niceeval 0.13.1 → 0.13.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/INDEX.md +25 -2
- package/README.md +9 -10
- package/README.zh.md +9 -10
- package/dist/agents/ai-sdk.cjs.map +1 -1
- package/dist/agents/ai-sdk.d.cts +1 -1
- package/dist/agents/ai-sdk.d.mts +1 -1
- package/dist/agents/ai-sdk.d.ts +1 -1
- package/dist/agents/bub-install-spec.cjs +19 -5
- package/dist/agents/bub-install-spec.cjs.map +1 -1
- package/dist/agents/bub-install-spec.d.cts +9 -2
- package/dist/agents/bub-install-spec.d.mts +9 -2
- package/dist/agents/bub-install-spec.d.ts +9 -2
- package/dist/agents/bub-install-spec.js +14 -12
- package/dist/agents/bub.cjs +10 -6
- package/dist/agents/bub.cjs.map +1 -1
- package/dist/agents/bub.d.cts +2 -2
- package/dist/agents/bub.d.mts +2 -2
- package/dist/agents/bub.d.ts +2 -2
- package/dist/agents/claude-code.cjs.map +1 -1
- package/dist/agents/claude-code.d.cts +2 -2
- package/dist/agents/claude-code.d.mts +2 -2
- package/dist/agents/claude-code.d.ts +2 -2
- package/dist/agents/codex.cjs +4 -4
- package/dist/agents/codex.cjs.map +1 -1
- package/dist/agents/codex.d.cts +3 -3
- package/dist/agents/codex.d.mts +3 -3
- package/dist/agents/codex.d.ts +3 -3
- package/dist/agents/coding-cli-versions.cjs +10 -1
- package/dist/agents/coding-cli-versions.cjs.map +1 -1
- package/dist/agents/coding-cli-versions.d.cts +8 -3
- package/dist/agents/coding-cli-versions.d.mts +8 -3
- package/dist/agents/coding-cli-versions.d.ts +8 -3
- package/dist/agents/coding-cli-versions.js +18 -12
- package/dist/agents/deepseek-harness.cjs +151 -0
- package/dist/agents/deepseek-harness.cjs.map +1 -0
- package/dist/agents/deepseek-harness.d.cts +14 -0
- package/dist/agents/deepseek-harness.d.mts +14 -0
- package/dist/agents/deepseek-harness.d.ts +14 -0
- package/dist/agents/deepseek-harness.js +7 -0
- package/dist/agents/deepseek-harness.js.map +1 -0
- package/dist/agents/exact-npm-plugins.cjs +28 -0
- package/dist/agents/exact-npm-plugins.cjs.map +1 -0
- package/dist/agents/exact-npm-plugins.d.cts +10 -0
- package/dist/agents/exact-npm-plugins.d.mts +10 -0
- package/dist/agents/exact-npm-plugins.d.ts +10 -0
- package/dist/agents/exact-npm-plugins.js +8 -0
- package/dist/agents/exact-npm-plugins.js.map +1 -0
- package/dist/agents/index.cjs +5 -1
- package/dist/agents/index.cjs.map +1 -1
- package/dist/agents/index.d.cts +4 -0
- package/dist/agents/index.d.mts +4 -0
- package/dist/agents/index.d.ts +4 -0
- package/dist/agents/index.js +54 -50
- package/dist/agents/index.mjs +54 -50
- package/dist/agents/omp.cjs +170 -0
- package/dist/agents/omp.cjs.map +1 -0
- package/dist/agents/omp.d.cts +14 -0
- package/dist/agents/omp.d.mts +14 -0
- package/dist/agents/omp.d.ts +14 -0
- package/dist/agents/omp.js +7 -0
- package/dist/agents/omp.js.map +1 -0
- package/dist/agents/openclaw.cjs +51 -2
- package/dist/agents/openclaw.cjs.map +1 -1
- package/dist/agents/openclaw.d.cts +2 -0
- package/dist/agents/openclaw.d.mts +2 -0
- package/dist/agents/openclaw.d.ts +2 -0
- package/dist/agents/types.cjs.map +1 -1
- package/dist/agents/types.d.cts +4 -4
- package/dist/agents/types.d.mts +4 -4
- package/dist/agents/types.d.ts +4 -4
- package/dist/agents/ui-message-stream.cjs +24 -7
- package/dist/agents/ui-message-stream.cjs.map +1 -1
- package/dist/analysis/api.cjs +28 -2
- package/dist/analysis/api.cjs.map +1 -1
- package/dist/analysis/api.d.cts +22 -1
- package/dist/analysis/api.d.mts +22 -1
- package/dist/analysis/api.d.ts +22 -1
- package/dist/analysis/api.js +8 -6
- package/dist/analysis/bindings.cjs +11 -18
- package/dist/analysis/bindings.cjs.map +1 -1
- package/dist/analysis/bindings.d.cts +177 -49
- package/dist/analysis/bindings.d.mts +177 -49
- package/dist/analysis/bindings.d.ts +177 -49
- package/dist/analysis/contracts.cjs +1 -0
- package/dist/analysis/contracts.cjs.map +1 -1
- package/dist/analysis/contracts.d.cts +43 -3
- package/dist/analysis/contracts.d.mts +43 -3
- package/dist/analysis/contracts.d.ts +43 -3
- package/dist/analysis/cost-projection.cjs +2 -1
- package/dist/analysis/cost-projection.cjs.map +1 -1
- package/dist/analysis/cost-projection.d.cts +1 -1
- package/dist/analysis/cost-projection.d.mts +1 -1
- package/dist/analysis/cost-projection.d.ts +1 -1
- package/dist/analysis/cost.cjs +13 -3
- package/dist/analysis/cost.cjs.map +1 -1
- package/dist/analysis/cost.d.cts +10 -3
- package/dist/analysis/cost.d.mts +10 -3
- package/dist/analysis/cost.d.ts +10 -3
- package/dist/analysis/experiment-groups.cjs +96 -0
- package/dist/analysis/experiment-groups.cjs.map +1 -0
- package/dist/analysis/experiment-groups.d.cts +9 -0
- package/dist/analysis/experiment-groups.d.mts +9 -0
- package/dist/analysis/experiment-groups.d.ts +9 -0
- package/dist/analysis/experiment-groups.js +10 -0
- package/dist/analysis/experiment-groups.js.map +1 -0
- package/dist/analysis/index.cjs +4 -1
- package/dist/analysis/index.cjs.map +1 -1
- package/dist/analysis/index.d.cts +4 -2
- package/dist/analysis/index.d.mts +4 -2
- package/dist/analysis/index.d.ts +4 -2
- package/dist/analysis/index.js +40 -36
- package/dist/analysis/index.mjs +40 -36
- package/dist/assertions/limits.cjs +4 -0
- package/dist/assertions/limits.cjs.map +1 -1
- package/dist/assertions/limits.d.cts +4 -0
- package/dist/assertions/limits.d.mts +4 -0
- package/dist/assertions/limits.d.ts +4 -0
- package/dist/assertions/match.cjs +63 -25
- package/dist/assertions/match.cjs.map +1 -1
- package/dist/assertions/match.d.cts +7 -0
- package/dist/assertions/match.d.mts +7 -0
- package/dist/assertions/match.d.ts +7 -0
- package/dist/assertions/record/attachment.cjs +32 -4
- package/dist/assertions/record/attachment.cjs.map +1 -1
- package/dist/assertions/record/diff.cjs +10 -4
- package/dist/assertions/record/diff.cjs.map +1 -1
- package/dist/assertions/record/producer.cjs +101 -4
- package/dist/assertions/record/producer.cjs.map +1 -1
- package/dist/assertions/record/producer.d.cts +4 -1
- package/dist/assertions/record/producer.d.mts +4 -1
- package/dist/assertions/record/producer.d.ts +4 -1
- package/dist/assertions/runtime.cjs +45 -5
- package/dist/assertions/runtime.cjs.map +1 -1
- package/dist/cli/record.cjs +113 -8
- package/dist/cli/record.cjs.map +1 -1
- package/dist/cli/record.d.cts +20 -1
- package/dist/cli/record.d.mts +20 -1
- package/dist/cli/record.d.ts +20 -1
- package/dist/cli/record.js +4 -2
- package/dist/cli.cjs +386 -31
- package/dist/cli.cjs.map +1 -1
- package/dist/cli.d.cts +10 -2
- package/dist/cli.d.mts +10 -2
- package/dist/cli.d.ts +10 -2
- package/dist/context/send-retry.cjs +4 -4
- package/dist/context/send-retry.cjs.map +1 -1
- package/dist/coordination/host/runtime.cjs +8 -30
- package/dist/coordination/host/runtime.cjs.map +1 -1
- package/dist/coordination/host/runtime.d.cts +3 -3
- package/dist/coordination/host/runtime.d.mts +3 -3
- package/dist/coordination/host/runtime.d.ts +3 -3
- package/dist/coordination/host/types.cjs.map +1 -1
- package/dist/coordination/host/types.d.cts +0 -5
- package/dist/coordination/host/types.d.mts +0 -5
- package/dist/coordination/host/types.d.ts +0 -5
- package/dist/coordination/platform/node.cjs +90 -9
- package/dist/coordination/platform/node.cjs.map +1 -1
- package/dist/define.cjs +25 -1
- package/dist/define.cjs.map +1 -1
- package/dist/experiment/host/index.cjs +1 -0
- package/dist/experiment/host/index.cjs.map +1 -1
- package/dist/i18n/en.cjs +26 -8
- package/dist/i18n/en.cjs.map +1 -1
- package/dist/i18n/en.d.cts +19 -6
- package/dist/i18n/en.d.mts +19 -6
- package/dist/i18n/en.d.ts +19 -6
- package/dist/index.cjs.map +1 -1
- package/dist/index.d.cts +1 -1
- package/dist/index.d.mts +1 -1
- package/dist/index.d.ts +1 -1
- package/dist/o11y/parsers/hermes.cjs +1 -1
- package/dist/o11y/parsers/hermes.cjs.map +1 -1
- package/dist/o11y/prices.json +1 -1
- package/dist/o11y/record/families.cjs +12 -3
- package/dist/o11y/record/families.cjs.map +1 -1
- package/dist/o11y/record/families.d.cts +8 -8
- package/dist/o11y/record/families.d.mts +8 -8
- package/dist/o11y/record/families.d.ts +8 -8
- package/dist/o11y/record/family-projectors.d.cts +66 -18
- package/dist/o11y/record/family-projectors.d.mts +66 -18
- package/dist/o11y/record/family-projectors.d.ts +66 -18
- package/dist/o11y/record/family-writers.d.cts +11 -3
- package/dist/o11y/record/family-writers.d.mts +11 -3
- package/dist/o11y/record/family-writers.d.ts +11 -3
- package/dist/o11y/record/runner-producer.cjs +14 -4
- package/dist/o11y/record/runner-producer.cjs.map +1 -1
- package/dist/record/codec/core.d.cts +2 -2
- package/dist/record/codec/core.d.mts +2 -2
- package/dist/record/codec/core.d.ts +2 -2
- package/dist/record/family/catalog.d.cts +44 -12
- package/dist/record/family/catalog.d.mts +44 -12
- package/dist/record/family/catalog.d.ts +44 -12
- package/dist/record/family/index.js +128 -124
- package/dist/record/family/observability-v1.cjs +35 -0
- package/dist/record/family/observability-v1.cjs.map +1 -0
- package/dist/record/family/observability-v1.d.cts +3 -0
- package/dist/record/family/observability-v1.d.mts +3 -0
- package/dist/record/family/observability-v1.d.ts +3 -0
- package/dist/record/family/observability-v1.js +6 -0
- package/dist/record/family/observability-v1.js.map +1 -0
- package/dist/record/family/observability.cjs +58 -9
- package/dist/record/family/observability.cjs.map +1 -1
- package/dist/record/family/observability.d.cts +1014 -15
- package/dist/record/family/observability.d.mts +1014 -15
- package/dist/record/family/observability.d.ts +1014 -15
- package/dist/record/family/observability.js +22 -18
- package/dist/record/host/runtime.cjs +600 -31
- package/dist/record/host/runtime.cjs.map +1 -1
- package/dist/record/host/types.cjs.map +1 -1
- package/dist/record/host/types.d.cts +33 -9
- package/dist/record/host/types.d.mts +33 -9
- package/dist/record/host/types.d.ts +33 -9
- package/dist/record/maintenance/runtime.cjs +1 -2
- package/dist/record/maintenance/runtime.cjs.map +1 -1
- package/dist/record/model/core.cjs.map +1 -1
- package/dist/record/model/core.d.cts +1 -1
- package/dist/record/model/core.d.mts +1 -1
- package/dist/record/model/core.d.ts +1 -1
- package/dist/record/model/definition.d.cts +6 -6
- package/dist/record/model/definition.d.mts +6 -6
- package/dist/record/model/definition.d.ts +6 -6
- package/dist/record/model/identifiers.cjs +1 -1
- package/dist/record/model/identifiers.cjs.map +1 -1
- package/dist/record/model/identifiers.d.cts +1 -1
- package/dist/record/model/identifiers.d.mts +1 -1
- package/dist/record/model/identifiers.d.ts +1 -1
- package/dist/record/platform/node.cjs +159 -23
- package/dist/record/platform/node.cjs.map +1 -1
- package/dist/record/platform/services.cjs.map +1 -1
- package/dist/record/platform/services.d.cts +15 -2
- package/dist/record/platform/services.d.mts +15 -2
- package/dist/record/platform/services.d.ts +15 -2
- package/dist/record/reader/errors.cjs +37 -1
- package/dist/record/reader/errors.cjs.map +1 -1
- package/dist/record/reader/errors.d.cts +49 -1
- package/dist/record/reader/errors.d.mts +49 -1
- package/dist/record/reader/errors.d.ts +49 -1
- package/dist/record/reader/errors.js +22 -12
- package/dist/record/reader/format.cjs +76 -17
- package/dist/record/reader/format.cjs.map +1 -1
- package/dist/record/reader/format.d.cts +11 -4
- package/dist/record/reader/format.d.mts +11 -4
- package/dist/record/reader/format.d.ts +11 -4
- package/dist/record/reader/format.js +2 -0
- package/dist/record/reader/index.js +32 -18
- package/dist/record/reader/runtime.cjs +64 -0
- package/dist/record/reader/runtime.cjs.map +1 -1
- package/dist/record/reader/runtime.d.cts +15 -0
- package/dist/record/reader/runtime.d.mts +15 -0
- package/dist/record/reader/runtime.d.ts +15 -0
- package/dist/record/reader/runtime.js +2 -0
- package/dist/report/assets/styles.css +53 -29
- package/dist/report/built-in/analysis-values.cjs +21 -13
- package/dist/report/built-in/analysis-values.cjs.map +1 -1
- package/dist/report/built-in/analysis-values.d.cts +11 -22
- package/dist/report/built-in/analysis-values.d.mts +11 -22
- package/dist/report/built-in/analysis-values.d.ts +11 -22
- package/dist/report/built-in/analysis-values.js +4 -6
- package/dist/report/built-in/index.cjs +1 -3
- package/dist/report/built-in/index.cjs.map +1 -1
- package/dist/report/built-in/index.d.cts +1 -1
- package/dist/report/built-in/index.d.mts +1 -1
- package/dist/report/built-in/index.d.ts +1 -1
- package/dist/report/built-in/index.js +4 -8
- package/dist/report/built-in/index.mjs +4 -8
- package/dist/report/built-in/machine.cjs +0 -2
- package/dist/report/built-in/machine.cjs.map +1 -1
- package/dist/report/built-in/machine.d.cts +0 -2
- package/dist/report/built-in/machine.d.mts +0 -2
- package/dist/report/built-in/machine.d.ts +0 -2
- package/dist/report/built-in/result-components.cjs +49 -51
- package/dist/report/built-in/result-components.cjs.map +1 -1
- package/dist/report/built-in/result-components.d.cts +16 -17
- package/dist/report/built-in/result-components.d.mts +16 -17
- package/dist/report/built-in/result-components.d.ts +16 -17
- package/dist/report/built-in/result-components.js +0 -4
- package/dist/report/built-in/run-membership-overview.cjs +126 -17
- package/dist/report/built-in/run-membership-overview.cjs.map +1 -1
- package/dist/report/built-in/run-membership-overview.d.cts +14 -0
- package/dist/report/built-in/run-membership-overview.d.mts +14 -0
- package/dist/report/built-in/run-membership-overview.d.ts +14 -0
- package/dist/report/built-in/run-membership-overview.js +4 -2
- package/dist/report/built-in/standard.cjs +80 -45
- package/dist/report/built-in/standard.cjs.map +1 -1
- package/dist/report/built-in/standard.d.cts +19 -41
- package/dist/report/built-in/standard.d.mts +19 -41
- package/dist/report/built-in/standard.d.ts +19 -41
- package/dist/report/built-in/standard.js +6 -6
- package/dist/report/components/attempt-detail/compute.cjs +78 -2
- package/dist/report/components/attempt-detail/compute.cjs.map +1 -1
- package/dist/report/components/attempt-detail/compute.d.cts +1 -1
- package/dist/report/components/attempt-detail/compute.d.mts +1 -1
- package/dist/report/components/attempt-detail/compute.d.ts +1 -1
- package/dist/report/components/attempt-detail/content.cjs +1 -1
- package/dist/report/components/attempt-detail/content.cjs.map +1 -1
- package/dist/report/components/entity-lists/compute.cjs +42 -5
- package/dist/report/components/entity-lists/compute.cjs.map +1 -1
- package/dist/report/components/entity-lists/compute.d.cts +10 -8
- package/dist/report/components/entity-lists/compute.d.mts +10 -8
- package/dist/report/components/entity-lists/compute.d.ts +10 -8
- package/dist/report/components/entity-lists/content.cjs +31 -6
- package/dist/report/components/entity-lists/content.cjs.map +1 -1
- package/dist/report/components/entity-lists/index.cjs +9 -4
- package/dist/report/components/entity-lists/index.cjs.map +1 -1
- package/dist/report/components/entity-lists/index.d.cts +4 -7
- package/dist/report/components/entity-lists/index.d.mts +4 -7
- package/dist/report/components/entity-lists/index.d.ts +4 -7
- package/dist/report/components/summaries/compute.cjs +8 -1
- package/dist/report/components/summaries/compute.cjs.map +1 -1
- package/dist/report/components/summaries/compute.d.cts +9 -1
- package/dist/report/components/summaries/compute.d.mts +9 -1
- package/dist/report/components/summaries/compute.d.ts +9 -1
- package/dist/report/components/summaries/index.cjs +48 -4
- package/dist/report/components/summaries/index.cjs.map +1 -1
- package/dist/report/components/summaries/index.d.cts +7 -4
- package/dist/report/components/summaries/index.d.mts +7 -4
- package/dist/report/components/summaries/index.d.ts +7 -4
- package/dist/report/definition/cell.cjs +3 -0
- package/dist/report/definition/cell.cjs.map +1 -1
- package/dist/report/definition/cell.d.cts +2 -0
- package/dist/report/definition/cell.d.mts +2 -0
- package/dist/report/definition/cell.d.ts +2 -0
- package/dist/report/definition/primitives/chart.cjs +8 -5
- package/dist/report/definition/primitives/chart.cjs.map +1 -1
- package/dist/report/definition/primitives/points-dataset.cjs +1 -0
- package/dist/report/definition/primitives/points-dataset.cjs.map +1 -1
- package/dist/report/definition/primitives.cjs +15 -2
- package/dist/report/definition/primitives.cjs.map +1 -1
- package/dist/report/definition/report.cjs +27 -3
- package/dist/report/definition/report.cjs.map +1 -1
- package/dist/report/definition/report.d.cts +8 -0
- package/dist/report/definition/report.d.mts +8 -0
- package/dist/report/definition/report.d.ts +8 -0
- package/dist/report/execution/machine.cjs +30 -14
- package/dist/report/execution/machine.cjs.map +1 -1
- package/dist/report/execution/machine.d.cts +20 -10
- package/dist/report/execution/machine.d.mts +20 -10
- package/dist/report/execution/machine.d.ts +20 -10
- package/dist/report/execution/machine.js +6 -6
- package/dist/report/host/execute.cjs +25 -16
- package/dist/report/host/execute.cjs.map +1 -1
- package/dist/report/host/execute.d.cts +4 -0
- package/dist/report/host/execute.d.mts +4 -0
- package/dist/report/host/execute.d.ts +4 -0
- package/dist/report/host/from-record.cjs +22 -5
- package/dist/report/host/from-record.cjs.map +1 -1
- package/dist/report/host/machine.cjs +53 -11
- package/dist/report/host/machine.cjs.map +1 -1
- package/dist/report/host/site-runtime.cjs +9 -8
- package/dist/report/host/site-runtime.cjs.map +1 -1
- package/dist/report/host/site-runtime.d.cts +1 -1
- package/dist/report/host/site-runtime.d.mts +1 -1
- package/dist/report/host/site-runtime.d.ts +1 -1
- package/dist/report/host/static.cjs +56 -9
- package/dist/report/host/static.cjs.map +1 -1
- package/dist/report/index.cjs.map +1 -1
- package/dist/report/index.d.cts +1 -1
- package/dist/report/index.d.mts +1 -1
- package/dist/report/index.d.ts +1 -1
- package/dist/report/model/dataset.cjs +7 -2
- package/dist/report/model/dataset.cjs.map +1 -1
- package/dist/report/model/locale.cjs +2 -0
- package/dist/report/model/locale.cjs.map +1 -1
- package/dist/report/model/locale.d.cts +1 -0
- package/dist/report/model/locale.d.mts +1 -0
- package/dist/report/model/locale.d.ts +1 -0
- package/dist/report/model/metrics.cjs +3 -1
- package/dist/report/model/metrics.cjs.map +1 -1
- package/dist/report/model/pricing.cjs +16 -11
- package/dist/report/model/pricing.cjs.map +1 -1
- package/dist/report/react/styles.css +53 -29
- package/dist/runner/adoption.cjs +1 -0
- package/dist/runner/adoption.cjs.map +1 -1
- package/dist/runner/attempt.cjs +44 -5
- package/dist/runner/attempt.cjs.map +1 -1
- package/dist/runner/attempt.d.cts +13 -1
- package/dist/runner/attempt.d.mts +13 -1
- package/dist/runner/attempt.d.ts +13 -1
- package/dist/runner/config-identity.cjs +10 -2
- package/dist/runner/config-identity.cjs.map +1 -1
- package/dist/runner/config-identity.d.cts +4 -2
- package/dist/runner/config-identity.d.mts +4 -2
- package/dist/runner/config-identity.d.ts +4 -2
- package/dist/runner/feedback/human.cjs +298 -59
- package/dist/runner/feedback/human.cjs.map +1 -1
- package/dist/runner/feedback/json.cjs +14 -6
- package/dist/runner/feedback/json.cjs.map +1 -1
- package/dist/runner/feedback/json.d.cts +3 -2
- package/dist/runner/feedback/json.d.mts +3 -2
- package/dist/runner/feedback/json.d.ts +3 -2
- package/dist/runner/feedback/reducer.cjs +2 -0
- package/dist/runner/feedback/reducer.cjs.map +1 -1
- package/dist/runner/judge-config.cjs +1 -1
- package/dist/runner/judge-config.cjs.map +1 -1
- package/dist/runner/ledger.cjs +124 -12
- package/dist/runner/ledger.cjs.map +1 -1
- package/dist/runner/project-current.cjs +13 -5
- package/dist/runner/project-current.cjs.map +1 -1
- package/dist/runner/run.cjs +564 -256
- package/dist/runner/run.cjs.map +1 -1
- package/dist/runner/sandbox-pool.cjs +161 -30
- package/dist/runner/sandbox-pool.cjs.map +1 -1
- package/dist/runner/sandbox-pool.d.cts +34 -2
- package/dist/runner/sandbox-pool.d.mts +34 -2
- package/dist/runner/sandbox-pool.d.ts +34 -2
- package/dist/runner/shared-state-lease.cjs +803 -0
- package/dist/runner/shared-state-lease.cjs.map +1 -0
- package/dist/runner/shared-state-lease.d.cts +121 -0
- package/dist/runner/shared-state-lease.d.mts +121 -0
- package/dist/runner/shared-state-lease.d.ts +121 -0
- package/dist/runner/shared-state-lease.js +24 -0
- package/dist/runner/shared-state-lease.js.map +1 -0
- package/dist/runner/teardown-registry.cjs +97 -34
- package/dist/runner/teardown-registry.cjs.map +1 -1
- package/dist/runner/teardown-registry.d.cts +27 -1
- package/dist/runner/teardown-registry.d.mts +27 -1
- package/dist/runner/teardown-registry.d.ts +27 -1
- package/dist/runner/teardown-registry.js +20 -16
- package/dist/runner/types.cjs.map +1 -1
- package/dist/runner/types.d.cts +31 -5
- package/dist/runner/types.d.mts +31 -5
- package/dist/runner/types.d.ts +31 -5
- package/dist/sample/analysis.cjs +249 -459
- package/dist/sample/analysis.cjs.map +1 -1
- package/dist/sample/capability.cjs +138 -7
- package/dist/sample/capability.cjs.map +1 -1
- package/dist/sample/capability.d.cts +3 -1
- package/dist/sample/capability.d.mts +3 -1
- package/dist/sample/capability.d.ts +3 -1
- package/dist/sample/capability.js +8 -6
- package/dist/sandbox/docker-agent-image.cjs +5 -1
- package/dist/sandbox/docker-agent-image.cjs.map +1 -1
- package/dist/sandbox/docker-agent-image.d.cts +2 -0
- package/dist/sandbox/docker-agent-image.d.mts +2 -0
- package/dist/sandbox/docker-agent-image.d.ts +2 -0
- package/dist/sandbox/docker-agent-image.js +14 -10
- package/dist/sandbox/docker-profile/schema.cjs +283 -534
- package/dist/sandbox/docker-profile/schema.cjs.map +1 -1
- package/dist/sandbox/docker-profile/schema.d.cts +10 -27
- package/dist/sandbox/docker-profile/schema.d.mts +10 -27
- package/dist/sandbox/docker-profile/schema.d.ts +10 -27
- package/dist/sandbox/e2b-agent-template.cjs +5 -2
- package/dist/sandbox/e2b-agent-template.cjs.map +1 -1
- package/dist/sandbox/index.cjs +4 -2
- package/dist/sandbox/index.cjs.map +1 -1
- package/dist/sandbox/index.d.cts +1 -1
- package/dist/sandbox/index.d.mts +1 -1
- package/dist/sandbox/index.d.ts +1 -1
- package/dist/sandbox/index.js +120 -116
- package/dist/sandbox/index.mjs +120 -116
- package/dist/sandbox/keep-registry.cjs +28 -66
- package/dist/sandbox/keep-registry.cjs.map +1 -1
- package/dist/sandbox/types.cjs.map +1 -1
- package/dist/sandbox/types.d.cts +1 -1
- package/dist/sandbox/types.d.mts +1 -1
- package/dist/sandbox/types.d.ts +1 -1
- package/dist/shared/aggregate.cjs +16 -6
- package/dist/shared/aggregate.cjs.map +1 -1
- package/dist/shared/aggregate.d.cts +12 -4
- package/dist/shared/aggregate.d.mts +12 -4
- package/dist/shared/aggregate.d.ts +12 -4
- package/dist/shared/turn-label.cjs +20 -2
- package/dist/shared/turn-label.cjs.map +1 -1
- package/dist/shared/turn-label.d.cts +2 -0
- package/dist/shared/turn-label.d.mts +2 -0
- package/dist/shared/turn-label.d.ts +2 -0
- package/dist/shared/turn-label.js +4 -2
- package/dist/vendor/remark/remark-gfm.cjs +7 -7
- package/dist/vendor/remark/remark-gfm.cjs.map +1 -1
- package/dist/vendor/remark/remark-parse.cjs +12 -12
- package/dist/vendor/remark/remark-parse.cjs.map +1 -1
- package/dist/vendor/remark/remark.cjs +13 -13
- package/dist/vendor/remark/remark.cjs.map +1 -1
- package/dist/view/server.cjs +6 -0
- package/dist/view/server.cjs.map +1 -1
- package/docs-site/images/hitl-handshake-zh.svg +3 -3
- package/docs-site/zh/examples/ai-agent-application.mdx +1 -1
- package/docs-site/zh/examples/coding-agent-extensions.mdx +1 -1
- package/docs-site/zh/examples/integrations/ai-sdk-v7.mdx +4 -3
- package/docs-site/zh/examples/integrations/claude-sdk.mdx +3 -3
- package/docs-site/zh/examples/integrations/codex-sdk.mdx +5 -5
- package/docs-site/zh/examples/integrations/langgraph.mdx +4 -4
- package/docs-site/zh/examples/integrations/pi-sdk.mdx +5 -5
- package/docs-site/zh/explanation/adapter.mdx +5 -5
- package/docs-site/zh/explanation/assert.mdx +7 -7
- package/docs-site/zh/explanation/drive.mdx +4 -2
- package/docs-site/zh/explanation/evals.mdx +10 -8
- package/docs-site/zh/explanation/experiment.mdx +2 -2
- package/docs-site/zh/explanation/hitl.mdx +9 -9
- package/docs-site/zh/explanation/judge.mdx +5 -5
- package/docs-site/zh/explanation/overview.mdx +3 -3
- package/docs-site/zh/explanation/runner.mdx +1 -1
- package/docs-site/zh/index.mdx +3 -3
- package/docs-site/zh/reference/builtin-agents.mdx +13 -13
- package/docs-site/zh/reference/capabilities.mdx +1 -1
- package/docs-site/zh/reference/cli.mdx +18 -12
- package/docs-site/zh/reference/define-agent.mdx +130 -5
- package/docs-site/zh/reference/define-config.mdx +2 -2
- package/docs-site/zh/reference/define-eval.mdx +4 -4
- package/docs-site/zh/reference/events.mdx +11 -11
- package/docs-site/zh/reference/official-adapters.mdx +6 -6
- package/docs-site/zh/reference/results-data.mdx +6 -0
- package/docs-site/zh/troubleshooting/debugging.mdx +1 -1
- package/docs-site/zh/troubleshooting/recover-after-kill.mdx +37 -5
- package/docs-site/zh/tutorials/accept-results.mdx +72 -0
- package/docs-site/zh/tutorials/agent-onboarding.mdx +6 -6
- package/docs-site/zh/tutorials/authoring.mdx +9 -13
- package/docs-site/zh/tutorials/ci-integration.mdx +6 -6
- package/docs-site/zh/tutorials/compare-runs.mdx +61 -0
- package/docs-site/zh/tutorials/concurrency.mdx +13 -12
- package/docs-site/zh/tutorials/configuration.mdx +4 -6
- package/docs-site/zh/tutorials/connect-your-agent.mdx +1 -1
- package/docs-site/zh/tutorials/control-run-cost.mdx +58 -0
- package/docs-site/zh/tutorials/criteria-files.mdx +2 -2
- package/docs-site/zh/tutorials/custom-reports.mdx +2 -2
- package/docs-site/zh/tutorials/debug-lifecycle-plan.mdx +46 -0
- package/docs-site/zh/tutorials/docker-in-docker.mdx +52 -43
- package/docs-site/zh/tutorials/eval-groups.mdx +6 -6
- package/docs-site/zh/tutorials/evaluation-kinds.mdx +3 -3
- package/docs-site/zh/tutorials/experiments.mdx +2 -3
- package/docs-site/zh/tutorials/handle-execution-failures.mdx +18 -10
- package/docs-site/zh/tutorials/install-custom-sandbox-agent.mdx +16 -8
- package/docs-site/zh/tutorials/nixos-managed-dind.mdx +160 -0
- package/docs-site/zh/tutorials/plugins.mdx +14 -0
- package/docs-site/zh/tutorials/rerun-and-cache.mdx +4 -2
- package/docs-site/zh/tutorials/sandbox-agent.mdx +17 -11
- package/docs-site/zh/tutorials/sandbox-providers.mdx +17 -13
- package/docs-site/zh/tutorials/sandbox-reuse.mdx +12 -19
- package/docs-site/zh/tutorials/select-and-preview.mdx +65 -0
- package/docs-site/zh/tutorials/verify-judge.mdx +78 -0
- package/docs-site/zh/tutorials/viewing-results.mdx +38 -1
- package/docs-site/zh/tutorials/write-experiment.mdx +7 -5
- package/docs-site/zh/tutorials/write-send.mdx +9 -8
- package/package.json +9 -5
- package/dist/report/built-in/overview.cjs +0 -44
- package/dist/report/built-in/overview.cjs.map +0 -1
- package/dist/report/built-in/overview.d.cts +0 -6
- package/dist/report/built-in/overview.d.mts +0 -6
- package/dist/report/built-in/overview.d.ts +0 -6
- package/dist/report/built-in/overview.js +0 -9
- package/dist/report/built-in/overview.js.map +0 -1
- package/dist/runner/gate-lease.cjs +0 -283
- package/dist/runner/gate-lease.cjs.map +0 -1
- package/dist/runner/gate-lease.d.cts +0 -62
- package/dist/runner/gate-lease.d.mts +0 -62
- package/dist/runner/gate-lease.d.ts +0 -62
- package/dist/runner/gate-lease.js +0 -22
- package/dist/runner/gate-lease.js.map +0 -1
|
@@ -2,9 +2,9 @@
|
|
|
2
2
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
3
3
|
exports.loadBuiltInSummaryRows = loadBuiltInSummaryRows;
|
|
4
4
|
exports.loadBuiltInExperimentRows = loadBuiltInExperimentRows;
|
|
5
|
-
exports.loadBuiltInAttemptRows = loadBuiltInAttemptRows;
|
|
6
5
|
const aggregate_ts_1 = require("../model/aggregate.cjs");
|
|
7
6
|
const metrics_ts_1 = require("../model/metrics.cjs");
|
|
7
|
+
const compute_ts_1 = require("../components/entity-lists/compute.cjs");
|
|
8
8
|
function loadBuiltInSummaryRows(sample) {
|
|
9
9
|
return (0, aggregate_ts_1.aggregate)(sample, {
|
|
10
10
|
by: {},
|
|
@@ -12,17 +12,25 @@ function loadBuiltInSummaryRows(sample) {
|
|
|
12
12
|
});
|
|
13
13
|
}
|
|
14
14
|
/** One closed row per selected Experiment for overview charts and tables. */
|
|
15
|
-
function loadBuiltInExperimentRows(sample) {
|
|
16
|
-
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
15
|
+
async function loadBuiltInExperimentRows(sample) {
|
|
16
|
+
const rows = await (0, compute_ts_1.experimentListData)(sample);
|
|
17
|
+
const output = rows.map((row) => Object.freeze({
|
|
18
|
+
key: String(row.experimentId),
|
|
19
|
+
experiment: String(row.experimentId),
|
|
20
|
+
evaluationKind: row.evaluationKind,
|
|
21
|
+
...(row.totalScore === undefined ? {} : { totalScore: row.totalScore }),
|
|
22
|
+
passRate: row.evaluationKind === "pass" ? row.endToEndPassRate : null,
|
|
23
|
+
durationMs: row.durationMs,
|
|
24
|
+
tokens: row.tokens,
|
|
25
|
+
}));
|
|
26
|
+
const issues = new Map();
|
|
27
|
+
for (const issue of output.flatMap((row) => [row.passRate, row.durationMs, row.tokens]
|
|
28
|
+
.filter((metric) => metric !== null)
|
|
29
|
+
.flatMap((metric) => metric.issues))) {
|
|
30
|
+
issues.set(`${issue.code}\u0000${issue.message}`, issue);
|
|
31
|
+
}
|
|
32
|
+
return Object.freeze(Object.assign(output, {
|
|
33
|
+
issues: Object.freeze([...issues.values()]),
|
|
34
|
+
}));
|
|
27
35
|
}
|
|
28
36
|
//# sourceMappingURL=analysis-values.cjs.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"analysis-values.js","sourceRoot":"","sources":["analysis-values.ts"],"names":[],"mappings":";;
|
|
1
|
+
{"version":3,"file":"analysis-values.js","sourceRoot":"","sources":["analysis-values.ts"],"names":[],"mappings":";;AAUA,wDAKC;AAID,8DAoBC;AAhCD,wDAAkD;AAClD,oDAAoE;AACpE,sEAA2E;AAC3E,SAAgB,sBAAsB,CAAC,MAAc;IACjD,OAAO,IAAA,wBAAS,EAAC,MAAM,EAAE;QACrB,EAAE,EAAE,EAAE;QACN,MAAM,EAAE,EAAE,QAAQ,EAAR,qBAAQ,EAAE,UAAU,EAAV,uBAAU,EAAE,MAAM,EAAN,mBAAM,EAAE;KAC3C,CAAC,CAAC;AACP,CAAC;AAGD,6EAA6E;AACtE,KAAK,UAAU,yBAAyB,CAAC,MAAc;IAC1D,MAAM,IAAI,GAAG,MAAM,IAAA,+BAAkB,EAAC,MAAM,CAAC,CAAC;IAC9C,MAAM,MAAM,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC;QAC3C,GAAG,EAAE,MAAM,CAAC,GAAG,CAAC,YAAY,CAAC;QAC7B,UAAU,EAAE,MAAM,CAAC,GAAG,CAAC,YAAY,CAAC;QACpC,cAAc,EAAE,GAAG,CAAC,cAAc;QAClC,GAAG,CAAC,GAAG,CAAC,UAAU,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,UAAU,EAAE,GAAG,CAAC,UAAU,EAAE,CAAC;QACvE,QAAQ,EAAE,GAAG,CAAC,cAAc,KAAK,MAAM,CAAC,CAAC,CAAC,GAAG,CAAC,gBAAgB,CAAC,CAAC,CAAC,IAAI;QACrE,UAAU,EAAE,GAAG,CAAC,UAAU;QAC1B,MAAM,EAAE,GAAG,CAAC,MAAM;KACrB,CAAC,CAAC,CAAC;IACJ,MAAM,MAAM,GAAG,IAAI,GAAG,EAAqE,CAAC;IAC5F,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,OAAO,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,QAAQ,EAAE,GAAG,CAAC,UAAU,EAAE,GAAG,CAAC,MAAM,CAAC;SACjF,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,KAAK,IAAI,CAAC;SACnC,OAAO,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,EAAE,CAAC;QACvC,MAAM,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,IAAI,SAAS,KAAK,CAAC,OAAO,EAAE,EAAE,KAAK,CAAC,CAAC;IAC7D,CAAC;IACD,OAAO,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,MAAM,EAAE;QACvC,MAAM,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,CAAC;KAC9C,CAAC,CAAC,CAAC;AACR,CAAC","sourcesContent":["/**\n * The standard report only composes Analysis-issued ClosedRows. In\n * particular, it does not create a second Measure catalog or copy an Analysis\n * reducer: all denominator, missingness, provenance, and cost semantics come\n * from the public Report/Analysis facade.\n */\nimport type { Sample } from \"../../analysis/index.ts\";\nimport { aggregate } from \"../model/aggregate.ts\";\nimport { durationMs, passRate, tokens, } from \"../model/metrics.ts\";\nimport { experimentListData } from \"../components/entity-lists/compute.ts\";\nexport function loadBuiltInSummaryRows(sample: Sample) {\n return aggregate(sample, {\n by: {},\n values: { passRate, durationMs, tokens },\n });\n}\nexport type BuiltInSummaryRows = Awaited<ReturnType<typeof loadBuiltInSummaryRows>>;\nexport type BuiltInSummaryRow = BuiltInSummaryRows[number];\n/** One closed row per selected Experiment for overview charts and tables. */\nexport async function loadBuiltInExperimentRows(sample: Sample) {\n const rows = await experimentListData(sample);\n const output = rows.map((row) => Object.freeze({\n key: String(row.experimentId),\n experiment: String(row.experimentId),\n evaluationKind: row.evaluationKind,\n ...(row.totalScore === undefined ? {} : { totalScore: row.totalScore }),\n passRate: row.evaluationKind === \"pass\" ? row.endToEndPassRate : null,\n durationMs: row.durationMs,\n tokens: row.tokens,\n }));\n const issues = new Map<string, typeof rows[number][\"endToEndPassRate\"][\"issues\"][number]>();\n for (const issue of output.flatMap((row) => [row.passRate, row.durationMs, row.tokens]\n .filter((metric) => metric !== null)\n .flatMap((metric) => metric.issues))) {\n issues.set(`${issue.code}\\u0000${issue.message}`, issue);\n }\n return Object.freeze(Object.assign(output, {\n issues: Object.freeze([...issues.values()]),\n }));\n}\nexport type BuiltInExperimentRows = Awaited<ReturnType<typeof loadBuiltInExperimentRows>>;\nexport type BuiltInExperimentRow = BuiltInExperimentRows[number];\n"]}
|
|
@@ -15,27 +15,16 @@ export declare function loadBuiltInSummaryRows(sample: Sample): Promise<import("
|
|
|
15
15
|
export type BuiltInSummaryRows = Awaited<ReturnType<typeof loadBuiltInSummaryRows>>;
|
|
16
16
|
export type BuiltInSummaryRow = BuiltInSummaryRows[number];
|
|
17
17
|
/** One closed row per selected Experiment for overview charts and tables. */
|
|
18
|
-
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
}
|
|
18
|
+
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<Readonly<Readonly<{
|
|
19
|
+
passRate: import("../../analysis/contracts.cjs").MetricValue<number> | null;
|
|
20
|
+
durationMs: import("../../analysis/contracts.cjs").MetricValue<number>;
|
|
21
|
+
tokens: import("../../analysis/contracts.cjs").MetricValue<number>;
|
|
22
|
+
totalScore?: number | undefined;
|
|
23
|
+
key: string;
|
|
24
|
+
experiment: string;
|
|
25
|
+
evaluationKind: import("../components/entity-lists/compute.cjs").EvaluationKindComposition;
|
|
26
|
+
}>[] & {
|
|
27
|
+
issues: readonly import("../../analysis/contracts.cjs").AnalysisIssue[];
|
|
28
|
+
}>>;
|
|
27
29
|
export type BuiltInExperimentRows = Awaited<ReturnType<typeof loadBuiltInExperimentRows>>;
|
|
28
30
|
export type BuiltInExperimentRow = BuiltInExperimentRows[number];
|
|
29
|
-
/** One closed row per selected logical Attempt; used only by the Attempts Page. */
|
|
30
|
-
export declare function loadBuiltInAttemptRows(sample: Sample): Promise<import("../../analysis/contracts.cjs").ClosedRows<Readonly<{
|
|
31
|
-
readonly key: string;
|
|
32
|
-
} & {
|
|
33
|
-
readonly experiment: import("../../analysis/index.cjs").ExperimentId;
|
|
34
|
-
readonly evalId: import("../../analysis/index.cjs").EvalId;
|
|
35
|
-
readonly attempt: import("../../attempt-locator.cjs").AttemptLocator | null;
|
|
36
|
-
} & {
|
|
37
|
-
readonly passRate: import("../../analysis/contracts.cjs").MetricValue<number>;
|
|
38
|
-
readonly durationMs: import("../../analysis/contracts.cjs").MetricValue<number>;
|
|
39
|
-
readonly tokens: import("../../analysis/contracts.cjs").MetricValue<number>;
|
|
40
|
-
}>>>;
|
|
41
|
-
export type BuiltInAttemptRows = Awaited<ReturnType<typeof loadBuiltInAttemptRows>>;
|
|
@@ -15,27 +15,16 @@ export declare function loadBuiltInSummaryRows(sample: Sample): Promise<import("
|
|
|
15
15
|
export type BuiltInSummaryRows = Awaited<ReturnType<typeof loadBuiltInSummaryRows>>;
|
|
16
16
|
export type BuiltInSummaryRow = BuiltInSummaryRows[number];
|
|
17
17
|
/** One closed row per selected Experiment for overview charts and tables. */
|
|
18
|
-
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
}
|
|
18
|
+
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<Readonly<Readonly<{
|
|
19
|
+
passRate: import("../../analysis/contracts.mjs").MetricValue<number> | null;
|
|
20
|
+
durationMs: import("../../analysis/contracts.mjs").MetricValue<number>;
|
|
21
|
+
tokens: import("../../analysis/contracts.mjs").MetricValue<number>;
|
|
22
|
+
totalScore?: number | undefined;
|
|
23
|
+
key: string;
|
|
24
|
+
experiment: string;
|
|
25
|
+
evaluationKind: import("../components/entity-lists/compute.mjs").EvaluationKindComposition;
|
|
26
|
+
}>[] & {
|
|
27
|
+
issues: readonly import("../../analysis/contracts.mjs").AnalysisIssue[];
|
|
28
|
+
}>>;
|
|
27
29
|
export type BuiltInExperimentRows = Awaited<ReturnType<typeof loadBuiltInExperimentRows>>;
|
|
28
30
|
export type BuiltInExperimentRow = BuiltInExperimentRows[number];
|
|
29
|
-
/** One closed row per selected logical Attempt; used only by the Attempts Page. */
|
|
30
|
-
export declare function loadBuiltInAttemptRows(sample: Sample): Promise<import("../../analysis/contracts.mjs").ClosedRows<Readonly<{
|
|
31
|
-
readonly key: string;
|
|
32
|
-
} & {
|
|
33
|
-
readonly experiment: import("../../analysis/index.mjs").ExperimentId;
|
|
34
|
-
readonly evalId: import("../../analysis/index.mjs").EvalId;
|
|
35
|
-
readonly attempt: import("../../attempt-locator.mjs").AttemptLocator | null;
|
|
36
|
-
} & {
|
|
37
|
-
readonly passRate: import("../../analysis/contracts.mjs").MetricValue<number>;
|
|
38
|
-
readonly durationMs: import("../../analysis/contracts.mjs").MetricValue<number>;
|
|
39
|
-
readonly tokens: import("../../analysis/contracts.mjs").MetricValue<number>;
|
|
40
|
-
}>>>;
|
|
41
|
-
export type BuiltInAttemptRows = Awaited<ReturnType<typeof loadBuiltInAttemptRows>>;
|
|
@@ -15,27 +15,16 @@ export declare function loadBuiltInSummaryRows(sample: Sample): Promise<import("
|
|
|
15
15
|
export type BuiltInSummaryRows = Awaited<ReturnType<typeof loadBuiltInSummaryRows>>;
|
|
16
16
|
export type BuiltInSummaryRow = BuiltInSummaryRows[number];
|
|
17
17
|
/** One closed row per selected Experiment for overview charts and tables. */
|
|
18
|
-
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
}
|
|
18
|
+
export declare function loadBuiltInExperimentRows(sample: Sample): Promise<Readonly<Readonly<{
|
|
19
|
+
passRate: import("../../analysis/contracts.js").MetricValue<number> | null;
|
|
20
|
+
durationMs: import("../../analysis/contracts.js").MetricValue<number>;
|
|
21
|
+
tokens: import("../../analysis/contracts.js").MetricValue<number>;
|
|
22
|
+
totalScore?: number | undefined;
|
|
23
|
+
key: string;
|
|
24
|
+
experiment: string;
|
|
25
|
+
evaluationKind: import("../components/entity-lists/compute.js").EvaluationKindComposition;
|
|
26
|
+
}>[] & {
|
|
27
|
+
issues: readonly import("../../analysis/contracts.js").AnalysisIssue[];
|
|
28
|
+
}>>;
|
|
27
29
|
export type BuiltInExperimentRows = Awaited<ReturnType<typeof loadBuiltInExperimentRows>>;
|
|
28
30
|
export type BuiltInExperimentRow = BuiltInExperimentRows[number];
|
|
29
|
-
/** One closed row per selected logical Attempt; used only by the Attempts Page. */
|
|
30
|
-
export declare function loadBuiltInAttemptRows(sample: Sample): Promise<import("../../analysis/contracts.js").ClosedRows<Readonly<{
|
|
31
|
-
readonly key: string;
|
|
32
|
-
} & {
|
|
33
|
-
readonly experiment: import("../../analysis/index.js").ExperimentId;
|
|
34
|
-
readonly evalId: import("../../analysis/index.js").EvalId;
|
|
35
|
-
readonly attempt: import("../../attempt-locator.js").AttemptLocator | null;
|
|
36
|
-
} & {
|
|
37
|
-
readonly passRate: import("../../analysis/contracts.js").MetricValue<number>;
|
|
38
|
-
readonly durationMs: import("../../analysis/contracts.js").MetricValue<number>;
|
|
39
|
-
readonly tokens: import("../../analysis/contracts.js").MetricValue<number>;
|
|
40
|
-
}>>>;
|
|
41
|
-
export type BuiltInAttemptRows = Awaited<ReturnType<typeof loadBuiltInAttemptRows>>;
|
|
@@ -1,10 +1,8 @@
|
|
|
1
1
|
import { createRequire as __niceevalCreateRequire } from "node:module";
|
|
2
2
|
const __niceevalRequire = __niceevalCreateRequire(import.meta.url);
|
|
3
3
|
const __niceevalCanonical = __niceevalRequire("./analysis-values.cjs");
|
|
4
|
-
const __niceevalExport0 = __niceevalCanonical["
|
|
5
|
-
export { __niceevalExport0 as
|
|
6
|
-
const __niceevalExport1 = __niceevalCanonical["
|
|
7
|
-
export { __niceevalExport1 as
|
|
8
|
-
const __niceevalExport2 = __niceevalCanonical["loadBuiltInSummaryRows"];
|
|
9
|
-
export { __niceevalExport2 as loadBuiltInSummaryRows };
|
|
4
|
+
const __niceevalExport0 = __niceevalCanonical["loadBuiltInExperimentRows"];
|
|
5
|
+
export { __niceevalExport0 as loadBuiltInExperimentRows };
|
|
6
|
+
const __niceevalExport1 = __niceevalCanonical["loadBuiltInSummaryRows"];
|
|
7
|
+
export { __niceevalExport1 as loadBuiltInSummaryRows };
|
|
10
8
|
//# sourceMappingURL=analysis-values.js.map
|
|
@@ -1,14 +1,12 @@
|
|
|
1
1
|
"use strict";
|
|
2
2
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
3
|
-
exports.libraryDetailRoute = exports.experimentDetailTarget = exports.experimentDetailRoute = exports.attemptDetailTarget = exports.attemptDetailRoute = exports.
|
|
3
|
+
exports.libraryDetailRoute = exports.experimentDetailTarget = exports.experimentDetailRoute = exports.attemptDetailTarget = exports.attemptDetailRoute = exports.standardOverviewPage = exports.standardExperimentPage = exports.standardAttemptPage = exports.standard = void 0;
|
|
4
4
|
/** The exact built-in facade fixed in docs/feature/reports/library.md. */
|
|
5
5
|
var standard_tsx_1 = require("./standard.cjs");
|
|
6
6
|
Object.defineProperty(exports, "standard", { enumerable: true, get: function () { return standard_tsx_1.standard; } });
|
|
7
7
|
Object.defineProperty(exports, "standardAttemptPage", { enumerable: true, get: function () { return standard_tsx_1.standardAttemptPage; } });
|
|
8
|
-
Object.defineProperty(exports, "standardAttemptsPage", { enumerable: true, get: function () { return standard_tsx_1.standardAttemptsPage; } });
|
|
9
8
|
Object.defineProperty(exports, "standardExperimentPage", { enumerable: true, get: function () { return standard_tsx_1.standardExperimentPage; } });
|
|
10
9
|
Object.defineProperty(exports, "standardOverviewPage", { enumerable: true, get: function () { return standard_tsx_1.standardOverviewPage; } });
|
|
11
|
-
Object.defineProperty(exports, "standardTracesPage", { enumerable: true, get: function () { return standard_tsx_1.standardTracesPage; } });
|
|
12
10
|
var details_ts_1 = require("../library/details.cjs");
|
|
13
11
|
Object.defineProperty(exports, "attemptDetailRoute", { enumerable: true, get: function () { return details_ts_1.attemptDetailRoute; } });
|
|
14
12
|
Object.defineProperty(exports, "attemptDetailTarget", { enumerable: true, get: function () { return details_ts_1.attemptDetailTarget; } });
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"index.js","sourceRoot":"","sources":["index.tsx"],"names":[],"mappings":";;;AAAA,0EAA0E;AAC1E,
|
|
1
|
+
{"version":3,"file":"index.js","sourceRoot":"","sources":["index.tsx"],"names":[],"mappings":";;;AAAA,0EAA0E;AAC1E,8CAA8G;AAArG,wGAAA,QAAQ,OAAA;AAAE,mHAAA,mBAAmB,OAAA;AAAE,sHAAA,sBAAsB,OAAA;AAAE,oHAAA,oBAAoB,OAAA;AACpF,oDAAoJ;AAA3I,gHAAA,kBAAkB,OAAA;AAAE,iHAAA,mBAAmB,OAAA;AAAE,mHAAA,qBAAqB,OAAA;AAAE,oHAAA,sBAAsB,OAAA;AAAE,gHAAA,kBAAkB,OAAA","sourcesContent":["/** The exact built-in facade fixed in docs/feature/reports/library.md. */\nexport { standard, standardAttemptPage, standardExperimentPage, standardOverviewPage, } from \"./standard.tsx\";\nexport { attemptDetailRoute, attemptDetailTarget, experimentDetailRoute, experimentDetailTarget, libraryDetailRoute, } from \"../library/details.ts\";\nexport type { AttemptDetailTarget, ExperimentDetailTarget, LibraryDetailTarget, } from \"../library/details.ts\";\n"]}
|
|
@@ -1,4 +1,4 @@
|
|
|
1
1
|
/** The exact built-in facade fixed in docs/feature/reports/library.md. */
|
|
2
|
-
export { standard, standardAttemptPage,
|
|
2
|
+
export { standard, standardAttemptPage, standardExperimentPage, standardOverviewPage, } from "./standard.cjs";
|
|
3
3
|
export { attemptDetailRoute, attemptDetailTarget, experimentDetailRoute, experimentDetailTarget, libraryDetailRoute, } from "../library/details.cjs";
|
|
4
4
|
export type { AttemptDetailTarget, ExperimentDetailTarget, LibraryDetailTarget, } from "../library/details.cjs";
|
|
@@ -1,4 +1,4 @@
|
|
|
1
1
|
/** The exact built-in facade fixed in docs/feature/reports/library.md. */
|
|
2
|
-
export { standard, standardAttemptPage,
|
|
2
|
+
export { standard, standardAttemptPage, standardExperimentPage, standardOverviewPage, } from "./standard.mjs";
|
|
3
3
|
export { attemptDetailRoute, attemptDetailTarget, experimentDetailRoute, experimentDetailTarget, libraryDetailRoute, } from "../library/details.mjs";
|
|
4
4
|
export type { AttemptDetailTarget, ExperimentDetailTarget, LibraryDetailTarget, } from "../library/details.mjs";
|
|
@@ -1,4 +1,4 @@
|
|
|
1
1
|
/** The exact built-in facade fixed in docs/feature/reports/library.md. */
|
|
2
|
-
export { standard, standardAttemptPage,
|
|
2
|
+
export { standard, standardAttemptPage, standardExperimentPage, standardOverviewPage, } from "./standard.js";
|
|
3
3
|
export { attemptDetailRoute, attemptDetailTarget, experimentDetailRoute, experimentDetailTarget, libraryDetailRoute, } from "../library/details.js";
|
|
4
4
|
export type { AttemptDetailTarget, ExperimentDetailTarget, LibraryDetailTarget, } from "../library/details.js";
|
|
@@ -15,12 +15,8 @@ const __niceevalExport5 = __niceevalCanonical["standard"];
|
|
|
15
15
|
export { __niceevalExport5 as standard };
|
|
16
16
|
const __niceevalExport6 = __niceevalCanonical["standardAttemptPage"];
|
|
17
17
|
export { __niceevalExport6 as standardAttemptPage };
|
|
18
|
-
const __niceevalExport7 = __niceevalCanonical["
|
|
19
|
-
export { __niceevalExport7 as
|
|
20
|
-
const __niceevalExport8 = __niceevalCanonical["
|
|
21
|
-
export { __niceevalExport8 as
|
|
22
|
-
const __niceevalExport9 = __niceevalCanonical["standardOverviewPage"];
|
|
23
|
-
export { __niceevalExport9 as standardOverviewPage };
|
|
24
|
-
const __niceevalExport10 = __niceevalCanonical["standardTracesPage"];
|
|
25
|
-
export { __niceevalExport10 as standardTracesPage };
|
|
18
|
+
const __niceevalExport7 = __niceevalCanonical["standardExperimentPage"];
|
|
19
|
+
export { __niceevalExport7 as standardExperimentPage };
|
|
20
|
+
const __niceevalExport8 = __niceevalCanonical["standardOverviewPage"];
|
|
21
|
+
export { __niceevalExport8 as standardOverviewPage };
|
|
26
22
|
//# sourceMappingURL=index.js.map
|
|
@@ -15,12 +15,8 @@ const __niceevalExport5 = __niceevalCanonical["standard"];
|
|
|
15
15
|
export { __niceevalExport5 as standard };
|
|
16
16
|
const __niceevalExport6 = __niceevalCanonical["standardAttemptPage"];
|
|
17
17
|
export { __niceevalExport6 as standardAttemptPage };
|
|
18
|
-
const __niceevalExport7 = __niceevalCanonical["
|
|
19
|
-
export { __niceevalExport7 as
|
|
20
|
-
const __niceevalExport8 = __niceevalCanonical["
|
|
21
|
-
export { __niceevalExport8 as
|
|
22
|
-
const __niceevalExport9 = __niceevalCanonical["standardOverviewPage"];
|
|
23
|
-
export { __niceevalExport9 as standardOverviewPage };
|
|
24
|
-
const __niceevalExport10 = __niceevalCanonical["standardTracesPage"];
|
|
25
|
-
export { __niceevalExport10 as standardTracesPage };
|
|
18
|
+
const __niceevalExport7 = __niceevalCanonical["standardExperimentPage"];
|
|
19
|
+
export { __niceevalExport7 as standardExperimentPage };
|
|
20
|
+
const __niceevalExport8 = __niceevalCanonical["standardOverviewPage"];
|
|
21
|
+
export { __niceevalExport8 as standardOverviewPage };
|
|
26
22
|
//# sourceMappingURL=index.mjs.map
|
|
@@ -15,7 +15,6 @@ const report_ts_1 = require("../definition/report.cjs");
|
|
|
15
15
|
const machine_ts_1 = require("../execution/machine.cjs");
|
|
16
16
|
/** Stable names consumed by the Host-owned built-in machine registry. */
|
|
17
17
|
exports.builtInMachineProducerIds = Object.freeze({
|
|
18
|
-
defaultOverview: "niceeval.report.default-overview@v2",
|
|
19
18
|
runMembershipOverview: "niceeval.report.run-membership-overview@v2",
|
|
20
19
|
attemptOverview: "niceeval.report.attempt-overview@v2",
|
|
21
20
|
executionEvidence: "niceeval.report.execution-evidence@v2",
|
|
@@ -23,7 +22,6 @@ exports.builtInMachineProducerIds = Object.freeze({
|
|
|
23
22
|
sourceEvidence: "niceeval.report.source-evidence@v2",
|
|
24
23
|
sandboxHistory: "niceeval.report.sandbox-history@v2",
|
|
25
24
|
standard: "niceeval.report.standard@v2",
|
|
26
|
-
classicOverview: "niceeval.report.classic-overview@v2",
|
|
27
25
|
});
|
|
28
26
|
/**
|
|
29
27
|
* Creates a built-in Report without ever mutating the frozen Report returned
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"machine.js","sourceRoot":"","sources":["machine.ts"],"names":[],"mappings":";;;
|
|
1
|
+
{"version":3,"file":"machine.js","sourceRoot":"","sources":["machine.ts"],"names":[],"mappings":";;;AA2BA,kDAOC;AAlCD;;;;;;;;GAQG;AACH,uDAA+E;AAC/E,wDAA0G;AAC1G,yEAAyE;AAC5D,QAAA,yBAAyB,GAAG,MAAM,CAAC,MAAM,CAAC;IACnD,qBAAqB,EAAE,4CAA4C;IACnE,eAAe,EAAE,qCAAqC;IACtD,iBAAiB,EAAE,uCAAuC;IAC1D,cAAc,EAAE,oCAAoC;IACpD,cAAc,EAAE,oCAAoC;IACpD,cAAc,EAAE,oCAAoC;IACpD,QAAQ,EAAE,6BAA6B;CACjC,CAAC,CAAC;AAEZ;;;;GAIG;AACH,SAAgB,mBAAmB,CAAC,UAAoC,EAAE,UAAkB;IACxF,MAAM,KAAK,GAAG,IAAA,2CAA8B,EAAC,EAAE,GAAG,UAAU,EAAE,EAAE,IAAA,2CAA8B,EAAC,UAAU,CAAC,CAAC,CAAC;IAC5G,0EAA0E;IAC1E,4EAA4E;IAC5E,2EAA2E;IAC3E,6DAA6D;IAC7D,OAAQ,wBAA+D,CAAC,KAAK,CAAC,CAAC;AACnF,CAAC","sourcesContent":["/**\n * Built-in Reports opt into Host-owned machine output with a portable,\n * data-only descriptor. The descriptor is installed on the mutable\n * definition input before `defineReport()` freezes and normalizes it; the\n * definition module validates and copies only its versioned producer id.\n *\n * This module intentionally has no producer callback or registry. The Host\n * maps these ids to its own registry.\n */\nimport { defineReport, type ReportDefinition, } from \"../definition/report.ts\";\nimport { attachBuiltInMachineDescriptor, defineBuiltInMachineDescriptor, } from \"../execution/machine.ts\";\n/** Stable names consumed by the Host-owned built-in machine registry. */\nexport const builtInMachineProducerIds = Object.freeze({\n runMembershipOverview: \"niceeval.report.run-membership-overview@v2\",\n attemptOverview: \"niceeval.report.attempt-overview@v2\",\n executionEvidence: \"niceeval.report.execution-evidence@v2\",\n timingEvidence: \"niceeval.report.timing-evidence@v2\",\n sourceEvidence: \"niceeval.report.source-evidence@v2\",\n sandboxHistory: \"niceeval.report.sandbox-history@v2\",\n standard: \"niceeval.report.standard@v2\",\n} as const);\nexport type BuiltInMachineProducerId = (typeof builtInMachineProducerIds)[keyof typeof builtInMachineProducerIds];\n/**\n * Creates a built-in Report without ever mutating the frozen Report returned\n * by `defineReport()`. Repeated installed copies recognize this descriptor\n * through the two versioned `Symbol.for` keys rather than object identity.\n */\nexport function defineBuiltInReport(producerId: BuiltInMachineProducerId, definition: object): ReportDefinition {\n const input = attachBuiltInMachineDescriptor({ ...definition }, defineBuiltInMachineDescriptor(producerId));\n // The public overload deliberately preserves each Page's Input inference.\n // This internal bridge merely lets this factory accept every valid built-in\n // Page shape; defineReport still validates the complete declaration at run\n // time and copies the descriptor before freezing its output.\n return (defineReport as unknown as (value: object) => ReportDefinition)(input);\n}\n"]}
|
|
@@ -10,7 +10,6 @@
|
|
|
10
10
|
import { type ReportDefinition } from "../definition/report.cjs";
|
|
11
11
|
/** Stable names consumed by the Host-owned built-in machine registry. */
|
|
12
12
|
export declare const builtInMachineProducerIds: Readonly<{
|
|
13
|
-
readonly defaultOverview: "niceeval.report.default-overview@v2";
|
|
14
13
|
readonly runMembershipOverview: "niceeval.report.run-membership-overview@v2";
|
|
15
14
|
readonly attemptOverview: "niceeval.report.attempt-overview@v2";
|
|
16
15
|
readonly executionEvidence: "niceeval.report.execution-evidence@v2";
|
|
@@ -18,7 +17,6 @@ export declare const builtInMachineProducerIds: Readonly<{
|
|
|
18
17
|
readonly sourceEvidence: "niceeval.report.source-evidence@v2";
|
|
19
18
|
readonly sandboxHistory: "niceeval.report.sandbox-history@v2";
|
|
20
19
|
readonly standard: "niceeval.report.standard@v2";
|
|
21
|
-
readonly classicOverview: "niceeval.report.classic-overview@v2";
|
|
22
20
|
}>;
|
|
23
21
|
export type BuiltInMachineProducerId = (typeof builtInMachineProducerIds)[keyof typeof builtInMachineProducerIds];
|
|
24
22
|
/**
|
|
@@ -10,7 +10,6 @@
|
|
|
10
10
|
import { type ReportDefinition } from "../definition/report.mjs";
|
|
11
11
|
/** Stable names consumed by the Host-owned built-in machine registry. */
|
|
12
12
|
export declare const builtInMachineProducerIds: Readonly<{
|
|
13
|
-
readonly defaultOverview: "niceeval.report.default-overview@v2";
|
|
14
13
|
readonly runMembershipOverview: "niceeval.report.run-membership-overview@v2";
|
|
15
14
|
readonly attemptOverview: "niceeval.report.attempt-overview@v2";
|
|
16
15
|
readonly executionEvidence: "niceeval.report.execution-evidence@v2";
|
|
@@ -18,7 +17,6 @@ export declare const builtInMachineProducerIds: Readonly<{
|
|
|
18
17
|
readonly sourceEvidence: "niceeval.report.source-evidence@v2";
|
|
19
18
|
readonly sandboxHistory: "niceeval.report.sandbox-history@v2";
|
|
20
19
|
readonly standard: "niceeval.report.standard@v2";
|
|
21
|
-
readonly classicOverview: "niceeval.report.classic-overview@v2";
|
|
22
20
|
}>;
|
|
23
21
|
export type BuiltInMachineProducerId = (typeof builtInMachineProducerIds)[keyof typeof builtInMachineProducerIds];
|
|
24
22
|
/**
|
|
@@ -10,7 +10,6 @@
|
|
|
10
10
|
import { type ReportDefinition } from "../definition/report.js";
|
|
11
11
|
/** Stable names consumed by the Host-owned built-in machine registry. */
|
|
12
12
|
export declare const builtInMachineProducerIds: Readonly<{
|
|
13
|
-
readonly defaultOverview: "niceeval.report.default-overview@v2";
|
|
14
13
|
readonly runMembershipOverview: "niceeval.report.run-membership-overview@v2";
|
|
15
14
|
readonly attemptOverview: "niceeval.report.attempt-overview@v2";
|
|
16
15
|
readonly executionEvidence: "niceeval.report.execution-evidence@v2";
|
|
@@ -18,7 +17,6 @@ export declare const builtInMachineProducerIds: Readonly<{
|
|
|
18
17
|
readonly sourceEvidence: "niceeval.report.source-evidence@v2";
|
|
19
18
|
readonly sandboxHistory: "niceeval.report.sandbox-history@v2";
|
|
20
19
|
readonly standard: "niceeval.report.standard@v2";
|
|
21
|
-
readonly classicOverview: "niceeval.report.classic-overview@v2";
|
|
22
20
|
}>;
|
|
23
21
|
export type BuiltInMachineProducerId = (typeof builtInMachineProducerIds)[keyof typeof builtInMachineProducerIds];
|
|
24
22
|
/**
|
|
@@ -1,12 +1,10 @@
|
|
|
1
1
|
"use strict";
|
|
2
2
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
3
|
-
exports.ExperimentDetailResultView = exports.AttemptDetailResultView = exports.RunMembershipResultView = exports.SandboxHistoryResultView = exports.SourcesResultView = exports.FileChangesTrajectory = exports.AttemptTrace = exports.AttemptEvidenceResultView =
|
|
3
|
+
exports.ExperimentDetailResultView = exports.AttemptDetailResultView = exports.RunMembershipResultView = exports.SandboxHistoryResultView = exports.SourcesResultView = exports.FileChangesTrajectory = exports.AttemptTrace = exports.AttemptEvidenceResultView = void 0;
|
|
4
4
|
const jsx_runtime_1 = require("react/jsx-runtime");
|
|
5
5
|
const index_tsx_1 = require("../components/site-components/index.cjs");
|
|
6
6
|
const tree_ts_1 = require("../definition/tree.cjs");
|
|
7
7
|
const primitives_tsx_1 = require("../definition/primitives.cjs");
|
|
8
|
-
const metrics_ts_1 = require("../model/metrics.cjs");
|
|
9
|
-
const details_ts_1 = require("../library/details.cjs");
|
|
10
8
|
const DETAIL_ROWS_MAX = 200;
|
|
11
9
|
const SOURCE_TEXT_MAX = 12000;
|
|
12
10
|
function metricCell(metric) {
|
|
@@ -29,62 +27,33 @@ function issueSection(view) {
|
|
|
29
27
|
const rows = issueRows(view);
|
|
30
28
|
return rows.length === 0 ? null : ((0, jsx_runtime_1.jsx)(primitives_tsx_1.Section, { title: "Analysis notes", meta: `${rows.length}`, children: (0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: rows, columns: ["code", "message"] }) }));
|
|
31
29
|
}
|
|
32
|
-
function
|
|
33
|
-
const
|
|
34
|
-
return ((0, jsx_runtime_1.jsxs)(jsx_runtime_1.Fragment, { children: [(0, jsx_runtime_1.jsx)(index_tsx_1.Hero, { data: input.hero }), (0, jsx_runtime_1.jsxs)(primitives_tsx_1.Grid, { children: [(0, jsx_runtime_1.jsx)(primitives_tsx_1.Stat, { label: "Pass rate", value: summary === undefined ? null : metricCell(summary.passRate) }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Stat, { label: "Mean duration", value: summary === undefined ? null : metricCell(summary.durationMs) }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Stat, { label: "Mean tokens", value: summary === undefined ? null : metricCell(summary.tokens) }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Stat, { label: "Experiments", value: input.counts.experiments }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Stat, { label: "Included attempts", value: input.counts.attempts })] })] }));
|
|
35
|
-
}
|
|
36
|
-
function experimentPoints(rows) {
|
|
37
|
-
return rows.map((row) => (0, metrics_ts_1.evidenceRow)({
|
|
38
|
-
experiment: String(row.experiment),
|
|
39
|
-
passRate: row.passRate,
|
|
40
|
-
durationMs: row.durationMs,
|
|
41
|
-
}));
|
|
42
|
-
}
|
|
43
|
-
function experimentTarget(experiment) {
|
|
44
|
-
return { page: "experiment", params: (0, details_ts_1.experimentDetailTarget)(experiment) };
|
|
45
|
-
}
|
|
46
|
-
function standardOverviewTree(input) {
|
|
47
|
-
const points = experimentPoints(input.experiments);
|
|
48
|
-
return ((0, jsx_runtime_1.jsxs)(primitives_tsx_1.Col, { children: [heroAndKpis(input), (0, jsx_runtime_1.jsxs)(primitives_tsx_1.Section, { title: "Experiment results", meta: `${input.experiments.length} experiments`, children: [points.length === 0 ? (0, jsx_runtime_1.jsx)(primitives_tsx_1.Text, { children: "No experiment rows are available for this selection." }) : ((0, jsx_runtime_1.jsx)(primitives_tsx_1.Col, { children: (0, jsx_runtime_1.jsx)(primitives_tsx_1.Bars, { points: points, x: "experiment", y: "passRate", point: "experiment", pointTarget: (point) => experimentTarget(point.key), sort: { field: "passRate", direction: "desc" }, layout: "horizontal" }) })), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: input.experiments, columns: [
|
|
49
|
-
{ field: "experiment", label: "Experiment" },
|
|
50
|
-
{ field: "passRate", label: "Pass rate" },
|
|
51
|
-
{ field: "durationMs", label: "Mean duration" },
|
|
52
|
-
{ field: "tokens", label: "Mean tokens" },
|
|
53
|
-
] })] }), issueSection(input.experiments)] }));
|
|
54
|
-
}
|
|
55
|
-
/** The default product overview: Hero, exactly six KPIs, Bars, Scatter, and Table. */
|
|
56
|
-
exports.StandardOverviewResultView = (0, tree_ts_1.defineComponent)((input) => standardOverviewTree(input));
|
|
57
|
-
exports.StandardOverviewResultView.displayName = "StandardOverviewResultView";
|
|
58
|
-
function standardAttemptsTree(input) {
|
|
59
|
-
const rows = input.attempts.map((row) => ({
|
|
30
|
+
function membershipTable(rows) {
|
|
31
|
+
const normalized = rows.map((row) => ({
|
|
60
32
|
...row,
|
|
61
|
-
|
|
62
|
-
|
|
63
|
-
|
|
33
|
+
outcome: row.outcome ?? null,
|
|
34
|
+
phase: row.phase ?? null,
|
|
35
|
+
error: row.error ?? null,
|
|
36
|
+
sharedFailure: row.sharedFailure ?? null,
|
|
64
37
|
}));
|
|
65
|
-
return ((0, jsx_runtime_1.
|
|
66
|
-
{ field: "experiment", label: "Experiment" },
|
|
67
|
-
{ field: "evalId", label: "Eval" },
|
|
68
|
-
{ field: "attempt", label: "Attempt" },
|
|
69
|
-
{ field: "passRate", label: "Pass rate" },
|
|
70
|
-
{ field: "durationMs", label: "Mean duration" },
|
|
71
|
-
{ field: "tokens", label: "Mean tokens" },
|
|
72
|
-
] }) }), issueSection(input.attempts)] }));
|
|
73
|
-
}
|
|
74
|
-
exports.StandardAttemptsResultView = (0, tree_ts_1.defineComponent)((input) => standardAttemptsTree(input));
|
|
75
|
-
exports.StandardAttemptsResultView.displayName = "StandardAttemptsResultView";
|
|
76
|
-
function membershipTable(rows) {
|
|
77
|
-
return ((0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: rows, columns: [
|
|
38
|
+
return ((0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: normalized, columns: [
|
|
78
39
|
{ field: "experiment", label: "Experiment" },
|
|
79
40
|
{ field: "eval", label: "Eval" },
|
|
80
41
|
{ field: "attempt", label: "Attempt" },
|
|
42
|
+
{ field: "state", label: "Membership" },
|
|
43
|
+
{ field: "outcome", label: "Outcome" },
|
|
44
|
+
{ field: "error", label: "Error" },
|
|
45
|
+
{ field: "phase", label: "Phase" },
|
|
46
|
+
{ field: "sharedFailure", label: "Shared failure" },
|
|
81
47
|
{ field: "locator", label: "Attempt locator" },
|
|
82
48
|
{ field: "selectedRun", label: "Selected run" },
|
|
83
49
|
{ field: "slot", label: "Slot" },
|
|
84
|
-
{ field: "state", label: "Membership" },
|
|
85
50
|
{ field: "relation", label: "Relation" },
|
|
86
51
|
] }));
|
|
87
52
|
}
|
|
53
|
+
function runErrorsSection(rows) {
|
|
54
|
+
const notStarted = rows.reduce((total, row) => total + row.affectedCount, 0);
|
|
55
|
+
return rows.length === 0 ? null : ((0, jsx_runtime_1.jsx)(primitives_tsx_1.Section, { title: "Run errors", meta: `${notStarted} attempt${notStarted === 1 ? "" : "s"} not started`, children: rows.map((row) => ((0, jsx_runtime_1.jsxs)(primitives_tsx_1.Section, { title: "Sandbox image build failed", children: [(0, jsx_runtime_1.jsx)(primitives_tsx_1.Text, { children: `affected: ${row.affected}` }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Text, { children: `error: ${row.message}` })] }, row.key))) }));
|
|
56
|
+
}
|
|
88
57
|
function evidenceEntryRows(view, locator) {
|
|
89
58
|
return view.entries
|
|
90
59
|
.filter((entry) => locator === undefined || entry.attempt.locator === locator)
|
|
@@ -301,8 +270,28 @@ function sandboxTree(props) {
|
|
|
301
270
|
exports.SandboxHistoryResultView = (0, tree_ts_1.defineComponent)((props) => sandboxTree(props));
|
|
302
271
|
exports.SandboxHistoryResultView.displayName = "SandboxHistoryResultView";
|
|
303
272
|
function runMembershipTree(input) {
|
|
304
|
-
const
|
|
305
|
-
|
|
273
|
+
const hasAttempts = input.members.some((member) => member.locator !== null);
|
|
274
|
+
const rows = input.members.map((member) => ({
|
|
275
|
+
key: member.key,
|
|
276
|
+
experiment: member.experiment,
|
|
277
|
+
eval: member.eval,
|
|
278
|
+
attempt: `#${member.attempt + 1}`,
|
|
279
|
+
result: member.locator === null
|
|
280
|
+
? member.outcome === "errored" ? "errored" : "not started"
|
|
281
|
+
: member.relation === "reference" || member.state === "carried" || member.state === "accepted"
|
|
282
|
+
? `using result ${member.locator}`
|
|
283
|
+
: member.verdict ?? member.outcome ?? member.state,
|
|
284
|
+
details: member.locator === null
|
|
285
|
+
? null
|
|
286
|
+
: { kind: "locator", locator: member.locator },
|
|
287
|
+
}));
|
|
288
|
+
return ((0, jsx_runtime_1.jsxs)(primitives_tsx_1.Col, { children: [(0, jsx_runtime_1.jsx)(index_tsx_1.Hero, { data: input.hero }), runErrorsSection(input.errors), (0, jsx_runtime_1.jsxs)(primitives_tsx_1.Section, { title: "Planned attempts", meta: `${input.members.length} attempt${input.members.length === 1 ? "" : "s"}`, children: [(0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: limited(rows), columns: [
|
|
289
|
+
{ field: "experiment", label: "Experiment" },
|
|
290
|
+
{ field: "eval", label: "Eval" },
|
|
291
|
+
{ field: "attempt", label: "Attempt" },
|
|
292
|
+
{ field: "result", label: "Result" },
|
|
293
|
+
{ field: "details", label: "Details" },
|
|
294
|
+
] }), omittedText(rows.length, Math.min(rows.length, DETAIL_ROWS_MAX), "planned attempts")] }), hasAttempts ? (0, jsx_runtime_1.jsx)(exports.AttemptEvidenceResultView, { view: input.evidence }) : null, hasAttempts ? issueSection(input.summary) : null] }));
|
|
306
295
|
}
|
|
307
296
|
exports.RunMembershipResultView = (0, tree_ts_1.defineComponent)((input) => runMembershipTree(input));
|
|
308
297
|
exports.RunMembershipResultView.displayName = "RunMembershipResultView";
|
|
@@ -314,7 +303,16 @@ exports.AttemptDetailResultView = (0, tree_ts_1.defineComponent)((input) => atte
|
|
|
314
303
|
exports.AttemptDetailResultView.displayName = "AttemptDetailResultView";
|
|
315
304
|
function experimentDetailTree(input) {
|
|
316
305
|
const rows = input.rows.filter((row) => String(row.experiment) === input.experiment);
|
|
317
|
-
|
|
306
|
+
const hasScore = rows.some((row) => row.evaluationKind !== "pass");
|
|
307
|
+
const hasPass = rows.some((row) => row.evaluationKind === "pass");
|
|
308
|
+
const columns = [
|
|
309
|
+
"experiment",
|
|
310
|
+
...(hasPass ? ["passRate"] : []),
|
|
311
|
+
...(hasScore ? ["totalScore"] : []),
|
|
312
|
+
"durationMs",
|
|
313
|
+
"tokens",
|
|
314
|
+
];
|
|
315
|
+
return ((0, jsx_runtime_1.jsxs)(primitives_tsx_1.Col, { children: [(0, jsx_runtime_1.jsx)(index_tsx_1.Hero, { data: input.hero }), (0, jsx_runtime_1.jsx)(primitives_tsx_1.Section, { title: `Experiment · ${input.experiment}`, children: (0, jsx_runtime_1.jsx)(primitives_tsx_1.Table, { rows: rows, columns: columns }) }), (0, jsx_runtime_1.jsxs)(primitives_tsx_1.Section, { title: "Experiment members", meta: `${input.members.length} slots`, children: [membershipTable(limited(input.members)), omittedText(input.members.length, Math.min(input.members.length, DETAIL_ROWS_MAX), "member rows")] }), issueSection(input.rows)] }));
|
|
318
316
|
}
|
|
319
317
|
exports.ExperimentDetailResultView = (0, tree_ts_1.defineComponent)((input) => experimentDetailTree(input));
|
|
320
318
|
exports.ExperimentDetailResultView.displayName = "ExperimentDetailResultView";
|