@mailwoman/corpus 9.3.0 → 10.0.0

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 (991) hide show
  1. package/README.md +16 -0
  2. package/lib/adapters/geonames/adapter.ts +3 -3
  3. package/lib/adapters/geonames/postal/adapter.ts +122 -0
  4. package/lib/adapters/index.ts +39 -32
  5. package/lib/adapters/openaddresses/adapter.ts +4 -7
  6. package/lib/adapters/osm/adapter.ts +278 -0
  7. package/lib/adapters/overture/adapter.ts +22 -7
  8. package/lib/adapters/utils/index.ts +1 -1
  9. package/lib/adapters/wof/admin/json/adapter.ts +239 -0
  10. package/lib/adapters/wof/json-rows.ts +181 -0
  11. package/lib/adapters/wof/postalcode-json/adapter.ts +196 -0
  12. package/lib/au/adapters/gnaf/adapter.ts +152 -0
  13. package/lib/au/adapters/gnaf/assemble.ts +225 -0
  14. package/lib/cz/recipes/pcfirst-preposition.ts +112 -0
  15. package/lib/de/recipes/locale.ts +360 -0
  16. package/lib/fr/adapters/ban/adapter.ts +159 -0
  17. package/lib/fr/recipes/admin-split.ts +243 -0
  18. package/lib/fr/recipes/bare-street.ts +171 -0
  19. package/lib/fr/recipes/fragment.ts +334 -0
  20. package/lib/fr/recipes/lieudit.ts +274 -0
  21. package/lib/fr/recipes/order.ts +246 -0
  22. package/lib/fr/tools/fetch/ban.ts +289 -0
  23. package/lib/index.ts +1 -1
  24. package/lib/international/recipes/locale.ts +598 -0
  25. package/lib/jp/tools/fetch/houjin.ts +141 -0
  26. package/lib/kr/adapters/juso/label-rows.ts +259 -0
  27. package/lib/kr/adapters/localdata/align.ts +299 -0
  28. package/lib/kr/adapters/localdata/key-index.ts +185 -0
  29. package/lib/kr/adapters/localdata/permit-rows.ts +186 -0
  30. package/lib/kr/tools/fetch/juso.ts +172 -0
  31. package/lib/kr/tools/fetch/localdata.ts +206 -0
  32. package/lib/nl/recipes/postcode.ts +117 -0
  33. package/lib/no/recipes/fragment.ts +213 -0
  34. package/lib/no/recipes/street/index.ts +91 -0
  35. package/lib/no/recipes/street/led.ts +135 -0
  36. package/lib/recipes/anchor-absorption.ts +2 -2
  37. package/lib/recipes/bare/country.ts +95 -0
  38. package/lib/recipes/bare/postcode/eval.ts +99 -0
  39. package/lib/recipes/bare/postcode/index.ts +327 -0
  40. package/lib/recipes/boundary-stress.ts +3 -1
  41. package/lib/recipes/country-balanced.ts +3 -2
  42. package/lib/recipes/house-venue.ts +2 -1
  43. package/lib/recipes/index.ts +28 -21
  44. package/lib/recipes/intersection.ts +3 -2
  45. package/lib/recipes/po/box/cedex/au-nz.ts +82 -0
  46. package/lib/recipes/po/box/cedex/ca.ts +58 -0
  47. package/lib/recipes/po/box/cedex/fr.ts +76 -0
  48. package/lib/recipes/po/box/cedex/phrases.ts +108 -0
  49. package/lib/recipes/po/box/cedex/recipe.ts +203 -0
  50. package/lib/recipes/po/box/cedex/sources.ts +189 -0
  51. package/lib/recipes/po/box/cedex/types.ts +35 -0
  52. package/lib/recipes/po/box/cedex/us.ts +69 -0
  53. package/lib/recipes/po/box/cedex/vocabulary.ts +106 -0
  54. package/lib/recipes/po/box/cedex.ts +9 -0
  55. package/lib/recipes/po/box/index.ts +128 -0
  56. package/lib/recipes/scaffold.ts +18 -42
  57. package/lib/recipes/street/affix.ts +807 -0
  58. package/lib/recipes/street/bare.ts +102 -0
  59. package/lib/recipes/street.ts +3 -1
  60. package/lib/recipes/sub/venue/context.ts +149 -0
  61. package/lib/recipes/sub/venue/index.ts +997 -0
  62. package/lib/recipes/sub/venue/sources.ts +739 -0
  63. package/lib/recipes/trailing-region.ts +9 -9
  64. package/lib/recipes/unit.ts +3 -3
  65. package/lib/sg/recipes/register.ts +270 -0
  66. package/lib/sg/tools/fetch/acra.ts +179 -0
  67. package/lib/si/recipes/bare-village.ts +99 -0
  68. package/lib/south-asia/recipes/register.ts +214 -0
  69. package/lib/synthesizers/anchor-absorption.ts +1 -1
  70. package/lib/synthesizers/locale.ts +261 -0
  71. package/lib/synthesizers/street.ts +1 -1
  72. package/lib/synthesizers/utils.ts +2 -2
  73. package/lib/tools/audit.ts +3 -2
  74. package/lib/tools/corpus-stats.ts +3 -3
  75. package/lib/tools/fetch/download/index.ts +10 -0
  76. package/lib/tools/fetch/download/manifest.ts +102 -0
  77. package/lib/tools/fetch/download/network.ts +226 -0
  78. package/lib/tools/fetch/download/retry.ts +173 -0
  79. package/lib/tools/fetch/geonames/dump.ts +257 -0
  80. package/lib/tools/fetch/geonames/postal.ts +180 -0
  81. package/lib/tools/fetch/index.ts +47 -20
  82. package/lib/tools/fetch/openaddresses.ts +2 -2
  83. package/lib/tools/fetch/ourairports.ts +2 -2
  84. package/lib/tools/fetch/ppd.ts +1 -3
  85. package/lib/tools/fetch/wikidata-subvenue.ts +2 -2
  86. package/lib/tools/golden/promote.ts +340 -0
  87. package/lib/tools/golden/relabel-street/directory.ts +340 -0
  88. package/lib/tools/golden/relabel-street/index.ts +9 -0
  89. package/lib/tools/golden/relabel-street/row.ts +361 -0
  90. package/lib/tools/index.ts +10 -10
  91. package/lib/tools/ingest-csv.ts +5 -7
  92. package/lib/tools/lint/slice/vocab.ts +376 -0
  93. package/lib/tools/overlay-manifest.ts +92 -25
  94. package/lib/tools/overture-subvenue.ts +1 -1
  95. package/lib/tools/postcode-triples.ts +138 -14
  96. package/lib/tools/sub/venue/harvest.ts +234 -0
  97. package/lib/tools/sub/venue/head-nouns.ts +245 -0
  98. package/lib/tools/sub/venue/lexicon.ts +429 -0
  99. package/lib/tools/sub/venue/surfaces.ts +95 -0
  100. package/lib/tools/sub/venue/table.ts +281 -0
  101. package/lib/tools/sub/venue/wikidata.ts +85 -0
  102. package/lib/tw/tools/fetch/gcis.ts +180 -0
  103. package/lib/types.ts +1 -1
  104. package/lib/us/adapters/fcc-bdc/adapter.ts +173 -0
  105. package/lib/us/adapters/state/hi-schools/adapter.ts +160 -0
  106. package/lib/us/adapters/state/ia-contractors/adapter.ts +136 -0
  107. package/lib/us/adapters/state/ny-notaries/adapter.ts +133 -0
  108. package/lib/us/adapters/state/tx-notaries/adapter.ts +149 -0
  109. package/lib/us/adapters/tiger/adapter.ts +250 -0
  110. package/lib/us/adapters/usgov/hrsa-fqhc/adapter.ts +156 -0
  111. package/lib/us/adapters/usgov/imls-pls/adapter.ts +134 -0
  112. package/lib/us/adapters/usgov/irs-bmf/adapter.ts +145 -0
  113. package/lib/us/adapters/usgov/nad/adapter.ts +324 -0
  114. package/lib/us/adapters/usgov/nppes/adapter.ts +143 -0
  115. package/lib/us/adapters/usgov/samhsa-treatment-locator/adapter.ts +182 -0
  116. package/lib/us/tools/fetch/hrsa.ts +59 -0
  117. package/lib/us/tools/fetch/imls-pls.ts +149 -0
  118. package/lib/us/tools/fetch/nad.ts +341 -0
  119. package/lib/us/tools/fetch/nppes.ts +165 -0
  120. package/lib/us/tools/fetch/state/hi-schools.ts +161 -0
  121. package/lib/us/tools/fetch/state/sources.ts +146 -0
  122. package/lib/us/tools/fetch/tiger-full.ts +308 -0
  123. package/lib/utils/align.ts +1 -1
  124. package/lib/utils/golden.ts +12 -8
  125. package/lib/utils/wof-json.ts +7 -13
  126. package/out/adapters/geonames/adapter.d.ts +1 -1
  127. package/out/adapters/geonames/adapter.js +3 -3
  128. package/out/adapters/geonames/adapter.js.map +1 -1
  129. package/out/adapters/geonames/postal/adapter.d.ts.map +1 -0
  130. package/out/adapters/geonames/postal/adapter.js +107 -0
  131. package/out/adapters/geonames/postal/adapter.js.map +1 -0
  132. package/out/adapters/index.d.ts +16 -16
  133. package/out/adapters/index.d.ts.map +1 -1
  134. package/out/adapters/index.js +34 -32
  135. package/out/adapters/index.js.map +1 -1
  136. package/out/adapters/openaddresses/adapter.d.ts.map +1 -1
  137. package/out/adapters/openaddresses/adapter.js +4 -6
  138. package/out/adapters/openaddresses/adapter.js.map +1 -1
  139. package/out/adapters/osm/adapter.d.ts +95 -0
  140. package/out/adapters/osm/adapter.d.ts.map +1 -0
  141. package/out/adapters/osm/adapter.js +211 -0
  142. package/out/adapters/osm/adapter.js.map +1 -0
  143. package/out/adapters/overture/adapter.d.ts +10 -0
  144. package/out/adapters/overture/adapter.d.ts.map +1 -1
  145. package/out/adapters/overture/adapter.js +20 -7
  146. package/out/adapters/overture/adapter.js.map +1 -1
  147. package/out/adapters/synth-po-box/adapter.js.map +1 -1
  148. package/out/adapters/utils/index.d.ts +1 -1
  149. package/out/adapters/utils/index.d.ts.map +1 -1
  150. package/out/adapters/utils/index.js.map +1 -1
  151. package/out/adapters/wof/admin/json/adapter.d.ts +79 -0
  152. package/out/adapters/wof/admin/json/adapter.d.ts.map +1 -0
  153. package/out/adapters/wof/admin/json/adapter.js +204 -0
  154. package/out/adapters/wof/admin/json/adapter.js.map +1 -0
  155. package/out/adapters/wof/json-rows.d.ts +73 -0
  156. package/out/adapters/wof/json-rows.d.ts.map +1 -0
  157. package/out/adapters/wof/json-rows.js +115 -0
  158. package/out/adapters/wof/json-rows.js.map +1 -0
  159. package/out/adapters/wof/postalcode-json/adapter.d.ts +65 -0
  160. package/out/adapters/wof/postalcode-json/adapter.d.ts.map +1 -0
  161. package/out/adapters/wof/postalcode-json/adapter.js +167 -0
  162. package/out/adapters/wof/postalcode-json/adapter.js.map +1 -0
  163. package/out/au/adapters/gnaf/adapter.d.ts.map +1 -0
  164. package/out/au/adapters/gnaf/adapter.js +127 -0
  165. package/out/au/adapters/gnaf/adapter.js.map +1 -0
  166. package/out/au/adapters/gnaf/assemble.d.ts +65 -0
  167. package/out/au/adapters/gnaf/assemble.d.ts.map +1 -0
  168. package/out/au/adapters/gnaf/assemble.js +154 -0
  169. package/out/au/adapters/gnaf/assemble.js.map +1 -0
  170. package/out/cn/recipes/organizational-units.d.ts +36 -0
  171. package/out/cn/recipes/organizational-units.d.ts.map +1 -0
  172. package/out/cn/recipes/organizational-units.js +159 -0
  173. package/out/cn/recipes/organizational-units.js.map +1 -0
  174. package/out/cz/recipes/pcfirst-preposition.d.ts +22 -0
  175. package/out/cz/recipes/pcfirst-preposition.d.ts.map +1 -0
  176. package/out/cz/recipes/pcfirst-preposition.js +101 -0
  177. package/out/cz/recipes/pcfirst-preposition.js.map +1 -0
  178. package/out/de/recipes/locale.d.ts +44 -0
  179. package/out/de/recipes/locale.d.ts.map +1 -0
  180. package/out/de/recipes/locale.js +289 -0
  181. package/out/de/recipes/locale.js.map +1 -0
  182. package/out/env.d.ts +1 -1
  183. package/out/fr/adapters/ban/adapter.d.ts.map +1 -0
  184. package/out/fr/adapters/ban/adapter.js +123 -0
  185. package/out/fr/adapters/ban/adapter.js.map +1 -0
  186. package/out/fr/adapters/ban/street-decompose.d.ts.map +1 -0
  187. package/out/fr/adapters/ban/street-decompose.js.map +1 -0
  188. package/out/fr/recipes/admin-split.d.ts +38 -0
  189. package/out/fr/recipes/admin-split.d.ts.map +1 -0
  190. package/out/fr/recipes/admin-split.js +192 -0
  191. package/out/fr/recipes/admin-split.js.map +1 -0
  192. package/out/fr/recipes/bare-street.d.ts +28 -0
  193. package/out/fr/recipes/bare-street.d.ts.map +1 -0
  194. package/out/fr/recipes/bare-street.js +144 -0
  195. package/out/fr/recipes/bare-street.js.map +1 -0
  196. package/out/fr/recipes/fragment.d.ts +77 -0
  197. package/out/fr/recipes/fragment.d.ts.map +1 -0
  198. package/out/fr/recipes/fragment.js +277 -0
  199. package/out/fr/recipes/fragment.js.map +1 -0
  200. package/out/fr/recipes/lieudit.d.ts +32 -0
  201. package/out/fr/recipes/lieudit.d.ts.map +1 -0
  202. package/out/fr/recipes/lieudit.js +211 -0
  203. package/out/fr/recipes/lieudit.js.map +1 -0
  204. package/out/fr/recipes/order.d.ts +38 -0
  205. package/out/fr/recipes/order.d.ts.map +1 -0
  206. package/out/fr/recipes/order.js +198 -0
  207. package/out/fr/recipes/order.js.map +1 -0
  208. package/out/fr/tools/fetch/ban.d.ts +31 -0
  209. package/out/fr/tools/fetch/ban.d.ts.map +1 -0
  210. package/out/fr/tools/fetch/ban.js +239 -0
  211. package/out/fr/tools/fetch/ban.js.map +1 -0
  212. package/out/index.d.ts +1 -1
  213. package/out/index.d.ts.map +1 -1
  214. package/out/index.js +1 -1
  215. package/out/index.js.map +1 -1
  216. package/out/international/recipes/locale.d.ts +136 -0
  217. package/out/international/recipes/locale.d.ts.map +1 -0
  218. package/out/international/recipes/locale.js +458 -0
  219. package/out/international/recipes/locale.js.map +1 -0
  220. package/out/jp/adapters/wof-admin/adapter.d.ts.map +1 -0
  221. package/out/jp/adapters/wof-admin/adapter.js.map +1 -0
  222. package/out/jp/tools/fetch/houjin.d.ts +24 -0
  223. package/out/jp/tools/fetch/houjin.d.ts.map +1 -0
  224. package/out/jp/tools/fetch/houjin.js +102 -0
  225. package/out/jp/tools/fetch/houjin.js.map +1 -0
  226. package/out/kr/adapters/juso/label-rows.d.ts +76 -0
  227. package/out/kr/adapters/juso/label-rows.d.ts.map +1 -0
  228. package/out/kr/adapters/juso/label-rows.js +194 -0
  229. package/out/kr/adapters/juso/label-rows.js.map +1 -0
  230. package/out/kr/adapters/localdata/align.d.ts +43 -0
  231. package/out/kr/adapters/localdata/align.d.ts.map +1 -0
  232. package/out/kr/adapters/localdata/align.js +223 -0
  233. package/out/kr/adapters/localdata/align.js.map +1 -0
  234. package/out/kr/adapters/localdata/key-index.d.ts +71 -0
  235. package/out/kr/adapters/localdata/key-index.d.ts.map +1 -0
  236. package/out/kr/adapters/localdata/key-index.js +133 -0
  237. package/out/kr/adapters/localdata/key-index.js.map +1 -0
  238. package/out/kr/adapters/localdata/permit-rows.d.ts +66 -0
  239. package/out/kr/adapters/localdata/permit-rows.d.ts.map +1 -0
  240. package/out/kr/adapters/localdata/permit-rows.js +131 -0
  241. package/out/kr/adapters/localdata/permit-rows.js.map +1 -0
  242. package/out/kr/tools/fetch/juso.d.ts +30 -0
  243. package/out/kr/tools/fetch/juso.d.ts.map +1 -0
  244. package/out/kr/tools/fetch/juso.js +120 -0
  245. package/out/kr/tools/fetch/juso.js.map +1 -0
  246. package/out/kr/tools/fetch/localdata.d.ts +30 -0
  247. package/out/kr/tools/fetch/localdata.d.ts.map +1 -0
  248. package/out/kr/tools/fetch/localdata.js +139 -0
  249. package/out/kr/tools/fetch/localdata.js.map +1 -0
  250. package/out/nl/recipes/postcode.d.ts +26 -0
  251. package/out/nl/recipes/postcode.d.ts.map +1 -0
  252. package/out/nl/recipes/postcode.js +103 -0
  253. package/out/nl/recipes/postcode.js.map +1 -0
  254. package/out/no/recipes/fragment.d.ts +43 -0
  255. package/out/no/recipes/fragment.d.ts.map +1 -0
  256. package/out/no/recipes/fragment.js +183 -0
  257. package/out/no/recipes/fragment.js.map +1 -0
  258. package/out/no/recipes/street/index.d.ts +18 -0
  259. package/out/no/recipes/street/index.d.ts.map +1 -0
  260. package/out/no/recipes/street/index.js +72 -0
  261. package/out/no/recipes/street/index.js.map +1 -0
  262. package/out/no/recipes/street/led.d.ts +24 -0
  263. package/out/no/recipes/street/led.d.ts.map +1 -0
  264. package/out/no/recipes/street/led.js +114 -0
  265. package/out/no/recipes/street/led.js.map +1 -0
  266. package/out/recipes/anchor-absorption.d.ts.map +1 -1
  267. package/out/recipes/anchor-absorption.js +2 -1
  268. package/out/recipes/anchor-absorption.js.map +1 -1
  269. package/out/recipes/bare/country.d.ts +25 -0
  270. package/out/recipes/bare/country.d.ts.map +1 -0
  271. package/out/recipes/bare/country.js +88 -0
  272. package/out/recipes/bare/country.js.map +1 -0
  273. package/out/recipes/bare/postcode/eval.d.ts +29 -0
  274. package/out/recipes/bare/postcode/eval.d.ts.map +1 -0
  275. package/out/recipes/bare/postcode/eval.js +85 -0
  276. package/out/recipes/bare/postcode/eval.js.map +1 -0
  277. package/out/recipes/bare/postcode/index.d.ts +62 -0
  278. package/out/recipes/bare/postcode/index.d.ts.map +1 -0
  279. package/out/recipes/bare/postcode/index.js +267 -0
  280. package/out/recipes/bare/postcode/index.js.map +1 -0
  281. package/out/recipes/boundary-stress.d.ts.map +1 -1
  282. package/out/recipes/boundary-stress.js +2 -1
  283. package/out/recipes/boundary-stress.js.map +1 -1
  284. package/out/recipes/country-balanced.d.ts.map +1 -1
  285. package/out/recipes/country-balanced.js +2 -1
  286. package/out/recipes/country-balanced.js.map +1 -1
  287. package/out/recipes/house-venue.d.ts.map +1 -1
  288. package/out/recipes/house-venue.js +2 -1
  289. package/out/recipes/house-venue.js.map +1 -1
  290. package/out/recipes/index.d.ts.map +1 -1
  291. package/out/recipes/index.js +28 -21
  292. package/out/recipes/index.js.map +1 -1
  293. package/out/recipes/intersection.d.ts.map +1 -1
  294. package/out/recipes/intersection.js +2 -1
  295. package/out/recipes/intersection.js.map +1 -1
  296. package/out/recipes/po/box/cedex/au-nz.d.ts +7 -0
  297. package/out/recipes/po/box/cedex/au-nz.d.ts.map +1 -0
  298. package/out/recipes/po/box/cedex/au-nz.js +56 -0
  299. package/out/recipes/po/box/cedex/au-nz.js.map +1 -0
  300. package/out/recipes/po/box/cedex/ca.d.ts +7 -0
  301. package/out/recipes/po/box/cedex/ca.d.ts.map +1 -0
  302. package/out/recipes/po/box/cedex/ca.js +38 -0
  303. package/out/recipes/po/box/cedex/ca.js.map +1 -0
  304. package/out/recipes/po/box/cedex/fr.d.ts +7 -0
  305. package/out/recipes/po/box/cedex/fr.d.ts.map +1 -0
  306. package/out/recipes/po/box/cedex/fr.js +57 -0
  307. package/out/recipes/po/box/cedex/fr.js.map +1 -0
  308. package/out/recipes/po/box/cedex/phrases.d.ts +8 -0
  309. package/out/recipes/po/box/cedex/phrases.d.ts.map +1 -0
  310. package/out/recipes/po/box/cedex/phrases.js +76 -0
  311. package/out/recipes/po/box/cedex/phrases.js.map +1 -0
  312. package/out/recipes/po/box/cedex/recipe.d.ts +9 -0
  313. package/out/recipes/po/box/cedex/recipe.d.ts.map +1 -0
  314. package/out/recipes/po/box/cedex/recipe.js +153 -0
  315. package/out/recipes/po/box/cedex/recipe.js.map +1 -0
  316. package/out/recipes/po/box/cedex/sources.d.ts +86 -0
  317. package/out/recipes/po/box/cedex/sources.d.ts.map +1 -0
  318. package/out/recipes/po/box/cedex/sources.js +143 -0
  319. package/out/recipes/po/box/cedex/sources.js.map +1 -0
  320. package/out/recipes/po/box/cedex/types.d.ts +31 -0
  321. package/out/recipes/po/box/cedex/types.d.ts.map +1 -0
  322. package/out/recipes/po/box/cedex/types.js +5 -0
  323. package/out/recipes/po/box/cedex/types.js.map +1 -0
  324. package/out/recipes/po/box/cedex/us.d.ts +7 -0
  325. package/out/recipes/po/box/cedex/us.d.ts.map +1 -0
  326. package/out/recipes/po/box/cedex/us.js +51 -0
  327. package/out/recipes/po/box/cedex/us.js.map +1 -0
  328. package/out/recipes/po/box/cedex/vocabulary.d.ts +68 -0
  329. package/out/recipes/po/box/cedex/vocabulary.d.ts.map +1 -0
  330. package/out/recipes/po/box/cedex/vocabulary.js +90 -0
  331. package/out/recipes/po/box/cedex/vocabulary.js.map +1 -0
  332. package/out/recipes/po/box/cedex.d.ts +9 -0
  333. package/out/recipes/po/box/cedex.d.ts.map +1 -0
  334. package/out/recipes/po/box/cedex.js +9 -0
  335. package/out/recipes/po/box/cedex.js.map +1 -0
  336. package/out/recipes/po/box/index.d.ts +17 -0
  337. package/out/recipes/po/box/index.d.ts.map +1 -0
  338. package/out/recipes/po/box/index.js +105 -0
  339. package/out/recipes/po/box/index.js.map +1 -0
  340. package/out/recipes/scaffold.d.ts +20 -1
  341. package/out/recipes/scaffold.d.ts.map +1 -1
  342. package/out/recipes/scaffold.js +18 -49
  343. package/out/recipes/scaffold.js.map +1 -1
  344. package/out/recipes/street/affix.d.ts +89 -0
  345. package/out/recipes/street/affix.d.ts.map +1 -0
  346. package/out/recipes/street/affix.js +565 -0
  347. package/out/recipes/street/affix.js.map +1 -0
  348. package/out/recipes/street/bare.d.ts +25 -0
  349. package/out/recipes/street/bare.d.ts.map +1 -0
  350. package/out/recipes/street/bare.js +87 -0
  351. package/out/recipes/street/bare.js.map +1 -0
  352. package/out/recipes/street.d.ts.map +1 -1
  353. package/out/recipes/street.js +2 -1
  354. package/out/recipes/street.js.map +1 -1
  355. package/out/recipes/sub/venue/context.d.ts +29 -0
  356. package/out/recipes/sub/venue/context.d.ts.map +1 -0
  357. package/out/recipes/sub/venue/context.js +106 -0
  358. package/out/recipes/sub/venue/context.js.map +1 -0
  359. package/out/recipes/sub/venue/index.d.ts +270 -0
  360. package/out/recipes/sub/venue/index.d.ts.map +1 -0
  361. package/out/recipes/sub/venue/index.js +728 -0
  362. package/out/recipes/sub/venue/index.js.map +1 -0
  363. package/out/recipes/sub/venue/sources.d.ts +209 -0
  364. package/out/recipes/sub/venue/sources.d.ts.map +1 -0
  365. package/out/recipes/sub/venue/sources.js +555 -0
  366. package/out/recipes/sub/venue/sources.js.map +1 -0
  367. package/out/recipes/trailing-region.d.ts.map +1 -1
  368. package/out/recipes/trailing-region.js +8 -2
  369. package/out/recipes/trailing-region.js.map +1 -1
  370. package/out/recipes/unit.d.ts.map +1 -1
  371. package/out/recipes/unit.js +2 -1
  372. package/out/recipes/unit.js.map +1 -1
  373. package/out/sg/recipes/register.d.ts +69 -0
  374. package/out/sg/recipes/register.d.ts.map +1 -0
  375. package/out/sg/recipes/register.js +211 -0
  376. package/out/sg/recipes/register.js.map +1 -0
  377. package/out/sg/tools/fetch/acra.d.ts +23 -0
  378. package/out/sg/tools/fetch/acra.d.ts.map +1 -0
  379. package/out/sg/tools/fetch/acra.js +126 -0
  380. package/out/sg/tools/fetch/acra.js.map +1 -0
  381. package/out/si/recipes/bare-village.d.ts +32 -0
  382. package/out/si/recipes/bare-village.d.ts.map +1 -0
  383. package/out/si/recipes/bare-village.js +89 -0
  384. package/out/si/recipes/bare-village.js.map +1 -0
  385. package/out/south-asia/recipes/register.d.ts +51 -0
  386. package/out/south-asia/recipes/register.d.ts.map +1 -0
  387. package/out/south-asia/recipes/register.js +154 -0
  388. package/out/south-asia/recipes/register.js.map +1 -0
  389. package/out/synthesizers/anchor-absorption.d.ts +1 -1
  390. package/out/synthesizers/anchor-absorption.d.ts.map +1 -1
  391. package/out/synthesizers/anchor-absorption.js.map +1 -1
  392. package/out/synthesizers/boundary-stress.js.map +1 -1
  393. package/out/synthesizers/house-venue.js.map +1 -1
  394. package/out/synthesizers/intersection.js.map +1 -1
  395. package/out/synthesizers/locale.d.ts +116 -0
  396. package/out/synthesizers/locale.d.ts.map +1 -0
  397. package/out/synthesizers/locale.js +150 -0
  398. package/out/synthesizers/locale.js.map +1 -0
  399. package/out/synthesizers/no-street.d.ts.map +1 -1
  400. package/out/synthesizers/no-street.js.map +1 -1
  401. package/out/synthesizers/po-box.js.map +1 -1
  402. package/out/synthesizers/street.d.ts.map +1 -1
  403. package/out/synthesizers/street.js +1 -1
  404. package/out/synthesizers/street.js.map +1 -1
  405. package/out/synthesizers/utils.js +1 -1
  406. package/out/synthesizers/utils.js.map +1 -1
  407. package/out/test-kit/corpus-recipe.d.ts.map +1 -1
  408. package/out/test-kit/corpus-recipe.js.map +1 -1
  409. package/out/tools/audit.d.ts.map +1 -1
  410. package/out/tools/audit.js +3 -2
  411. package/out/tools/audit.js.map +1 -1
  412. package/out/tools/corpus-stats.js +3 -3
  413. package/out/tools/corpus-stats.js.map +1 -1
  414. package/out/tools/fetch/download/index.d.ts +10 -0
  415. package/out/tools/fetch/download/index.d.ts.map +1 -0
  416. package/out/tools/fetch/download/index.js +10 -0
  417. package/out/tools/fetch/download/index.js.map +1 -0
  418. package/out/tools/fetch/download/manifest.d.ts +47 -0
  419. package/out/tools/fetch/download/manifest.d.ts.map +1 -0
  420. package/out/tools/fetch/download/manifest.js +74 -0
  421. package/out/tools/fetch/download/manifest.js.map +1 -0
  422. package/out/tools/fetch/download/network.d.ts +105 -0
  423. package/out/tools/fetch/download/network.d.ts.map +1 -0
  424. package/out/tools/fetch/download/network.js +130 -0
  425. package/out/tools/fetch/download/network.js.map +1 -0
  426. package/out/tools/fetch/download/retry.d.ts +45 -0
  427. package/out/tools/fetch/download/retry.d.ts.map +1 -0
  428. package/out/tools/fetch/download/retry.js +136 -0
  429. package/out/tools/fetch/download/retry.js.map +1 -0
  430. package/out/tools/fetch/geonames/dump.d.ts +91 -0
  431. package/out/tools/fetch/geonames/dump.d.ts.map +1 -0
  432. package/out/tools/fetch/geonames/dump.js +171 -0
  433. package/out/tools/fetch/geonames/dump.js.map +1 -0
  434. package/out/tools/fetch/geonames/postal.d.ts +76 -0
  435. package/out/tools/fetch/geonames/postal.d.ts.map +1 -0
  436. package/out/tools/fetch/geonames/postal.js +126 -0
  437. package/out/tools/fetch/geonames/postal.js.map +1 -0
  438. package/out/tools/fetch/index.d.ts +47 -20
  439. package/out/tools/fetch/index.d.ts.map +1 -1
  440. package/out/tools/fetch/index.js +47 -20
  441. package/out/tools/fetch/index.js.map +1 -1
  442. package/out/tools/fetch/openaddresses.d.ts +1 -1
  443. package/out/tools/fetch/openaddresses.d.ts.map +1 -1
  444. package/out/tools/fetch/openaddresses.js +1 -1
  445. package/out/tools/fetch/openaddresses.js.map +1 -1
  446. package/out/tools/fetch/ourairports.d.ts +1 -1
  447. package/out/tools/fetch/ourairports.d.ts.map +1 -1
  448. package/out/tools/fetch/ourairports.js +1 -1
  449. package/out/tools/fetch/ourairports.js.map +1 -1
  450. package/out/tools/fetch/ppd.js +1 -3
  451. package/out/tools/fetch/ppd.js.map +1 -1
  452. package/out/tools/fetch/wikidata-subvenue.d.ts +1 -1
  453. package/out/tools/fetch/wikidata-subvenue.d.ts.map +1 -1
  454. package/out/tools/fetch/wikidata-subvenue.js +1 -1
  455. package/out/tools/fetch/wikidata-subvenue.js.map +1 -1
  456. package/out/tools/golden/expand.d.ts +90 -0
  457. package/out/tools/golden/expand.d.ts.map +1 -0
  458. package/out/tools/golden/expand.js +406 -0
  459. package/out/tools/golden/expand.js.map +1 -0
  460. package/out/tools/golden/promote.d.ts +72 -0
  461. package/out/tools/golden/promote.d.ts.map +1 -0
  462. package/out/tools/golden/promote.js +243 -0
  463. package/out/tools/golden/promote.js.map +1 -0
  464. package/out/tools/golden/relabel-street/directory.d.ts +83 -0
  465. package/out/tools/golden/relabel-street/directory.d.ts.map +1 -0
  466. package/out/tools/golden/relabel-street/directory.js +216 -0
  467. package/out/tools/golden/relabel-street/directory.js.map +1 -0
  468. package/out/tools/golden/relabel-street/index.d.ts +9 -0
  469. package/out/tools/golden/relabel-street/index.d.ts.map +1 -0
  470. package/out/tools/golden/relabel-street/index.js +9 -0
  471. package/out/tools/golden/relabel-street/index.js.map +1 -0
  472. package/out/tools/golden/relabel-street/row.d.ts +121 -0
  473. package/out/tools/golden/relabel-street/row.d.ts.map +1 -0
  474. package/out/tools/golden/relabel-street/row.js +232 -0
  475. package/out/tools/golden/relabel-street/row.js.map +1 -0
  476. package/out/tools/index.d.ts +10 -10
  477. package/out/tools/index.d.ts.map +1 -1
  478. package/out/tools/index.js +10 -10
  479. package/out/tools/index.js.map +1 -1
  480. package/out/tools/ingest-csv.d.ts.map +1 -1
  481. package/out/tools/ingest-csv.js +5 -7
  482. package/out/tools/ingest-csv.js.map +1 -1
  483. package/out/tools/lint/slice/index.d.ts +93 -0
  484. package/out/tools/lint/slice/index.d.ts.map +1 -0
  485. package/out/tools/lint/slice/index.js +322 -0
  486. package/out/tools/lint/slice/index.js.map +1 -0
  487. package/out/tools/lint/slice/vocab.d.ts +90 -0
  488. package/out/tools/lint/slice/vocab.d.ts.map +1 -0
  489. package/out/tools/lint/slice/vocab.js +254 -0
  490. package/out/tools/lint/slice/vocab.js.map +1 -0
  491. package/out/tools/overlay-manifest.d.ts +56 -6
  492. package/out/tools/overlay-manifest.d.ts.map +1 -1
  493. package/out/tools/overlay-manifest.js +56 -15
  494. package/out/tools/overlay-manifest.js.map +1 -1
  495. package/out/tools/overture-subvenue.d.ts +1 -1
  496. package/out/tools/postcode-triples.d.ts +36 -0
  497. package/out/tools/postcode-triples.d.ts.map +1 -1
  498. package/out/tools/postcode-triples.js +91 -12
  499. package/out/tools/postcode-triples.js.map +1 -1
  500. package/out/tools/slice/kryptonite.d.ts +32 -0
  501. package/out/tools/slice/kryptonite.d.ts.map +1 -0
  502. package/out/tools/slice/kryptonite.js +99 -0
  503. package/out/tools/slice/kryptonite.js.map +1 -0
  504. package/out/tools/slice/translit.d.ts +51 -0
  505. package/out/tools/slice/translit.d.ts.map +1 -0
  506. package/out/tools/slice/translit.js +166 -0
  507. package/out/tools/slice/translit.js.map +1 -0
  508. package/out/tools/sub/venue/harvest.d.ts +100 -0
  509. package/out/tools/sub/venue/harvest.d.ts.map +1 -0
  510. package/out/tools/sub/venue/harvest.js +168 -0
  511. package/out/tools/sub/venue/harvest.js.map +1 -0
  512. package/out/tools/sub/venue/head-nouns.d.ts +50 -0
  513. package/out/tools/sub/venue/head-nouns.d.ts.map +1 -0
  514. package/out/tools/sub/venue/head-nouns.js.map +1 -0
  515. package/out/tools/sub/venue/lexicon.d.ts +156 -0
  516. package/out/tools/sub/venue/lexicon.d.ts.map +1 -0
  517. package/out/tools/sub/venue/lexicon.js +283 -0
  518. package/out/tools/sub/venue/lexicon.js.map +1 -0
  519. package/out/tools/sub/venue/promotions.d.ts +94 -0
  520. package/out/tools/sub/venue/promotions.d.ts.map +1 -0
  521. package/out/tools/sub/venue/promotions.js +266 -0
  522. package/out/tools/sub/venue/promotions.js.map +1 -0
  523. package/out/tools/sub/venue/surfaces.d.ts +45 -0
  524. package/out/tools/sub/venue/surfaces.d.ts.map +1 -0
  525. package/out/tools/sub/venue/surfaces.js.map +1 -0
  526. package/out/tools/sub/venue/table.d.ts +229 -0
  527. package/out/tools/sub/venue/table.d.ts.map +1 -0
  528. package/out/tools/sub/venue/table.js.map +1 -0
  529. package/out/tools/sub/venue/wikidata.d.ts +26 -0
  530. package/out/tools/sub/venue/wikidata.d.ts.map +1 -0
  531. package/out/tools/sub/venue/wikidata.js +62 -0
  532. package/out/tools/sub/venue/wikidata.js.map +1 -0
  533. package/out/tw/tools/fetch/gcis.d.ts +23 -0
  534. package/out/tw/tools/fetch/gcis.d.ts.map +1 -0
  535. package/out/tw/tools/fetch/gcis.js +135 -0
  536. package/out/tw/tools/fetch/gcis.js.map +1 -0
  537. package/out/types.d.ts +1 -1
  538. package/out/types.d.ts.map +1 -1
  539. package/out/us/adapters/fcc-bdc/adapter.d.ts.map +1 -0
  540. package/out/us/adapters/fcc-bdc/adapter.js +142 -0
  541. package/out/us/adapters/fcc-bdc/adapter.js.map +1 -0
  542. package/out/us/adapters/state/hi-schools/adapter.d.ts.map +1 -0
  543. package/out/us/adapters/state/hi-schools/adapter.js +136 -0
  544. package/out/us/adapters/state/hi-schools/adapter.js.map +1 -0
  545. package/out/us/adapters/state/hi-schools/workbook.d.ts.map +1 -0
  546. package/out/us/adapters/state/hi-schools/workbook.js.map +1 -0
  547. package/out/us/adapters/state/ia-contractors/adapter.d.ts.map +1 -0
  548. package/out/us/adapters/state/ia-contractors/adapter.js +104 -0
  549. package/out/us/adapters/state/ia-contractors/adapter.js.map +1 -0
  550. package/out/us/adapters/state/ny-notaries/adapter.d.ts.map +1 -0
  551. package/out/us/adapters/state/ny-notaries/adapter.js +112 -0
  552. package/out/us/adapters/state/ny-notaries/adapter.js.map +1 -0
  553. package/out/us/adapters/state/tx-notaries/adapter.d.ts.map +1 -0
  554. package/out/us/adapters/state/tx-notaries/adapter.js +118 -0
  555. package/out/us/adapters/state/tx-notaries/adapter.js.map +1 -0
  556. package/out/us/adapters/tiger/adapter.d.ts +55 -0
  557. package/out/us/adapters/tiger/adapter.d.ts.map +1 -0
  558. package/out/us/adapters/tiger/adapter.js +197 -0
  559. package/out/us/adapters/tiger/adapter.js.map +1 -0
  560. package/out/us/adapters/tiger/street-decompose.d.ts.map +1 -0
  561. package/out/us/adapters/tiger/street-decompose.js.map +1 -0
  562. package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts.map +1 -0
  563. package/out/us/adapters/usgov/hrsa-fqhc/adapter.js +121 -0
  564. package/out/us/adapters/usgov/hrsa-fqhc/adapter.js.map +1 -0
  565. package/out/us/adapters/usgov/imls-pls/adapter.d.ts.map +1 -0
  566. package/out/us/adapters/usgov/imls-pls/adapter.js +106 -0
  567. package/out/us/adapters/usgov/imls-pls/adapter.js.map +1 -0
  568. package/out/us/adapters/usgov/irs-bmf/adapter.d.ts.map +1 -0
  569. package/out/us/adapters/usgov/irs-bmf/adapter.js +116 -0
  570. package/out/us/adapters/usgov/irs-bmf/adapter.js.map +1 -0
  571. package/out/us/adapters/usgov/nad/adapter.d.ts.map +1 -0
  572. package/out/us/adapters/usgov/nad/adapter.js +239 -0
  573. package/out/us/adapters/usgov/nad/adapter.js.map +1 -0
  574. package/out/us/adapters/usgov/nppes/adapter.d.ts.map +1 -0
  575. package/out/us/adapters/usgov/nppes/adapter.js +110 -0
  576. package/out/us/adapters/usgov/nppes/adapter.js.map +1 -0
  577. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts.map +1 -0
  578. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js +143 -0
  579. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js.map +1 -0
  580. package/out/us/fips-state.d.ts +41 -0
  581. package/out/us/fips-state.d.ts.map +1 -0
  582. package/out/us/fips-state.js +100 -0
  583. package/out/us/fips-state.js.map +1 -0
  584. package/out/us/tools/fetch/hrsa.d.ts +18 -0
  585. package/out/us/tools/fetch/hrsa.d.ts.map +1 -0
  586. package/out/us/tools/fetch/hrsa.js +46 -0
  587. package/out/us/tools/fetch/hrsa.js.map +1 -0
  588. package/out/us/tools/fetch/imls-pls.d.ts +25 -0
  589. package/out/us/tools/fetch/imls-pls.d.ts.map +1 -0
  590. package/out/us/tools/fetch/imls-pls.js +108 -0
  591. package/out/us/tools/fetch/imls-pls.js.map +1 -0
  592. package/out/us/tools/fetch/nad.d.ts +70 -0
  593. package/out/us/tools/fetch/nad.d.ts.map +1 -0
  594. package/out/us/tools/fetch/nad.js +234 -0
  595. package/out/us/tools/fetch/nad.js.map +1 -0
  596. package/out/us/tools/fetch/nppes.d.ts +26 -0
  597. package/out/us/tools/fetch/nppes.d.ts.map +1 -0
  598. package/out/us/tools/fetch/nppes.js +126 -0
  599. package/out/us/tools/fetch/nppes.js.map +1 -0
  600. package/out/us/tools/fetch/state/hi-schools.d.ts +26 -0
  601. package/out/us/tools/fetch/state/hi-schools.d.ts.map +1 -0
  602. package/out/us/tools/fetch/state/hi-schools.js +112 -0
  603. package/out/us/tools/fetch/state/hi-schools.js.map +1 -0
  604. package/out/us/tools/fetch/state/sources.d.ts +26 -0
  605. package/out/us/tools/fetch/state/sources.d.ts.map +1 -0
  606. package/out/us/tools/fetch/state/sources.js +111 -0
  607. package/out/us/tools/fetch/state/sources.js.map +1 -0
  608. package/out/us/tools/fetch/tiger-full.d.ts +22 -0
  609. package/out/us/tools/fetch/tiger-full.d.ts.map +1 -0
  610. package/out/us/tools/fetch/tiger-full.js +217 -0
  611. package/out/us/tools/fetch/tiger-full.js.map +1 -0
  612. package/out/utils/align.d.ts +1 -1
  613. package/out/utils/align.d.ts.map +1 -1
  614. package/out/utils/align.js.map +1 -1
  615. package/out/utils/golden.d.ts +7 -3
  616. package/out/utils/golden.d.ts.map +1 -1
  617. package/out/utils/golden.js +12 -8
  618. package/out/utils/golden.js.map +1 -1
  619. package/out/utils/parquet.d.ts.map +1 -1
  620. package/out/utils/split.js.map +1 -1
  621. package/out/utils/wof-json.d.ts.map +1 -1
  622. package/out/utils/wof-json.js +7 -12
  623. package/out/utils/wof-json.js.map +1 -1
  624. package/out/ve/recipes/reviewed-postcode-tail.d.ts.map +1 -0
  625. package/out/ve/recipes/reviewed-postcode-tail.js.map +1 -0
  626. package/package.json +196 -46
  627. package/lib/adapters/ban/adapter.ts +0 -190
  628. package/lib/adapters/fcc-bdc/adapter.ts +0 -174
  629. package/lib/adapters/geonames-postal/adapter.ts +0 -122
  630. package/lib/adapters/gnaf/adapter.ts +0 -149
  631. package/lib/adapters/gnaf/assemble.ts +0 -225
  632. package/lib/adapters/state-hi-schools/adapter.ts +0 -170
  633. package/lib/adapters/state-ia-contractors/adapter.ts +0 -144
  634. package/lib/adapters/state-ny-notaries/adapter.ts +0 -141
  635. package/lib/adapters/state-tx-notaries/adapter.ts +0 -157
  636. package/lib/adapters/tiger/adapter.ts +0 -252
  637. package/lib/adapters/usgov-hrsa-fqhc/adapter.ts +0 -167
  638. package/lib/adapters/usgov-imls-pls/adapter.ts +0 -143
  639. package/lib/adapters/usgov-irs-bmf/adapter.ts +0 -151
  640. package/lib/adapters/usgov-nad/adapter.ts +0 -334
  641. package/lib/adapters/usgov-nppes/adapter.ts +0 -151
  642. package/lib/adapters/usgov-samhsa-treatment-locator/adapter.ts +0 -193
  643. package/lib/adapters/wof-admin-json/adapter.ts +0 -231
  644. package/lib/adapters/wof-json-rows.ts +0 -129
  645. package/lib/adapters/wof-postalcode-json/adapter.ts +0 -176
  646. package/lib/name-prone-us-suffixes.ts +0 -12
  647. package/lib/recipes/bare-country.ts +0 -95
  648. package/lib/recipes/cz-pcfirst-preposition.ts +0 -111
  649. package/lib/recipes/fr-admin-split.ts +0 -242
  650. package/lib/recipes/fr-bare-street.ts +0 -170
  651. package/lib/recipes/fr-fragment.ts +0 -333
  652. package/lib/recipes/fr-lieudit.ts +0 -301
  653. package/lib/recipes/fr-order.ts +0 -245
  654. package/lib/recipes/german.ts +0 -359
  655. package/lib/recipes/locale.ts +0 -599
  656. package/lib/recipes/nl-postcode.ts +0 -115
  657. package/lib/recipes/no-fragment.ts +0 -219
  658. package/lib/recipes/no-street-led.ts +0 -141
  659. package/lib/recipes/no-street.ts +0 -90
  660. package/lib/recipes/po-box-cedex.ts +0 -976
  661. package/lib/recipes/po-box.ts +0 -123
  662. package/lib/recipes/si-bare-village.ts +0 -97
  663. package/lib/recipes/street-affix.ts +0 -804
  664. package/lib/recipes/street-bare.ts +0 -100
  665. package/lib/recipes/sub-venue-sources.ts +0 -877
  666. package/lib/recipes/sub-venue.ts +0 -997
  667. package/lib/synthesizers/german.ts +0 -260
  668. package/lib/tools/fetch/ban.ts +0 -289
  669. package/lib/tools/fetch/download.ts +0 -262
  670. package/lib/tools/fetch/geonames-dump.ts +0 -257
  671. package/lib/tools/fetch/geonames-postal.ts +0 -180
  672. package/lib/tools/fetch/hrsa.ts +0 -59
  673. package/lib/tools/fetch/imls-pls.ts +0 -149
  674. package/lib/tools/fetch/nad.ts +0 -341
  675. package/lib/tools/fetch/nppes.ts +0 -165
  676. package/lib/tools/fetch/state-hi-schools.ts +0 -161
  677. package/lib/tools/fetch/state-sources.ts +0 -146
  678. package/lib/tools/fetch/tiger-full.ts +0 -308
  679. package/lib/tools/golden-promote.ts +0 -337
  680. package/lib/tools/golden-relabel-street.ts +0 -683
  681. package/lib/tools/lint-slice-vocab.ts +0 -388
  682. package/lib/tools/sub-venue/harvest.ts +0 -234
  683. package/lib/tools/sub-venue/head-nouns.ts +0 -245
  684. package/lib/tools/sub-venue/surfaces.ts +0 -95
  685. package/lib/tools/sub-venue/table.ts +0 -281
  686. package/lib/tools/sub-venue/wikidata.ts +0 -85
  687. package/lib/tools/sub-venue-lexicon.ts +0 -429
  688. package/out/adapters/ban/adapter.d.ts.map +0 -1
  689. package/out/adapters/ban/adapter.js +0 -149
  690. package/out/adapters/ban/adapter.js.map +0 -1
  691. package/out/adapters/ban/street-decompose.d.ts.map +0 -1
  692. package/out/adapters/ban/street-decompose.js.map +0 -1
  693. package/out/adapters/fcc-bdc/adapter.d.ts.map +0 -1
  694. package/out/adapters/fcc-bdc/adapter.js +0 -144
  695. package/out/adapters/fcc-bdc/adapter.js.map +0 -1
  696. package/out/adapters/geonames-postal/adapter.d.ts.map +0 -1
  697. package/out/adapters/geonames-postal/adapter.js +0 -107
  698. package/out/adapters/geonames-postal/adapter.js.map +0 -1
  699. package/out/adapters/gnaf/adapter.d.ts.map +0 -1
  700. package/out/adapters/gnaf/adapter.js +0 -124
  701. package/out/adapters/gnaf/adapter.js.map +0 -1
  702. package/out/adapters/gnaf/assemble.d.ts +0 -65
  703. package/out/adapters/gnaf/assemble.d.ts.map +0 -1
  704. package/out/adapters/gnaf/assemble.js +0 -154
  705. package/out/adapters/gnaf/assemble.js.map +0 -1
  706. package/out/adapters/state-hi-schools/adapter.d.ts.map +0 -1
  707. package/out/adapters/state-hi-schools/adapter.js +0 -145
  708. package/out/adapters/state-hi-schools/adapter.js.map +0 -1
  709. package/out/adapters/state-hi-schools/workbook.d.ts.map +0 -1
  710. package/out/adapters/state-hi-schools/workbook.js.map +0 -1
  711. package/out/adapters/state-ia-contractors/adapter.d.ts.map +0 -1
  712. package/out/adapters/state-ia-contractors/adapter.js +0 -111
  713. package/out/adapters/state-ia-contractors/adapter.js.map +0 -1
  714. package/out/adapters/state-ny-notaries/adapter.d.ts.map +0 -1
  715. package/out/adapters/state-ny-notaries/adapter.js +0 -119
  716. package/out/adapters/state-ny-notaries/adapter.js.map +0 -1
  717. package/out/adapters/state-tx-notaries/adapter.d.ts.map +0 -1
  718. package/out/adapters/state-tx-notaries/adapter.js +0 -125
  719. package/out/adapters/state-tx-notaries/adapter.js.map +0 -1
  720. package/out/adapters/tiger/adapter.d.ts +0 -55
  721. package/out/adapters/tiger/adapter.d.ts.map +0 -1
  722. package/out/adapters/tiger/adapter.js +0 -201
  723. package/out/adapters/tiger/adapter.js.map +0 -1
  724. package/out/adapters/tiger/street-decompose.d.ts.map +0 -1
  725. package/out/adapters/tiger/street-decompose.js.map +0 -1
  726. package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts.map +0 -1
  727. package/out/adapters/usgov-hrsa-fqhc/adapter.js +0 -132
  728. package/out/adapters/usgov-hrsa-fqhc/adapter.js.map +0 -1
  729. package/out/adapters/usgov-imls-pls/adapter.d.ts.map +0 -1
  730. package/out/adapters/usgov-imls-pls/adapter.js +0 -114
  731. package/out/adapters/usgov-imls-pls/adapter.js.map +0 -1
  732. package/out/adapters/usgov-irs-bmf/adapter.d.ts.map +0 -1
  733. package/out/adapters/usgov-irs-bmf/adapter.js +0 -120
  734. package/out/adapters/usgov-irs-bmf/adapter.js.map +0 -1
  735. package/out/adapters/usgov-nad/adapter.d.ts.map +0 -1
  736. package/out/adapters/usgov-nad/adapter.js +0 -249
  737. package/out/adapters/usgov-nad/adapter.js.map +0 -1
  738. package/out/adapters/usgov-nppes/adapter.d.ts.map +0 -1
  739. package/out/adapters/usgov-nppes/adapter.js +0 -118
  740. package/out/adapters/usgov-nppes/adapter.js.map +0 -1
  741. package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts.map +0 -1
  742. package/out/adapters/usgov-samhsa-treatment-locator/adapter.js +0 -154
  743. package/out/adapters/usgov-samhsa-treatment-locator/adapter.js.map +0 -1
  744. package/out/adapters/wof-admin-jp/adapter.d.ts.map +0 -1
  745. package/out/adapters/wof-admin-jp/adapter.js.map +0 -1
  746. package/out/adapters/wof-admin-json/adapter.d.ts +0 -79
  747. package/out/adapters/wof-admin-json/adapter.d.ts.map +0 -1
  748. package/out/adapters/wof-admin-json/adapter.js +0 -197
  749. package/out/adapters/wof-admin-json/adapter.js.map +0 -1
  750. package/out/adapters/wof-json-rows.d.ts +0 -62
  751. package/out/adapters/wof-json-rows.d.ts.map +0 -1
  752. package/out/adapters/wof-json-rows.js +0 -83
  753. package/out/adapters/wof-json-rows.js.map +0 -1
  754. package/out/adapters/wof-postalcode-json/adapter.d.ts +0 -65
  755. package/out/adapters/wof-postalcode-json/adapter.d.ts.map +0 -1
  756. package/out/adapters/wof-postalcode-json/adapter.js +0 -151
  757. package/out/adapters/wof-postalcode-json/adapter.js.map +0 -1
  758. package/out/codex/us-fips-state.d.ts +0 -41
  759. package/out/codex/us-fips-state.d.ts.map +0 -1
  760. package/out/codex/us-fips-state.js +0 -100
  761. package/out/codex/us-fips-state.js.map +0 -1
  762. package/out/name-prone-us-suffixes.d.ts +0 -12
  763. package/out/name-prone-us-suffixes.d.ts.map +0 -1
  764. package/out/name-prone-us-suffixes.js +0 -12
  765. package/out/name-prone-us-suffixes.js.map +0 -1
  766. package/out/recipes/bare-country.d.ts +0 -25
  767. package/out/recipes/bare-country.d.ts.map +0 -1
  768. package/out/recipes/bare-country.js +0 -87
  769. package/out/recipes/bare-country.js.map +0 -1
  770. package/out/recipes/cn-organizational-units.d.ts +0 -36
  771. package/out/recipes/cn-organizational-units.d.ts.map +0 -1
  772. package/out/recipes/cn-organizational-units.js +0 -159
  773. package/out/recipes/cn-organizational-units.js.map +0 -1
  774. package/out/recipes/cz-pcfirst-preposition.d.ts +0 -22
  775. package/out/recipes/cz-pcfirst-preposition.d.ts.map +0 -1
  776. package/out/recipes/cz-pcfirst-preposition.js +0 -100
  777. package/out/recipes/cz-pcfirst-preposition.js.map +0 -1
  778. package/out/recipes/fr-admin-split.d.ts +0 -38
  779. package/out/recipes/fr-admin-split.d.ts.map +0 -1
  780. package/out/recipes/fr-admin-split.js +0 -192
  781. package/out/recipes/fr-admin-split.js.map +0 -1
  782. package/out/recipes/fr-bare-street.d.ts +0 -28
  783. package/out/recipes/fr-bare-street.d.ts.map +0 -1
  784. package/out/recipes/fr-bare-street.js +0 -143
  785. package/out/recipes/fr-bare-street.js.map +0 -1
  786. package/out/recipes/fr-fragment.d.ts +0 -77
  787. package/out/recipes/fr-fragment.d.ts.map +0 -1
  788. package/out/recipes/fr-fragment.js +0 -276
  789. package/out/recipes/fr-fragment.js.map +0 -1
  790. package/out/recipes/fr-lieudit.d.ts +0 -32
  791. package/out/recipes/fr-lieudit.d.ts.map +0 -1
  792. package/out/recipes/fr-lieudit.js +0 -227
  793. package/out/recipes/fr-lieudit.js.map +0 -1
  794. package/out/recipes/fr-order.d.ts +0 -38
  795. package/out/recipes/fr-order.d.ts.map +0 -1
  796. package/out/recipes/fr-order.js +0 -197
  797. package/out/recipes/fr-order.js.map +0 -1
  798. package/out/recipes/german.d.ts +0 -44
  799. package/out/recipes/german.d.ts.map +0 -1
  800. package/out/recipes/german.js +0 -288
  801. package/out/recipes/german.js.map +0 -1
  802. package/out/recipes/locale.d.ts +0 -136
  803. package/out/recipes/locale.d.ts.map +0 -1
  804. package/out/recipes/locale.js +0 -460
  805. package/out/recipes/locale.js.map +0 -1
  806. package/out/recipes/nl-postcode.d.ts +0 -26
  807. package/out/recipes/nl-postcode.d.ts.map +0 -1
  808. package/out/recipes/nl-postcode.js +0 -102
  809. package/out/recipes/nl-postcode.js.map +0 -1
  810. package/out/recipes/no-fragment.d.ts +0 -43
  811. package/out/recipes/no-fragment.d.ts.map +0 -1
  812. package/out/recipes/no-fragment.js +0 -182
  813. package/out/recipes/no-fragment.js.map +0 -1
  814. package/out/recipes/no-street-led.d.ts +0 -24
  815. package/out/recipes/no-street-led.d.ts.map +0 -1
  816. package/out/recipes/no-street-led.js +0 -113
  817. package/out/recipes/no-street-led.js.map +0 -1
  818. package/out/recipes/no-street.d.ts +0 -18
  819. package/out/recipes/no-street.d.ts.map +0 -1
  820. package/out/recipes/no-street.js +0 -71
  821. package/out/recipes/no-street.js.map +0 -1
  822. package/out/recipes/po-box-cedex.d.ts +0 -38
  823. package/out/recipes/po-box-cedex.d.ts.map +0 -1
  824. package/out/recipes/po-box-cedex.js +0 -767
  825. package/out/recipes/po-box-cedex.js.map +0 -1
  826. package/out/recipes/po-box.d.ts +0 -17
  827. package/out/recipes/po-box.d.ts.map +0 -1
  828. package/out/recipes/po-box.js +0 -100
  829. package/out/recipes/po-box.js.map +0 -1
  830. package/out/recipes/reviewed-postcode-tail.d.ts.map +0 -1
  831. package/out/recipes/reviewed-postcode-tail.js.map +0 -1
  832. package/out/recipes/si-bare-village.d.ts +0 -32
  833. package/out/recipes/si-bare-village.d.ts.map +0 -1
  834. package/out/recipes/si-bare-village.js +0 -88
  835. package/out/recipes/si-bare-village.js.map +0 -1
  836. package/out/recipes/street-affix.d.ts +0 -89
  837. package/out/recipes/street-affix.d.ts.map +0 -1
  838. package/out/recipes/street-affix.js +0 -563
  839. package/out/recipes/street-affix.js.map +0 -1
  840. package/out/recipes/street-bare.d.ts +0 -25
  841. package/out/recipes/street-bare.d.ts.map +0 -1
  842. package/out/recipes/street-bare.js +0 -86
  843. package/out/recipes/street-bare.js.map +0 -1
  844. package/out/recipes/sub-venue-sources.d.ts +0 -229
  845. package/out/recipes/sub-venue-sources.d.ts.map +0 -1
  846. package/out/recipes/sub-venue-sources.js +0 -653
  847. package/out/recipes/sub-venue-sources.js.map +0 -1
  848. package/out/recipes/sub-venue.d.ts +0 -269
  849. package/out/recipes/sub-venue.d.ts.map +0 -1
  850. package/out/recipes/sub-venue.js +0 -725
  851. package/out/recipes/sub-venue.js.map +0 -1
  852. package/out/synthesizers/german.d.ts +0 -115
  853. package/out/synthesizers/german.d.ts.map +0 -1
  854. package/out/synthesizers/german.js +0 -149
  855. package/out/synthesizers/german.js.map +0 -1
  856. package/out/tools/fetch/ban.d.ts +0 -31
  857. package/out/tools/fetch/ban.d.ts.map +0 -1
  858. package/out/tools/fetch/ban.js +0 -239
  859. package/out/tools/fetch/ban.js.map +0 -1
  860. package/out/tools/fetch/download.d.ts +0 -117
  861. package/out/tools/fetch/download.d.ts.map +0 -1
  862. package/out/tools/fetch/download.js +0 -160
  863. package/out/tools/fetch/download.js.map +0 -1
  864. package/out/tools/fetch/geonames-dump.d.ts +0 -91
  865. package/out/tools/fetch/geonames-dump.d.ts.map +0 -1
  866. package/out/tools/fetch/geonames-dump.js +0 -171
  867. package/out/tools/fetch/geonames-dump.js.map +0 -1
  868. package/out/tools/fetch/geonames-postal.d.ts +0 -76
  869. package/out/tools/fetch/geonames-postal.d.ts.map +0 -1
  870. package/out/tools/fetch/geonames-postal.js +0 -126
  871. package/out/tools/fetch/geonames-postal.js.map +0 -1
  872. package/out/tools/fetch/hrsa.d.ts +0 -18
  873. package/out/tools/fetch/hrsa.d.ts.map +0 -1
  874. package/out/tools/fetch/hrsa.js +0 -46
  875. package/out/tools/fetch/hrsa.js.map +0 -1
  876. package/out/tools/fetch/imls-pls.d.ts +0 -25
  877. package/out/tools/fetch/imls-pls.d.ts.map +0 -1
  878. package/out/tools/fetch/imls-pls.js +0 -108
  879. package/out/tools/fetch/imls-pls.js.map +0 -1
  880. package/out/tools/fetch/nad.d.ts +0 -70
  881. package/out/tools/fetch/nad.d.ts.map +0 -1
  882. package/out/tools/fetch/nad.js +0 -234
  883. package/out/tools/fetch/nad.js.map +0 -1
  884. package/out/tools/fetch/nppes.d.ts +0 -26
  885. package/out/tools/fetch/nppes.d.ts.map +0 -1
  886. package/out/tools/fetch/nppes.js +0 -126
  887. package/out/tools/fetch/nppes.js.map +0 -1
  888. package/out/tools/fetch/state-hi-schools.d.ts +0 -26
  889. package/out/tools/fetch/state-hi-schools.d.ts.map +0 -1
  890. package/out/tools/fetch/state-hi-schools.js +0 -112
  891. package/out/tools/fetch/state-hi-schools.js.map +0 -1
  892. package/out/tools/fetch/state-sources.d.ts +0 -26
  893. package/out/tools/fetch/state-sources.d.ts.map +0 -1
  894. package/out/tools/fetch/state-sources.js +0 -111
  895. package/out/tools/fetch/state-sources.js.map +0 -1
  896. package/out/tools/fetch/tiger-full.d.ts +0 -22
  897. package/out/tools/fetch/tiger-full.d.ts.map +0 -1
  898. package/out/tools/fetch/tiger-full.js +0 -217
  899. package/out/tools/fetch/tiger-full.js.map +0 -1
  900. package/out/tools/golden-expand.d.ts +0 -90
  901. package/out/tools/golden-expand.d.ts.map +0 -1
  902. package/out/tools/golden-expand.js +0 -406
  903. package/out/tools/golden-expand.js.map +0 -1
  904. package/out/tools/golden-promote.d.ts +0 -72
  905. package/out/tools/golden-promote.d.ts.map +0 -1
  906. package/out/tools/golden-promote.js +0 -240
  907. package/out/tools/golden-promote.js.map +0 -1
  908. package/out/tools/golden-relabel-street.d.ts +0 -196
  909. package/out/tools/golden-relabel-street.d.ts.map +0 -1
  910. package/out/tools/golden-relabel-street.js +0 -440
  911. package/out/tools/golden-relabel-street.js.map +0 -1
  912. package/out/tools/lint-slice-vocab.d.ts +0 -90
  913. package/out/tools/lint-slice-vocab.d.ts.map +0 -1
  914. package/out/tools/lint-slice-vocab.js +0 -267
  915. package/out/tools/lint-slice-vocab.js.map +0 -1
  916. package/out/tools/lint-slice.d.ts +0 -93
  917. package/out/tools/lint-slice.d.ts.map +0 -1
  918. package/out/tools/lint-slice.js +0 -322
  919. package/out/tools/lint-slice.js.map +0 -1
  920. package/out/tools/slice-kryptonite.d.ts +0 -32
  921. package/out/tools/slice-kryptonite.d.ts.map +0 -1
  922. package/out/tools/slice-kryptonite.js +0 -99
  923. package/out/tools/slice-kryptonite.js.map +0 -1
  924. package/out/tools/slice-translit.d.ts +0 -51
  925. package/out/tools/slice-translit.d.ts.map +0 -1
  926. package/out/tools/slice-translit.js +0 -166
  927. package/out/tools/slice-translit.js.map +0 -1
  928. package/out/tools/sub-venue/harvest.d.ts +0 -100
  929. package/out/tools/sub-venue/harvest.d.ts.map +0 -1
  930. package/out/tools/sub-venue/harvest.js +0 -168
  931. package/out/tools/sub-venue/harvest.js.map +0 -1
  932. package/out/tools/sub-venue/head-nouns.d.ts +0 -50
  933. package/out/tools/sub-venue/head-nouns.d.ts.map +0 -1
  934. package/out/tools/sub-venue/head-nouns.js.map +0 -1
  935. package/out/tools/sub-venue/surfaces.d.ts +0 -45
  936. package/out/tools/sub-venue/surfaces.d.ts.map +0 -1
  937. package/out/tools/sub-venue/surfaces.js.map +0 -1
  938. package/out/tools/sub-venue/table.d.ts +0 -229
  939. package/out/tools/sub-venue/table.d.ts.map +0 -1
  940. package/out/tools/sub-venue/table.js.map +0 -1
  941. package/out/tools/sub-venue/wikidata.d.ts +0 -26
  942. package/out/tools/sub-venue/wikidata.d.ts.map +0 -1
  943. package/out/tools/sub-venue/wikidata.js +0 -62
  944. package/out/tools/sub-venue/wikidata.js.map +0 -1
  945. package/out/tools/sub-venue-lexicon.d.ts +0 -156
  946. package/out/tools/sub-venue-lexicon.d.ts.map +0 -1
  947. package/out/tools/sub-venue-lexicon.js +0 -283
  948. package/out/tools/sub-venue-lexicon.js.map +0 -1
  949. package/out/tools/sub-venue-promotions.d.ts +0 -94
  950. package/out/tools/sub-venue-promotions.d.ts.map +0 -1
  951. package/out/tools/sub-venue-promotions.js +0 -266
  952. package/out/tools/sub-venue-promotions.js.map +0 -1
  953. /package/lib/{recipes/cn-organizational-units.ts → cn/recipes/organizational-units.ts} +0 -0
  954. /package/lib/{adapters → fr/adapters}/ban/street-decompose.ts +0 -0
  955. /package/lib/{adapters/wof-admin-jp → jp/adapters/wof-admin}/adapter.ts +0 -0
  956. /package/lib/tools/{golden-expand.ts → golden/expand.ts} +0 -0
  957. /package/lib/tools/{lint-slice.ts → lint/slice/index.ts} +0 -0
  958. /package/lib/tools/{slice-kryptonite.ts → slice/kryptonite.ts} +0 -0
  959. /package/lib/tools/{slice-translit.ts → slice/translit.ts} +0 -0
  960. /package/lib/tools/{sub-venue-promotions.ts → sub/venue/promotions.ts} +0 -0
  961. /package/lib/{adapters/state-hi-schools → us/adapters/state/hi-schools}/workbook.ts +0 -0
  962. /package/lib/{adapters → us/adapters}/tiger/street-decompose.ts +0 -0
  963. /package/lib/{codex/us-fips-state.ts → us/fips-state.ts} +0 -0
  964. /package/lib/{recipes → ve/recipes}/reviewed-postcode-tail.ts +0 -0
  965. /package/out/adapters/{geonames-postal → geonames/postal}/adapter.d.ts +0 -0
  966. /package/out/{adapters → au/adapters}/gnaf/adapter.d.ts +0 -0
  967. /package/out/{adapters → fr/adapters}/ban/adapter.d.ts +0 -0
  968. /package/out/{adapters → fr/adapters}/ban/street-decompose.d.ts +0 -0
  969. /package/out/{adapters → fr/adapters}/ban/street-decompose.js +0 -0
  970. /package/out/{adapters/wof-admin-jp → jp/adapters/wof-admin}/adapter.d.ts +0 -0
  971. /package/out/{adapters/wof-admin-jp → jp/adapters/wof-admin}/adapter.js +0 -0
  972. /package/out/tools/{sub-venue → sub/venue}/head-nouns.js +0 -0
  973. /package/out/tools/{sub-venue → sub/venue}/surfaces.js +0 -0
  974. /package/out/tools/{sub-venue → sub/venue}/table.js +0 -0
  975. /package/out/{adapters → us/adapters}/fcc-bdc/adapter.d.ts +0 -0
  976. /package/out/{adapters/state-hi-schools → us/adapters/state/hi-schools}/adapter.d.ts +0 -0
  977. /package/out/{adapters/state-hi-schools → us/adapters/state/hi-schools}/workbook.d.ts +0 -0
  978. /package/out/{adapters/state-hi-schools → us/adapters/state/hi-schools}/workbook.js +0 -0
  979. /package/out/{adapters/state-ia-contractors → us/adapters/state/ia-contractors}/adapter.d.ts +0 -0
  980. /package/out/{adapters/state-ny-notaries → us/adapters/state/ny-notaries}/adapter.d.ts +0 -0
  981. /package/out/{adapters/state-tx-notaries → us/adapters/state/tx-notaries}/adapter.d.ts +0 -0
  982. /package/out/{adapters → us/adapters}/tiger/street-decompose.d.ts +0 -0
  983. /package/out/{adapters → us/adapters}/tiger/street-decompose.js +0 -0
  984. /package/out/{adapters/usgov-hrsa-fqhc → us/adapters/usgov/hrsa-fqhc}/adapter.d.ts +0 -0
  985. /package/out/{adapters/usgov-imls-pls → us/adapters/usgov/imls-pls}/adapter.d.ts +0 -0
  986. /package/out/{adapters/usgov-irs-bmf → us/adapters/usgov/irs-bmf}/adapter.d.ts +0 -0
  987. /package/out/{adapters/usgov-nad → us/adapters/usgov/nad}/adapter.d.ts +0 -0
  988. /package/out/{adapters/usgov-nppes → us/adapters/usgov/nppes}/adapter.d.ts +0 -0
  989. /package/out/{adapters/usgov-samhsa-treatment-locator → us/adapters/usgov/samhsa-treatment-locator}/adapter.d.ts +0 -0
  990. /package/out/{recipes → ve/recipes}/reviewed-postcode-tail.d.ts +0 -0
  991. /package/out/{recipes → ve/recipes}/reviewed-postcode-tail.js +0 -0
@@ -1,322 +0,0 @@
1
- /**
2
- * @copyright Sister Software
3
- * @license AGPL-3.0
4
- * @author Teffen Ellis, et al.
5
- *
6
- * Corpus linter. Compares a new slice against pre-computed corpus statistics and flags patterns
7
- * that would cause the class of failure we hit with v0.6.2's "5th Avenue Theatre" adversarial
8
- * venue templates.
9
- *
10
- * Per DeepSeek turn 9 design (2026-05-29). v1 checks:
11
- *
12
- * 1. **Token-label distribution outliers.** For each token in the new slice, compare the slice's
13
- * majority label to the corpus's majority label. Flag when the corpus has a
14
- * confidently-established majority (>66%) AND the slice's majority differs AND both have
15
- * non-trivial counts (slice ≥ 50, corpus ≥ 200).
16
- * 2. **Label-vacuum tokens.** Token labeled with a tag that has ZERO instances in the corpus for that
17
- * token, despite the token being well-represented in the corpus. Stronger signal than #1 —
18
- * we're introducing a novel association, not shifting a distribution.
19
- * 3. **Bigram-label collisions.** Identical (token_bigram, label_bigram) appears in slice while the
20
- * same token_bigram has a DIFFERENT majority label_bigram in the corpus. The "5th Avenue"
21
- * with [B-venue, I-venue] vs corpus's [B-house_number, I-street] case.
22
- * 4. **Common-form anti-pattern rules.** Applies `lint-rules.json` — token-regex → forbidden-labels
23
- * mappings — flagging matches.
24
- * 5. **Basic sanity.** Truncated rows (tokens.length !== labels.length), all-O rows >90% of slice.
25
- *
26
- * Output: markdown report on stdout, optional JSON sidecar via `outJSON`. The command exits 0 if
27
- * no errors, 1 if any errors (warnings don't check). Per the design, the MANIFEST entry for a
28
- * flagged slice should require `lint_acknowledged: true` before training consumes it.
29
- *
30
- * Usage: mailwoman dev lint corpus-slice\
31
- * --slice <new-slice.parquet>\
32
- * --stats <corpus-stats.json>\
33
- * [--rules <rules.json>]\
34
- * [--out-md /tmp/lint-report.md]\
35
- * [--out-json /tmp/lint-report.json]
36
- */
37
- import { readLocalJSONFile } from "@mailwoman/core/fs/readers";
38
- import { writeLocalFile, writeLocalJSONFile } from "@mailwoman/core/fs/writers";
39
- import { resolvePackagePath } from "@mailwoman/core/module/resolvers";
40
- import { accumulateCooccurrences, createCooccurrenceStats, SLICE_STATS_SEP as SEP, streamTokenLabelRows, } from "#utils/slice-stats";
41
- /**
42
- * Occurrences of a forbidden label before it is reported — one or two are noise, five is a pattern.
43
- */
44
- const FORBIDDEN_LABEL_REPORT_THRESHOLD = 5;
45
- /**
46
- * Examples printed per finding before the list is truncated.
47
- */
48
- const MAX_LISTED_EXAMPLES = 20;
49
- /**
50
- * Calibrated thresholds (DeepSeek turn 9). These can be tuned over time if new failure modes surface that the current
51
- * numbers miss.
52
- */
53
- const CORPUS_CONFIDENCE_FLOOR = 0.66;
54
- const SLICE_MIN_COUNT = 50;
55
- const CORPUS_MIN_COUNT = 200;
56
- const VACUUM_SLICE_MIN_COUNT = 20;
57
- const VACUUM_CORPUS_MIN_COUNT = 100;
58
- const BIGRAM_MIN_COUNT = 10;
59
- const ALL_O_RATIO_CEILING = 0.9;
60
- /**
61
- * Default `lint-rules.json` path — the rules ship beside this module in the source tree. tsc does not emit
62
- * readFileSync'd JSON into `out/`, so the compiled tree falls back to the source-tree copy (corpus/out/src/tools/ →
63
- * corpus/src/tools/). In-repo the `node` exports condition loads this module from source anyway, so the sibling URL is
64
- * the common path.
65
- */
66
- function defaultRulesPath() {
67
- return resolvePackagePath("@mailwoman/corpus", "lib", "tools", "lint-rules.json");
68
- }
69
- async function statsFromSlice(rows) {
70
- const co = createCooccurrenceStats();
71
- const out = {
72
- rowCount: 0,
73
- tokens: co.tokens,
74
- bigrams: co.bigrams,
75
- truncatedRows: 0,
76
- allORows: 0,
77
- };
78
- for await (const row of rows) {
79
- out.rowCount++;
80
- if (row.tokens.length !== row.labels.length) {
81
- out.truncatedRows++;
82
- continue;
83
- }
84
- if (row.labels.every((l) => l === "O")) {
85
- out.allORows++;
86
- }
87
- accumulateCooccurrences(co, row.tokens, row.labels);
88
- }
89
- return out;
90
- }
91
- function majorityLabel(distribution) {
92
- const entries = distribution instanceof Map ? [...distribution.entries()] : Object.entries(distribution);
93
- let bestLabel = "";
94
- let bestCount = 0;
95
- let total = 0;
96
- for (const [label, count] of entries) {
97
- total += count;
98
- if (count > bestCount) {
99
- bestCount = count;
100
- bestLabel = label;
101
- }
102
- }
103
- return { label: bestLabel, count: bestCount, total, confidence: total === 0 ? 0 : bestCount / total };
104
- }
105
- function checkDistributionOutliers(slice, corpus) {
106
- const flags = [];
107
- for (const [token, sliceLabelMap] of slice.tokens) {
108
- const corpusLabelMap = corpus.tokens[token];
109
- if (!corpusLabelMap)
110
- continue;
111
- const sliceMaj = majorityLabel(sliceLabelMap);
112
- const corpusMaj = majorityLabel(corpusLabelMap);
113
- if (corpusMaj.confidence >= CORPUS_CONFIDENCE_FLOOR &&
114
- sliceMaj.label !== corpusMaj.label &&
115
- sliceMaj.count >= SLICE_MIN_COUNT &&
116
- corpusMaj.total >= CORPUS_MIN_COUNT) {
117
- flags.push({
118
- check: "distribution-outlier",
119
- severity: "error",
120
- token,
121
- sliceLabel: sliceMaj.label,
122
- corpusLabel: corpusMaj.label,
123
- sliceCount: sliceMaj.count,
124
- corpusCount: corpusMaj.count,
125
- detail: `Token "${token}": slice majority is ${sliceMaj.label} (${sliceMaj.count}/${sliceMaj.total}, ${(sliceMaj.confidence * 100).toFixed(0)}%), corpus majority is ${corpusMaj.label} (${corpusMaj.count}/${corpusMaj.total}, ${(corpusMaj.confidence * 100).toFixed(0)}%).`,
126
- });
127
- }
128
- }
129
- return flags;
130
- }
131
- function checkLabelVacuum(slice, corpus) {
132
- const flags = [];
133
- for (const [token, sliceLabelMap] of slice.tokens) {
134
- const corpusLabelMap = corpus.tokens[token];
135
- if (!corpusLabelMap)
136
- continue;
137
- const corpusTotal = Object.values(corpusLabelMap).reduce((a, b) => a + b, 0);
138
- if (corpusTotal < VACUUM_CORPUS_MIN_COUNT)
139
- continue;
140
- for (const [label, sliceCount] of sliceLabelMap) {
141
- if (sliceCount < VACUUM_SLICE_MIN_COUNT)
142
- continue;
143
- if (corpusLabelMap[label] === undefined || corpusLabelMap[label] === 0) {
144
- flags.push({
145
- check: "label-vacuum",
146
- severity: "error",
147
- token,
148
- sliceLabel: label,
149
- sliceCount,
150
- corpusCount: corpusTotal,
151
- detail: `Token "${token}": slice labels it ${label} ${sliceCount} times, but the corpus (${corpusTotal} instances of this token) has ZERO instances of this label.`,
152
- });
153
- }
154
- }
155
- }
156
- return flags;
157
- }
158
- function checkBigramCollisions(slice, corpus) {
159
- const flags = [];
160
- for (const [bigram, sliceLabelMap] of slice.bigrams) {
161
- const corpusLabelMap = corpus.bigrams[bigram];
162
- if (!corpusLabelMap)
163
- continue;
164
- const sliceMaj = majorityLabel(sliceLabelMap);
165
- const corpusMaj = majorityLabel(corpusLabelMap);
166
- if (sliceMaj.label !== corpusMaj.label &&
167
- sliceMaj.count >= BIGRAM_MIN_COUNT &&
168
- corpusMaj.count >= BIGRAM_MIN_COUNT) {
169
- const renderBigram = bigram.split(SEP).join(" ");
170
- const renderSliceLabel = sliceMaj.label.split(SEP).join(" → ");
171
- const renderCorpusLabel = corpusMaj.label.split(SEP).join(" → ");
172
- flags.push({
173
- check: "bigram-collision",
174
- severity: "error",
175
- bigram: renderBigram,
176
- sliceLabel: renderSliceLabel,
177
- corpusLabel: renderCorpusLabel,
178
- sliceCount: sliceMaj.count,
179
- corpusCount: corpusMaj.count,
180
- detail: `Bigram "${renderBigram}": slice label-bigram is [${renderSliceLabel}] (${sliceMaj.count}×), corpus label-bigram is [${renderCorpusLabel}] (${corpusMaj.count}×). Same surface text, different structural reading.`,
181
- });
182
- }
183
- }
184
- return flags;
185
- }
186
- function checkRules(slice, rulesFile) {
187
- const flags = [];
188
- const compiled = rulesFile.rules.map((r) => ({
189
- rule: r,
190
- regex: new RegExp(r.pattern, r.pattern_case_sensitive ? "" : "i"),
191
- }));
192
- for (const [token, labelMap] of slice.tokens) {
193
- for (const { rule, regex } of compiled) {
194
- if (!regex.test(token))
195
- continue;
196
- for (const [label, count] of labelMap) {
197
- if (rule.forbidden_labels.includes(label) && count >= FORBIDDEN_LABEL_REPORT_THRESHOLD) {
198
- flags.push({
199
- check: "anti-pattern-rule",
200
- severity: rule.severity,
201
- ruleID: rule.id,
202
- token,
203
- sliceLabel: label,
204
- sliceCount: count,
205
- detail: `Token "${token}" matched rule ${rule.id} and is labeled ${label} ${count} time(s). Rule message: ${rule.message}`,
206
- });
207
- }
208
- }
209
- }
210
- }
211
- return flags;
212
- }
213
- function checkSanity(slice) {
214
- const flags = [];
215
- if (slice.truncatedRows > 0) {
216
- flags.push({
217
- check: "truncated-rows",
218
- severity: "error",
219
- detail: `${slice.truncatedRows} row(s) have tokens.length !== labels.length. Pipeline alignment bug.`,
220
- });
221
- }
222
- const allORatio = slice.allORows / Math.max(1, slice.rowCount);
223
- if (allORatio >= ALL_O_RATIO_CEILING) {
224
- flags.push({
225
- check: "all-O-slice",
226
- severity: "warn",
227
- detail: `${slice.allORows}/${slice.rowCount} rows (${(allORatio * 100).toFixed(0)}%) are entirely O-labeled. Slice contributes no signal.`,
228
- });
229
- }
230
- return flags;
231
- }
232
- function renderReport(opts, slice, flags) {
233
- const errors = flags.filter((f) => f.severity === "error");
234
- const warns = flags.filter((f) => f.severity === "warn");
235
- const verdict = !errors.length ? "**PASS** ✓" : "**FLAGGED** ⚠";
236
- const lines = [
237
- `# Corpus Lint: ${verdict}`,
238
- "",
239
- `- **Slice:** \`${opts.slicePath}\``,
240
- `- **Corpus stats:** \`${opts.statsPath}\``,
241
- `- **Rules:** \`${opts.rulesPath}\``,
242
- `- **Slice rows:** ${slice.rowCount}`,
243
- `- **Unique tokens:** ${slice.tokens.size}`,
244
- `- **Unique bigrams:** ${slice.bigrams.size}`,
245
- "",
246
- `**Errors:** ${errors.length} (checks the slice's inclusion unless MANIFEST sets \`lint_acknowledged: true\`)`,
247
- `**Warnings:** ${warns.length} (advisory)`,
248
- "",
249
- ];
250
- if (!flags.length) {
251
- lines.push("No anomalies detected.");
252
- return lines.join("\n");
253
- }
254
- const byCheck = new Map();
255
- for (const f of flags) {
256
- const arr = byCheck.get(f.check) ?? [];
257
- arr.push(f);
258
- byCheck.set(f.check, arr);
259
- }
260
- for (const [check, list] of byCheck) {
261
- lines.push(`## ${check} (${list.length})`);
262
- lines.push("");
263
- // Sort by sliceCount desc — highest-volume issues first
264
- list.sort((a, b) => (b.sliceCount ?? 0) - (a.sliceCount ?? 0));
265
- for (const f of list.slice(0, 20)) {
266
- lines.push(`- **[${f.severity.toUpperCase()}]** ${f.detail}`);
267
- }
268
- if (list.length > MAX_LISTED_EXAMPLES) {
269
- lines.push(`- ... and ${list.length - 20} more`);
270
- }
271
- lines.push("");
272
- }
273
- return lines.join("\n");
274
- }
275
- /**
276
- * Lint a slice against corpus stats + the anti-pattern rules; print the markdown report to stdout.
277
- */
278
- export async function lintCorpusSlice(options, report) {
279
- const rulesPath = options.rulesPath ?? defaultRulesPath();
280
- report?.(`Reading corpus stats from ${options.statsPath}...`);
281
- const corpus = await readLocalJSONFile(options.statsPath);
282
- report?.(` ${corpus.row_count} rows from ${corpus.slice_paths.length} slice(s); ${Object.keys(corpus.tokens).length} tokens, ${Object.keys(corpus.bigrams).length} bigrams`);
283
- report?.(`Reading slice from ${options.slicePath}...`);
284
- const slice = await statsFromSlice(streamTokenLabelRows(options.slicePath));
285
- report?.(` ${slice.rowCount} rows`);
286
- report?.(`Loading rules from ${rulesPath}...`);
287
- const rulesFile = await readLocalJSONFile(rulesPath);
288
- report?.(`Running checks...`);
289
- const flags = [
290
- ...checkDistributionOutliers(slice, corpus),
291
- ...checkLabelVacuum(slice, corpus),
292
- ...checkBigramCollisions(slice, corpus),
293
- ...checkRules(slice, rulesFile),
294
- ...checkSanity(slice),
295
- ];
296
- const rendered = renderReport({ slicePath: options.slicePath, statsPath: options.statsPath, rulesPath }, slice, flags);
297
- console.log(rendered);
298
- if (options.outMd) {
299
- await writeLocalFile(rendered, options.outMd);
300
- }
301
- if (options.outJSON) {
302
- await writeLocalJSONFile({
303
- slice: options.slicePath,
304
- stats: options.statsPath,
305
- flags,
306
- summary: {
307
- errors: flags.filter((f) => f.severity === "error").length,
308
- warnings: flags.filter((f) => f.severity === "warn").length,
309
- },
310
- }, options.outJSON);
311
- }
312
- const errorCount = flags.filter((f) => f.severity === "error").length;
313
- const warningCount = flags.filter((f) => f.severity === "warn").length;
314
- if (errorCount > 0) {
315
- report?.(`LINT FAILED: ${errorCount} error(s).`);
316
- }
317
- else {
318
- report?.("LINT PASSED.");
319
- }
320
- return { errors: errorCount, warnings: warningCount, findings: flags, report: rendered };
321
- }
322
- //# sourceMappingURL=lint-slice.js.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"lint-slice.js","sourceRoot":"","sources":["../../lib/tools/lint-slice.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAA;AAC9D,OAAO,EAAE,cAAc,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AAC/E,OAAO,EAAE,kBAAkB,EAAE,MAAM,kCAAkC,CAAA;AAErE,OAAO,EACN,uBAAuB,EACvB,uBAAuB,EACvB,eAAe,IAAI,GAAG,EACtB,oBAAoB,GAEpB,MAAM,oBAAoB,CAAA;AAE3B;;GAEG;AACH,MAAM,gCAAgC,GAAG,CAAC,CAAA;AAE1C;;GAEG;AACH,MAAM,mBAAmB,GAAG,EAAE,CAAA;AAE9B;;;GAGG;AACH,MAAM,uBAAuB,GAAG,IAAI,CAAA;AACpC,MAAM,eAAe,GAAG,EAAE,CAAA;AAC1B,MAAM,gBAAgB,GAAG,GAAG,CAAA;AAC5B,MAAM,sBAAsB,GAAG,EAAE,CAAA;AACjC,MAAM,uBAAuB,GAAG,GAAG,CAAA;AACnC,MAAM,gBAAgB,GAAG,EAAE,CAAA;AAC3B,MAAM,mBAAmB,GAAG,GAAG,CAAA;AAE/B;;;;;GAKG;AACH,SAAS,gBAAgB;IACxB,OAAO,kBAAkB,CAAC,mBAAmB,EAAE,KAAK,EAAE,OAAO,EAAE,iBAAiB,CAAC,CAAA;AAClF,CAAC;AAwDD,KAAK,UAAU,cAAc,CAAC,IAA6B;IAC1D,MAAM,EAAE,GAAG,uBAAuB,EAAE,CAAA;IAEpC,MAAM,GAAG,GAAe;QACvB,QAAQ,EAAE,CAAC;QACX,MAAM,EAAE,EAAE,CAAC,MAAM;QACjB,OAAO,EAAE,EAAE,CAAC,OAAO;QACnB,aAAa,EAAE,CAAC;QAChB,QAAQ,EAAE,CAAC;KACX,CAAA;IAED,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QAC9B,GAAG,CAAC,QAAQ,EAAE,CAAA;QAEd,IAAI,GAAG,CAAC,MAAM,CAAC,MAAM,KAAK,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,CAAC;YAC7C,GAAG,CAAC,aAAa,EAAE,CAAA;YAEnB,SAAQ;QACT,CAAC;QAED,IAAI,GAAG,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,KAAK,GAAG,CAAC,EAAE,CAAC;YACxC,GAAG,CAAC,QAAQ,EAAE,CAAA;QACf,CAAC;QAED,uBAAuB,CAAC,EAAE,EAAE,GAAG,CAAC,MAAM,EAAE,GAAG,CAAC,MAAM,CAAC,CAAA;IACpD,CAAC;IAED,OAAO,GAAG,CAAA;AACX,CAAC;AAED,SAAS,aAAa,CAAC,YAA0D;IAMhF,MAAM,OAAO,GAAG,YAAY,YAAY,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,YAAY,CAAC,OAAO,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,YAAY,CAAC,CAAA;IACxG,IAAI,SAAS,GAAG,EAAE,CAAA;IAClB,IAAI,SAAS,GAAG,CAAC,CAAA;IACjB,IAAI,KAAK,GAAG,CAAC,CAAA;IAEb,KAAK,MAAM,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,OAAO,EAAE,CAAC;QACtC,KAAK,IAAI,KAAK,CAAA;QAEd,IAAI,KAAK,GAAG,SAAS,EAAE,CAAC;YACvB,SAAS,GAAG,KAAK,CAAA;YACjB,SAAS,GAAG,KAAK,CAAA;QAClB,CAAC;IACF,CAAC;IAED,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,KAAK,EAAE,SAAS,EAAE,KAAK,EAAE,UAAU,EAAE,KAAK,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,GAAG,KAAK,EAAE,CAAA;AACtG,CAAC;AA+BD,SAAS,yBAAyB,CAAC,KAAiB,EAAE,MAAmB;IACxE,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,KAAK,MAAM,CAAC,KAAK,EAAE,aAAa,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC;QACnD,MAAM,cAAc,GAAG,MAAM,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;QAE3C,IAAI,CAAC,cAAc;YAAE,SAAQ;QAC7B,MAAM,QAAQ,GAAG,aAAa,CAAC,aAAa,CAAC,CAAA;QAC7C,MAAM,SAAS,GAAG,aAAa,CAAC,cAAc,CAAC,CAAA;QAE/C,IACC,SAAS,CAAC,UAAU,IAAI,uBAAuB;YAC/C,QAAQ,CAAC,KAAK,KAAK,SAAS,CAAC,KAAK;YAClC,QAAQ,CAAC,KAAK,IAAI,eAAe;YACjC,SAAS,CAAC,KAAK,IAAI,gBAAgB,EAClC,CAAC;YACF,KAAK,CAAC,IAAI,CAAC;gBACV,KAAK,EAAE,sBAAsB;gBAC7B,QAAQ,EAAE,OAAO;gBACjB,KAAK;gBACL,UAAU,EAAE,QAAQ,CAAC,KAAK;gBAC1B,WAAW,EAAE,SAAS,CAAC,KAAK;gBAC5B,UAAU,EAAE,QAAQ,CAAC,KAAK;gBAC1B,WAAW,EAAE,SAAS,CAAC,KAAK;gBAC5B,MAAM,EAAE,UAAU,KAAK,wBAAwB,QAAQ,CAAC,KAAK,KAAK,QAAQ,CAAC,KAAK,IAAI,QAAQ,CAAC,KAAK,KAAK,CAAC,QAAQ,CAAC,UAAU,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,0BAA0B,SAAS,CAAC,KAAK,KAAK,SAAS,CAAC,KAAK,IAAI,SAAS,CAAC,KAAK,KAAK,CAAC,SAAS,CAAC,UAAU,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,KAAK;aAC9Q,CAAC,CAAA;QACH,CAAC;IACF,CAAC;IAED,OAAO,KAAK,CAAA;AACb,CAAC;AAED,SAAS,gBAAgB,CAAC,KAAiB,EAAE,MAAmB;IAC/D,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,KAAK,MAAM,CAAC,KAAK,EAAE,aAAa,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC;QACnD,MAAM,cAAc,GAAG,MAAM,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;QAE3C,IAAI,CAAC,cAAc;YAAE,SAAQ;QAC7B,MAAM,WAAW,GAAG,MAAM,CAAC,MAAM,CAAC,cAAc,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC,CAAA;QAE5E,IAAI,WAAW,GAAG,uBAAuB;YAAE,SAAQ;QAEnD,KAAK,MAAM,CAAC,KAAK,EAAE,UAAU,CAAC,IAAI,aAAa,EAAE,CAAC;YACjD,IAAI,UAAU,GAAG,sBAAsB;gBAAE,SAAQ;YAEjD,IAAI,cAAc,CAAC,KAAK,CAAC,KAAK,SAAS,IAAI,cAAc,CAAC,KAAK,CAAC,KAAK,CAAC,EAAE,CAAC;gBACxE,KAAK,CAAC,IAAI,CAAC;oBACV,KAAK,EAAE,cAAc;oBACrB,QAAQ,EAAE,OAAO;oBACjB,KAAK;oBACL,UAAU,EAAE,KAAK;oBACjB,UAAU;oBACV,WAAW,EAAE,WAAW;oBACxB,MAAM,EAAE,UAAU,KAAK,sBAAsB,KAAK,IAAI,UAAU,2BAA2B,WAAW,6DAA6D;iBACnK,CAAC,CAAA;YACH,CAAC;QACF,CAAC;IACF,CAAC;IAED,OAAO,KAAK,CAAA;AACb,CAAC;AAED,SAAS,qBAAqB,CAAC,KAAiB,EAAE,MAAmB;IACpE,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,KAAK,MAAM,CAAC,MAAM,EAAE,aAAa,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,CAAC;QACrD,MAAM,cAAc,GAAG,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;QAE7C,IAAI,CAAC,cAAc;YAAE,SAAQ;QAC7B,MAAM,QAAQ,GAAG,aAAa,CAAC,aAAa,CAAC,CAAA;QAC7C,MAAM,SAAS,GAAG,aAAa,CAAC,cAAc,CAAC,CAAA;QAE/C,IACC,QAAQ,CAAC,KAAK,KAAK,SAAS,CAAC,KAAK;YAClC,QAAQ,CAAC,KAAK,IAAI,gBAAgB;YAClC,SAAS,CAAC,KAAK,IAAI,gBAAgB,EAClC,CAAC;YACF,MAAM,YAAY,GAAG,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;YAChD,MAAM,gBAAgB,GAAG,QAAQ,CAAC,KAAK,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9D,MAAM,iBAAiB,GAAG,SAAS,CAAC,KAAK,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAEhE,KAAK,CAAC,IAAI,CAAC;gBACV,KAAK,EAAE,kBAAkB;gBACzB,QAAQ,EAAE,OAAO;gBACjB,MAAM,EAAE,YAAY;gBACpB,UAAU,EAAE,gBAAgB;gBAC5B,WAAW,EAAE,iBAAiB;gBAC9B,UAAU,EAAE,QAAQ,CAAC,KAAK;gBAC1B,WAAW,EAAE,SAAS,CAAC,KAAK;gBAC5B,MAAM,EAAE,WAAW,YAAY,6BAA6B,gBAAgB,MAAM,QAAQ,CAAC,KAAK,+BAA+B,iBAAiB,MAAM,SAAS,CAAC,KAAK,sDAAsD;aAC3N,CAAC,CAAA;QACH,CAAC;IACF,CAAC;IAED,OAAO,KAAK,CAAA;AACb,CAAC;AAED,SAAS,UAAU,CAAC,KAAiB,EAAE,SAAwB;IAC9D,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,MAAM,QAAQ,GAAG,SAAS,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;QAC5C,IAAI,EAAE,CAAC;QACP,KAAK,EAAE,IAAI,MAAM,CAAC,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC,sBAAsB,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,GAAG,CAAC;KACjE,CAAC,CAAC,CAAA;IAEH,KAAK,MAAM,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC;QAC9C,KAAK,MAAM,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,QAAQ,EAAE,CAAC;YACxC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC;gBAAE,SAAQ;YAEhC,KAAK,MAAM,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,QAAQ,EAAE,CAAC;gBACvC,IAAI,IAAI,CAAC,gBAAgB,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,IAAI,gCAAgC,EAAE,CAAC;oBACxF,KAAK,CAAC,IAAI,CAAC;wBACV,KAAK,EAAE,mBAAmB;wBAC1B,QAAQ,EAAE,IAAI,CAAC,QAAQ;wBACvB,MAAM,EAAE,IAAI,CAAC,EAAE;wBACf,KAAK;wBACL,UAAU,EAAE,KAAK;wBACjB,UAAU,EAAE,KAAK;wBACjB,MAAM,EAAE,UAAU,KAAK,kBAAkB,IAAI,CAAC,EAAE,mBAAmB,KAAK,IAAI,KAAK,2BAA2B,IAAI,CAAC,OAAO,EAAE;qBAC1H,CAAC,CAAA;gBACH,CAAC;YACF,CAAC;QACF,CAAC;IACF,CAAC;IAED,OAAO,KAAK,CAAA;AACb,CAAC;AAED,SAAS,WAAW,CAAC,KAAiB;IACrC,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,IAAI,KAAK,CAAC,aAAa,GAAG,CAAC,EAAE,CAAC;QAC7B,KAAK,CAAC,IAAI,CAAC;YACV,KAAK,EAAE,gBAAgB;YACvB,QAAQ,EAAE,OAAO;YACjB,MAAM,EAAE,GAAG,KAAK,CAAC,aAAa,uEAAuE;SACrG,CAAC,CAAA;IACH,CAAC;IAED,MAAM,SAAS,GAAG,KAAK,CAAC,QAAQ,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,KAAK,CAAC,QAAQ,CAAC,CAAA;IAE9D,IAAI,SAAS,IAAI,mBAAmB,EAAE,CAAC;QACtC,KAAK,CAAC,IAAI,CAAC;YACV,KAAK,EAAE,aAAa;YACpB,QAAQ,EAAE,MAAM;YAChB,MAAM,EAAE,GAAG,KAAK,CAAC,QAAQ,IAAI,KAAK,CAAC,QAAQ,UAAU,CAAC,SAAS,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,yDAAyD;SAC1I,CAAC,CAAA;IACH,CAAC;IAED,OAAO,KAAK,CAAA;AACb,CAAC;AAED,SAAS,YAAY,CACpB,IAAiE,EACjE,KAAiB,EACjB,KAAsB;IAEtB,MAAM,MAAM,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,OAAO,CAAC,CAAA;IAC1D,MAAM,KAAK,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,MAAM,CAAC,CAAA;IACxD,MAAM,OAAO,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,eAAe,CAAA;IAE/D,MAAM,KAAK,GAAa;QACvB,kBAAkB,OAAO,EAAE;QAC3B,EAAE;QACF,kBAAkB,IAAI,CAAC,SAAS,IAAI;QACpC,yBAAyB,IAAI,CAAC,SAAS,IAAI;QAC3C,kBAAkB,IAAI,CAAC,SAAS,IAAI;QACpC,qBAAqB,KAAK,CAAC,QAAQ,EAAE;QACrC,wBAAwB,KAAK,CAAC,MAAM,CAAC,IAAI,EAAE;QAC3C,yBAAyB,KAAK,CAAC,OAAO,CAAC,IAAI,EAAE;QAC7C,EAAE;QACF,eAAe,MAAM,CAAC,MAAM,kFAAkF;QAC9G,iBAAiB,KAAK,CAAC,MAAM,aAAa;QAC1C,EAAE;KACF,CAAA;IAED,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,CAAC;QACnB,KAAK,CAAC,IAAI,CAAC,wBAAwB,CAAC,CAAA;QAEpC,OAAO,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;IACxB,CAAC;IAED,MAAM,OAAO,GAAG,IAAI,GAAG,EAA2B,CAAA;IAElD,KAAK,MAAM,CAAC,IAAI,KAAK,EAAE,CAAC;QACvB,MAAM,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,IAAI,EAAE,CAAA;QACtC,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;QACX,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,EAAE,GAAG,CAAC,CAAA;IAC1B,CAAC;IAED,KAAK,MAAM,CAAC,KAAK,EAAE,IAAI,CAAC,IAAI,OAAO,EAAE,CAAC;QACrC,KAAK,CAAC,IAAI,CAAC,MAAM,KAAK,KAAK,IAAI,CAAC,MAAM,GAAG,CAAC,CAAA;QAC1C,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;QACd,wDAAwD;QACxD,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,UAAU,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,UAAU,IAAI,CAAC,CAAC,CAAC,CAAA;QAE9D,KAAK,MAAM,CAAC,IAAI,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,EAAE,CAAC,EAAE,CAAC;YACnC,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC,QAAQ,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC,MAAM,EAAE,CAAC,CAAA;QAC9D,CAAC;QAED,IAAI,IAAI,CAAC,MAAM,GAAG,mBAAmB,EAAE,CAAC;YACvC,KAAK,CAAC,IAAI,CAAC,aAAa,IAAI,CAAC,MAAM,GAAG,EAAE,OAAO,CAAC,CAAA;QACjD,CAAC;QAED,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;IACf,CAAC;IAED,OAAO,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;AACxB,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,KAAK,UAAU,eAAe,CACpC,OAA+B,EAC/B,MAA+B;IAE/B,MAAM,SAAS,GAAG,OAAO,CAAC,SAAS,IAAI,gBAAgB,EAAE,CAAA;IACzD,MAAM,EAAE,CAAC,6BAA6B,OAAO,CAAC,SAAS,KAAK,CAAC,CAAA;IAC7D,MAAM,MAAM,GAAG,MAAM,iBAAiB,CAAc,OAAO,CAAC,SAAS,CAAC,CAAA;IAEtE,MAAM,EAAE,CACP,KAAK,MAAM,CAAC,SAAS,cAAc,MAAM,CAAC,WAAW,CAAC,MAAM,cAAc,MAAM,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,MAAM,YAAY,MAAM,CAAC,IAAI,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,MAAM,UAAU,CACnK,CAAA;IAED,MAAM,EAAE,CAAC,sBAAsB,OAAO,CAAC,SAAS,KAAK,CAAC,CAAA;IAEtD,MAAM,KAAK,GAAG,MAAM,cAAc,CAAC,oBAAoB,CAAC,OAAO,CAAC,SAAS,CAAC,CAAC,CAAA;IAE3E,MAAM,EAAE,CAAC,KAAK,KAAK,CAAC,QAAQ,OAAO,CAAC,CAAA;IAEpC,MAAM,EAAE,CAAC,sBAAsB,SAAS,KAAK,CAAC,CAAA;IAC9C,MAAM,SAAS,GAAG,MAAM,iBAAiB,CAAgB,SAAS,CAAC,CAAA;IAEnE,MAAM,EAAE,CAAC,mBAAmB,CAAC,CAAA;IAE7B,MAAM,KAAK,GAAoB;QAC9B,GAAG,yBAAyB,CAAC,KAAK,EAAE,MAAM,CAAC;QAC3C,GAAG,gBAAgB,CAAC,KAAK,EAAE,MAAM,CAAC;QAClC,GAAG,qBAAqB,CAAC,KAAK,EAAE,MAAM,CAAC;QACvC,GAAG,UAAU,CAAC,KAAK,EAAE,SAAS,CAAC;QAC/B,GAAG,WAAW,CAAC,KAAK,CAAC;KACrB,CAAA;IAED,MAAM,QAAQ,GAAG,YAAY,CAAC,EAAE,SAAS,EAAE,OAAO,CAAC,SAAS,EAAE,SAAS,EAAE,OAAO,CAAC,SAAS,EAAE,SAAS,EAAE,EAAE,KAAK,EAAE,KAAK,CAAC,CAAA;IAEtH,OAAO,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAA;IAErB,IAAI,OAAO,CAAC,KAAK,EAAE,CAAC;QACnB,MAAM,cAAc,CAAC,QAAQ,EAAE,OAAO,CAAC,KAAK,CAAC,CAAA;IAC9C,CAAC;IAED,IAAI,OAAO,CAAC,OAAO,EAAE,CAAC;QACrB,MAAM,kBAAkB,CACvB;YACC,KAAK,EAAE,OAAO,CAAC,SAAS;YACxB,KAAK,EAAE,OAAO,CAAC,SAAS;YACxB,KAAK;YACL,OAAO,EAAE;gBACR,MAAM,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,OAAO,CAAC,CAAC,MAAM;gBAC1D,QAAQ,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,MAAM,CAAC,CAAC,MAAM;aAC3D;SACD,EACD,OAAO,CAAC,OAAO,CACf,CAAA;IACF,CAAC;IAED,MAAM,UAAU,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,OAAO,CAAC,CAAC,MAAM,CAAA;IACrE,MAAM,YAAY,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,KAAK,MAAM,CAAC,CAAC,MAAM,CAAA;IAEtE,IAAI,UAAU,GAAG,CAAC,EAAE,CAAC;QACpB,MAAM,EAAE,CAAC,gBAAgB,UAAU,YAAY,CAAC,CAAA;IACjD,CAAC;SAAM,CAAC;QACP,MAAM,EAAE,CAAC,cAAc,CAAC,CAAA;IACzB,CAAC;IAED,OAAO,EAAE,MAAM,EAAE,UAAU,EAAE,QAAQ,EAAE,YAAY,EAAE,QAAQ,EAAE,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,CAAA;AACzF,CAAC"}
@@ -1,32 +0,0 @@
1
- /**
2
- * @copyright Sister Software
3
- * @license AGPL-3.0
4
- * @author Teffen Ellis, et al.
5
- *
6
- * Build a parquet slice from the DeepSeek-generated kryptonite JSONL and emit the corpus-v0.4.0
7
- * MANIFEST. corpus-v0.4.0 is a pure adapter-addition revision: it points at every slice from
8
- * v0.3.0 plus the new kryptonite slice(s). No v0.3.0 bytes are touched or re-shuffled.
9
- *
10
- * See docs/engineering/reference/CORPUS_V0_4_0_GENERATION.md for the why; that doc also pins the
11
- * DeepSeek model version + prompt versions used to produce the JSONL.
12
- *
13
- * Invoke via `mailwoman corpus slice kryptonite \
14
- * --jsonl /data/corpus/versioned/v0.4.0/kryptonite/canonical-kryptonite.jsonl \
15
- * --base-manifest /data/corpus/versioned/v0.3.0/corpus-v0.3.0/MANIFEST.json \
16
- * --out-dir /data/corpus/versioned/v0.4.0`
17
- */
18
- export interface SliceKryptoniteOptions {
19
- jsonl: string;
20
- baseManifest: string;
21
- outDir: string;
22
- /**
23
- * Default `"0.4.0"`.
24
- */
25
- corpusVersion?: string;
26
- /**
27
- * Default `"deepseek-kryptonite"`.
28
- */
29
- source?: string;
30
- }
31
- export declare function buildKryptoniteSlice(options: SliceKryptoniteOptions, report?: (line: string) => void): Promise<void>;
32
- //# sourceMappingURL=slice-kryptonite.d.ts.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"slice-kryptonite.d.ts","sourceRoot":"","sources":["../../lib/tools/slice-kryptonite.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AAWH,MAAM,WAAW,sBAAsB;IACtC,KAAK,EAAE,MAAM,CAAA;IACb,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB;;OAEG;IACH,MAAM,CAAC,EAAE,MAAM,CAAA;CACf;AAiCD,wBAAsB,oBAAoB,CACzC,OAAO,EAAE,sBAAsB,EAC/B,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAC7B,OAAO,CAAC,IAAI,CAAC,CA6Df"}
@@ -1,99 +0,0 @@
1
- /**
2
- * @copyright Sister Software
3
- * @license AGPL-3.0
4
- * @author Teffen Ellis, et al.
5
- *
6
- * Build a parquet slice from the DeepSeek-generated kryptonite JSONL and emit the corpus-v0.4.0
7
- * MANIFEST. corpus-v0.4.0 is a pure adapter-addition revision: it points at every slice from
8
- * v0.3.0 plus the new kryptonite slice(s). No v0.3.0 bytes are touched or re-shuffled.
9
- *
10
- * See docs/engineering/reference/CORPUS_V0_4_0_GENERATION.md for the why; that doc also pins the
11
- * DeepSeek model version + prompt versions used to produce the JSONL.
12
- *
13
- * Invoke via `mailwoman corpus slice kryptonite \
14
- * --jsonl /data/corpus/versioned/v0.4.0/kryptonite/canonical-kryptonite.jsonl \
15
- * --base-manifest /data/corpus/versioned/v0.3.0/corpus-v0.3.0/MANIFEST.json \
16
- * --out-dir /data/corpus/versioned/v0.4.0`
17
- */
18
- import { pathExists, readLocalJSONFile } from "@mailwoman/core/fs/readers";
19
- import { writeLocalJSONFile, writeLocalTextFile, makeDirectories } from "@mailwoman/core/fs/writers";
20
- import { join } from "path-ts";
21
- import { JSONSpliterator } from "spliterator";
22
- import { alignRow, PARQUET_COLUMNS, ROW_GROUP_SIZE, SLICE_COMPRESSION, writeSlices } from "#utils";
23
- async function* canonicalRows(jsonl, corpusVersion) {
24
- for await (const raw of JSONSpliterator.fromAsync(jsonl)) {
25
- // Strip sidecar underscore-prefixed fields the generator left behind for debugging.
26
- const components = raw["components"];
27
- yield {
28
- raw: raw["raw"],
29
- components,
30
- country: raw["country"] ?? "US",
31
- locale: raw["locale"] ?? undefined,
32
- source: raw["source"] ?? "deepseek-kryptonite",
33
- source_id: raw["source_id"],
34
- corpus_version: corpusVersion,
35
- license: raw["license"] ?? "Synthetic (DeepSeek-v4-flash, AGPL-compatible)",
36
- synth: raw["synth"],
37
- };
38
- }
39
- }
40
- async function* labeledRows(jsonl, corpusVersion, quarantineLog) {
41
- for await (const row of canonicalRows(jsonl, corpusVersion)) {
42
- const result = alignRow(row);
43
- if (result.kind === "labeled") {
44
- yield result.row;
45
- }
46
- else {
47
- quarantineLog.push(`${row.source_id}\t${result.row.reason}`);
48
- }
49
- }
50
- }
51
- export async function buildKryptoniteSlice(options, report) {
52
- const corpusVersion = options.corpusVersion ?? "0.4.0";
53
- const source = options.source ?? "deepseek-kryptonite";
54
- if (!(await pathExists(options.jsonl)))
55
- throw new Error(`jsonl not found: ${options.jsonl}`);
56
- if (!(await pathExists(options.baseManifest)))
57
- throw new Error(`base-manifest not found: ${options.baseManifest}`);
58
- await makeDirectories(options.outDir);
59
- const quarantine = [];
60
- const newManifest = await writeSlices({ train: labeledRows(options.jsonl, corpusVersion, quarantine) }, { outputDir: options.outDir, corpusVersion });
61
- report?.(`wrote ${newManifest.total_rows} rows into ${newManifest.slices.length} slice(s); ` +
62
- `quarantined ${quarantine.length}`);
63
- if (quarantine.length) {
64
- const qPath = join(options.outDir, `corpus-v${corpusVersion}`, "quarantine-kryptonite.tsv");
65
- await writeLocalTextFile(quarantine.join("\n") + "\n", qPath);
66
- report?.(`quarantine log → ${qPath}`);
67
- }
68
- // Stamp the new slice's source field for audit.ts (which prefers slice.source over
69
- // first_source_id-prefix inference). Without this, deepseek-kryptonite IDs would have
70
- // to match a prefix in KNOWN_SOURCE_PREFIXES — we add it there too as a belt-and-braces.
71
- for (const sh of newManifest.slices) {
72
- sh.source = source;
73
- }
74
- // Compose the final corpus-v0.4.0 manifest: every slice from base + the new slice(s).
75
- const base = await readLocalJSONFile(options.baseManifest);
76
- const combined = {
77
- corpus_version: corpusVersion,
78
- schema: PARQUET_COLUMNS,
79
- rows_per_slice: base.rows_per_slice,
80
- row_group_size: base.row_group_size ?? ROW_GROUP_SIZE,
81
- slices: [...base.slices, ...newManifest.slices],
82
- counts: {
83
- train: base.counts.train + (newManifest.counts.train ?? 0),
84
- val: base.counts.val,
85
- test: base.counts.test,
86
- },
87
- total_rows: base.total_rows + newManifest.total_rows,
88
- };
89
- // Stamp source on the legacy v0.3.0 slices too, so audit's slice.source path is the
90
- // authoritative one. v0.3.0 slices mix sources; we use the first_source_id-prefix
91
- // inference for them (audit.ts will re-derive on its own when slice.source is absent).
92
- const combinedPath = join(options.outDir, `corpus-v${corpusVersion}`, "MANIFEST.json");
93
- await writeLocalJSONFile(combined, combinedPath);
94
- report?.(`wrote combined manifest → ${combinedPath}`);
95
- report?.(` total_rows=${combined.total_rows} (base=${base.total_rows}, added=${newManifest.total_rows})`);
96
- report?.(` slices=${combined.slices.length} (base=${base.slices.length}, added=${newManifest.slices.length})`);
97
- report?.(` compression=${SLICE_COMPRESSION}`);
98
- }
99
- //# sourceMappingURL=slice-kryptonite.js.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"slice-kryptonite.js","sourceRoot":"","sources":["../../lib/tools/slice-kryptonite.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AAEH,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAA;AAC1E,OAAO,EAAE,kBAAkB,EAAE,kBAAkB,EAAE,eAAe,EAAE,MAAM,4BAA4B,CAAA;AACpG,OAAO,EAAE,IAAI,EAAE,MAAM,SAAS,CAAA;AAC9B,OAAO,EAAE,eAAe,EAAE,MAAM,aAAa,CAAA;AAG7C,OAAO,EAAE,QAAQ,EAAE,eAAe,EAAE,cAAc,EAAE,iBAAiB,EAAE,WAAW,EAAE,MAAM,QAAQ,CAAA;AAiBlG,KAAK,SAAS,CAAC,CAAC,aAAa,CAAC,KAAa,EAAE,aAAqB;IACjE,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,eAAe,CAAC,SAAS,CAA0B,KAAK,CAAC,EAAE,CAAC;QACnF,oFAAoF;QACpF,MAAM,UAAU,GAAG,GAAG,CAAC,YAAY,CAA2B,CAAA;QAE9D,MAAM;YACL,GAAG,EAAE,GAAG,CAAC,KAAK,CAAW;YACzB,UAAU;YACV,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,IAAI;YAC3C,MAAM,EAAG,GAAG,CAAC,QAAQ,CAAY,IAAI,SAAS;YAC9C,MAAM,EAAG,GAAG,CAAC,QAAQ,CAAY,IAAI,qBAAqB;YAC1D,SAAS,EAAE,GAAG,CAAC,WAAW,CAAW;YACrC,cAAc,EAAE,aAAa;YAC7B,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,gDAAgD;YACvF,KAAK,EAAE,GAAG,CAAC,OAAO,CAA0B;SAC5C,CAAA;IACF,CAAC;AACF,CAAC;AAED,KAAK,SAAS,CAAC,CAAC,WAAW,CAAC,KAAa,EAAE,aAAqB,EAAE,aAAuB;IACxF,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,aAAa,CAAC,KAAK,EAAE,aAAa,CAAC,EAAE,CAAC;QAC7D,MAAM,MAAM,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAA;QAE5B,IAAI,MAAM,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;YAC/B,MAAM,MAAM,CAAC,GAAG,CAAA;QACjB,CAAC;aAAM,CAAC;YACP,aAAa,CAAC,IAAI,CAAC,GAAG,GAAG,CAAC,SAAS,KAAK,MAAM,CAAC,GAAG,CAAC,MAAM,EAAE,CAAC,CAAA;QAC7D,CAAC;IACF,CAAC;AACF,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,oBAAoB,CACzC,OAA+B,EAC/B,MAA+B;IAE/B,MAAM,aAAa,GAAG,OAAO,CAAC,aAAa,IAAI,OAAO,CAAA;IACtD,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,IAAI,qBAAqB,CAAA;IAEtD,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,oBAAoB,OAAO,CAAC,KAAK,EAAE,CAAC,CAAA;IAE5F,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,4BAA4B,OAAO,CAAC,YAAY,EAAE,CAAC,CAAA;IAElH,MAAM,eAAe,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;IAErC,MAAM,UAAU,GAAa,EAAE,CAAA;IAE/B,MAAM,WAAW,GAAG,MAAM,WAAW,CACpC,EAAE,KAAK,EAAE,WAAW,CAAC,OAAO,CAAC,KAAK,EAAE,aAAa,EAAE,UAAU,CAAC,EAAE,EAChE,EAAE,SAAS,EAAE,OAAO,CAAC,MAAM,EAAE,aAAa,EAAE,CAC5C,CAAA;IAED,MAAM,EAAE,CACP,SAAS,WAAW,CAAC,UAAU,cAAc,WAAW,CAAC,MAAM,CAAC,MAAM,aAAa;QAClF,eAAe,UAAU,CAAC,MAAM,EAAE,CACnC,CAAA;IAED,IAAI,UAAU,CAAC,MAAM,EAAE,CAAC;QACvB,MAAM,KAAK,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,WAAW,aAAa,EAAE,EAAE,2BAA2B,CAAC,CAAA;QAC3F,MAAM,kBAAkB,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,IAAI,EAAE,KAAK,CAAC,CAAA;QAC7D,MAAM,EAAE,CAAC,oBAAoB,KAAK,EAAE,CAAC,CAAA;IACtC,CAAC;IAED,mFAAmF;IACnF,sFAAsF;IACtF,yFAAyF;IACzF,KAAK,MAAM,EAAE,IAAI,WAAW,CAAC,MAAM,EAAE,CAAC;QACrC,EAAE,CAAC,MAAM,GAAG,MAAM,CAAA;IACnB,CAAC;IAED,sFAAsF;IACtF,MAAM,IAAI,GAAG,MAAM,iBAAiB,CAAgB,OAAO,CAAC,YAAY,CAAC,CAAA;IAEzE,MAAM,QAAQ,GAAkB;QAC/B,cAAc,EAAE,aAAa;QAC7B,MAAM,EAAE,eAAe;QACvB,cAAc,EAAE,IAAI,CAAC,cAAc;QACnC,cAAc,EAAE,IAAI,CAAC,cAAc,IAAI,cAAc;QACrD,MAAM,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,GAAG,WAAW,CAAC,MAAM,CAAC;QAC/C,MAAM,EAAE;YACP,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,GAAG,CAAC,WAAW,CAAC,MAAM,CAAC,KAAK,IAAI,CAAC,CAAC;YAC1D,GAAG,EAAE,IAAI,CAAC,MAAM,CAAC,GAAG;YACpB,IAAI,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;SACtB;QACD,UAAU,EAAE,IAAI,CAAC,UAAU,GAAG,WAAW,CAAC,UAAU;KACpD,CAAA;IAED,oFAAoF;IACpF,kFAAkF;IAClF,uFAAuF;IACvF,MAAM,YAAY,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,WAAW,aAAa,EAAE,EAAE,eAAe,CAAC,CAAA;IACtF,MAAM,kBAAkB,CAAC,QAAQ,EAAE,YAAY,CAAC,CAAA;IAChD,MAAM,EAAE,CAAC,6BAA6B,YAAY,EAAE,CAAC,CAAA;IACrD,MAAM,EAAE,CAAC,gBAAgB,QAAQ,CAAC,UAAU,UAAU,IAAI,CAAC,UAAU,WAAW,WAAW,CAAC,UAAU,GAAG,CAAC,CAAA;IAC1G,MAAM,EAAE,CAAC,YAAY,QAAQ,CAAC,MAAM,CAAC,MAAM,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,WAAW,WAAW,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAA;IAC/G,MAAM,EAAE,CAAC,iBAAiB,iBAAiB,EAAE,CAAC,CAAA;AAC/C,CAAC"}
@@ -1,51 +0,0 @@
1
- /**
2
- * @copyright Sister Software
3
- * @license AGPL-3.0
4
- * @author Teffen Ellis, et al.
5
- *
6
- * Build per-script parquet slices from the DeepSeek-generated transliteration JSONL and emit the
7
- * corpus-v0.4.0 MANIFEST that combines them with the existing kryptonite + v0.3.0 slices.
8
- *
9
- * Sibling to `slice-kryptonite.ts`. The two modules share the same composition pattern: take a
10
- * base MANIFEST, append new slices, write a combined MANIFEST. Differences specific to
11
- * transliteration:
12
- *
13
- * - One JSONL contains rows from N target scripts (source = `deepseek-translit-<slug>`). We bucket by
14
- * `source` and write one slice per script so `audit.ts` can attribute each slice to its
15
- * synthetic source without relying on filename-prefix inference.
16
- * - Each slice is written to `train/part-translit-<slug>.parquet` (distinct from kryptonite's
17
- * `part-0000.parquet`, which v0.4.0's first builder already produced).
18
- * - Inherits the path-canonicalization fix flagged in Thread B's postmortem: v0.3.0 slice paths are
19
- * rewritten from `$MAILWOMAN_DATA_ROOT/...` to `/data/...` in the combined MANIFEST so
20
- * all paths share one container-friendly form.
21
- *
22
- * See docs/engineering/reference/CORPUS_V0_4_0_GENERATION.md for prompts, model, and the
23
- * reproducibility contract.
24
- *
25
- * Invoke via `mailwoman corpus slice translit \
26
- * --jsonl /data/corpus/versioned/v0.4.0/transliteration/canonical-transliteration.jsonl \
27
- * --base-manifest /data/corpus/versioned/v0.4.0/corpus-v0.4.0/MANIFEST.json \
28
- * --out-dir /data/corpus/versioned/v0.4.0`
29
- */
30
- export interface SliceTranslitOptions {
31
- jsonl: string;
32
- baseManifest: string;
33
- outDir: string;
34
- /**
35
- * Default `"0.4.0"`.
36
- */
37
- corpusVersion?: string;
38
- /**
39
- * Default `"/data/"`.
40
- */
41
- canonicalPathPrefix?: string;
42
- /**
43
- * Prefix the base manifest's slice paths currently carry, to be rewritten to
44
- * {@link SliceTranslitOptions.canonicalPathPrefix}. Defaults to `mailwomanDataRoot()` with a trailing slash — the
45
- * root that WROTE those paths. Pass it explicitly when translating a manifest generated under a different
46
- * `$MAILWOMAN_DATA_ROOT` than the one you are running with.
47
- */
48
- legacyPathPrefix?: string;
49
- }
50
- export declare function buildTranslitSlice(options: SliceTranslitOptions, report?: (line: string) => void): Promise<void>;
51
- //# sourceMappingURL=slice-translit.d.ts.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"slice-translit.d.ts","sourceRoot":"","sources":["../../lib/tools/slice-translit.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAaH,MAAM,WAAW,oBAAoB;IACpC,KAAK,EAAE,MAAM,CAAA;IACb,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB;;OAEG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAA;IAC5B;;;;;OAKG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAA;CACzB;AA0DD,wBAAsB,kBAAkB,CACvC,OAAO,EAAE,oBAAoB,EAC7B,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAC7B,OAAO,CAAC,IAAI,CAAC,CAgGf"}