niceeval 0.5.3 → 0.6.0

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.
Files changed (344) hide show
  1. package/INDEX.md +55 -0
  2. package/dist/agents/types.d.ts +317 -0
  3. package/dist/context/types.d.ts +335 -0
  4. package/dist/i18n/core.d.ts +11 -0
  5. package/dist/i18n/en.d.ts +181 -0
  6. package/dist/i18n/index.d.ts +5 -0
  7. package/dist/i18n/zh-CN.d.ts +183 -0
  8. package/dist/o11y/execution-tree.d.ts +103 -0
  9. package/dist/o11y/otlp/parse.d.ts +3 -0
  10. package/dist/o11y/otlp/receiver.d.ts +21 -0
  11. package/dist/o11y/otlp/select.d.ts +22 -0
  12. package/dist/o11y/otlp/turn-otel.d.ts +43 -0
  13. package/dist/o11y/types.d.ts +190 -0
  14. package/dist/report/aggregate.d.ts +57 -0
  15. package/dist/report/aggregate.js +212 -0
  16. package/dist/report/built-ins/cost-pass-rate-comparison.d.ts +1 -0
  17. package/dist/report/built-ins/cost-pass-rate-comparison.js +17 -0
  18. package/dist/report/built-ins/index.d.ts +1 -0
  19. package/dist/report/built-ins/index.js +2 -0
  20. package/dist/report/components.d.ts +178 -0
  21. package/dist/report/components.js +132 -0
  22. package/dist/report/compute.d.ts +106 -0
  23. package/dist/report/compute.js +618 -0
  24. package/dist/report/flag.d.ts +6 -0
  25. package/dist/report/flag.js +8 -0
  26. package/dist/report/format.d.ts +32 -0
  27. package/dist/report/format.js +126 -0
  28. package/dist/report/index.d.ts +19 -0
  29. package/dist/report/index.js +30 -0
  30. package/dist/report/load.d.ts +12 -0
  31. package/dist/report/load.js +56 -0
  32. package/dist/report/locale.d.ts +83 -0
  33. package/dist/report/locale.js +160 -0
  34. package/dist/report/metrics.d.ts +21 -0
  35. package/dist/report/metrics.js +115 -0
  36. package/dist/report/primitives.d.ts +64 -0
  37. package/dist/report/primitives.js +113 -0
  38. package/dist/report/react/AttemptList.d.ts +23 -0
  39. package/dist/report/react/AttemptList.js +20 -0
  40. package/dist/report/react/DeltaTable.d.ts +8 -0
  41. package/dist/report/react/DeltaTable.js +24 -0
  42. package/dist/report/react/EvalList.d.ts +10 -0
  43. package/dist/report/react/EvalList.js +0 -0
  44. package/dist/report/react/ExperimentList.d.ts +10 -0
  45. package/dist/report/react/ExperimentList.js +15 -0
  46. package/dist/report/react/GroupSummary.d.ts +8 -0
  47. package/dist/report/react/GroupSummary.js +8 -0
  48. package/dist/report/react/MetricBars.d.ts +10 -0
  49. package/dist/report/react/MetricBars.js +48 -0
  50. package/dist/report/react/MetricLine.d.ts +9 -0
  51. package/dist/report/react/MetricLine.js +68 -0
  52. package/dist/report/react/MetricMatrix.d.ts +10 -0
  53. package/dist/report/react/MetricMatrix.js +23 -0
  54. package/dist/report/react/MetricScatter.d.ts +9 -0
  55. package/dist/report/react/MetricScatter.js +100 -0
  56. package/dist/report/react/MetricTable.d.ts +11 -0
  57. package/dist/report/react/MetricTable.js +26 -0
  58. package/dist/report/react/RunOverview.d.ts +8 -0
  59. package/dist/report/react/RunOverview.js +12 -0
  60. package/dist/report/react/Scoreboard.d.ts +8 -0
  61. package/dist/report/react/Scoreboard.js +26 -0
  62. package/dist/report/react/cell.d.ts +9 -0
  63. package/dist/report/react/cell.js +9 -0
  64. package/dist/report/react/chart-math.d.ts +14 -0
  65. package/dist/report/react/chart-math.js +68 -0
  66. package/dist/report/react/colors.d.ts +14 -0
  67. package/dist/report/react/colors.js +45 -0
  68. package/dist/report/react/fixtures.d.ts +19 -0
  69. package/dist/report/react/fixtures.js +393 -0
  70. package/dist/report/react/format.d.ts +3 -0
  71. package/dist/report/react/format.js +7 -0
  72. package/dist/report/react/index.d.ts +6 -0
  73. package/dist/report/react/index.js +14 -0
  74. package/dist/report/report.d.ts +24 -0
  75. package/dist/report/report.js +49 -0
  76. package/dist/report/text/faces.d.ts +29 -0
  77. package/dist/report/text/faces.js +503 -0
  78. package/dist/report/text/layout.d.ts +75 -0
  79. package/dist/report/text/layout.js +173 -0
  80. package/dist/report/text/plot.d.ts +37 -0
  81. package/dist/report/text/plot.js +81 -0
  82. package/dist/report/text/table.d.ts +8 -0
  83. package/dist/report/text/table.js +86 -0
  84. package/dist/report/tree.d.ts +99 -0
  85. package/dist/report/tree.js +228 -0
  86. package/dist/report/types.d.ts +420 -0
  87. package/dist/report/types.js +4 -0
  88. package/dist/report/web.d.ts +15 -0
  89. package/dist/report/web.js +37 -0
  90. package/dist/results/annotated-source.d.ts +61 -0
  91. package/dist/results/attempt-evidence.d.ts +69 -0
  92. package/dist/results/attempt-source.d.ts +15 -0
  93. package/dist/results/locator.d.ts +75 -0
  94. package/dist/results/locator.js +0 -0
  95. package/dist/results/select.d.ts +45 -0
  96. package/dist/results/select.js +230 -0
  97. package/dist/results/source-hash.d.ts +9 -0
  98. package/dist/results/types.d.ts +204 -0
  99. package/dist/runner/eval-source.d.ts +22 -0
  100. package/dist/runner/fingerprint.d.ts +23 -0
  101. package/dist/runner/types.d.ts +332 -0
  102. package/dist/sandbox/docker-stream.d.ts +24 -0
  103. package/dist/sandbox/docker.d.ts +84 -0
  104. package/dist/sandbox/e2b.d.ts +31 -0
  105. package/dist/sandbox/errors.d.ts +4 -0
  106. package/dist/sandbox/local-files.d.ts +6 -0
  107. package/dist/sandbox/paths.d.ts +4 -0
  108. package/dist/sandbox/registry.d.ts +13 -0
  109. package/dist/sandbox/resolve.d.ts +40 -0
  110. package/dist/sandbox/retry.d.ts +10 -0
  111. package/dist/sandbox/shell.d.ts +11 -0
  112. package/dist/sandbox/source-files.d.ts +13 -0
  113. package/dist/sandbox/types.d.ts +203 -0
  114. package/dist/sandbox/vercel.d.ts +37 -0
  115. package/dist/scoring/types.d.ts +70 -0
  116. package/dist/shared/aggregate.d.ts +29 -0
  117. package/dist/shared/aggregate.js +55 -0
  118. package/dist/shared/types.d.ts +27 -0
  119. package/dist/shared/verdict.d.ts +27 -0
  120. package/dist/shared/verdict.js +34 -0
  121. package/dist/tty-line.d.ts +6 -0
  122. package/dist/types.d.ts +7 -0
  123. package/dist/util.d.ts +22 -0
  124. package/docs-site/zh/concepts/adapter.mdx +9 -4
  125. package/docs-site/zh/concepts/assert.mdx +8 -8
  126. package/docs-site/zh/concepts/evals.mdx +5 -5
  127. package/docs-site/zh/concepts/experiment.mdx +3 -1
  128. package/docs-site/zh/concepts/judge.mdx +9 -9
  129. package/docs-site/zh/concepts/overview.mdx +2 -2
  130. package/docs-site/zh/concepts/tier.mdx +6 -6
  131. package/docs-site/zh/example/claude-code-codex-plugin.mdx +9 -5
  132. package/docs-site/zh/example/claude-code-codex-skill.mdx +12 -7
  133. package/docs-site/zh/guides/agent-feedback-loop.mdx +235 -78
  134. package/docs-site/zh/guides/connect-otel.mdx +3 -3
  135. package/docs-site/zh/guides/connect-your-agent.mdx +2 -2
  136. package/docs-site/zh/guides/custom-reports.mdx +196 -36
  137. package/docs-site/zh/guides/experiments.mdx +7 -0
  138. package/docs-site/zh/guides/fixtures.mdx +1 -1
  139. package/docs-site/zh/guides/official-adapters.mdx +31 -9
  140. package/docs-site/zh/guides/publish-report.mdx +4 -4
  141. package/docs-site/zh/guides/report-components.mdx +194 -55
  142. package/docs-site/zh/guides/reporters.mdx +2 -2
  143. package/docs-site/zh/guides/results-data.mdx +35 -34
  144. package/docs-site/zh/guides/runner.mdx +2 -2
  145. package/docs-site/zh/guides/sandbox-agent.mdx +10 -5
  146. package/docs-site/zh/guides/sandbox-providers.mdx +102 -0
  147. package/docs-site/zh/guides/viewing-results.mdx +199 -79
  148. package/docs-site/zh/guides/write-experiment.mdx +23 -1
  149. package/docs-site/zh/guides/write-send.mdx +2 -2
  150. package/docs-site/zh/index.mdx +4 -3
  151. package/docs-site/zh/introduction.mdx +2 -1
  152. package/docs-site/zh/quickstart.mdx +5 -3
  153. package/docs-site/zh/reference/builtin-agents.mdx +38 -13
  154. package/docs-site/zh/reference/capabilities.mdx +1 -1
  155. package/docs-site/zh/reference/cli.mdx +36 -13
  156. package/docs-site/zh/reference/define-agent.mdx +24 -12
  157. package/docs-site/zh/reference/define-config.mdx +1 -1
  158. package/docs-site/zh/reference/define-eval.mdx +3 -3
  159. package/docs-site/zh/reference/events.mdx +7 -1
  160. package/docs-site/zh/reference/expect.mdx +3 -3
  161. package/package.json +10 -6
  162. package/src/agents/bub.ts +134 -43
  163. package/src/agents/claude-code.test.ts +152 -0
  164. package/src/agents/claude-code.ts +127 -13
  165. package/src/agents/codex.test.ts +192 -0
  166. package/src/agents/codex.ts +151 -12
  167. package/src/agents/index.ts +11 -3
  168. package/src/agents/manifest.ts +41 -0
  169. package/src/agents/plugin-config.test.ts +94 -0
  170. package/src/agents/shared.ts +33 -2
  171. package/src/agents/skills.test.ts +214 -0
  172. package/src/agents/skills.ts +304 -0
  173. package/src/agents/streaming.ts +5 -5
  174. package/src/agents/types.ts +62 -2
  175. package/src/cli.ts +74 -47
  176. package/src/context/context.test.ts +20 -2
  177. package/src/context/context.ts +3 -0
  178. package/src/context/control-flow.ts +1 -1
  179. package/src/context/session.test.ts +92 -1
  180. package/src/context/session.ts +26 -1
  181. package/src/context/types.ts +4 -4
  182. package/src/define.test.ts +82 -0
  183. package/src/define.ts +72 -11
  184. package/src/expect/index.ts +5 -5
  185. package/src/i18n/en.ts +36 -15
  186. package/src/i18n/zh-CN.ts +39 -21
  187. package/src/index.ts +1 -1
  188. package/src/o11y/execution-tree.test.ts +452 -0
  189. package/src/o11y/execution-tree.ts +367 -0
  190. package/src/o11y/otlp/receiver.ts +1 -1
  191. package/src/o11y/otlp/select.ts +7 -3
  192. package/src/o11y/parsers/claude-code.test.ts +175 -0
  193. package/src/o11y/parsers/claude-code.ts +32 -0
  194. package/src/o11y/types.ts +6 -1
  195. package/src/report/aggregate.ts +26 -20
  196. package/src/report/built-in-user-parity.test.tsx +643 -0
  197. package/src/report/built-ins/cost-pass-rate-comparison.tsx +23 -0
  198. package/src/report/built-ins/index.ts +2 -0
  199. package/src/report/components.tsx +223 -50
  200. package/src/report/compute.ts +358 -53
  201. package/src/report/dual-render.test.tsx +1125 -0
  202. package/src/report/flag.ts +1 -1
  203. package/src/report/format.ts +50 -2
  204. package/src/report/index.ts +48 -14
  205. package/src/report/load.ts +1 -1
  206. package/src/report/locale.ts +206 -0
  207. package/src/report/metrics.ts +39 -15
  208. package/src/report/primitives.tsx +100 -2
  209. package/src/report/react/AttemptList.tsx +115 -0
  210. package/src/report/react/DeltaTable.tsx +9 -6
  211. package/src/report/react/EvalList.tsx +0 -0
  212. package/src/report/react/ExperimentList.tsx +108 -0
  213. package/src/report/react/GroupSummary.tsx +66 -0
  214. package/src/report/react/MetricBars.tsx +24 -13
  215. package/src/report/react/MetricLine.tsx +21 -21
  216. package/src/report/react/MetricMatrix.tsx +8 -4
  217. package/src/report/react/MetricScatter.tsx +136 -96
  218. package/src/report/react/MetricTable.tsx +115 -15
  219. package/src/report/react/RunOverview.tsx +48 -26
  220. package/src/report/react/Scoreboard.tsx +15 -7
  221. package/src/report/react/cell.tsx +18 -9
  222. package/src/{view/app/lib/chart.ts → report/react/chart-math.ts} +2 -6
  223. package/src/report/react/colors.ts +24 -15
  224. package/src/report/react/enhance.js +129 -0
  225. package/src/report/react/fixtures.ts +163 -34
  226. package/src/report/react/format.ts +1 -1
  227. package/src/report/react/index.tsx +22 -6
  228. package/src/report/react/render.test.tsx +164 -31
  229. package/src/report/react/styles.css +405 -213
  230. package/src/report/report.test.ts +403 -94
  231. package/src/report/report.ts +29 -11
  232. package/src/report/text/faces.ts +303 -95
  233. package/src/report/text/layout.ts +76 -11
  234. package/src/report/text/table.ts +98 -0
  235. package/src/report/tree.ts +103 -20
  236. package/src/report/types.ts +196 -25
  237. package/src/report/web.ts +46 -19
  238. package/src/results/annotated-source.test.ts +146 -0
  239. package/src/results/annotated-source.ts +128 -0
  240. package/src/results/attempt-evidence.test.ts +158 -0
  241. package/src/results/attempt-evidence.ts +158 -0
  242. package/src/results/attempt-source.ts +48 -0
  243. package/src/results/copy.ts +154 -139
  244. package/src/results/format.ts +54 -47
  245. package/src/results/host-equivalence.test.ts +527 -0
  246. package/src/results/index.ts +56 -14
  247. package/src/results/locator.test.ts +198 -0
  248. package/src/results/locator.ts +0 -0
  249. package/src/results/open.ts +392 -175
  250. package/src/results/results.test.ts +943 -494
  251. package/src/results/select.ts +129 -23
  252. package/src/results/skipped-notice.ts +0 -0
  253. package/src/results/source-hash.ts +28 -0
  254. package/src/results/types.ts +98 -62
  255. package/src/results/writer.ts +304 -176
  256. package/src/runner/attempt.test.ts +170 -0
  257. package/src/runner/attempt.ts +97 -15
  258. package/src/runner/discover.test.ts +65 -0
  259. package/src/runner/discover.ts +6 -2
  260. package/src/runner/eval-source.test.ts +121 -0
  261. package/src/runner/eval-source.ts +45 -0
  262. package/src/runner/fingerprint.ts +54 -1
  263. package/src/runner/remote-sandbox.ts +1 -1
  264. package/src/runner/report.test.ts +6 -8
  265. package/src/runner/report.ts +5 -4
  266. package/src/runner/reporters/artifacts.ts +27 -41
  267. package/src/runner/reporters/braintrust.test.ts +3 -3
  268. package/src/runner/reporters/braintrust.ts +1 -1
  269. package/src/runner/reporters/console.ts +6 -6
  270. package/src/runner/reporters/json.ts +4 -4
  271. package/src/runner/reporters/live.test.ts +56 -0
  272. package/src/runner/reporters/live.ts +36 -10
  273. package/src/runner/reporters/quiet.test.ts +66 -0
  274. package/src/runner/reporters/quiet.ts +49 -0
  275. package/src/runner/reporters/shared.ts +5 -5
  276. package/src/runner/reporters/table.ts +17 -17
  277. package/src/runner/run.ts +61 -102
  278. package/src/runner/types.ts +43 -32
  279. package/src/sandbox/checkpoint.ts +1 -1
  280. package/src/sandbox/docker.ts +15 -2
  281. package/src/sandbox/e2b.ts +11 -5
  282. package/src/sandbox/errors.ts +13 -0
  283. package/src/sandbox/index.ts +3 -3
  284. package/src/sandbox/local-files.ts +1 -1
  285. package/src/sandbox/registry.ts +6 -5
  286. package/src/sandbox/resolve.ts +38 -23
  287. package/src/sandbox/retry.test.ts +68 -0
  288. package/src/sandbox/retry.ts +44 -0
  289. package/src/sandbox/types.ts +69 -28
  290. package/src/sandbox/vercel.ts +17 -3
  291. package/src/scoring/collector.ts +3 -3
  292. package/src/scoring/judge.ts +2 -2
  293. package/src/scoring/scoped.ts +13 -1
  294. package/src/scoring/types.ts +7 -7
  295. package/src/scoring/verdict.ts +4 -4
  296. package/src/shared/aggregate.ts +23 -3
  297. package/src/shared/types.ts +2 -2
  298. package/src/shared/{outcome.ts → verdict.ts} +15 -15
  299. package/src/show/compose.ts +12 -132
  300. package/src/show/index.ts +125 -94
  301. package/src/show/render.ts +660 -160
  302. package/src/show/show.test.ts +472 -253
  303. package/src/source-loc.ts +1 -1
  304. package/src/tty-line.ts +32 -0
  305. package/src/types.ts +2 -2
  306. package/src/util.test.ts +21 -1
  307. package/src/util.ts +5 -1
  308. package/src/view/app/App.tsx +33 -198
  309. package/src/view/app/components/AttemptModal.tsx +6 -6
  310. package/src/view/app/components/CodeView.tsx +2 -2
  311. package/src/view/app/components/CopyControls.tsx +32 -36
  312. package/src/view/app/components/SkippedRunsBanner.tsx +140 -0
  313. package/src/view/app/components/ui/badge.tsx +1 -1
  314. package/src/view/app/components/ui/dialog.tsx +2 -2
  315. package/src/view/app/i18n.ts +40 -31
  316. package/src/view/app/lib/artifact-url.ts +1 -1
  317. package/src/view/app/lib/attempt-route.test.ts +37 -47
  318. package/src/view/app/lib/attempt-route.ts +26 -26
  319. package/src/view/app/lib/rows.ts +9 -112
  320. package/src/view/app/lib/verdict.ts +25 -0
  321. package/src/view/app/main.tsx +9 -12
  322. package/src/view/app/pages/RunsPage.tsx +4 -4
  323. package/src/view/app/pages/TracesPage.tsx +3 -3
  324. package/src/view/app/shared.ts +0 -1
  325. package/src/view/app/types.ts +6 -51
  326. package/src/view/artifact-serving.test.ts +136 -0
  327. package/src/view/client-dist/app.css +1 -1
  328. package/src/view/client-dist/app.js +21 -21
  329. package/src/view/data.test.ts +214 -143
  330. package/src/view/data.ts +166 -120
  331. package/src/view/index.ts +23 -11
  332. package/src/view/server.ts +52 -17
  333. package/src/view/shared/types.ts +22 -31
  334. package/src/view/styles.css +117 -327
  335. package/src/view/view-report.test.ts +146 -98
  336. package/docs-site/zh/guides/sandbox-backends.mdx +0 -75
  337. package/src/report/default-report.tsx +0 -222
  338. package/src/report/dual-face.test.tsx +0 -527
  339. package/src/report/react/CaseList.tsx +0 -70
  340. package/src/view/app/components/CostScoreChart.tsx +0 -147
  341. package/src/view/app/components/ExperimentTable.tsx +0 -284
  342. package/src/view/app/components/GroupSelector.tsx +0 -61
  343. package/src/view/app/components/primitives.tsx +0 -43
  344. package/src/view/app/lib/outcome.ts +0 -82
@@ -1,222 +0,0 @@
1
- // DefaultReport:官方水位整块 —— 官方两扇门裸跑时渲染的就是它,零 props、纯声明。
2
- // 「渲染面纯同步」与它不冲突,靠的是一个数据事实:官方水位(overview、榜单、失败清单)
3
- // 只读瘦身条目、不碰任何懒加载工件,宿主对着已挑好的选集总是把这份数据备好、经上下文
4
- // 注入(renderReportToText / renderReportToStaticHtml 里 prepareDefaultReportData)。
5
- // 它渲染的口径钉死为宿主注入的那份选集 —— 零 props 意味着没有跟随的通道,这是锚点语义:
6
- // 官方口径与自定义口径并排对照。默认无特权:数据全部来自公开计算函数。
7
- //
8
- // text 面即 `niceeval show` 的榜单(docs-site/zh/guides/viewing-results.mdx 的示例块是
9
- // 行为规范):`Current verdicts` 头标注合成自几个 run、experiment 表带 eval 级折叠的
10
- // evals 列、Failing 清单每条带判决时间与下钻命令。
11
-
12
- import { basename } from "node:path";
13
- import type { Selection } from "../results/index.ts";
14
- import { foldEvalOutcome } from "../shared/outcome.ts";
15
- import { defineComponent } from "./tree.ts";
16
- import type { CaseListData, OverviewData, TableData } from "./types.ts";
17
- import { caseListData, overviewData, tableData } from "./compute.ts";
18
- import { attemptCostUSD, costUSD, durationMs, passRate } from "./metrics.ts";
19
- import { RunOverview, MetricTable, CaseList } from "./components.tsx";
20
- import { cellText } from "./text/faces.ts";
21
- import { renderAlignedRows } from "./text/layout.ts";
22
-
23
- /** 榜单的合成标注与 eval 级折叠(text 面用;与 verdicts/cases 出自同一选集)。 */
24
- export interface VerdictBoardData {
25
- /** 判决合成自几个物理 run。 */
26
- composedFromRuns: number;
27
- /** 参与合成的最新 run(目录名,即时间戳)。 */
28
- latestRun?: string;
29
- /** experiment → eval 级折叠计票(evals 列的 13/15)。 */
30
- tallies: Record<string, { passedEvals: number; totalEvals: number }>;
31
- /** eval 级折叠后失败/出错的题,带判决时间;新失败在前。 */
32
- failing: {
33
- evalId: string;
34
- experimentId: string;
35
- outcome: "failed" | "errored";
36
- /** 最新 attempt 的第一条失败断言("gate calledTool(...)")或错误摘要。 */
37
- reason?: string;
38
- /** 判决产生的时刻(最新 attempt 的 startedAt,缺失退快照时刻)。 */
39
- verdictAt?: string;
40
- }[];
41
- /** limit 之外还有几条,如实报。 */
42
- failingTruncated: number;
43
- }
44
-
45
- export interface DefaultReportData {
46
- overview: OverviewData;
47
- /** 现刻榜单:experiment × (passRate, costUSD, durationMs)。 */
48
- verdicts: TableData;
49
- cases: CaseListData;
50
- board: VerdictBoardData;
51
- }
52
-
53
- /** 失败清单的出厂截断;完整清单自己摆 <CaseList>(截断如实报剩余)。 */
54
- const DEFAULT_CASE_LIMIT = 10;
55
-
56
- function buildBoard(selection: Selection): VerdictBoardData {
57
- const tallies: VerdictBoardData["tallies"] = {};
58
- const failing: VerdictBoardData["failing"] = [];
59
- const runs = new Set<string>();
60
- let latestRunDir = "";
61
-
62
- for (const snapshot of selection.snapshots) {
63
- let passedEvals = 0;
64
- for (const ev of snapshot.evals) {
65
- for (const attempt of ev.attempts) {
66
- runs.add(attempt.runDir.dir);
67
- if (attempt.runDir.dir > latestRunDir) latestRunDir = attempt.runDir.dir;
68
- }
69
- const outcome = foldEvalOutcome(ev.attempts.map((a) => a.result));
70
- if (outcome === "passed") {
71
- passedEvals += 1;
72
- continue;
73
- }
74
- if (outcome !== "failed" && outcome !== "errored") continue;
75
- const latest = ev.attempts.reduce((a, b) => (b.result.attempt >= a.result.attempt ? b : a));
76
- const failedAssertion = latest.result.assertions.find((a) => !a.passed);
77
- const reason = failedAssertion
78
- ? `${failedAssertion.severity} ${failedAssertion.name}`
79
- : latest.result.error !== undefined
80
- ? `error: ${latest.result.error}`
81
- : undefined;
82
- failing.push({
83
- evalId: ev.id,
84
- experimentId: snapshot.experimentId,
85
- outcome,
86
- ...(reason !== undefined ? { reason } : {}),
87
- ...(latest.result.startedAt !== undefined || snapshot.startedAt
88
- ? { verdictAt: latest.result.startedAt ?? snapshot.startedAt }
89
- : {}),
90
- });
91
- }
92
- tallies[snapshot.experimentId] = { passedEvals, totalEvals: snapshot.evals.length };
93
- }
94
-
95
- failing.sort((a, b) => (b.verdictAt ?? "").localeCompare(a.verdictAt ?? ""));
96
- const shown = failing.slice(0, DEFAULT_CASE_LIMIT);
97
- return {
98
- composedFromRuns: runs.size,
99
- ...(latestRunDir ? { latestRun: basename(latestRunDir) } : {}),
100
- tallies,
101
- failing: shown,
102
- failingTruncated: failing.length - shown.length,
103
- };
104
- }
105
-
106
- /** 宿主渲染前备好官方水位:只读瘦身条目,代价可忽略。 */
107
- export async function prepareDefaultReportData(selection: Selection): Promise<DefaultReportData> {
108
- return {
109
- overview: await overviewData(selection),
110
- verdicts: await tableData(selection, {
111
- rows: "experiment",
112
- columns: [passRate, costUSD, durationMs],
113
- sort: passRate,
114
- }),
115
- cases: await caseListData(selection, { limit: DEFAULT_CASE_LIMIT }),
116
- board: buildBoard(selection),
117
- };
118
- }
119
-
120
- let activeData: DefaultReportData | null = null;
121
-
122
- /** 宿主(与渲染入口)用:在注入好的官方水位下同步渲染。 */
123
- export function runWithDefaultReportData<T>(data: DefaultReportData, fn: () => T): T {
124
- const prev = activeData;
125
- activeData = data;
126
- try {
127
- return fn();
128
- } finally {
129
- activeData = prev;
130
- }
131
- }
132
-
133
- function requireData(): DefaultReportData {
134
- if (!activeData) {
135
- throw new Error(
136
- "<DefaultReport /> renders the host-injected selection; render the report via " +
137
- "`niceeval show --report` / `niceeval view --report` (or renderReportToText / renderReportToStaticHtml). " +
138
- "Outside a host, compose the same blocks yourself: RunOverview, MetricTable, CaseList.",
139
- );
140
- }
141
- return activeData;
142
- }
143
-
144
- /** 判决时间的相对标注(text 面的 "41s ago";渲染面纯同步,Date.now 不算 IO)。 */
145
- function agoText(iso: string | undefined, now: number): string {
146
- if (!iso) return "";
147
- const ms = now - Date.parse(iso);
148
- if (!Number.isFinite(ms) || ms < 10_000) return "just now";
149
- const seconds = Math.floor(ms / 1000);
150
- if (seconds < 120) return `${seconds}s ago`;
151
- const minutes = Math.floor(seconds / 60);
152
- if (minutes < 120) return `${minutes}m ago`;
153
- const hours = Math.floor(minutes / 60);
154
- if (hours < 48) return `${hours}h ago`;
155
- return `${Math.floor(hours / 24)}d ago`;
156
- }
157
-
158
- /** show 榜单(viewing-results.mdx「榜单」块):头 + experiment 表 + Failing 清单。 */
159
- function boardText(data: DefaultReportData): string {
160
- const { verdicts, board, overview } = data;
161
- const experiments = verdicts.rows.length;
162
- const head = [
163
- "Current verdicts",
164
- `${experiments} ${experiments === 1 ? "experiment" : "experiments"}`,
165
- `composed from ${board.composedFromRuns} ${board.composedFromRuns === 1 ? "run" : "runs"}`,
166
- ...(board.latestRun ? [`latest ${board.latestRun}`] : []),
167
- ].join(" · ");
168
- const headBlock = [head, ...overview.warnings.map((w) => `! ${w.message}`)].join("\n");
169
-
170
- // 表:experiment | evals(eval 级折叠计票)| pass | cost | duration(顺序随 verdicts 预排)
171
- const header = [verdicts.dimension, "evals", "pass", "cost", "duration"];
172
- const rows = verdicts.rows.map((row) => {
173
- const tally = board.tallies[row.key];
174
- const cells = row.cells as Record<string, TableData["rows"][number]["cells"][string]>;
175
- return [
176
- row.key,
177
- tally ? `${tally.passedEvals}/${tally.totalEvals}` : "—",
178
- cells[passRate.name] ? cellText(cells[passRate.name]) : "—",
179
- cells[costUSD.name] ? cellText(cells[costUSD.name]) : "—",
180
- cells[durationMs.name] ? cellText(cells[durationMs.name]) : "—",
181
- ];
182
- });
183
- const table = renderAlignedRows([header, ...rows]);
184
-
185
- const blocks = [headBlock, table];
186
- if (board.failing.length > 0) {
187
- const now = Date.now();
188
- const aligned = renderAlignedRows(
189
- board.failing.map((f) => [
190
- `✗ ${f.evalId}`,
191
- f.experimentId,
192
- f.reason ?? f.outcome,
193
- agoText(f.verdictAt, now),
194
- ]),
195
- ).split("\n");
196
- const lines: string[] = ["Failing:"];
197
- board.failing.forEach((f, i) => {
198
- lines.push(` ${aligned[i]}`);
199
- lines.push(` → niceeval show ${f.evalId}`);
200
- });
201
- if (board.failingTruncated > 0) lines.push(` (${board.failingTruncated} more not shown)`);
202
- blocks.push(lines.join("\n"));
203
- }
204
- return blocks.join("\n\n");
205
- }
206
-
207
- export const DefaultReport = defineComponent<Record<string, never>>({
208
- web() {
209
- const data = requireData();
210
- return (
211
- <div className="nre nre-default-report">
212
- <RunOverview data={data.overview} />
213
- <MetricTable data={data.verdicts} />
214
- {data.cases.rows.length > 0 && <CaseList data={data.cases} />}
215
- </div>
216
- );
217
- },
218
- text() {
219
- return boardText(requireData());
220
- },
221
- });
222
- DefaultReport.displayName = "DefaultReport";