. The Dock activation-policy change remains outside this adoption because this isolated proof does not measure LaunchServices or live scrolling.

Verification

Synthetic 365-day history, debug build, one warm-up excluded:

Measurement	Before	After
Warm view-layout median, 3 samples	248.7 ms	41.1 ms
Model-build median, 10 samples	18.2 ms	19.9 ms
Initial content height	15,675 pt	1,958 pt
Total spend / tokens / requests	$365 / 365,000 / 365	unchanged

The rendered-height regression failed all four baseline iterations and passes after the change. These are layout measurements on a shared host, not scroll-FPS or view-body invocation measurements. Expanded mode still exposes every row.

CODEXBAR_SUPPRESS_TEST_KEYCHAIN_ACCESS=1 CODEXBAR_TEST_CODEX_FILE_ISOLATION=1 CODEXBAR_TEST_SESSION_FILE_ISOLATION=1 CODEXBAR_LONG_RANGE_PROOF_DIR=/tmp/perf-ui-m-proof/after CODEXBAR_SPEND_PROOF_DIR=/tmp/perf-ui-m-proof/after CODEXBAR_COST_PRIVACY_PROOF_DIR=/tmp/perf-ui-m-proof/after/privacy swift test --skip-build --no-parallel --filter 'SpendDashboard|SpendStackedBarChartTests|MenuBarLayout|MenuBarStatusItem|StatusItemLayoutPaceSignatureTests|StatusItemConditionalSignatureTests|StatusItemCombinedMetricPlaceholderTests|SyntheticMenuCardTests|ProviderArchitectureGatekeeperTests|UserFacingLocalizationCoverageTests'
make check

Results: Test run with 540 tests in 41 suites passed; XCTest ran 7 cases with 4 opt-in cases skipped and 0 failures; make check passed with 0 SwiftLint violations. An initial parallel focused run had timing failures in the existing concurrency suite; the repository runner's --no-parallel mode passes the complete selection.

Synthetic before/after renders were inspected. The 30-day, hourly, selected-day hourly, project, session, and overview PNGs are byte-identical. Proof images are retained locally for the coordinator; this lane's permitted REST operations do not include uploads. No running app was relaunched or real account probed.

Independent review reported one missing-import finding; rejected after verifying the existing @testable import CodexBar at the top of the test and the successful build. No accepted actionable findings remain.

The v0.65.0 status-item renderer is unchanged; added width goldens retain 56/49 pt for the default synthetic 16-pt-icon, 50% fixture and 26/19 pt for icon-only (Regular/Tight). The spacing report needs its actual layout settings. PR #3997 is not adopted: its session titles/project labels bypass Hide personal info, and its hourly chart changes from the full range to the latest active day.

Closes #3998
Refs #3997
Refs #3993

[clawsweeper]

Codex review: blocked before merge. Reviewed September 25, 2026, 3:42 PM ET / 19:42 UTC.

ClawSweeper review

What this changes

The PR initially displays the newest 30 daily spend rows with an expand control, groups provider tooltips, shares chart styling, and adds layout and regression coverage.

Merge readiness

⛔ Blocked before merge - 2 items remain

Current main still lays out every daily ledger row. This focused PR addresses that behavior, and the reported native layout measurements support its narrower initial-layout claim. The full test suite remains unreported under repository handoff guidance.

Priority: P2
Reviewed head: 0fb7e87b65e7fae16261ba5539b81c4f206af49c

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	The patch is focused and its native layout measurements support the stated improvement, with full-suite handoff validation still outstanding.
Proof confidence	🐚 platinum hermit (4/6)	Not applicable: The owner-authored PR is exempt from the contributor proof gate. Its reported macOS NSHostingView run exercises the production spend section with 365 synthetic days and observes a shorter, faster initial layout after the change; it does not establish live scroll-frame performance. No stored-data contract changes.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Not applicable	Not applicable: The owner-authored PR is exempt from the contributor proof gate. Its reported macOS NSHostingView run exercises the production spend section with 365 synthetic days and observes a shorter, faster initial layout after the change; it does not establish live scroll-frame performance. No stored-data contract changes.
Evidence reviewed	7 items	Introduced ledger limit: The pinned introduced patch selects the newest 30 summaries initially and keeps complete expansion available. Current main behavior: The pinned main revision renders the entire reversed daily-summary sequence, so the central change is not already on main. Full-period model: The model builds one summary per covered day and feeds it separately from group totals and chart points.
Findings	None	None.
Security	None	None.

How this fits together

CodexBar turns provider usage snapshots into daily spend summaries for the Settings window. Charts, totals, and the ledger consume those summaries; this change limits the rows laid out initially while preserving the full-period data.

flowchart LR
A[Provider usage] --> B[Daily spend summaries]
B --> C[Selected range]
C --> D[Charts and totals]
C --> E[Initial ledger row limit]
E --> F[Settings display]
F --> G[Show all control]

Loading

Before merge

• Resolve merge risk (P1) - The repository requests a full make test run before handoff; the PR reports focused tests and make check, while full-suite completion is not established.
• Complete next step (P2) - Report a full make test run from a keychain-safe macOS setup before merge.

Agent review details

Security

None.

Review metrics

Metric	Value	Why it matters
Synthetic native layout	248.7 → 41.1 ms median; 15,675 → 1,958 pt	The reported before-and-after measurements directly target initial long-range layout.
Code and test delta	production +71/−72 lines; tests +129 lines	The shared chart-style refactor offsets the ledger change's production growth.

Merge-risk options

Maintainer options:

1. Decide the mitigation before merge
   Land the bounded ledger after full-suite validation, with complete accounting retained and the separate Dock performance question handled on its own merits.
2. Pause or close
   Do not merge this PR until maintainers decide whether the risk is worth taking.

Technical review

Best possible solution:

Land the bounded ledger after full-suite validation, with complete accounting retained and the separate Dock performance question handled on its own merits.

Do we have a high-confidence way to reproduce the issue?

Yes: a 365-day synthetic history can exercise the production Settings view in an AppKit host, and current main visibly renders every row. The reviewer did not independently run the macOS layout test.

Is this the best way to solve the issue?

Yes: the existing expand-control pattern gives users access to every row while reducing initial layout work; model totals and charts retain the full range.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against a2cf7face10b.

Labels

Label changes:

• add P2: Long-range Settings layout is a user-facing performance improvement with a limited affected surface.
• add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: The owner-authored PR is exempt from the contributor proof gate. Its reported macOS NSHostingView run exercises the production spend section with 365 synthetic days and observes a shorter, faster initial layout after the change; it does not establish live scroll-frame performance. No stored-data contract changes.

Label justifications:

• P2: Long-range Settings layout is a user-facing performance improvement with a limited affected surface.
• rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: The owner-authored PR is exempt from the contributor proof gate. Its reported macOS NSHostingView run exercises the production spend section with 365 synthetic days and observes a shorter, faster initial layout after the change; it does not establish live scroll-frame performance. No stored-data contract changes.

Evidence

What I checked:

• Introduced ledger limit: The pinned introduced patch selects the newest 30 summaries initially and keeps complete expansion available. (Sources/CodexBar/PreferencesSpendDashboardPane.swift:1126, 0fb7e87b65e7)
• Current main behavior: The pinned main revision renders the entire reversed daily-summary sequence, so the central change is not already on main. (Sources/CodexBar/PreferencesSpendDashboardPane.swift:1165, a2cf7face10b)
• Full-period model: The model builds one summary per covered day and feeds it separately from group totals and chart points. (Sources/CodexBar/SpendDashboardModel.swift:1013, 0fb7e87b65e7)
• Native layout validation: The PR body reports a synthetic 365-day macOS layout median of 248.7 ms before and 41.1 ms after, with initial height falling from 15,675 to 1,958 pt. The opt-in test hosts the production currency section in NSHostingView and checks full-period totals; it does not measure live scrolling. (Tests/CodexBarTests/SpendDashboardLongRangeRenderTests.swift:64, 0fb7e87b65e7)
• Related work: The PR body explicitly adopts the ledger portion of Fix Usage & Spend scroll lag on long ranges #3998 while leaving its Dock change outside this patch. The other related spend PR is closed unmerged, and the menu-spacing report concerns a different behavior.
• Feature history: Path history identifies the original daily-ledger addition and later spend-area contributions; older source-line introduction could not be verified from the available partial Git objects. (Sources/CodexBar/PreferencesSpendDashboardPane.swift, 5e5a9003efb2)

Likely related people:

• Sahil Imtiaz: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)
• Peter Steinberger: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)
• Yuxin-Qiao: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)

Rank-up moves

Optional improvements that raise the rating; they are not merge blockers.

• Report the repository's full make test result from a keychain-safe macOS setup.

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.