The relative gate calibrated once before the scenarios, so load that
started afterwards (the push and pull_request runs of one commit overlap
on the same runner, next to the test, build and image jobs) slowed the
scenarios but not the yardstick. A pull_request run failed `two years`
at 1653.1 ms against a 1648.7 ms budget this way.
The spec now repeats the calibration after the protocol and uses the
slower of the two values. The budgets can only loosen, never tighten,
and the log reprints the table with the budgets actually applied when
they change.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
After #302, the development machine's protocol takes about 41 seconds
instead of 232. The budget multipliers now add about 1.5 times headroom
to the new ratios (0.1 for typed obligations, which leaves room for
runner jitter), and the hang guard drops from 600 to 240 seconds, with
a 210-second test timeout: about three times the expected run on
srvhome, which is about 1.9 times slower than nessar-desktop.
Part of #286 and #302. The gate still needs a passing CI benchmark on
two runners.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The differential check's frozen bundle will be rebuilt for #302; the
relative gate keeps timing the unchanged a871f8f bundle so its ratios
stay comparable with the recorded history.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Replace Vitest with bun test across the whole suite. A preload compiles
components through Angular's JIT transform with inlined templates, sets
up jsdom, TestBed and zoneless change detection, and a small shim covers
the Vitest APIs bun:test lacks. Files run isolated, real-worker specs
spawn the running Bun, and line coverage from lcov is converted to the
existing HTML and Cobertura reports. The CI benchmark replaces its heap
cap with a container memory limit. Vitest, Jest, Nx and their Vite
configuration are removed.
Refs #291
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Print the calibration and a Markdown table with one aligned row per
scenario as soon as it finishes: both planning times, mean, budget, ratio,
multiplier, outcome, placements, determinism and budget status. Write
directly to stdout so Vitest adds no per-line headers. The complete TSV
report still follows the table.
Refs #281, #285
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The first PR run landed on srvhome, about 1.9 times slower than
nessar-desktop. The fixed 300-second timeout killed its protocol during the
loaded year, although every completed scenario was within its relative
budget. Raise the CI timeout to 600 seconds and the test timeout to 570
seconds so they only guard against hangs, and record the first CI-runner
ratios.
Refs #285
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Before the protocol, time the frozen engine on the two-year scenario, best
of three, and fail any scenario whose mean planning time exceeds its
multiplier times that calibration. Multipliers come from development-machine
ratios with about 1.5x headroom until #286 tightens them. The 300-second
timeout stays only as a hang guard. The benchmark keeps running on push and
pull_request events because the push run gates image publication.
This replaces the absolute 300-second acceptance protocol named in #78.
Refs #285
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Print the report header before the measured runs and each scenario's rows
as soon as its repetitions complete, so a timed-out job still shows the
finished scenarios. The final report, scenarios, warm-ups, repetitions,
assertions, heap limit and timeout are unchanged.
Refs #281
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reduce objective evaluation and calendar allocation work while preserving planner outcomes and deterministic explanations. Restore required benchmark gating and record the unchanged protocol, differential checks, and browser evidence for #78.
Evaluate explicit, ordered optimization configurations through one shared
planner model. Add six editable presets, durable one-time legacy migration,
and consistent completed and cancelled plan provenance.
Include functional regressions and the agreed trace-removal and benchmark
follow-up specs. Performance acceptance remains in #78 under unchanged gates.
Refs: #70, #71, #72, #73, #74, #75, #76, #77, #78
- Introduced calculationDate in various service interfaces and snapshots to track the date of planning calculations.
- Added futureEntitlement and includedEntitlement fields to the translation service and respective language files for better user feedback.
- Updated leave-planner.worker to handle calculationDate and declaredDays in requests.
- Enhanced tests to ensure proper handling of calculation dates and artifact retention during planning processes.
- Improved snapshot and explanation outputs to include calculation date for clarity in planning results.
- Implemented PlanningDurationService to provide an indicative wait time before calculation based on user input and calendar settings.
- Introduced duration estimation logic that considers horizon dates, balances, and rules without triggering a calculation.
- Enhanced user input component to display estimated duration using the new service.
- Updated translation files to include new duration estimate strings in multiple languages.
- Added tests for PlanningDurationService to ensure accurate duration estimation and calibration.
- Refactored leave planner service to remove hardcoded deadlines and improve worker responsiveness.
Implement plan explanations, original decision traces, calendar replay, and worker counterfactual analysis. Preserve incomplete checkpoints and conflict reasons, with accessible navigation and localized comparison details.
Keep planning outcomes independent of tracing and bound worker delivery with receiver acknowledgements. Retain complete candidate history while releasing temporary buffers and benchmark references.
Validated with 201 tests, app/spec type checks, lint, production build, baseline outcome comparisons, and repeated loaded browser measurements. Document performance and remaining limits.
nessarandClaude Sonnet 5.5