@mailwoman/corpus 9.4.0 → 10.1.0

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 (1724) hide show
  1. package/README.md +134 -32
  2. package/data/PROVENANCE.md +623 -17
  3. package/data/address-source-register.json +14249 -0
  4. package/data/builds/v0.7.0-de-holdout/inputs.json +22 -0
  5. package/data/builds/v0.8.0-registers/inputs.json +92 -0
  6. package/data/license-decisions.json +1603 -0
  7. package/data/licenses/PROVENANCE.md +65 -0
  8. package/data/licenses/cdla-permissive-2.0.md +43 -0
  9. package/data/licenses/geonames-publication.md +34 -0
  10. package/data/licenses/licence-ouverte-2.0.md +36 -0
  11. package/data/licenses/national-address-database.md +46 -0
  12. package/data/licenses/nlod-2.0.md +90 -0
  13. package/data/licenses/sirene-publication.md +38 -0
  14. package/data/licenses/whosonfirst-licenses.md +65 -0
  15. package/data/source-resolutions.json +804 -0
  16. package/data/sub-venue-lexicon.json +1 -1
  17. package/data/training-manifests/v0.7.0-de-holdout.effective.json +477 -0
  18. package/data/training-manifests/v0.7.0-de-holdout.json +78 -0
  19. package/lib/adapters/cordis/adapter.ts +785 -0
  20. package/lib/adapters/dedup-key.ts +40 -0
  21. package/lib/adapters/errors.ts +21 -0
  22. package/lib/adapters/geonames/adapter.ts +43 -53
  23. package/lib/adapters/geonames/postal/adapter.ts +116 -0
  24. package/lib/adapters/gleif/adapter.ts +828 -0
  25. package/lib/adapters/libpostal-dictionary.ts +39 -0
  26. package/lib/adapters/openaddresses/adapter.ts +54 -61
  27. package/lib/adapters/osm/adapter.ts +356 -0
  28. package/lib/adapters/overture/adapter.ts +65 -41
  29. package/lib/adapters/po-box/adapter.ts +202 -0
  30. package/lib/adapters/registry.ts +79 -0
  31. package/lib/adapters/sha256-stream.ts +49 -0
  32. package/lib/adapters/source-id.ts +49 -0
  33. package/lib/adapters/street-line.ts +135 -0
  34. package/lib/adapters/wof/admin/json/adapter.ts +254 -0
  35. package/lib/adapters/wof/json-rows.ts +211 -0
  36. package/lib/adapters/wof/postalcode-json/adapter.ts +194 -0
  37. package/lib/adapters.ts +230 -0
  38. package/lib/au/adapters/gnaf/adapter.ts +144 -0
  39. package/lib/au/adapters/gnaf/assemble.ts +205 -0
  40. package/lib/be/adapters/brussels/adapter.ts +566 -0
  41. package/lib/be/adapters/vlaanderen/adapter.ts +670 -0
  42. package/lib/be/adapters/wallonie/adapter.ts +537 -0
  43. package/lib/be/tools/fetch/brussels.ts +239 -0
  44. package/lib/be/tools/fetch/wallonie.ts +274 -0
  45. package/lib/build/adapters.ts +110 -0
  46. package/lib/build/align.ts +364 -0
  47. package/lib/build/checkpoint.ts +131 -0
  48. package/lib/build/eligibility.ts +96 -0
  49. package/lib/build/heap.ts +53 -0
  50. package/lib/build/inputs.ts +81 -0
  51. package/lib/build/rows.ts +21 -0
  52. package/lib/build/types.ts +232 -0
  53. package/lib/build.ts +127 -315
  54. package/lib/cn/recipes/organizational-units.ts +178 -0
  55. package/lib/cz/adapters/cuzk/adapter.ts +317 -0
  56. package/lib/cz/recipes/pcfirst-preposition.ts +125 -0
  57. package/lib/cz/tools/fetch/cuzk.ts +699 -0
  58. package/lib/de/recipes/locale.ts +370 -0
  59. package/lib/directory.ts +31 -0
  60. package/lib/dk/adapters/inspire/adapter.ts +345 -0
  61. package/lib/dk/tools/fetch/inspire.ts +321 -0
  62. package/lib/ee/adapters/ads/adapter.ts +246 -0
  63. package/lib/ee/tools/fetch/ads.ts +270 -0
  64. package/lib/env.ts +6 -2
  65. package/lib/es/adapters/bizkaia/adapter.ts +388 -0
  66. package/lib/es/adapters/catastro/adapter.ts +343 -0
  67. package/lib/es/adapters/gipuzkoa/adapter.ts +263 -0
  68. package/lib/es/adapters/navarra/adapter.ts +289 -0
  69. package/lib/es/cadastre.ts +194 -0
  70. package/lib/es/tools/fetch/bizkaia.ts +748 -0
  71. package/lib/es/tools/fetch/catastro.ts +754 -0
  72. package/lib/es/tools/fetch/gipuzkoa.ts +336 -0
  73. package/lib/es/tools/fetch/navarra.ts +488 -0
  74. package/lib/fi/adapters/ryhti/adapter.ts +323 -0
  75. package/lib/fi/tools/fetch/ryhti.ts +476 -0
  76. package/lib/fingerprints.ts +154 -0
  77. package/lib/fr/adapters/annuaire-education/adapter.ts +598 -0
  78. package/lib/fr/adapters/ban/adapter.ts +187 -0
  79. package/lib/fr/adapters/finess/adapter.ts +618 -0
  80. package/lib/fr/adapters/finess/postal-lines.ts +157 -0
  81. package/lib/fr/adapters/finess/voie-type.ts +200 -0
  82. package/lib/fr/insee-country.ts +62 -0
  83. package/lib/fr/recipes/admin-split.ts +221 -0
  84. package/lib/fr/recipes/bare-street.ts +158 -0
  85. package/lib/fr/recipes/fragment.ts +251 -0
  86. package/lib/fr/recipes/lieudit.ts +310 -0
  87. package/lib/fr/recipes/order.ts +219 -0
  88. package/lib/fr/street-decompose.ts +67 -0
  89. package/lib/fr/tools/fetch/annuaire-education.ts +306 -0
  90. package/lib/fr/tools/fetch/ban.ts +298 -0
  91. package/lib/fr/tools/fetch/finess.ts +352 -0
  92. package/lib/gauntlet-inputs.ts +92 -0
  93. package/lib/index.ts +10 -4
  94. package/lib/inspire/address.ts +457 -0
  95. package/lib/inspire/archive.ts +44 -0
  96. package/lib/inspire/errors.ts +71 -0
  97. package/lib/inspire/stream.ts +262 -0
  98. package/lib/international/recipes/locale.ts +556 -0
  99. package/lib/it/adapters/anac/adapter.ts +522 -0
  100. package/lib/it/tools/fetch/anac.ts +394 -0
  101. package/lib/jp/adapters/wof-admin/adapter.ts +224 -0
  102. package/lib/jp/tools/fetch/houjin.ts +139 -0
  103. package/lib/kr/adapters/juso/label-rows.ts +239 -0
  104. package/lib/kr/adapters/localdata/align.ts +288 -0
  105. package/lib/kr/adapters/localdata/key-index.ts +177 -0
  106. package/lib/kr/adapters/localdata/permit-rows.ts +193 -0
  107. package/lib/kr/tools/fetch/juso.ts +170 -0
  108. package/lib/kr/tools/fetch/localdata.ts +205 -0
  109. package/lib/lu/adapters/bd-adresses/adapter.ts +240 -0
  110. package/lib/lu/tools/fetch/bd-adresses.ts +351 -0
  111. package/lib/lv/adapters/varis/adapter.ts +499 -0
  112. package/lib/lv/tools/fetch/varis.ts +214 -0
  113. package/lib/nl/adapters/kadaster/adapter.ts +386 -0
  114. package/lib/nl/recipes/postcode.ts +139 -0
  115. package/lib/nl/tools/fetch/kadaster.ts +298 -0
  116. package/lib/no/adapters/matrikkelen/adapter.ts +187 -0
  117. package/lib/no/recipes/fragment.ts +189 -0
  118. package/lib/no/recipes/street/led.ts +154 -0
  119. package/lib/no/recipes/street.ts +109 -0
  120. package/lib/no/tools/fetch/matrikkelen.ts +519 -0
  121. package/lib/parquet/duckdb.ts +144 -0
  122. package/lib/parquet/readers.ts +86 -0
  123. package/lib/parquet/schema.ts +214 -0
  124. package/lib/parquet/shuffle.ts +59 -0
  125. package/lib/parquet/span-sql.ts +54 -0
  126. package/lib/parquet/streams.ts +113 -0
  127. package/lib/parquet/writers.ts +451 -0
  128. package/lib/pl/adapters/emuia/adapter.ts +202 -0
  129. package/lib/pl/tools/fetch/emuia.ts +360 -0
  130. package/lib/recipes/anchor-absorption.ts +30 -19
  131. package/lib/recipes/bare/country.ts +87 -0
  132. package/lib/recipes/bare/postcode/eval.ts +99 -0
  133. package/lib/recipes/bare/postcode.ts +346 -0
  134. package/lib/recipes/boundary-stress.ts +36 -19
  135. package/lib/recipes/country-balanced.ts +67 -87
  136. package/lib/recipes/house-venue.ts +31 -13
  137. package/lib/recipes/intersection.ts +83 -61
  138. package/lib/recipes/po/box/cedex/au-nz.ts +82 -0
  139. package/lib/recipes/po/box/cedex/ca.ts +59 -0
  140. package/lib/recipes/po/box/cedex/fr.ts +77 -0
  141. package/lib/recipes/po/box/cedex/phrases.ts +109 -0
  142. package/lib/recipes/po/box/cedex/recipe.ts +215 -0
  143. package/lib/recipes/po/box/cedex/sources.ts +189 -0
  144. package/lib/recipes/po/box/cedex/types.ts +35 -0
  145. package/lib/recipes/po/box/cedex/us.ts +74 -0
  146. package/lib/recipes/po/box/cedex/vocabulary.ts +106 -0
  147. package/lib/recipes/po/box/cedex.ts +9 -0
  148. package/lib/recipes/po/box.ts +149 -0
  149. package/lib/recipes/scaffold.ts +258 -182
  150. package/lib/recipes/sources.ts +399 -0
  151. package/lib/recipes/street/affix.ts +707 -0
  152. package/lib/recipes/street/bare.ts +115 -0
  153. package/lib/recipes/street.ts +32 -8
  154. package/lib/recipes/sub/venue/context.ts +156 -0
  155. package/lib/recipes/sub/venue/render.ts +170 -0
  156. package/lib/recipes/sub/venue/sources.ts +588 -0
  157. package/lib/recipes/sub/venue.ts +801 -0
  158. package/lib/recipes/trailing-region.ts +70 -81
  159. package/lib/recipes/unit/sources.ts +44 -0
  160. package/lib/recipes/unit.ts +102 -90
  161. package/lib/recipes.ts +106 -0
  162. package/lib/registers.ts +278 -0
  163. package/lib/ro/adapters/retea-scolara/adapter.ts +264 -0
  164. package/lib/ro/tools/fetch/retea-scolara.ts +187 -0
  165. package/lib/runner.ts +220 -84
  166. package/lib/sg/recipes/register.ts +264 -0
  167. package/lib/sg/tools/fetch/acra.ts +174 -0
  168. package/lib/si/recipes/bare-village.ts +116 -0
  169. package/lib/sk/adapters/inspire/adapter.ts +395 -0
  170. package/lib/sk/tools/fetch/inspire.ts +402 -0
  171. package/lib/source-register/effective-manifest.ts +659 -0
  172. package/lib/source-register/register.ts +539 -0
  173. package/lib/source-register/training-manifest.ts +218 -0
  174. package/lib/source-register/types.ts +590 -0
  175. package/lib/source-register.ts +71 -0
  176. package/lib/south-asia/recipes/register.ts +233 -0
  177. package/lib/subvenue/harvest.ts +251 -0
  178. package/lib/subvenue/promotions.ts +293 -0
  179. package/lib/subvenue/surfaces.ts +94 -0
  180. package/lib/subvenue/table.ts +264 -0
  181. package/lib/surfaces/locale.ts +246 -0
  182. package/lib/synthesizers/anchor-absorption.ts +55 -91
  183. package/lib/synthesizers/boundary-stress.ts +49 -131
  184. package/lib/synthesizers/house-venue.ts +44 -102
  185. package/lib/synthesizers/intersection.ts +37 -26
  186. package/lib/synthesizers/no-street.ts +70 -119
  187. package/lib/synthesizers/po-box.ts +90 -105
  188. package/lib/synthesizers/street.ts +47 -35
  189. package/lib/synthesizers/utils.ts +65 -219
  190. package/lib/test-kit/corpus-recipe.ts +45 -28
  191. package/lib/test-kit.ts +116 -0
  192. package/lib/tools/align-canonical.ts +64 -0
  193. package/lib/tools/audit.ts +114 -118
  194. package/lib/tools/corpus-stats.ts +46 -32
  195. package/lib/tools/fetch/atom.ts +164 -0
  196. package/lib/tools/fetch/cordis.ts +318 -0
  197. package/lib/tools/fetch/download/manifest.ts +114 -0
  198. package/lib/tools/fetch/download/network.ts +250 -0
  199. package/lib/tools/fetch/download/retry.ts +174 -0
  200. package/lib/tools/fetch/download.ts +4 -471
  201. package/lib/tools/fetch/geonames/dump.ts +275 -0
  202. package/lib/tools/fetch/geonames/postal.ts +145 -0
  203. package/lib/tools/fetch/gleif.ts +360 -0
  204. package/lib/tools/fetch/header.ts +69 -0
  205. package/lib/tools/fetch/inspire-addresses.ts +495 -0
  206. package/lib/tools/fetch/openaddresses.ts +20 -109
  207. package/lib/tools/fetch/ourairports.ts +16 -49
  208. package/lib/tools/fetch/ppd.ts +10 -10
  209. package/lib/tools/fetch/vlaanderen-ad.ts +435 -0
  210. package/lib/tools/fetch/wfs-harvest.ts +943 -0
  211. package/lib/tools/fetch/wikidata-subvenue.ts +43 -143
  212. package/lib/tools/fetch/zip-archive.ts +100 -0
  213. package/lib/tools/fetch.ts +143 -0
  214. package/lib/tools/golden/expand.ts +605 -0
  215. package/lib/tools/golden/promote.ts +349 -0
  216. package/lib/tools/golden/relabel-street/directory.ts +367 -0
  217. package/lib/tools/golden/relabel-street/row.ts +283 -0
  218. package/lib/tools/golden/relabel-street.ts +9 -0
  219. package/lib/tools/ingest-csv.ts +45 -50
  220. package/lib/tools/jsonl-to-parquet.ts +59 -98
  221. package/lib/tools/lint/recipe-output/vocab.ts +366 -0
  222. package/lib/tools/lint/recipe-output.ts +482 -0
  223. package/lib/tools/migrate/overlay-parquet.ts +173 -0
  224. package/lib/tools/migrate/recipe-output.ts +153 -0
  225. package/lib/tools/migrate/source-names.ts +161 -0
  226. package/lib/tools/overlay/kryptonite.ts +144 -0
  227. package/lib/tools/overlay/manifest.ts +407 -0
  228. package/lib/tools/overlay/merge-source.ts +138 -0
  229. package/lib/tools/overlay/split-slice.ts +144 -0
  230. package/lib/tools/overlay/translit.ts +234 -0
  231. package/lib/tools/overture-subvenue.ts +18 -91
  232. package/lib/tools/postcode-triples.ts +362 -167
  233. package/lib/tools/source-register/build.ts +819 -0
  234. package/lib/tools/sub/venue/head-nouns.ts +208 -0
  235. package/lib/tools/sub/venue/lexicon.ts +394 -0
  236. package/lib/tools/sub/venue/wikidata.ts +88 -0
  237. package/lib/tools.ts +31 -0
  238. package/lib/tw/tools/fetch/gcis.ts +181 -0
  239. package/lib/types.ts +304 -101
  240. package/lib/us/adapters/fcc-bdc/adapter.ts +158 -0
  241. package/lib/us/adapters/state/hi-schools/adapter.ts +171 -0
  242. package/lib/us/adapters/state/hi-schools/workbook.ts +40 -0
  243. package/lib/us/adapters/state/ia-contractors/adapter.ts +148 -0
  244. package/lib/us/adapters/state/ny-notaries/adapter.ts +132 -0
  245. package/lib/us/adapters/state/tx-notaries/adapter.ts +155 -0
  246. package/lib/us/adapters/tiger/adapter.ts +225 -0
  247. package/lib/us/adapters/tiger/street-decompose.ts +73 -0
  248. package/lib/us/adapters/usgov/hrsa-fqhc/adapter.ts +171 -0
  249. package/lib/us/adapters/usgov/imls-pls/adapter.ts +140 -0
  250. package/lib/us/adapters/usgov/irs-bmf/adapter.ts +154 -0
  251. package/lib/us/adapters/usgov/nad/adapter.ts +263 -0
  252. package/lib/us/adapters/usgov/nppes/adapter.ts +180 -0
  253. package/lib/us/adapters/usgov/samhsa-treatment-locator/adapter.ts +195 -0
  254. package/lib/us/fips-state.ts +122 -0
  255. package/lib/us/tools/fetch/hrsa.ts +58 -0
  256. package/lib/us/tools/fetch/imls-pls.ts +131 -0
  257. package/lib/us/tools/fetch/nad.ts +359 -0
  258. package/lib/us/tools/fetch/nppes.ts +153 -0
  259. package/lib/us/tools/fetch/state/hi-schools.ts +167 -0
  260. package/lib/us/tools/fetch/state/sources.ts +146 -0
  261. package/lib/us/tools/fetch/tiger-full.ts +256 -0
  262. package/lib/utils/align.ts +43 -53
  263. package/lib/utils/cooccurrence-stats.ts +101 -0
  264. package/lib/utils/golden.ts +39 -50
  265. package/lib/utils/license.ts +238 -26
  266. package/lib/utils/split.ts +205 -88
  267. package/lib/utils/tokenize.ts +22 -34
  268. package/lib/utils/wof-json.ts +134 -87
  269. package/lib/utils.ts +13 -0
  270. package/lib/ve/recipes/reviewed-postcode-tail.ts +217 -0
  271. package/out/adapters/cordis/adapter.d.ts +324 -0
  272. package/out/adapters/cordis/adapter.d.ts.map +1 -0
  273. package/out/adapters/cordis/adapter.js +616 -0
  274. package/out/adapters/cordis/adapter.js.map +1 -0
  275. package/out/adapters/dedup-key.d.ts +30 -0
  276. package/out/adapters/dedup-key.d.ts.map +1 -0
  277. package/out/adapters/dedup-key.js +35 -0
  278. package/out/adapters/dedup-key.js.map +1 -0
  279. package/out/adapters/errors.d.ts +17 -0
  280. package/out/adapters/errors.d.ts.map +1 -0
  281. package/out/adapters/errors.js +20 -0
  282. package/out/adapters/errors.js.map +1 -0
  283. package/out/adapters/geonames/adapter.d.ts +26 -28
  284. package/out/adapters/geonames/adapter.d.ts.map +1 -1
  285. package/out/adapters/geonames/adapter.js +40 -50
  286. package/out/adapters/geonames/adapter.js.map +1 -1
  287. package/out/adapters/geonames/postal/adapter.d.ts +41 -0
  288. package/out/adapters/geonames/postal/adapter.d.ts.map +1 -0
  289. package/out/adapters/geonames/postal/adapter.js +102 -0
  290. package/out/adapters/geonames/postal/adapter.js.map +1 -0
  291. package/out/adapters/gleif/adapter.d.ts +293 -0
  292. package/out/adapters/gleif/adapter.d.ts.map +1 -0
  293. package/out/adapters/gleif/adapter.js +684 -0
  294. package/out/adapters/gleif/adapter.js.map +1 -0
  295. package/out/adapters/libpostal-dictionary.d.ts +22 -0
  296. package/out/adapters/libpostal-dictionary.d.ts.map +1 -0
  297. package/out/adapters/libpostal-dictionary.js +36 -0
  298. package/out/adapters/libpostal-dictionary.js.map +1 -0
  299. package/out/adapters/openaddresses/adapter.d.ts +28 -43
  300. package/out/adapters/openaddresses/adapter.d.ts.map +1 -1
  301. package/out/adapters/openaddresses/adapter.js +37 -51
  302. package/out/adapters/openaddresses/adapter.js.map +1 -1
  303. package/out/adapters/osm/adapter.d.ts +101 -0
  304. package/out/adapters/osm/adapter.d.ts.map +1 -0
  305. package/out/adapters/osm/adapter.js +260 -0
  306. package/out/adapters/osm/adapter.js.map +1 -0
  307. package/out/adapters/overture/adapter.d.ts +14 -30
  308. package/out/adapters/overture/adapter.d.ts.map +1 -1
  309. package/out/adapters/overture/adapter.js +57 -41
  310. package/out/adapters/overture/adapter.js.map +1 -1
  311. package/out/adapters/po-box/adapter.d.ts +63 -0
  312. package/out/adapters/po-box/adapter.d.ts.map +1 -0
  313. package/out/adapters/po-box/adapter.js +139 -0
  314. package/out/adapters/po-box/adapter.js.map +1 -0
  315. package/out/adapters/registry.d.ts +55 -0
  316. package/out/adapters/registry.d.ts.map +1 -0
  317. package/out/adapters/registry.js +43 -0
  318. package/out/adapters/registry.js.map +1 -0
  319. package/out/adapters/sha256-stream.d.ts +26 -0
  320. package/out/adapters/sha256-stream.d.ts.map +1 -0
  321. package/out/adapters/sha256-stream.js +34 -0
  322. package/out/adapters/sha256-stream.js.map +1 -0
  323. package/out/adapters/source-id.d.ts +33 -0
  324. package/out/adapters/source-id.d.ts.map +1 -0
  325. package/out/adapters/source-id.js +42 -0
  326. package/out/adapters/source-id.js.map +1 -0
  327. package/out/adapters/street-line.d.ts +102 -0
  328. package/out/adapters/street-line.d.ts.map +1 -0
  329. package/out/adapters/street-line.js +117 -0
  330. package/out/adapters/street-line.js.map +1 -0
  331. package/out/adapters/wof/admin/json/adapter.d.ts +88 -0
  332. package/out/adapters/wof/admin/json/adapter.d.ts.map +1 -0
  333. package/out/adapters/wof/admin/json/adapter.js +220 -0
  334. package/out/adapters/wof/admin/json/adapter.js.map +1 -0
  335. package/out/adapters/wof/json-rows.d.ts +89 -0
  336. package/out/adapters/wof/json-rows.d.ts.map +1 -0
  337. package/out/adapters/wof/json-rows.js +135 -0
  338. package/out/adapters/wof/json-rows.js.map +1 -0
  339. package/out/adapters/wof/postalcode-json/adapter.d.ts +49 -0
  340. package/out/adapters/wof/postalcode-json/adapter.d.ts.map +1 -0
  341. package/out/adapters/wof/postalcode-json/adapter.js +165 -0
  342. package/out/adapters/wof/postalcode-json/adapter.js.map +1 -0
  343. package/out/adapters.d.ts +75 -0
  344. package/out/adapters.d.ts.map +1 -0
  345. package/out/adapters.js +167 -0
  346. package/out/adapters.js.map +1 -0
  347. package/out/au/adapters/gnaf/adapter.d.ts +37 -0
  348. package/out/au/adapters/gnaf/adapter.d.ts.map +1 -0
  349. package/out/au/adapters/gnaf/adapter.js +119 -0
  350. package/out/au/adapters/gnaf/adapter.js.map +1 -0
  351. package/out/au/adapters/gnaf/assemble.d.ts +54 -0
  352. package/out/au/adapters/gnaf/assemble.d.ts.map +1 -0
  353. package/out/au/adapters/gnaf/assemble.js +134 -0
  354. package/out/au/adapters/gnaf/assemble.js.map +1 -0
  355. package/out/be/adapters/brussels/adapter.d.ts +212 -0
  356. package/out/be/adapters/brussels/adapter.d.ts.map +1 -0
  357. package/out/be/adapters/brussels/adapter.js +435 -0
  358. package/out/be/adapters/brussels/adapter.js.map +1 -0
  359. package/out/be/adapters/vlaanderen/adapter.d.ts +233 -0
  360. package/out/be/adapters/vlaanderen/adapter.d.ts.map +1 -0
  361. package/out/be/adapters/vlaanderen/adapter.js +416 -0
  362. package/out/be/adapters/vlaanderen/adapter.js.map +1 -0
  363. package/out/be/adapters/wallonie/adapter.d.ts +196 -0
  364. package/out/be/adapters/wallonie/adapter.d.ts.map +1 -0
  365. package/out/be/adapters/wallonie/adapter.js +437 -0
  366. package/out/be/adapters/wallonie/adapter.js.map +1 -0
  367. package/out/be/tools/fetch/brussels.d.ts +119 -0
  368. package/out/be/tools/fetch/brussels.d.ts.map +1 -0
  369. package/out/be/tools/fetch/brussels.js +146 -0
  370. package/out/be/tools/fetch/brussels.js.map +1 -0
  371. package/out/be/tools/fetch/wallonie.d.ts +119 -0
  372. package/out/be/tools/fetch/wallonie.d.ts.map +1 -0
  373. package/out/be/tools/fetch/wallonie.js +167 -0
  374. package/out/be/tools/fetch/wallonie.js.map +1 -0
  375. package/out/build/adapters.d.ts +33 -0
  376. package/out/build/adapters.d.ts.map +1 -0
  377. package/out/build/adapters.js +68 -0
  378. package/out/build/adapters.js.map +1 -0
  379. package/out/build/align.d.ts +73 -0
  380. package/out/build/align.d.ts.map +1 -0
  381. package/out/build/align.js +226 -0
  382. package/out/build/align.js.map +1 -0
  383. package/out/build/checkpoint.d.ts +94 -0
  384. package/out/build/checkpoint.d.ts.map +1 -0
  385. package/out/build/checkpoint.js +59 -0
  386. package/out/build/checkpoint.js.map +1 -0
  387. package/out/build/eligibility.d.ts +47 -0
  388. package/out/build/eligibility.d.ts.map +1 -0
  389. package/out/build/eligibility.js +80 -0
  390. package/out/build/eligibility.js.map +1 -0
  391. package/out/build/heap.d.ts +28 -0
  392. package/out/build/heap.d.ts.map +1 -0
  393. package/out/build/heap.js +47 -0
  394. package/out/build/heap.js.map +1 -0
  395. package/out/build/inputs.d.ts +52 -0
  396. package/out/build/inputs.d.ts.map +1 -0
  397. package/out/build/inputs.js +49 -0
  398. package/out/build/inputs.js.map +1 -0
  399. package/out/build/rows.d.ts +16 -0
  400. package/out/build/rows.d.ts.map +1 -0
  401. package/out/build/rows.js +19 -0
  402. package/out/build/rows.js.map +1 -0
  403. package/out/build/types.d.ts +220 -0
  404. package/out/build/types.d.ts.map +1 -0
  405. package/out/build/types.js +34 -0
  406. package/out/build/types.js.map +1 -0
  407. package/out/build.d.ts +6 -121
  408. package/out/build.d.ts.map +1 -1
  409. package/out/build.js +112 -207
  410. package/out/build.js.map +1 -1
  411. package/out/cn/recipes/organizational-units.d.ts +19 -0
  412. package/out/cn/recipes/organizational-units.d.ts.map +1 -0
  413. package/out/cn/recipes/organizational-units.js +138 -0
  414. package/out/cn/recipes/organizational-units.js.map +1 -0
  415. package/out/cz/adapters/cuzk/adapter.d.ts +52 -0
  416. package/out/cz/adapters/cuzk/adapter.d.ts.map +1 -0
  417. package/out/cz/adapters/cuzk/adapter.js +266 -0
  418. package/out/cz/adapters/cuzk/adapter.js.map +1 -0
  419. package/out/cz/recipes/pcfirst-preposition.d.ts +18 -0
  420. package/out/cz/recipes/pcfirst-preposition.d.ts.map +1 -0
  421. package/out/cz/recipes/pcfirst-preposition.js +110 -0
  422. package/out/cz/recipes/pcfirst-preposition.js.map +1 -0
  423. package/out/cz/tools/fetch/cuzk.d.ts +272 -0
  424. package/out/cz/tools/fetch/cuzk.d.ts.map +1 -0
  425. package/out/cz/tools/fetch/cuzk.js +446 -0
  426. package/out/cz/tools/fetch/cuzk.js.map +1 -0
  427. package/out/de/recipes/locale.d.ts +42 -0
  428. package/out/de/recipes/locale.d.ts.map +1 -0
  429. package/out/de/recipes/locale.js +299 -0
  430. package/out/de/recipes/locale.js.map +1 -0
  431. package/out/directory.d.ts +20 -0
  432. package/out/directory.d.ts.map +1 -0
  433. package/out/directory.js +27 -0
  434. package/out/directory.js.map +1 -0
  435. package/out/dk/adapters/inspire/adapter.d.ts +154 -0
  436. package/out/dk/adapters/inspire/adapter.d.ts.map +1 -0
  437. package/out/dk/adapters/inspire/adapter.js +258 -0
  438. package/out/dk/adapters/inspire/adapter.js.map +1 -0
  439. package/out/dk/tools/fetch/inspire.d.ts +100 -0
  440. package/out/dk/tools/fetch/inspire.d.ts.map +1 -0
  441. package/out/dk/tools/fetch/inspire.js +230 -0
  442. package/out/dk/tools/fetch/inspire.js.map +1 -0
  443. package/out/ee/adapters/ads/adapter.d.ts +109 -0
  444. package/out/ee/adapters/ads/adapter.d.ts.map +1 -0
  445. package/out/ee/adapters/ads/adapter.js +191 -0
  446. package/out/ee/adapters/ads/adapter.js.map +1 -0
  447. package/out/ee/tools/fetch/ads.d.ts +123 -0
  448. package/out/ee/tools/fetch/ads.d.ts.map +1 -0
  449. package/out/ee/tools/fetch/ads.js +195 -0
  450. package/out/ee/tools/fetch/ads.js.map +1 -0
  451. package/out/env.d.ts +7 -3
  452. package/out/env.d.ts.map +1 -1
  453. package/out/env.js +6 -2
  454. package/out/env.js.map +1 -1
  455. package/out/es/adapters/bizkaia/adapter.d.ts +143 -0
  456. package/out/es/adapters/bizkaia/adapter.d.ts.map +1 -0
  457. package/out/es/adapters/bizkaia/adapter.js +332 -0
  458. package/out/es/adapters/bizkaia/adapter.js.map +1 -0
  459. package/out/es/adapters/catastro/adapter.d.ts +104 -0
  460. package/out/es/adapters/catastro/adapter.d.ts.map +1 -0
  461. package/out/es/adapters/catastro/adapter.js +288 -0
  462. package/out/es/adapters/catastro/adapter.js.map +1 -0
  463. package/out/es/adapters/gipuzkoa/adapter.d.ts +91 -0
  464. package/out/es/adapters/gipuzkoa/adapter.d.ts.map +1 -0
  465. package/out/es/adapters/gipuzkoa/adapter.js +225 -0
  466. package/out/es/adapters/gipuzkoa/adapter.js.map +1 -0
  467. package/out/es/adapters/navarra/adapter.d.ts +105 -0
  468. package/out/es/adapters/navarra/adapter.d.ts.map +1 -0
  469. package/out/es/adapters/navarra/adapter.js +241 -0
  470. package/out/es/adapters/navarra/adapter.js.map +1 -0
  471. package/out/es/cadastre.d.ts +109 -0
  472. package/out/es/cadastre.d.ts.map +1 -0
  473. package/out/es/cadastre.js +130 -0
  474. package/out/es/cadastre.js.map +1 -0
  475. package/out/es/tools/fetch/bizkaia.d.ts +278 -0
  476. package/out/es/tools/fetch/bizkaia.d.ts.map +1 -0
  477. package/out/es/tools/fetch/bizkaia.js +477 -0
  478. package/out/es/tools/fetch/bizkaia.js.map +1 -0
  479. package/out/es/tools/fetch/catastro.d.ts +272 -0
  480. package/out/es/tools/fetch/catastro.d.ts.map +1 -0
  481. package/out/es/tools/fetch/catastro.js +486 -0
  482. package/out/es/tools/fetch/catastro.js.map +1 -0
  483. package/out/es/tools/fetch/gipuzkoa.d.ts +162 -0
  484. package/out/es/tools/fetch/gipuzkoa.d.ts.map +1 -0
  485. package/out/es/tools/fetch/gipuzkoa.js +209 -0
  486. package/out/es/tools/fetch/gipuzkoa.js.map +1 -0
  487. package/out/es/tools/fetch/navarra.d.ts +189 -0
  488. package/out/es/tools/fetch/navarra.d.ts.map +1 -0
  489. package/out/es/tools/fetch/navarra.js +316 -0
  490. package/out/es/tools/fetch/navarra.js.map +1 -0
  491. package/out/fi/adapters/ryhti/adapter.d.ts +115 -0
  492. package/out/fi/adapters/ryhti/adapter.d.ts.map +1 -0
  493. package/out/fi/adapters/ryhti/adapter.js +244 -0
  494. package/out/fi/adapters/ryhti/adapter.js.map +1 -0
  495. package/out/fi/tools/fetch/ryhti.d.ts +206 -0
  496. package/out/fi/tools/fetch/ryhti.d.ts.map +1 -0
  497. package/out/fi/tools/fetch/ryhti.js +327 -0
  498. package/out/fi/tools/fetch/ryhti.js.map +1 -0
  499. package/out/fingerprints.d.ts +68 -0
  500. package/out/fingerprints.d.ts.map +1 -0
  501. package/out/fingerprints.js +130 -0
  502. package/out/fingerprints.js.map +1 -0
  503. package/out/fr/adapters/annuaire-education/adapter.d.ts +236 -0
  504. package/out/fr/adapters/annuaire-education/adapter.d.ts.map +1 -0
  505. package/out/fr/adapters/annuaire-education/adapter.js +475 -0
  506. package/out/fr/adapters/annuaire-education/adapter.js.map +1 -0
  507. package/out/fr/adapters/ban/adapter.d.ts +38 -0
  508. package/out/fr/adapters/ban/adapter.d.ts.map +1 -0
  509. package/out/fr/adapters/ban/adapter.js +134 -0
  510. package/out/fr/adapters/ban/adapter.js.map +1 -0
  511. package/out/fr/adapters/finess/adapter.d.ts +276 -0
  512. package/out/fr/adapters/finess/adapter.d.ts.map +1 -0
  513. package/out/fr/adapters/finess/adapter.js +519 -0
  514. package/out/fr/adapters/finess/adapter.js.map +1 -0
  515. package/out/fr/adapters/finess/postal-lines.d.ts +61 -0
  516. package/out/fr/adapters/finess/postal-lines.d.ts.map +1 -0
  517. package/out/fr/adapters/finess/postal-lines.js +114 -0
  518. package/out/fr/adapters/finess/postal-lines.js.map +1 -0
  519. package/out/fr/adapters/finess/voie-type.d.ts +77 -0
  520. package/out/fr/adapters/finess/voie-type.d.ts.map +1 -0
  521. package/out/fr/adapters/finess/voie-type.js +172 -0
  522. package/out/fr/adapters/finess/voie-type.js.map +1 -0
  523. package/out/fr/insee-country.d.ts +45 -0
  524. package/out/fr/insee-country.d.ts.map +1 -0
  525. package/out/fr/insee-country.js +60 -0
  526. package/out/fr/insee-country.js.map +1 -0
  527. package/out/fr/recipes/admin-split.d.ts +15 -0
  528. package/out/fr/recipes/admin-split.d.ts.map +1 -0
  529. package/out/fr/recipes/admin-split.js +170 -0
  530. package/out/fr/recipes/admin-split.js.map +1 -0
  531. package/out/fr/recipes/bare-street.d.ts +13 -0
  532. package/out/fr/recipes/bare-street.d.ts.map +1 -0
  533. package/out/fr/recipes/bare-street.js +129 -0
  534. package/out/fr/recipes/bare-street.js.map +1 -0
  535. package/out/fr/recipes/fragment.d.ts +18 -0
  536. package/out/fr/recipes/fragment.d.ts.map +1 -0
  537. package/out/fr/recipes/fragment.js +197 -0
  538. package/out/fr/recipes/fragment.js.map +1 -0
  539. package/out/fr/recipes/lieudit.d.ts +30 -0
  540. package/out/fr/recipes/lieudit.d.ts.map +1 -0
  541. package/out/fr/recipes/lieudit.js +231 -0
  542. package/out/fr/recipes/lieudit.js.map +1 -0
  543. package/out/fr/recipes/order.d.ts +15 -0
  544. package/out/fr/recipes/order.d.ts.map +1 -0
  545. package/out/fr/recipes/order.js +171 -0
  546. package/out/fr/recipes/order.js.map +1 -0
  547. package/out/fr/street-decompose.d.ts +29 -0
  548. package/out/fr/street-decompose.d.ts.map +1 -0
  549. package/out/fr/street-decompose.js +52 -0
  550. package/out/fr/street-decompose.js.map +1 -0
  551. package/out/fr/tools/fetch/annuaire-education.d.ts +130 -0
  552. package/out/fr/tools/fetch/annuaire-education.d.ts.map +1 -0
  553. package/out/fr/tools/fetch/annuaire-education.js +196 -0
  554. package/out/fr/tools/fetch/annuaire-education.js.map +1 -0
  555. package/out/fr/tools/fetch/ban.d.ts +33 -0
  556. package/out/fr/tools/fetch/ban.d.ts.map +1 -0
  557. package/out/fr/tools/fetch/ban.js +248 -0
  558. package/out/fr/tools/fetch/ban.js.map +1 -0
  559. package/out/fr/tools/fetch/finess.d.ts +145 -0
  560. package/out/fr/tools/fetch/finess.d.ts.map +1 -0
  561. package/out/fr/tools/fetch/finess.js +228 -0
  562. package/out/fr/tools/fetch/finess.js.map +1 -0
  563. package/out/gauntlet-inputs.d.ts +42 -0
  564. package/out/gauntlet-inputs.d.ts.map +1 -0
  565. package/out/gauntlet-inputs.js +78 -0
  566. package/out/gauntlet-inputs.js.map +1 -0
  567. package/out/index.d.ts +10 -4
  568. package/out/index.d.ts.map +1 -1
  569. package/out/index.js +10 -4
  570. package/out/index.js.map +1 -1
  571. package/out/inspire/address.d.ts +251 -0
  572. package/out/inspire/address.d.ts.map +1 -0
  573. package/out/inspire/address.js +384 -0
  574. package/out/inspire/address.js.map +1 -0
  575. package/out/inspire/archive.d.ts +36 -0
  576. package/out/inspire/archive.d.ts.map +1 -0
  577. package/out/inspire/archive.js +42 -0
  578. package/out/inspire/archive.js.map +1 -0
  579. package/out/inspire/errors.d.ts +55 -0
  580. package/out/inspire/errors.d.ts.map +1 -0
  581. package/out/inspire/errors.js +64 -0
  582. package/out/inspire/errors.js.map +1 -0
  583. package/out/inspire/stream.d.ts +137 -0
  584. package/out/inspire/stream.d.ts.map +1 -0
  585. package/out/inspire/stream.js +125 -0
  586. package/out/inspire/stream.js.map +1 -0
  587. package/out/international/recipes/locale.d.ts +90 -0
  588. package/out/international/recipes/locale.d.ts.map +1 -0
  589. package/out/international/recipes/locale.js +419 -0
  590. package/out/international/recipes/locale.js.map +1 -0
  591. package/out/it/adapters/anac/adapter.d.ts +267 -0
  592. package/out/it/adapters/anac/adapter.d.ts.map +1 -0
  593. package/out/it/adapters/anac/adapter.js +429 -0
  594. package/out/it/adapters/anac/adapter.js.map +1 -0
  595. package/out/it/tools/fetch/anac.d.ts +174 -0
  596. package/out/it/tools/fetch/anac.d.ts.map +1 -0
  597. package/out/it/tools/fetch/anac.js +272 -0
  598. package/out/it/tools/fetch/anac.js.map +1 -0
  599. package/out/jp/adapters/wof-admin/adapter.d.ts +68 -0
  600. package/out/jp/adapters/wof-admin/adapter.d.ts.map +1 -0
  601. package/out/jp/adapters/wof-admin/adapter.js +172 -0
  602. package/out/jp/adapters/wof-admin/adapter.js.map +1 -0
  603. package/out/jp/tools/fetch/houjin.d.ts +27 -0
  604. package/out/jp/tools/fetch/houjin.d.ts.map +1 -0
  605. package/out/jp/tools/fetch/houjin.js +106 -0
  606. package/out/jp/tools/fetch/houjin.js.map +1 -0
  607. package/out/kr/adapters/juso/label-rows.d.ts +62 -0
  608. package/out/kr/adapters/juso/label-rows.d.ts.map +1 -0
  609. package/out/kr/adapters/juso/label-rows.js +172 -0
  610. package/out/kr/adapters/juso/label-rows.js.map +1 -0
  611. package/out/kr/adapters/localdata/align.d.ts +31 -0
  612. package/out/kr/adapters/localdata/align.d.ts.map +1 -0
  613. package/out/kr/adapters/localdata/align.js +212 -0
  614. package/out/kr/adapters/localdata/align.js.map +1 -0
  615. package/out/kr/adapters/localdata/key-index.d.ts +62 -0
  616. package/out/kr/adapters/localdata/key-index.d.ts.map +1 -0
  617. package/out/kr/adapters/localdata/key-index.js +127 -0
  618. package/out/kr/adapters/localdata/key-index.js.map +1 -0
  619. package/out/kr/adapters/localdata/permit-rows.d.ts +73 -0
  620. package/out/kr/adapters/localdata/permit-rows.d.ts.map +1 -0
  621. package/out/kr/adapters/localdata/permit-rows.js +136 -0
  622. package/out/kr/adapters/localdata/permit-rows.js.map +1 -0
  623. package/out/kr/tools/fetch/juso.d.ts +33 -0
  624. package/out/kr/tools/fetch/juso.d.ts.map +1 -0
  625. package/out/kr/tools/fetch/juso.js +121 -0
  626. package/out/kr/tools/fetch/juso.js.map +1 -0
  627. package/out/kr/tools/fetch/localdata.d.ts +37 -0
  628. package/out/kr/tools/fetch/localdata.d.ts.map +1 -0
  629. package/out/kr/tools/fetch/localdata.js +147 -0
  630. package/out/kr/tools/fetch/localdata.js.map +1 -0
  631. package/out/lu/adapters/bd-adresses/adapter.d.ts +109 -0
  632. package/out/lu/adapters/bd-adresses/adapter.d.ts.map +1 -0
  633. package/out/lu/adapters/bd-adresses/adapter.js +196 -0
  634. package/out/lu/adapters/bd-adresses/adapter.js.map +1 -0
  635. package/out/lu/tools/fetch/bd-adresses.d.ts +163 -0
  636. package/out/lu/tools/fetch/bd-adresses.d.ts.map +1 -0
  637. package/out/lu/tools/fetch/bd-adresses.js +223 -0
  638. package/out/lu/tools/fetch/bd-adresses.js.map +1 -0
  639. package/out/lv/adapters/varis/adapter.d.ts +283 -0
  640. package/out/lv/adapters/varis/adapter.d.ts.map +1 -0
  641. package/out/lv/adapters/varis/adapter.js +404 -0
  642. package/out/lv/adapters/varis/adapter.js.map +1 -0
  643. package/out/lv/tools/fetch/varis.d.ts +80 -0
  644. package/out/lv/tools/fetch/varis.d.ts.map +1 -0
  645. package/out/lv/tools/fetch/varis.js +133 -0
  646. package/out/lv/tools/fetch/varis.js.map +1 -0
  647. package/out/nl/adapters/kadaster/adapter.d.ts +125 -0
  648. package/out/nl/adapters/kadaster/adapter.d.ts.map +1 -0
  649. package/out/nl/adapters/kadaster/adapter.js +310 -0
  650. package/out/nl/adapters/kadaster/adapter.js.map +1 -0
  651. package/out/nl/recipes/postcode.d.ts +24 -0
  652. package/out/nl/recipes/postcode.d.ts.map +1 -0
  653. package/out/nl/recipes/postcode.js +121 -0
  654. package/out/nl/recipes/postcode.js.map +1 -0
  655. package/out/nl/tools/fetch/kadaster.d.ts +155 -0
  656. package/out/nl/tools/fetch/kadaster.d.ts.map +1 -0
  657. package/out/nl/tools/fetch/kadaster.js +179 -0
  658. package/out/nl/tools/fetch/kadaster.js.map +1 -0
  659. package/out/no/adapters/matrikkelen/adapter.d.ts +44 -0
  660. package/out/no/adapters/matrikkelen/adapter.d.ts.map +1 -0
  661. package/out/no/adapters/matrikkelen/adapter.js +143 -0
  662. package/out/no/adapters/matrikkelen/adapter.js.map +1 -0
  663. package/out/no/recipes/fragment.d.ts +11 -0
  664. package/out/no/recipes/fragment.d.ts.map +1 -0
  665. package/out/no/recipes/fragment.js +149 -0
  666. package/out/no/recipes/fragment.js.map +1 -0
  667. package/out/no/recipes/street/led.d.ts +21 -0
  668. package/out/no/recipes/street/led.d.ts.map +1 -0
  669. package/out/no/recipes/street/led.js +124 -0
  670. package/out/no/recipes/street/led.js.map +1 -0
  671. package/out/no/recipes/street.d.ts +19 -0
  672. package/out/no/recipes/street.d.ts.map +1 -0
  673. package/out/no/recipes/street.js +88 -0
  674. package/out/no/recipes/street.js.map +1 -0
  675. package/out/no/tools/fetch/matrikkelen.d.ts +251 -0
  676. package/out/no/tools/fetch/matrikkelen.d.ts.map +1 -0
  677. package/out/no/tools/fetch/matrikkelen.js +345 -0
  678. package/out/no/tools/fetch/matrikkelen.js.map +1 -0
  679. package/out/parquet/duckdb.d.ts +100 -0
  680. package/out/parquet/duckdb.d.ts.map +1 -0
  681. package/out/parquet/duckdb.js +92 -0
  682. package/out/parquet/duckdb.js.map +1 -0
  683. package/out/parquet/readers.d.ts +44 -0
  684. package/out/parquet/readers.d.ts.map +1 -0
  685. package/out/parquet/readers.js +69 -0
  686. package/out/parquet/readers.js.map +1 -0
  687. package/out/parquet/schema.d.ts +95 -0
  688. package/out/parquet/schema.d.ts.map +1 -0
  689. package/out/parquet/schema.js +157 -0
  690. package/out/parquet/schema.js.map +1 -0
  691. package/out/parquet/shuffle.d.ts +25 -0
  692. package/out/parquet/shuffle.d.ts.map +1 -0
  693. package/out/parquet/shuffle.js +44 -0
  694. package/out/parquet/shuffle.js.map +1 -0
  695. package/out/parquet/span-sql.d.ts +49 -0
  696. package/out/parquet/span-sql.d.ts.map +1 -0
  697. package/out/parquet/span-sql.js +53 -0
  698. package/out/parquet/span-sql.js.map +1 -0
  699. package/out/parquet/streams.d.ts +47 -0
  700. package/out/parquet/streams.d.ts.map +1 -0
  701. package/out/parquet/streams.js +78 -0
  702. package/out/parquet/streams.js.map +1 -0
  703. package/out/parquet/writers.d.ts +124 -0
  704. package/out/parquet/writers.d.ts.map +1 -0
  705. package/out/parquet/writers.js +251 -0
  706. package/out/parquet/writers.js.map +1 -0
  707. package/out/pl/adapters/emuia/adapter.d.ts +81 -0
  708. package/out/pl/adapters/emuia/adapter.d.ts.map +1 -0
  709. package/out/pl/adapters/emuia/adapter.js +178 -0
  710. package/out/pl/adapters/emuia/adapter.js.map +1 -0
  711. package/out/pl/tools/fetch/emuia.d.ts +165 -0
  712. package/out/pl/tools/fetch/emuia.d.ts.map +1 -0
  713. package/out/pl/tools/fetch/emuia.js +260 -0
  714. package/out/pl/tools/fetch/emuia.js.map +1 -0
  715. package/out/recipes/anchor-absorption.d.ts +9 -9
  716. package/out/recipes/anchor-absorption.d.ts.map +1 -1
  717. package/out/recipes/anchor-absorption.js +29 -19
  718. package/out/recipes/anchor-absorption.js.map +1 -1
  719. package/out/recipes/bare/country.d.ts +7 -0
  720. package/out/recipes/bare/country.d.ts.map +1 -0
  721. package/out/recipes/bare/country.js +77 -0
  722. package/out/recipes/bare/country.js.map +1 -0
  723. package/out/recipes/bare/postcode/eval.d.ts +29 -0
  724. package/out/recipes/bare/postcode/eval.d.ts.map +1 -0
  725. package/out/recipes/bare/postcode/eval.js +85 -0
  726. package/out/recipes/bare/postcode/eval.js.map +1 -0
  727. package/out/recipes/bare/postcode.d.ts +41 -0
  728. package/out/recipes/bare/postcode.d.ts.map +1 -0
  729. package/out/recipes/bare/postcode.js +279 -0
  730. package/out/recipes/bare/postcode.js.map +1 -0
  731. package/out/recipes/boundary-stress.d.ts +8 -8
  732. package/out/recipes/boundary-stress.d.ts.map +1 -1
  733. package/out/recipes/boundary-stress.js +36 -19
  734. package/out/recipes/boundary-stress.js.map +1 -1
  735. package/out/recipes/country-balanced.d.ts +3 -21
  736. package/out/recipes/country-balanced.d.ts.map +1 -1
  737. package/out/recipes/country-balanced.js +58 -76
  738. package/out/recipes/country-balanced.js.map +1 -1
  739. package/out/recipes/house-venue.d.ts +8 -6
  740. package/out/recipes/house-venue.d.ts.map +1 -1
  741. package/out/recipes/house-venue.js +29 -13
  742. package/out/recipes/house-venue.js.map +1 -1
  743. package/out/recipes/intersection.d.ts +20 -20
  744. package/out/recipes/intersection.d.ts.map +1 -1
  745. package/out/recipes/intersection.js +65 -53
  746. package/out/recipes/intersection.js.map +1 -1
  747. package/out/recipes/po/box/cedex/au-nz.d.ts +7 -0
  748. package/out/recipes/po/box/cedex/au-nz.d.ts.map +1 -0
  749. package/out/recipes/po/box/cedex/au-nz.js +56 -0
  750. package/out/recipes/po/box/cedex/au-nz.js.map +1 -0
  751. package/out/recipes/po/box/cedex/ca.d.ts +7 -0
  752. package/out/recipes/po/box/cedex/ca.d.ts.map +1 -0
  753. package/out/recipes/po/box/cedex/ca.js +38 -0
  754. package/out/recipes/po/box/cedex/ca.js.map +1 -0
  755. package/out/recipes/po/box/cedex/fr.d.ts +7 -0
  756. package/out/recipes/po/box/cedex/fr.d.ts.map +1 -0
  757. package/out/recipes/po/box/cedex/fr.js +57 -0
  758. package/out/recipes/po/box/cedex/fr.js.map +1 -0
  759. package/out/recipes/po/box/cedex/phrases.d.ts +8 -0
  760. package/out/recipes/po/box/cedex/phrases.d.ts.map +1 -0
  761. package/out/recipes/po/box/cedex/phrases.js +77 -0
  762. package/out/recipes/po/box/cedex/phrases.js.map +1 -0
  763. package/out/recipes/po/box/cedex/recipe.d.ts +9 -0
  764. package/out/recipes/po/box/cedex/recipe.d.ts.map +1 -0
  765. package/out/recipes/po/box/cedex/recipe.js +163 -0
  766. package/out/recipes/po/box/cedex/recipe.js.map +1 -0
  767. package/out/recipes/po/box/cedex/sources.d.ts +86 -0
  768. package/out/recipes/po/box/cedex/sources.d.ts.map +1 -0
  769. package/out/recipes/po/box/cedex/sources.js +143 -0
  770. package/out/recipes/po/box/cedex/sources.js.map +1 -0
  771. package/out/recipes/po/box/cedex/types.d.ts +31 -0
  772. package/out/recipes/po/box/cedex/types.d.ts.map +1 -0
  773. package/out/recipes/po/box/cedex/types.js +5 -0
  774. package/out/recipes/po/box/cedex/types.js.map +1 -0
  775. package/out/recipes/po/box/cedex/us.d.ts +7 -0
  776. package/out/recipes/po/box/cedex/us.d.ts.map +1 -0
  777. package/out/recipes/po/box/cedex/us.js +54 -0
  778. package/out/recipes/po/box/cedex/us.js.map +1 -0
  779. package/out/recipes/po/box/cedex/vocabulary.d.ts +68 -0
  780. package/out/recipes/po/box/cedex/vocabulary.d.ts.map +1 -0
  781. package/out/recipes/po/box/cedex/vocabulary.js +90 -0
  782. package/out/recipes/po/box/cedex/vocabulary.js.map +1 -0
  783. package/out/recipes/po/box/cedex.d.ts +9 -0
  784. package/out/recipes/po/box/cedex.d.ts.map +1 -0
  785. package/out/recipes/po/box/cedex.js +9 -0
  786. package/out/recipes/po/box/cedex.js.map +1 -0
  787. package/out/recipes/po/box.d.ts +18 -0
  788. package/out/recipes/po/box.d.ts.map +1 -0
  789. package/out/recipes/po/box.js +123 -0
  790. package/out/recipes/po/box.js.map +1 -0
  791. package/out/recipes/scaffold.d.ts +214 -121
  792. package/out/recipes/scaffold.d.ts.map +1 -1
  793. package/out/recipes/scaffold.js +78 -88
  794. package/out/recipes/scaffold.js.map +1 -1
  795. package/out/recipes/sources.d.ts +178 -0
  796. package/out/recipes/sources.d.ts.map +1 -0
  797. package/out/recipes/sources.js +348 -0
  798. package/out/recipes/sources.js.map +1 -0
  799. package/out/recipes/street/affix.d.ts +59 -0
  800. package/out/recipes/street/affix.d.ts.map +1 -0
  801. package/out/recipes/street/affix.js +477 -0
  802. package/out/recipes/street/affix.js.map +1 -0
  803. package/out/recipes/street/bare.d.ts +27 -0
  804. package/out/recipes/street/bare.d.ts.map +1 -0
  805. package/out/recipes/street/bare.js +98 -0
  806. package/out/recipes/street/bare.js.map +1 -0
  807. package/out/recipes/street.d.ts +6 -4
  808. package/out/recipes/street.d.ts.map +1 -1
  809. package/out/recipes/street.js +29 -8
  810. package/out/recipes/street.js.map +1 -1
  811. package/out/recipes/sub/venue/context.d.ts +32 -0
  812. package/out/recipes/sub/venue/context.d.ts.map +1 -0
  813. package/out/recipes/sub/venue/context.js +110 -0
  814. package/out/recipes/sub/venue/context.js.map +1 -0
  815. package/out/recipes/sub/venue/render.d.ts +58 -0
  816. package/out/recipes/sub/venue/render.d.ts.map +1 -0
  817. package/out/recipes/sub/venue/render.js +118 -0
  818. package/out/recipes/sub/venue/render.js.map +1 -0
  819. package/out/recipes/sub/venue/sources.d.ts +166 -0
  820. package/out/recipes/sub/venue/sources.d.ts.map +1 -0
  821. package/out/recipes/sub/venue/sources.js +419 -0
  822. package/out/recipes/sub/venue/sources.js.map +1 -0
  823. package/out/recipes/sub/venue.d.ts +176 -0
  824. package/out/recipes/sub/venue.d.ts.map +1 -0
  825. package/out/recipes/sub/venue.js +562 -0
  826. package/out/recipes/sub/venue.js.map +1 -0
  827. package/out/recipes/trailing-region.d.ts +9 -61
  828. package/out/recipes/trailing-region.d.ts.map +1 -1
  829. package/out/recipes/trailing-region.js +56 -81
  830. package/out/recipes/trailing-region.js.map +1 -1
  831. package/out/recipes/unit/sources.d.ts +28 -0
  832. package/out/recipes/unit/sources.d.ts.map +1 -0
  833. package/out/recipes/unit/sources.js +30 -0
  834. package/out/recipes/unit/sources.js.map +1 -0
  835. package/out/recipes/unit.d.ts +30 -17
  836. package/out/recipes/unit.d.ts.map +1 -1
  837. package/out/recipes/unit.js +90 -79
  838. package/out/recipes/unit.js.map +1 -1
  839. package/out/recipes.d.ts +26 -0
  840. package/out/recipes.d.ts.map +1 -0
  841. package/out/recipes.js +100 -0
  842. package/out/recipes.js.map +1 -0
  843. package/out/registers.d.ts +277 -0
  844. package/out/registers.d.ts.map +1 -0
  845. package/out/registers.js +273 -0
  846. package/out/registers.js.map +1 -0
  847. package/out/ro/adapters/retea-scolara/adapter.d.ts +114 -0
  848. package/out/ro/adapters/retea-scolara/adapter.d.ts.map +1 -0
  849. package/out/ro/adapters/retea-scolara/adapter.js +209 -0
  850. package/out/ro/adapters/retea-scolara/adapter.js.map +1 -0
  851. package/out/ro/tools/fetch/retea-scolara.d.ts +69 -0
  852. package/out/ro/tools/fetch/retea-scolara.d.ts.map +1 -0
  853. package/out/ro/tools/fetch/retea-scolara.js +117 -0
  854. package/out/ro/tools/fetch/retea-scolara.js.map +1 -0
  855. package/out/runner.d.ts +120 -60
  856. package/out/runner.d.ts.map +1 -1
  857. package/out/runner.js +130 -56
  858. package/out/runner.js.map +1 -1
  859. package/out/sg/recipes/register.d.ts +55 -0
  860. package/out/sg/recipes/register.d.ts.map +1 -0
  861. package/out/sg/recipes/register.js +206 -0
  862. package/out/sg/recipes/register.js.map +1 -0
  863. package/out/sg/tools/fetch/acra.d.ts +20 -0
  864. package/out/sg/tools/fetch/acra.d.ts.map +1 -0
  865. package/out/sg/tools/fetch/acra.js +123 -0
  866. package/out/sg/tools/fetch/acra.js.map +1 -0
  867. package/out/si/recipes/bare-village.d.ts +31 -0
  868. package/out/si/recipes/bare-village.d.ts.map +1 -0
  869. package/out/si/recipes/bare-village.js +103 -0
  870. package/out/si/recipes/bare-village.js.map +1 -0
  871. package/out/sk/adapters/inspire/adapter.d.ts +119 -0
  872. package/out/sk/adapters/inspire/adapter.d.ts.map +1 -0
  873. package/out/sk/adapters/inspire/adapter.js +318 -0
  874. package/out/sk/adapters/inspire/adapter.js.map +1 -0
  875. package/out/sk/tools/fetch/inspire.d.ts +195 -0
  876. package/out/sk/tools/fetch/inspire.d.ts.map +1 -0
  877. package/out/sk/tools/fetch/inspire.js +280 -0
  878. package/out/sk/tools/fetch/inspire.js.map +1 -0
  879. package/out/source-register/effective-manifest.d.ts +355 -0
  880. package/out/source-register/effective-manifest.d.ts.map +1 -0
  881. package/out/source-register/effective-manifest.js +355 -0
  882. package/out/source-register/effective-manifest.js.map +1 -0
  883. package/out/source-register/register.d.ts +79 -0
  884. package/out/source-register/register.d.ts.map +1 -0
  885. package/out/source-register/register.js +400 -0
  886. package/out/source-register/register.js.map +1 -0
  887. package/out/source-register/training-manifest.d.ts +111 -0
  888. package/out/source-register/training-manifest.d.ts.map +1 -0
  889. package/out/source-register/training-manifest.js +120 -0
  890. package/out/source-register/training-manifest.js.map +1 -0
  891. package/out/source-register/types.d.ts +539 -0
  892. package/out/source-register/types.d.ts.map +1 -0
  893. package/out/source-register/types.js +263 -0
  894. package/out/source-register/types.js.map +1 -0
  895. package/out/source-register.d.ts +19 -0
  896. package/out/source-register.d.ts.map +1 -0
  897. package/out/source-register.js +16 -0
  898. package/out/source-register.js.map +1 -0
  899. package/out/south-asia/recipes/register.d.ts +54 -0
  900. package/out/south-asia/recipes/register.d.ts.map +1 -0
  901. package/out/south-asia/recipes/register.js +174 -0
  902. package/out/south-asia/recipes/register.js.map +1 -0
  903. package/out/subvenue/harvest.d.ts +117 -0
  904. package/out/subvenue/harvest.d.ts.map +1 -0
  905. package/out/subvenue/harvest.js +172 -0
  906. package/out/subvenue/harvest.js.map +1 -0
  907. package/out/subvenue/promotions.d.ts +65 -0
  908. package/out/subvenue/promotions.d.ts.map +1 -0
  909. package/out/subvenue/promotions.js +226 -0
  910. package/out/subvenue/promotions.js.map +1 -0
  911. package/out/subvenue/surfaces.d.ts +39 -0
  912. package/out/subvenue/surfaces.d.ts.map +1 -0
  913. package/out/subvenue/surfaces.js +75 -0
  914. package/out/subvenue/surfaces.js.map +1 -0
  915. package/out/subvenue/table.d.ts +212 -0
  916. package/out/subvenue/table.d.ts.map +1 -0
  917. package/out/subvenue/table.js +94 -0
  918. package/out/subvenue/table.js.map +1 -0
  919. package/out/surfaces/locale.d.ts +113 -0
  920. package/out/surfaces/locale.d.ts.map +1 -0
  921. package/out/surfaces/locale.js +136 -0
  922. package/out/surfaces/locale.js.map +1 -0
  923. package/out/synthesizers/anchor-absorption.d.ts +22 -31
  924. package/out/synthesizers/anchor-absorption.d.ts.map +1 -1
  925. package/out/synthesizers/anchor-absorption.js +30 -85
  926. package/out/synthesizers/anchor-absorption.js.map +1 -1
  927. package/out/synthesizers/boundary-stress.d.ts +17 -48
  928. package/out/synthesizers/boundary-stress.d.ts.map +1 -1
  929. package/out/synthesizers/boundary-stress.js +27 -121
  930. package/out/synthesizers/boundary-stress.js.map +1 -1
  931. package/out/synthesizers/house-venue.d.ts +22 -25
  932. package/out/synthesizers/house-venue.d.ts.map +1 -1
  933. package/out/synthesizers/house-venue.js +28 -97
  934. package/out/synthesizers/house-venue.js.map +1 -1
  935. package/out/synthesizers/intersection.d.ts +10 -8
  936. package/out/synthesizers/intersection.d.ts.map +1 -1
  937. package/out/synthesizers/intersection.js +33 -26
  938. package/out/synthesizers/intersection.js.map +1 -1
  939. package/out/synthesizers/no-street.d.ts +7 -37
  940. package/out/synthesizers/no-street.d.ts.map +1 -1
  941. package/out/synthesizers/no-street.js +55 -111
  942. package/out/synthesizers/no-street.js.map +1 -1
  943. package/out/synthesizers/po-box.d.ts +46 -45
  944. package/out/synthesizers/po-box.d.ts.map +1 -1
  945. package/out/synthesizers/po-box.js +57 -97
  946. package/out/synthesizers/po-box.js.map +1 -1
  947. package/out/synthesizers/street.d.ts +18 -12
  948. package/out/synthesizers/street.d.ts.map +1 -1
  949. package/out/synthesizers/street.js +37 -30
  950. package/out/synthesizers/street.js.map +1 -1
  951. package/out/synthesizers/utils.d.ts +32 -129
  952. package/out/synthesizers/utils.d.ts.map +1 -1
  953. package/out/synthesizers/utils.js +54 -166
  954. package/out/synthesizers/utils.js.map +1 -1
  955. package/out/test-kit/corpus-recipe.d.ts +29 -23
  956. package/out/test-kit/corpus-recipe.d.ts.map +1 -1
  957. package/out/test-kit/corpus-recipe.js +28 -13
  958. package/out/test-kit/corpus-recipe.js.map +1 -1
  959. package/out/test-kit.d.ts +63 -0
  960. package/out/test-kit.d.ts.map +1 -0
  961. package/out/test-kit.js +88 -0
  962. package/out/test-kit.js.map +1 -0
  963. package/out/tools/align-canonical.d.ts +7 -0
  964. package/out/tools/align-canonical.d.ts.map +1 -0
  965. package/out/tools/align-canonical.js +52 -0
  966. package/out/tools/align-canonical.js.map +1 -0
  967. package/out/tools/audit.d.ts +18 -11
  968. package/out/tools/audit.d.ts.map +1 -1
  969. package/out/tools/audit.js +87 -95
  970. package/out/tools/audit.js.map +1 -1
  971. package/out/tools/corpus-stats.d.ts +24 -15
  972. package/out/tools/corpus-stats.d.ts.map +1 -1
  973. package/out/tools/corpus-stats.js +35 -31
  974. package/out/tools/corpus-stats.js.map +1 -1
  975. package/out/tools/fetch/atom.d.ts +115 -0
  976. package/out/tools/fetch/atom.d.ts.map +1 -0
  977. package/out/tools/fetch/atom.js +84 -0
  978. package/out/tools/fetch/atom.js.map +1 -0
  979. package/out/tools/fetch/cordis.d.ts +123 -0
  980. package/out/tools/fetch/cordis.d.ts.map +1 -0
  981. package/out/tools/fetch/cordis.js +208 -0
  982. package/out/tools/fetch/cordis.js.map +1 -0
  983. package/out/tools/fetch/download/manifest.d.ts +56 -0
  984. package/out/tools/fetch/download/manifest.d.ts.map +1 -0
  985. package/out/tools/fetch/download/manifest.js +81 -0
  986. package/out/tools/fetch/download/manifest.js.map +1 -0
  987. package/out/tools/fetch/download/network.d.ts +126 -0
  988. package/out/tools/fetch/download/network.d.ts.map +1 -0
  989. package/out/tools/fetch/download/network.js +140 -0
  990. package/out/tools/fetch/download/network.js.map +1 -0
  991. package/out/tools/fetch/download/retry.d.ts +46 -0
  992. package/out/tools/fetch/download/retry.d.ts.map +1 -0
  993. package/out/tools/fetch/download/retry.js +136 -0
  994. package/out/tools/fetch/download/retry.js.map +1 -0
  995. package/out/tools/fetch/download.d.ts +4 -176
  996. package/out/tools/fetch/download.d.ts.map +1 -1
  997. package/out/tools/fetch/download.js +4 -312
  998. package/out/tools/fetch/download.js.map +1 -1
  999. package/out/tools/fetch/geonames/dump.d.ts +109 -0
  1000. package/out/tools/fetch/geonames/dump.d.ts.map +1 -0
  1001. package/out/tools/fetch/geonames/dump.js +181 -0
  1002. package/out/tools/fetch/geonames/dump.js.map +1 -0
  1003. package/out/tools/fetch/geonames/postal.d.ts +41 -0
  1004. package/out/tools/fetch/geonames/postal.d.ts.map +1 -0
  1005. package/out/tools/fetch/geonames/postal.js +87 -0
  1006. package/out/tools/fetch/geonames/postal.js.map +1 -0
  1007. package/out/tools/fetch/gleif.d.ts +137 -0
  1008. package/out/tools/fetch/gleif.d.ts.map +1 -0
  1009. package/out/tools/fetch/gleif.js +220 -0
  1010. package/out/tools/fetch/gleif.js.map +1 -0
  1011. package/out/tools/fetch/header.d.ts +41 -0
  1012. package/out/tools/fetch/header.d.ts.map +1 -0
  1013. package/out/tools/fetch/header.js +59 -0
  1014. package/out/tools/fetch/header.js.map +1 -0
  1015. package/out/tools/fetch/inspire-addresses.d.ts +249 -0
  1016. package/out/tools/fetch/inspire-addresses.d.ts.map +1 -0
  1017. package/out/tools/fetch/inspire-addresses.js +276 -0
  1018. package/out/tools/fetch/inspire-addresses.js.map +1 -0
  1019. package/out/tools/fetch/openaddresses.d.ts +0 -3
  1020. package/out/tools/fetch/openaddresses.d.ts.map +1 -1
  1021. package/out/tools/fetch/openaddresses.js +18 -97
  1022. package/out/tools/fetch/openaddresses.js.map +1 -1
  1023. package/out/tools/fetch/ourairports.d.ts +5 -32
  1024. package/out/tools/fetch/ourairports.d.ts.map +1 -1
  1025. package/out/tools/fetch/ourairports.js +13 -46
  1026. package/out/tools/fetch/ourairports.js.map +1 -1
  1027. package/out/tools/fetch/ppd.d.ts +4 -3
  1028. package/out/tools/fetch/ppd.d.ts.map +1 -1
  1029. package/out/tools/fetch/ppd.js +10 -9
  1030. package/out/tools/fetch/ppd.js.map +1 -1
  1031. package/out/tools/fetch/vlaanderen-ad.d.ts +131 -0
  1032. package/out/tools/fetch/vlaanderen-ad.d.ts.map +1 -0
  1033. package/out/tools/fetch/vlaanderen-ad.js +293 -0
  1034. package/out/tools/fetch/vlaanderen-ad.js.map +1 -0
  1035. package/out/tools/fetch/wfs-harvest.d.ts +410 -0
  1036. package/out/tools/fetch/wfs-harvest.d.ts.map +1 -0
  1037. package/out/tools/fetch/wfs-harvest.js +522 -0
  1038. package/out/tools/fetch/wfs-harvest.js.map +1 -0
  1039. package/out/tools/fetch/wikidata-subvenue.d.ts +31 -99
  1040. package/out/tools/fetch/wikidata-subvenue.d.ts.map +1 -1
  1041. package/out/tools/fetch/wikidata-subvenue.js +26 -121
  1042. package/out/tools/fetch/wikidata-subvenue.js.map +1 -1
  1043. package/out/tools/fetch/zip-archive.d.ts +54 -0
  1044. package/out/tools/fetch/zip-archive.d.ts.map +1 -0
  1045. package/out/tools/fetch/zip-archive.js +58 -0
  1046. package/out/tools/fetch/zip-archive.js.map +1 -0
  1047. package/out/tools/fetch.d.ts +140 -0
  1048. package/out/tools/fetch.d.ts.map +1 -0
  1049. package/out/tools/fetch.js +136 -0
  1050. package/out/tools/fetch.js.map +1 -0
  1051. package/out/tools/golden/expand.d.ts +105 -0
  1052. package/out/tools/golden/expand.d.ts.map +1 -0
  1053. package/out/tools/golden/expand.js +420 -0
  1054. package/out/tools/golden/expand.js.map +1 -0
  1055. package/out/tools/golden/promote.d.ts +77 -0
  1056. package/out/tools/golden/promote.d.ts.map +1 -0
  1057. package/out/tools/golden/promote.js +248 -0
  1058. package/out/tools/golden/promote.js.map +1 -0
  1059. package/out/tools/golden/relabel-street/directory.d.ts +98 -0
  1060. package/out/tools/golden/relabel-street/directory.d.ts.map +1 -0
  1061. package/out/tools/golden/relabel-street/directory.js +227 -0
  1062. package/out/tools/golden/relabel-street/directory.js.map +1 -0
  1063. package/out/tools/golden/relabel-street/row.d.ts +69 -0
  1064. package/out/tools/golden/relabel-street/row.d.ts.map +1 -0
  1065. package/out/tools/golden/relabel-street/row.js +174 -0
  1066. package/out/tools/golden/relabel-street/row.js.map +1 -0
  1067. package/out/tools/golden/relabel-street.d.ts +9 -0
  1068. package/out/tools/golden/relabel-street.d.ts.map +1 -0
  1069. package/out/tools/golden/relabel-street.js +9 -0
  1070. package/out/tools/golden/relabel-street.js.map +1 -0
  1071. package/out/tools/ingest-csv.d.ts +19 -15
  1072. package/out/tools/ingest-csv.d.ts.map +1 -1
  1073. package/out/tools/ingest-csv.js +39 -45
  1074. package/out/tools/ingest-csv.js.map +1 -1
  1075. package/out/tools/jsonl-to-parquet.d.ts +15 -28
  1076. package/out/tools/jsonl-to-parquet.d.ts.map +1 -1
  1077. package/out/tools/jsonl-to-parquet.js +49 -91
  1078. package/out/tools/jsonl-to-parquet.js.map +1 -1
  1079. package/out/tools/lint/recipe-output/vocab.d.ts +77 -0
  1080. package/out/tools/lint/recipe-output/vocab.d.ts.map +1 -0
  1081. package/out/tools/lint/recipe-output/vocab.js +239 -0
  1082. package/out/tools/lint/recipe-output/vocab.js.map +1 -0
  1083. package/out/tools/lint/recipe-output.d.ts +73 -0
  1084. package/out/tools/lint/recipe-output.d.ts.map +1 -0
  1085. package/out/tools/lint/recipe-output.js +302 -0
  1086. package/out/tools/lint/recipe-output.js.map +1 -0
  1087. package/out/tools/migrate/overlay-parquet.d.ts +30 -0
  1088. package/out/tools/migrate/overlay-parquet.d.ts.map +1 -0
  1089. package/out/tools/migrate/overlay-parquet.js +141 -0
  1090. package/out/tools/migrate/overlay-parquet.js.map +1 -0
  1091. package/out/tools/migrate/recipe-output.d.ts +35 -0
  1092. package/out/tools/migrate/recipe-output.d.ts.map +1 -0
  1093. package/out/tools/migrate/recipe-output.js +124 -0
  1094. package/out/tools/migrate/recipe-output.js.map +1 -0
  1095. package/out/tools/migrate/source-names.d.ts +52 -0
  1096. package/out/tools/migrate/source-names.d.ts.map +1 -0
  1097. package/out/tools/migrate/source-names.js +105 -0
  1098. package/out/tools/migrate/source-names.js.map +1 -0
  1099. package/out/tools/overlay/kryptonite.d.ts +33 -0
  1100. package/out/tools/overlay/kryptonite.d.ts.map +1 -0
  1101. package/out/tools/overlay/kryptonite.js +108 -0
  1102. package/out/tools/overlay/kryptonite.js.map +1 -0
  1103. package/out/tools/overlay/manifest.d.ts +142 -0
  1104. package/out/tools/overlay/manifest.d.ts.map +1 -0
  1105. package/out/tools/overlay/manifest.js +260 -0
  1106. package/out/tools/overlay/manifest.js.map +1 -0
  1107. package/out/tools/overlay/merge-source.d.ts +73 -0
  1108. package/out/tools/overlay/merge-source.d.ts.map +1 -0
  1109. package/out/tools/overlay/merge-source.js +63 -0
  1110. package/out/tools/overlay/merge-source.js.map +1 -0
  1111. package/out/tools/overlay/split-slice.d.ts +66 -0
  1112. package/out/tools/overlay/split-slice.d.ts.map +1 -0
  1113. package/out/tools/overlay/split-slice.js +78 -0
  1114. package/out/tools/overlay/split-slice.js.map +1 -0
  1115. package/out/tools/overlay/translit.d.ts +54 -0
  1116. package/out/tools/overlay/translit.d.ts.map +1 -0
  1117. package/out/tools/overlay/translit.js +177 -0
  1118. package/out/tools/overlay/translit.js.map +1 -0
  1119. package/out/tools/overture-subvenue.d.ts +17 -76
  1120. package/out/tools/overture-subvenue.d.ts.map +1 -1
  1121. package/out/tools/overture-subvenue.js +7 -74
  1122. package/out/tools/overture-subvenue.js.map +1 -1
  1123. package/out/tools/postcode-triples.d.ts +96 -109
  1124. package/out/tools/postcode-triples.d.ts.map +1 -1
  1125. package/out/tools/postcode-triples.js +237 -152
  1126. package/out/tools/postcode-triples.js.map +1 -1
  1127. package/out/tools/source-register/build.d.ts +145 -0
  1128. package/out/tools/source-register/build.d.ts.map +1 -0
  1129. package/out/tools/source-register/build.js +515 -0
  1130. package/out/tools/source-register/build.js.map +1 -0
  1131. package/out/tools/sub/venue/head-nouns.d.ts +21 -0
  1132. package/out/tools/sub/venue/head-nouns.d.ts.map +1 -0
  1133. package/out/tools/sub/venue/head-nouns.js +176 -0
  1134. package/out/tools/sub/venue/head-nouns.js.map +1 -0
  1135. package/out/tools/sub/venue/lexicon.d.ts +129 -0
  1136. package/out/tools/sub/venue/lexicon.d.ts.map +1 -0
  1137. package/out/tools/sub/venue/lexicon.js +239 -0
  1138. package/out/tools/sub/venue/lexicon.js.map +1 -0
  1139. package/out/tools/sub/venue/wikidata.d.ts +28 -0
  1140. package/out/tools/sub/venue/wikidata.d.ts.map +1 -0
  1141. package/out/tools/sub/venue/wikidata.js +65 -0
  1142. package/out/tools/sub/venue/wikidata.js.map +1 -0
  1143. package/out/tools.d.ts +31 -0
  1144. package/out/tools.d.ts.map +1 -0
  1145. package/out/tools.js +31 -0
  1146. package/out/tools.js.map +1 -0
  1147. package/out/tw/tools/fetch/gcis.d.ts +29 -0
  1148. package/out/tw/tools/fetch/gcis.d.ts.map +1 -0
  1149. package/out/tw/tools/fetch/gcis.js +141 -0
  1150. package/out/tw/tools/fetch/gcis.js.map +1 -0
  1151. package/out/types.d.ts +270 -101
  1152. package/out/types.d.ts.map +1 -1
  1153. package/out/types.js +133 -9
  1154. package/out/types.js.map +1 -1
  1155. package/out/us/adapters/fcc-bdc/adapter.d.ts +50 -0
  1156. package/out/us/adapters/fcc-bdc/adapter.d.ts.map +1 -0
  1157. package/out/us/adapters/fcc-bdc/adapter.js +129 -0
  1158. package/out/us/adapters/fcc-bdc/adapter.js.map +1 -0
  1159. package/out/us/adapters/state/hi-schools/adapter.d.ts +47 -0
  1160. package/out/us/adapters/state/hi-schools/adapter.d.ts.map +1 -0
  1161. package/out/us/adapters/state/hi-schools/adapter.js +146 -0
  1162. package/out/us/adapters/state/hi-schools/adapter.js.map +1 -0
  1163. package/out/us/adapters/state/hi-schools/workbook.d.ts +34 -0
  1164. package/out/us/adapters/state/hi-schools/workbook.d.ts.map +1 -0
  1165. package/out/us/adapters/state/hi-schools/workbook.js +22 -0
  1166. package/out/us/adapters/state/hi-schools/workbook.js.map +1 -0
  1167. package/out/us/adapters/state/ia-contractors/adapter.d.ts +41 -0
  1168. package/out/us/adapters/state/ia-contractors/adapter.d.ts.map +1 -0
  1169. package/out/us/adapters/state/ia-contractors/adapter.js +117 -0
  1170. package/out/us/adapters/state/ia-contractors/adapter.js.map +1 -0
  1171. package/out/us/adapters/state/ny-notaries/adapter.d.ts +33 -0
  1172. package/out/us/adapters/state/ny-notaries/adapter.d.ts.map +1 -0
  1173. package/out/us/adapters/state/ny-notaries/adapter.js +114 -0
  1174. package/out/us/adapters/state/ny-notaries/adapter.js.map +1 -0
  1175. package/out/us/adapters/state/tx-notaries/adapter.d.ts +34 -0
  1176. package/out/us/adapters/state/tx-notaries/adapter.d.ts.map +1 -0
  1177. package/out/us/adapters/state/tx-notaries/adapter.js +125 -0
  1178. package/out/us/adapters/state/tx-notaries/adapter.js.map +1 -0
  1179. package/out/us/adapters/tiger/adapter.d.ts +31 -0
  1180. package/out/us/adapters/tiger/adapter.d.ts.map +1 -0
  1181. package/out/us/adapters/tiger/adapter.js +173 -0
  1182. package/out/us/adapters/tiger/adapter.js.map +1 -0
  1183. package/out/us/adapters/tiger/street-decompose.d.ts +23 -0
  1184. package/out/us/adapters/tiger/street-decompose.d.ts.map +1 -0
  1185. package/out/us/adapters/tiger/street-decompose.js +58 -0
  1186. package/out/us/adapters/tiger/street-decompose.js.map +1 -0
  1187. package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts +62 -0
  1188. package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts.map +1 -0
  1189. package/out/us/adapters/usgov/hrsa-fqhc/adapter.js +133 -0
  1190. package/out/us/adapters/usgov/hrsa-fqhc/adapter.js.map +1 -0
  1191. package/out/us/adapters/usgov/imls-pls/adapter.d.ts +37 -0
  1192. package/out/us/adapters/usgov/imls-pls/adapter.d.ts.map +1 -0
  1193. package/out/us/adapters/usgov/imls-pls/adapter.js +113 -0
  1194. package/out/us/adapters/usgov/imls-pls/adapter.js.map +1 -0
  1195. package/out/us/adapters/usgov/irs-bmf/adapter.d.ts +38 -0
  1196. package/out/us/adapters/usgov/irs-bmf/adapter.d.ts.map +1 -0
  1197. package/out/us/adapters/usgov/irs-bmf/adapter.js +126 -0
  1198. package/out/us/adapters/usgov/irs-bmf/adapter.js.map +1 -0
  1199. package/out/us/adapters/usgov/nad/adapter.d.ts +28 -0
  1200. package/out/us/adapters/usgov/nad/adapter.d.ts.map +1 -0
  1201. package/out/us/adapters/usgov/nad/adapter.js +185 -0
  1202. package/out/us/adapters/usgov/nad/adapter.js.map +1 -0
  1203. package/out/us/adapters/usgov/nppes/adapter.d.ts +29 -0
  1204. package/out/us/adapters/usgov/nppes/adapter.d.ts.map +1 -0
  1205. package/out/us/adapters/usgov/nppes/adapter.js +140 -0
  1206. package/out/us/adapters/usgov/nppes/adapter.js.map +1 -0
  1207. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts +47 -0
  1208. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts.map +1 -0
  1209. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js +153 -0
  1210. package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js.map +1 -0
  1211. package/out/us/fips-state.d.ts +50 -0
  1212. package/out/us/fips-state.d.ts.map +1 -0
  1213. package/out/us/fips-state.js +109 -0
  1214. package/out/us/fips-state.js.map +1 -0
  1215. package/out/us/tools/fetch/hrsa.d.ts +18 -0
  1216. package/out/us/tools/fetch/hrsa.d.ts.map +1 -0
  1217. package/out/us/tools/fetch/hrsa.js +45 -0
  1218. package/out/us/tools/fetch/hrsa.js.map +1 -0
  1219. package/out/us/tools/fetch/imls-pls.d.ts +14 -0
  1220. package/out/us/tools/fetch/imls-pls.d.ts.map +1 -0
  1221. package/out/us/tools/fetch/imls-pls.js +91 -0
  1222. package/out/us/tools/fetch/imls-pls.js.map +1 -0
  1223. package/out/us/tools/fetch/nad.d.ts +82 -0
  1224. package/out/us/tools/fetch/nad.d.ts.map +1 -0
  1225. package/out/us/tools/fetch/nad.js +239 -0
  1226. package/out/us/tools/fetch/nad.js.map +1 -0
  1227. package/out/us/tools/fetch/nppes.d.ts +16 -0
  1228. package/out/us/tools/fetch/nppes.d.ts.map +1 -0
  1229. package/out/us/tools/fetch/nppes.js +114 -0
  1230. package/out/us/tools/fetch/nppes.js.map +1 -0
  1231. package/out/us/tools/fetch/state/hi-schools.d.ts +29 -0
  1232. package/out/us/tools/fetch/state/hi-schools.d.ts.map +1 -0
  1233. package/out/us/tools/fetch/state/hi-schools.js +114 -0
  1234. package/out/us/tools/fetch/state/hi-schools.js.map +1 -0
  1235. package/out/us/tools/fetch/state/sources.d.ts +27 -0
  1236. package/out/us/tools/fetch/state/sources.d.ts.map +1 -0
  1237. package/out/us/tools/fetch/state/sources.js +110 -0
  1238. package/out/us/tools/fetch/state/sources.js.map +1 -0
  1239. package/out/us/tools/fetch/tiger-full.d.ts +9 -0
  1240. package/out/us/tools/fetch/tiger-full.d.ts.map +1 -0
  1241. package/out/us/tools/fetch/tiger-full.js +178 -0
  1242. package/out/us/tools/fetch/tiger-full.js.map +1 -0
  1243. package/out/utils/align.d.ts +25 -20
  1244. package/out/utils/align.d.ts.map +1 -1
  1245. package/out/utils/align.js +27 -42
  1246. package/out/utils/align.js.map +1 -1
  1247. package/out/utils/cooccurrence-stats.d.ts +58 -0
  1248. package/out/utils/cooccurrence-stats.d.ts.map +1 -0
  1249. package/out/utils/cooccurrence-stats.js +62 -0
  1250. package/out/utils/cooccurrence-stats.js.map +1 -0
  1251. package/out/utils/golden.d.ts +23 -29
  1252. package/out/utils/golden.d.ts.map +1 -1
  1253. package/out/utils/golden.js +33 -45
  1254. package/out/utils/golden.js.map +1 -1
  1255. package/out/utils/license.d.ts +158 -21
  1256. package/out/utils/license.d.ts.map +1 -1
  1257. package/out/utils/license.js +156 -26
  1258. package/out/utils/license.js.map +1 -1
  1259. package/out/utils/split.d.ts +66 -70
  1260. package/out/utils/split.d.ts.map +1 -1
  1261. package/out/utils/split.js +158 -69
  1262. package/out/utils/split.js.map +1 -1
  1263. package/out/utils/tokenize.d.ts +16 -28
  1264. package/out/utils/tokenize.d.ts.map +1 -1
  1265. package/out/utils/tokenize.js +18 -31
  1266. package/out/utils/tokenize.js.map +1 -1
  1267. package/out/utils/wof-json.d.ts +88 -62
  1268. package/out/utils/wof-json.d.ts.map +1 -1
  1269. package/out/utils/wof-json.js +81 -65
  1270. package/out/utils/wof-json.js.map +1 -1
  1271. package/out/utils.d.ts +13 -0
  1272. package/out/utils.d.ts.map +1 -0
  1273. package/out/utils.js +13 -0
  1274. package/out/utils.js.map +1 -0
  1275. package/out/ve/recipes/reviewed-postcode-tail.d.ts +65 -0
  1276. package/out/ve/recipes/reviewed-postcode-tail.d.ts.map +1 -0
  1277. package/out/ve/recipes/reviewed-postcode-tail.js +146 -0
  1278. package/out/ve/recipes/reviewed-postcode-tail.js.map +1 -0
  1279. package/package.json +46 -330
  1280. package/lib/adapters/ban/adapter.ts +0 -190
  1281. package/lib/adapters/ban/street-decompose.ts +0 -66
  1282. package/lib/adapters/fcc-bdc/adapter.ts +0 -174
  1283. package/lib/adapters/geonames-postal/adapter.ts +0 -122
  1284. package/lib/adapters/gnaf/adapter.ts +0 -149
  1285. package/lib/adapters/gnaf/assemble.ts +0 -225
  1286. package/lib/adapters/index.ts +0 -153
  1287. package/lib/adapters/state-hi-schools/adapter.ts +0 -170
  1288. package/lib/adapters/state-hi-schools/workbook.ts +0 -35
  1289. package/lib/adapters/state-ia-contractors/adapter.ts +0 -144
  1290. package/lib/adapters/state-ny-notaries/adapter.ts +0 -141
  1291. package/lib/adapters/state-tx-notaries/adapter.ts +0 -157
  1292. package/lib/adapters/synth-po-box/adapter.ts +0 -204
  1293. package/lib/adapters/tiger/adapter.ts +0 -252
  1294. package/lib/adapters/tiger/street-decompose.ts +0 -83
  1295. package/lib/adapters/usgov-hrsa-fqhc/adapter.ts +0 -167
  1296. package/lib/adapters/usgov-imls-pls/adapter.ts +0 -143
  1297. package/lib/adapters/usgov-irs-bmf/adapter.ts +0 -151
  1298. package/lib/adapters/usgov-nad/adapter.ts +0 -334
  1299. package/lib/adapters/usgov-nppes/adapter.ts +0 -151
  1300. package/lib/adapters/usgov-samhsa-treatment-locator/adapter.ts +0 -193
  1301. package/lib/adapters/utils/index.ts +0 -307
  1302. package/lib/adapters/wof-admin-jp/adapter.ts +0 -214
  1303. package/lib/adapters/wof-admin-json/adapter.ts +0 -231
  1304. package/lib/adapters/wof-json-rows.ts +0 -129
  1305. package/lib/adapters/wof-postalcode-json/adapter.ts +0 -176
  1306. package/lib/codex/us-fips-state.ts +0 -113
  1307. package/lib/recipes/bare-country.ts +0 -95
  1308. package/lib/recipes/cn-organizational-units.ts +0 -199
  1309. package/lib/recipes/cz-pcfirst-preposition.ts +0 -112
  1310. package/lib/recipes/fr-admin-split.ts +0 -243
  1311. package/lib/recipes/fr-bare-street.ts +0 -171
  1312. package/lib/recipes/fr-fragment.ts +0 -334
  1313. package/lib/recipes/fr-lieudit.ts +0 -301
  1314. package/lib/recipes/fr-order.ts +0 -246
  1315. package/lib/recipes/german.ts +0 -360
  1316. package/lib/recipes/index.ts +0 -97
  1317. package/lib/recipes/locale.ts +0 -600
  1318. package/lib/recipes/nl-postcode.ts +0 -117
  1319. package/lib/recipes/no-fragment.ts +0 -213
  1320. package/lib/recipes/no-street-led.ts +0 -135
  1321. package/lib/recipes/no-street.ts +0 -91
  1322. package/lib/recipes/po-box-cedex.ts +0 -976
  1323. package/lib/recipes/po-box.ts +0 -124
  1324. package/lib/recipes/reviewed-postcode-tail.ts +0 -190
  1325. package/lib/recipes/si-bare-village.ts +0 -99
  1326. package/lib/recipes/street-affix.ts +0 -805
  1327. package/lib/recipes/street-bare.ts +0 -102
  1328. package/lib/recipes/sub-venue-sources.ts +0 -878
  1329. package/lib/recipes/sub-venue.ts +0 -998
  1330. package/lib/synthesizers/german.ts +0 -260
  1331. package/lib/test-kit/index.ts +0 -100
  1332. package/lib/tools/align-slice.ts +0 -68
  1333. package/lib/tools/fetch/acra-sg.ts +0 -174
  1334. package/lib/tools/fetch/ban.ts +0 -289
  1335. package/lib/tools/fetch/gcis-tw.ts +0 -175
  1336. package/lib/tools/fetch/geonames-dump.ts +0 -257
  1337. package/lib/tools/fetch/geonames-postal.ts +0 -180
  1338. package/lib/tools/fetch/houjin-jp.ts +0 -135
  1339. package/lib/tools/fetch/hrsa.ts +0 -59
  1340. package/lib/tools/fetch/imls-pls.ts +0 -149
  1341. package/lib/tools/fetch/index.ts +0 -174
  1342. package/lib/tools/fetch/juso-kr.ts +0 -165
  1343. package/lib/tools/fetch/localdata-kr.ts +0 -194
  1344. package/lib/tools/fetch/nad.ts +0 -341
  1345. package/lib/tools/fetch/nppes.ts +0 -165
  1346. package/lib/tools/fetch/state-hi-schools.ts +0 -161
  1347. package/lib/tools/fetch/state-sources.ts +0 -146
  1348. package/lib/tools/fetch/tiger-full.ts +0 -308
  1349. package/lib/tools/gb-title-case.ts +0 -69
  1350. package/lib/tools/golden-expand.ts +0 -574
  1351. package/lib/tools/golden-promote.ts +0 -337
  1352. package/lib/tools/golden-relabel-street.ts +0 -686
  1353. package/lib/tools/index.ts +0 -28
  1354. package/lib/tools/lint-slice-vocab.ts +0 -388
  1355. package/lib/tools/lint-slice.ts +0 -496
  1356. package/lib/tools/overlay-manifest.ts +0 -148
  1357. package/lib/tools/slice-kryptonite.ts +0 -137
  1358. package/lib/tools/slice-translit.ts +0 -218
  1359. package/lib/tools/sub-venue/harvest.ts +0 -234
  1360. package/lib/tools/sub-venue/head-nouns.ts +0 -245
  1361. package/lib/tools/sub-venue/surfaces.ts +0 -95
  1362. package/lib/tools/sub-venue/table.ts +0 -281
  1363. package/lib/tools/sub-venue/wikidata.ts +0 -85
  1364. package/lib/tools/sub-venue-lexicon.ts +0 -429
  1365. package/lib/tools/sub-venue-promotions.ts +0 -330
  1366. package/lib/utils/index.ts +0 -14
  1367. package/lib/utils/parquet.ts +0 -582
  1368. package/lib/utils/slice-stats.ts +0 -95
  1369. package/out/adapters/ban/adapter.d.ts +0 -44
  1370. package/out/adapters/ban/adapter.d.ts.map +0 -1
  1371. package/out/adapters/ban/adapter.js +0 -149
  1372. package/out/adapters/ban/adapter.js.map +0 -1
  1373. package/out/adapters/ban/street-decompose.d.ts +0 -28
  1374. package/out/adapters/ban/street-decompose.d.ts.map +0 -1
  1375. package/out/adapters/ban/street-decompose.js +0 -51
  1376. package/out/adapters/ban/street-decompose.js.map +0 -1
  1377. package/out/adapters/fcc-bdc/adapter.d.ts +0 -68
  1378. package/out/adapters/fcc-bdc/adapter.d.ts.map +0 -1
  1379. package/out/adapters/fcc-bdc/adapter.js +0 -144
  1380. package/out/adapters/fcc-bdc/adapter.js.map +0 -1
  1381. package/out/adapters/geonames-postal/adapter.d.ts +0 -54
  1382. package/out/adapters/geonames-postal/adapter.d.ts.map +0 -1
  1383. package/out/adapters/geonames-postal/adapter.js +0 -107
  1384. package/out/adapters/geonames-postal/adapter.js.map +0 -1
  1385. package/out/adapters/gnaf/adapter.d.ts +0 -49
  1386. package/out/adapters/gnaf/adapter.d.ts.map +0 -1
  1387. package/out/adapters/gnaf/adapter.js +0 -124
  1388. package/out/adapters/gnaf/adapter.js.map +0 -1
  1389. package/out/adapters/gnaf/assemble.d.ts +0 -65
  1390. package/out/adapters/gnaf/assemble.d.ts.map +0 -1
  1391. package/out/adapters/gnaf/assemble.js +0 -154
  1392. package/out/adapters/gnaf/assemble.js.map +0 -1
  1393. package/out/adapters/index.d.ts +0 -51
  1394. package/out/adapters/index.d.ts.map +0 -1
  1395. package/out/adapters/index.js +0 -95
  1396. package/out/adapters/index.js.map +0 -1
  1397. package/out/adapters/state-hi-schools/adapter.d.ts +0 -46
  1398. package/out/adapters/state-hi-schools/adapter.d.ts.map +0 -1
  1399. package/out/adapters/state-hi-schools/adapter.js +0 -145
  1400. package/out/adapters/state-hi-schools/adapter.js.map +0 -1
  1401. package/out/adapters/state-hi-schools/workbook.d.ts +0 -29
  1402. package/out/adapters/state-hi-schools/workbook.d.ts.map +0 -1
  1403. package/out/adapters/state-hi-schools/workbook.js +0 -20
  1404. package/out/adapters/state-hi-schools/workbook.js.map +0 -1
  1405. package/out/adapters/state-ia-contractors/adapter.d.ts +0 -34
  1406. package/out/adapters/state-ia-contractors/adapter.d.ts.map +0 -1
  1407. package/out/adapters/state-ia-contractors/adapter.js +0 -111
  1408. package/out/adapters/state-ia-contractors/adapter.js.map +0 -1
  1409. package/out/adapters/state-ny-notaries/adapter.d.ts +0 -32
  1410. package/out/adapters/state-ny-notaries/adapter.d.ts.map +0 -1
  1411. package/out/adapters/state-ny-notaries/adapter.js +0 -119
  1412. package/out/adapters/state-ny-notaries/adapter.js.map +0 -1
  1413. package/out/adapters/state-tx-notaries/adapter.d.ts +0 -33
  1414. package/out/adapters/state-tx-notaries/adapter.d.ts.map +0 -1
  1415. package/out/adapters/state-tx-notaries/adapter.js +0 -125
  1416. package/out/adapters/state-tx-notaries/adapter.js.map +0 -1
  1417. package/out/adapters/synth-po-box/adapter.d.ts +0 -67
  1418. package/out/adapters/synth-po-box/adapter.d.ts.map +0 -1
  1419. package/out/adapters/synth-po-box/adapter.js +0 -140
  1420. package/out/adapters/synth-po-box/adapter.js.map +0 -1
  1421. package/out/adapters/tiger/adapter.d.ts +0 -55
  1422. package/out/adapters/tiger/adapter.d.ts.map +0 -1
  1423. package/out/adapters/tiger/adapter.js +0 -201
  1424. package/out/adapters/tiger/adapter.js.map +0 -1
  1425. package/out/adapters/tiger/street-decompose.d.ts +0 -30
  1426. package/out/adapters/tiger/street-decompose.d.ts.map +0 -1
  1427. package/out/adapters/tiger/street-decompose.js +0 -68
  1428. package/out/adapters/tiger/street-decompose.js.map +0 -1
  1429. package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts +0 -56
  1430. package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts.map +0 -1
  1431. package/out/adapters/usgov-hrsa-fqhc/adapter.js +0 -132
  1432. package/out/adapters/usgov-hrsa-fqhc/adapter.js.map +0 -1
  1433. package/out/adapters/usgov-imls-pls/adapter.d.ts +0 -36
  1434. package/out/adapters/usgov-imls-pls/adapter.d.ts.map +0 -1
  1435. package/out/adapters/usgov-imls-pls/adapter.js +0 -114
  1436. package/out/adapters/usgov-imls-pls/adapter.js.map +0 -1
  1437. package/out/adapters/usgov-irs-bmf/adapter.d.ts +0 -37
  1438. package/out/adapters/usgov-irs-bmf/adapter.d.ts.map +0 -1
  1439. package/out/adapters/usgov-irs-bmf/adapter.js +0 -120
  1440. package/out/adapters/usgov-irs-bmf/adapter.js.map +0 -1
  1441. package/out/adapters/usgov-nad/adapter.d.ts +0 -48
  1442. package/out/adapters/usgov-nad/adapter.d.ts.map +0 -1
  1443. package/out/adapters/usgov-nad/adapter.js +0 -249
  1444. package/out/adapters/usgov-nad/adapter.js.map +0 -1
  1445. package/out/adapters/usgov-nppes/adapter.d.ts +0 -39
  1446. package/out/adapters/usgov-nppes/adapter.d.ts.map +0 -1
  1447. package/out/adapters/usgov-nppes/adapter.js +0 -118
  1448. package/out/adapters/usgov-nppes/adapter.js.map +0 -1
  1449. package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts +0 -46
  1450. package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts.map +0 -1
  1451. package/out/adapters/usgov-samhsa-treatment-locator/adapter.js +0 -154
  1452. package/out/adapters/usgov-samhsa-treatment-locator/adapter.js.map +0 -1
  1453. package/out/adapters/utils/index.d.ts +0 -191
  1454. package/out/adapters/utils/index.d.ts.map +0 -1
  1455. package/out/adapters/utils/index.js +0 -206
  1456. package/out/adapters/utils/index.js.map +0 -1
  1457. package/out/adapters/wof-admin-jp/adapter.d.ts +0 -65
  1458. package/out/adapters/wof-admin-jp/adapter.d.ts.map +0 -1
  1459. package/out/adapters/wof-admin-jp/adapter.js +0 -161
  1460. package/out/adapters/wof-admin-jp/adapter.js.map +0 -1
  1461. package/out/adapters/wof-admin-json/adapter.d.ts +0 -79
  1462. package/out/adapters/wof-admin-json/adapter.d.ts.map +0 -1
  1463. package/out/adapters/wof-admin-json/adapter.js +0 -197
  1464. package/out/adapters/wof-admin-json/adapter.js.map +0 -1
  1465. package/out/adapters/wof-json-rows.d.ts +0 -62
  1466. package/out/adapters/wof-json-rows.d.ts.map +0 -1
  1467. package/out/adapters/wof-json-rows.js +0 -83
  1468. package/out/adapters/wof-json-rows.js.map +0 -1
  1469. package/out/adapters/wof-postalcode-json/adapter.d.ts +0 -65
  1470. package/out/adapters/wof-postalcode-json/adapter.d.ts.map +0 -1
  1471. package/out/adapters/wof-postalcode-json/adapter.js +0 -151
  1472. package/out/adapters/wof-postalcode-json/adapter.js.map +0 -1
  1473. package/out/codex/us-fips-state.d.ts +0 -41
  1474. package/out/codex/us-fips-state.d.ts.map +0 -1
  1475. package/out/codex/us-fips-state.js +0 -100
  1476. package/out/codex/us-fips-state.js.map +0 -1
  1477. package/out/recipes/bare-country.d.ts +0 -25
  1478. package/out/recipes/bare-country.d.ts.map +0 -1
  1479. package/out/recipes/bare-country.js +0 -88
  1480. package/out/recipes/bare-country.js.map +0 -1
  1481. package/out/recipes/cn-organizational-units.d.ts +0 -36
  1482. package/out/recipes/cn-organizational-units.d.ts.map +0 -1
  1483. package/out/recipes/cn-organizational-units.js +0 -159
  1484. package/out/recipes/cn-organizational-units.js.map +0 -1
  1485. package/out/recipes/cz-pcfirst-preposition.d.ts +0 -22
  1486. package/out/recipes/cz-pcfirst-preposition.d.ts.map +0 -1
  1487. package/out/recipes/cz-pcfirst-preposition.js +0 -101
  1488. package/out/recipes/cz-pcfirst-preposition.js.map +0 -1
  1489. package/out/recipes/fr-admin-split.d.ts +0 -38
  1490. package/out/recipes/fr-admin-split.d.ts.map +0 -1
  1491. package/out/recipes/fr-admin-split.js +0 -192
  1492. package/out/recipes/fr-admin-split.js.map +0 -1
  1493. package/out/recipes/fr-bare-street.d.ts +0 -28
  1494. package/out/recipes/fr-bare-street.d.ts.map +0 -1
  1495. package/out/recipes/fr-bare-street.js +0 -144
  1496. package/out/recipes/fr-bare-street.js.map +0 -1
  1497. package/out/recipes/fr-fragment.d.ts +0 -77
  1498. package/out/recipes/fr-fragment.d.ts.map +0 -1
  1499. package/out/recipes/fr-fragment.js +0 -277
  1500. package/out/recipes/fr-fragment.js.map +0 -1
  1501. package/out/recipes/fr-lieudit.d.ts +0 -32
  1502. package/out/recipes/fr-lieudit.d.ts.map +0 -1
  1503. package/out/recipes/fr-lieudit.js +0 -227
  1504. package/out/recipes/fr-lieudit.js.map +0 -1
  1505. package/out/recipes/fr-order.d.ts +0 -38
  1506. package/out/recipes/fr-order.d.ts.map +0 -1
  1507. package/out/recipes/fr-order.js +0 -198
  1508. package/out/recipes/fr-order.js.map +0 -1
  1509. package/out/recipes/german.d.ts +0 -44
  1510. package/out/recipes/german.d.ts.map +0 -1
  1511. package/out/recipes/german.js +0 -289
  1512. package/out/recipes/german.js.map +0 -1
  1513. package/out/recipes/index.d.ts +0 -24
  1514. package/out/recipes/index.d.ts.map +0 -1
  1515. package/out/recipes/index.js +0 -91
  1516. package/out/recipes/index.js.map +0 -1
  1517. package/out/recipes/locale.d.ts +0 -136
  1518. package/out/recipes/locale.d.ts.map +0 -1
  1519. package/out/recipes/locale.js +0 -460
  1520. package/out/recipes/locale.js.map +0 -1
  1521. package/out/recipes/nl-postcode.d.ts +0 -26
  1522. package/out/recipes/nl-postcode.d.ts.map +0 -1
  1523. package/out/recipes/nl-postcode.js +0 -103
  1524. package/out/recipes/nl-postcode.js.map +0 -1
  1525. package/out/recipes/no-fragment.d.ts +0 -43
  1526. package/out/recipes/no-fragment.d.ts.map +0 -1
  1527. package/out/recipes/no-fragment.js +0 -183
  1528. package/out/recipes/no-fragment.js.map +0 -1
  1529. package/out/recipes/no-street-led.d.ts +0 -24
  1530. package/out/recipes/no-street-led.d.ts.map +0 -1
  1531. package/out/recipes/no-street-led.js +0 -114
  1532. package/out/recipes/no-street-led.js.map +0 -1
  1533. package/out/recipes/no-street.d.ts +0 -18
  1534. package/out/recipes/no-street.d.ts.map +0 -1
  1535. package/out/recipes/no-street.js +0 -72
  1536. package/out/recipes/no-street.js.map +0 -1
  1537. package/out/recipes/po-box-cedex.d.ts +0 -38
  1538. package/out/recipes/po-box-cedex.d.ts.map +0 -1
  1539. package/out/recipes/po-box-cedex.js +0 -768
  1540. package/out/recipes/po-box-cedex.js.map +0 -1
  1541. package/out/recipes/po-box.d.ts +0 -17
  1542. package/out/recipes/po-box.d.ts.map +0 -1
  1543. package/out/recipes/po-box.js +0 -101
  1544. package/out/recipes/po-box.js.map +0 -1
  1545. package/out/recipes/reviewed-postcode-tail.d.ts +0 -47
  1546. package/out/recipes/reviewed-postcode-tail.d.ts.map +0 -1
  1547. package/out/recipes/reviewed-postcode-tail.js +0 -125
  1548. package/out/recipes/reviewed-postcode-tail.js.map +0 -1
  1549. package/out/recipes/si-bare-village.d.ts +0 -32
  1550. package/out/recipes/si-bare-village.d.ts.map +0 -1
  1551. package/out/recipes/si-bare-village.js +0 -89
  1552. package/out/recipes/si-bare-village.js.map +0 -1
  1553. package/out/recipes/street-affix.d.ts +0 -89
  1554. package/out/recipes/street-affix.d.ts.map +0 -1
  1555. package/out/recipes/street-affix.js +0 -563
  1556. package/out/recipes/street-affix.js.map +0 -1
  1557. package/out/recipes/street-bare.d.ts +0 -25
  1558. package/out/recipes/street-bare.d.ts.map +0 -1
  1559. package/out/recipes/street-bare.js +0 -87
  1560. package/out/recipes/street-bare.js.map +0 -1
  1561. package/out/recipes/sub-venue-sources.d.ts +0 -229
  1562. package/out/recipes/sub-venue-sources.d.ts.map +0 -1
  1563. package/out/recipes/sub-venue-sources.js +0 -654
  1564. package/out/recipes/sub-venue-sources.js.map +0 -1
  1565. package/out/recipes/sub-venue.d.ts +0 -269
  1566. package/out/recipes/sub-venue.d.ts.map +0 -1
  1567. package/out/recipes/sub-venue.js +0 -726
  1568. package/out/recipes/sub-venue.js.map +0 -1
  1569. package/out/synthesizers/german.d.ts +0 -115
  1570. package/out/synthesizers/german.d.ts.map +0 -1
  1571. package/out/synthesizers/german.js +0 -149
  1572. package/out/synthesizers/german.js.map +0 -1
  1573. package/out/test-kit/index.d.ts +0 -48
  1574. package/out/test-kit/index.d.ts.map +0 -1
  1575. package/out/test-kit/index.js +0 -78
  1576. package/out/test-kit/index.js.map +0 -1
  1577. package/out/tools/align-slice.d.ts +0 -7
  1578. package/out/tools/align-slice.d.ts.map +0 -1
  1579. package/out/tools/align-slice.js +0 -55
  1580. package/out/tools/align-slice.js.map +0 -1
  1581. package/out/tools/fetch/acra-sg.d.ts +0 -23
  1582. package/out/tools/fetch/acra-sg.d.ts.map +0 -1
  1583. package/out/tools/fetch/acra-sg.js +0 -126
  1584. package/out/tools/fetch/acra-sg.js.map +0 -1
  1585. package/out/tools/fetch/ban.d.ts +0 -31
  1586. package/out/tools/fetch/ban.d.ts.map +0 -1
  1587. package/out/tools/fetch/ban.js +0 -239
  1588. package/out/tools/fetch/ban.js.map +0 -1
  1589. package/out/tools/fetch/gcis-tw.d.ts +0 -23
  1590. package/out/tools/fetch/gcis-tw.d.ts.map +0 -1
  1591. package/out/tools/fetch/gcis-tw.js +0 -135
  1592. package/out/tools/fetch/gcis-tw.js.map +0 -1
  1593. package/out/tools/fetch/geonames-dump.d.ts +0 -91
  1594. package/out/tools/fetch/geonames-dump.d.ts.map +0 -1
  1595. package/out/tools/fetch/geonames-dump.js +0 -171
  1596. package/out/tools/fetch/geonames-dump.js.map +0 -1
  1597. package/out/tools/fetch/geonames-postal.d.ts +0 -76
  1598. package/out/tools/fetch/geonames-postal.d.ts.map +0 -1
  1599. package/out/tools/fetch/geonames-postal.js +0 -126
  1600. package/out/tools/fetch/geonames-postal.js.map +0 -1
  1601. package/out/tools/fetch/houjin-jp.d.ts +0 -24
  1602. package/out/tools/fetch/houjin-jp.d.ts.map +0 -1
  1603. package/out/tools/fetch/houjin-jp.js +0 -102
  1604. package/out/tools/fetch/houjin-jp.js.map +0 -1
  1605. package/out/tools/fetch/hrsa.d.ts +0 -18
  1606. package/out/tools/fetch/hrsa.d.ts.map +0 -1
  1607. package/out/tools/fetch/hrsa.js +0 -46
  1608. package/out/tools/fetch/hrsa.js.map +0 -1
  1609. package/out/tools/fetch/imls-pls.d.ts +0 -25
  1610. package/out/tools/fetch/imls-pls.d.ts.map +0 -1
  1611. package/out/tools/fetch/imls-pls.js +0 -108
  1612. package/out/tools/fetch/imls-pls.js.map +0 -1
  1613. package/out/tools/fetch/index.d.ts +0 -171
  1614. package/out/tools/fetch/index.d.ts.map +0 -1
  1615. package/out/tools/fetch/index.js +0 -170
  1616. package/out/tools/fetch/index.js.map +0 -1
  1617. package/out/tools/fetch/juso-kr.d.ts +0 -30
  1618. package/out/tools/fetch/juso-kr.d.ts.map +0 -1
  1619. package/out/tools/fetch/juso-kr.js +0 -120
  1620. package/out/tools/fetch/juso-kr.js.map +0 -1
  1621. package/out/tools/fetch/localdata-kr.d.ts +0 -30
  1622. package/out/tools/fetch/localdata-kr.d.ts.map +0 -1
  1623. package/out/tools/fetch/localdata-kr.js +0 -139
  1624. package/out/tools/fetch/localdata-kr.js.map +0 -1
  1625. package/out/tools/fetch/nad.d.ts +0 -70
  1626. package/out/tools/fetch/nad.d.ts.map +0 -1
  1627. package/out/tools/fetch/nad.js +0 -234
  1628. package/out/tools/fetch/nad.js.map +0 -1
  1629. package/out/tools/fetch/nppes.d.ts +0 -26
  1630. package/out/tools/fetch/nppes.d.ts.map +0 -1
  1631. package/out/tools/fetch/nppes.js +0 -126
  1632. package/out/tools/fetch/nppes.js.map +0 -1
  1633. package/out/tools/fetch/state-hi-schools.d.ts +0 -26
  1634. package/out/tools/fetch/state-hi-schools.d.ts.map +0 -1
  1635. package/out/tools/fetch/state-hi-schools.js +0 -112
  1636. package/out/tools/fetch/state-hi-schools.js.map +0 -1
  1637. package/out/tools/fetch/state-sources.d.ts +0 -26
  1638. package/out/tools/fetch/state-sources.d.ts.map +0 -1
  1639. package/out/tools/fetch/state-sources.js +0 -111
  1640. package/out/tools/fetch/state-sources.js.map +0 -1
  1641. package/out/tools/fetch/tiger-full.d.ts +0 -22
  1642. package/out/tools/fetch/tiger-full.d.ts.map +0 -1
  1643. package/out/tools/fetch/tiger-full.js +0 -217
  1644. package/out/tools/fetch/tiger-full.js.map +0 -1
  1645. package/out/tools/gb-title-case.d.ts +0 -2
  1646. package/out/tools/gb-title-case.d.ts.map +0 -1
  1647. package/out/tools/gb-title-case.js +0 -65
  1648. package/out/tools/gb-title-case.js.map +0 -1
  1649. package/out/tools/golden-expand.d.ts +0 -90
  1650. package/out/tools/golden-expand.d.ts.map +0 -1
  1651. package/out/tools/golden-expand.js +0 -406
  1652. package/out/tools/golden-expand.js.map +0 -1
  1653. package/out/tools/golden-promote.d.ts +0 -72
  1654. package/out/tools/golden-promote.d.ts.map +0 -1
  1655. package/out/tools/golden-promote.js +0 -240
  1656. package/out/tools/golden-promote.js.map +0 -1
  1657. package/out/tools/golden-relabel-street.d.ts +0 -196
  1658. package/out/tools/golden-relabel-street.d.ts.map +0 -1
  1659. package/out/tools/golden-relabel-street.js +0 -439
  1660. package/out/tools/golden-relabel-street.js.map +0 -1
  1661. package/out/tools/index.d.ts +0 -28
  1662. package/out/tools/index.d.ts.map +0 -1
  1663. package/out/tools/index.js +0 -28
  1664. package/out/tools/index.js.map +0 -1
  1665. package/out/tools/lint-slice-vocab.d.ts +0 -90
  1666. package/out/tools/lint-slice-vocab.d.ts.map +0 -1
  1667. package/out/tools/lint-slice-vocab.js +0 -267
  1668. package/out/tools/lint-slice-vocab.js.map +0 -1
  1669. package/out/tools/lint-slice.d.ts +0 -93
  1670. package/out/tools/lint-slice.d.ts.map +0 -1
  1671. package/out/tools/lint-slice.js +0 -322
  1672. package/out/tools/lint-slice.js.map +0 -1
  1673. package/out/tools/overlay-manifest.d.ts +0 -39
  1674. package/out/tools/overlay-manifest.d.ts.map +0 -1
  1675. package/out/tools/overlay-manifest.js +0 -90
  1676. package/out/tools/overlay-manifest.js.map +0 -1
  1677. package/out/tools/slice-kryptonite.d.ts +0 -32
  1678. package/out/tools/slice-kryptonite.d.ts.map +0 -1
  1679. package/out/tools/slice-kryptonite.js +0 -99
  1680. package/out/tools/slice-kryptonite.js.map +0 -1
  1681. package/out/tools/slice-translit.d.ts +0 -51
  1682. package/out/tools/slice-translit.d.ts.map +0 -1
  1683. package/out/tools/slice-translit.js +0 -166
  1684. package/out/tools/slice-translit.js.map +0 -1
  1685. package/out/tools/sub-venue/harvest.d.ts +0 -100
  1686. package/out/tools/sub-venue/harvest.d.ts.map +0 -1
  1687. package/out/tools/sub-venue/harvest.js +0 -168
  1688. package/out/tools/sub-venue/harvest.js.map +0 -1
  1689. package/out/tools/sub-venue/head-nouns.d.ts +0 -50
  1690. package/out/tools/sub-venue/head-nouns.d.ts.map +0 -1
  1691. package/out/tools/sub-venue/head-nouns.js +0 -213
  1692. package/out/tools/sub-venue/head-nouns.js.map +0 -1
  1693. package/out/tools/sub-venue/surfaces.d.ts +0 -45
  1694. package/out/tools/sub-venue/surfaces.d.ts.map +0 -1
  1695. package/out/tools/sub-venue/surfaces.js +0 -77
  1696. package/out/tools/sub-venue/surfaces.js.map +0 -1
  1697. package/out/tools/sub-venue/table.d.ts +0 -229
  1698. package/out/tools/sub-venue/table.d.ts.map +0 -1
  1699. package/out/tools/sub-venue/table.js +0 -108
  1700. package/out/tools/sub-venue/table.js.map +0 -1
  1701. package/out/tools/sub-venue/wikidata.d.ts +0 -26
  1702. package/out/tools/sub-venue/wikidata.d.ts.map +0 -1
  1703. package/out/tools/sub-venue/wikidata.js +0 -62
  1704. package/out/tools/sub-venue/wikidata.js.map +0 -1
  1705. package/out/tools/sub-venue-lexicon.d.ts +0 -156
  1706. package/out/tools/sub-venue-lexicon.d.ts.map +0 -1
  1707. package/out/tools/sub-venue-lexicon.js +0 -283
  1708. package/out/tools/sub-venue-lexicon.js.map +0 -1
  1709. package/out/tools/sub-venue-promotions.d.ts +0 -94
  1710. package/out/tools/sub-venue-promotions.d.ts.map +0 -1
  1711. package/out/tools/sub-venue-promotions.js +0 -266
  1712. package/out/tools/sub-venue-promotions.js.map +0 -1
  1713. package/out/utils/index.d.ts +0 -14
  1714. package/out/utils/index.d.ts.map +0 -1
  1715. package/out/utils/index.js +0 -14
  1716. package/out/utils/index.js.map +0 -1
  1717. package/out/utils/parquet.d.ts +0 -178
  1718. package/out/utils/parquet.d.ts.map +0 -1
  1719. package/out/utils/parquet.js +0 -371
  1720. package/out/utils/parquet.js.map +0 -1
  1721. package/out/utils/slice-stats.d.ts +0 -52
  1722. package/out/utils/slice-stats.d.ts.map +0 -1
  1723. package/out/utils/slice-stats.js +0 -58
  1724. package/out/utils/slice-stats.js.map +0 -1
@@ -0,0 +1,260 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ *
6
+ * Assembles a corpus manifest that keeps every file of a base corpus and adds new parquet files.
7
+ */
8
+ import { dataRootPath } from "@mailwoman/core/data-root";
9
+ import { readLocalBuffer, readLocalJSONFile, statPath } from "@mailwoman/core/fs/readers";
10
+ import { writeLocalJSONFile } from "@mailwoman/core/fs/writers";
11
+ import { sha256Hex } from "@mailwoman/core/hash";
12
+ import { stringifyJSON } from "@mailwoman/core/json";
13
+ import { basename, dirname, PathBuilder } from "path-ts";
14
+ import { escapeSQLString, openDuckDB } from "#parquet/duckdb";
15
+ /**
16
+ * The legacy manifest keys for the file list and the rows per file.
17
+ *
18
+ * Earlier corpora are immutable and still use these keys, so readers accept both
19
+ * spellings and writers emit only the current ones.
20
+ * The keys are built by concatenation because a repository check bans the legacy word.
21
+ */
22
+ const PRE_RENAME_FILES_KEY = `sh${"ards"}`;
23
+ const PRE_RENAME_ROWS_PER_FILE_KEY = `rows_per_sh${"ard"}`;
24
+ /**
25
+ * Returns a manifest's file list from `slices` or from the legacy key.
26
+ *
27
+ * A non-empty `slices` wins, then a legacy list, then an empty `slices`.
28
+ * An empty list is a valid answer, but a manifest with neither key throws, because an
29
+ * empty result would misreport a missing list as a corpus without files.
30
+ */
31
+ export function baseManifestFiles(manifest) {
32
+ const slices = manifest.slices;
33
+ const preRename = manifest[PRE_RENAME_FILES_KEY];
34
+ if (Array.isArray(slices) && slices.length)
35
+ return slices;
36
+ if (Array.isArray(preRename))
37
+ return preRename;
38
+ if (Array.isArray(slices))
39
+ return slices;
40
+ throw new Error(`manifest names no file list: neither "slices" nor the pre-rename key holds an array. An empty list here ` +
41
+ `would read as a corpus of no files, which is a measurement this manifest does not support.`);
42
+ }
43
+ /**
44
+ * Returns the base's rows-per-file value under either key.
45
+ */
46
+ function baseRowsPerFile(base) {
47
+ return base.rows_per_slice ?? base[PRE_RENAME_ROWS_PER_FILE_KEY];
48
+ }
49
+ /**
50
+ * Returns the split encoded in a `<stem>.<split>.parquet` filename, or `null` for any other name.
51
+ *
52
+ * Only `splitOverlaySlice` writes names with this suffix.
53
+ */
54
+ export function splitFromFilename(parquet) {
55
+ const match = /\.(train|val|test)\.parquet$/.exec(parquet);
56
+ return match ? match[1] : null;
57
+ }
58
+ /**
59
+ * Reads one overlay parquet and describes it for the manifest.
60
+ *
61
+ * The loader groups rows by `file_source_counts`.
62
+ * The manifest records the `source` value supplied by the caller.
63
+ *
64
+ * A stale argument therefore writes a manifest that disagrees with the rows.
65
+ * The run trains on the rows.
66
+ *
67
+ * The next plan then repeats the stale label from the manifest.
68
+ *
69
+ * This reads the column and refuses the mismatch rather than recording the claim.
70
+ *
71
+ * `source_id` is materialized rather than aggregated because `first_source_id`
72
+ * and `last_source_id` mean the first and last row of the file.
73
+ * DuckDB's `first` and `last` aggregates read whichever row a parallel scan reaches first.
74
+ *
75
+ * That result can differ from the first and last rows in the file.
76
+ */
77
+ async function descriptor(localPath, modalPath, split, source) {
78
+ // `assembleOverlayManifest` calls this once for each file.
79
+ using db = await openDuckDB();
80
+ const result = await db.runAndReadAll(`SELECT source_id, source FROM read_parquet('${escapeSQLString(localPath)}')`);
81
+ const rows = result.getRowObjects();
82
+ const sids = rows.map((r) => r.source_id);
83
+ const stored = new Set(rows.map((r) => r.source));
84
+ if (stored.size !== 1 || !stored.has(source)) {
85
+ const found = [...stored].toSorted().join(", ");
86
+ throw new Error(`${localPath} holds source ${stored.size === 1 ? `"${found}"` : `${stored.size} values (${found})`}, ` +
87
+ `and the manifest was asked to record "${source}". The loader groups rows by the column, so a ` +
88
+ `manifest that disagrees with it is a label nothing reads and a plan derived from it carries forward.`);
89
+ }
90
+ return {
91
+ split,
92
+ path: modalPath,
93
+ format: "parquet",
94
+ compression: "SNAPPY",
95
+ rows: sids.length,
96
+ bytes: (await statPath(localPath)).size,
97
+ sha256: sha256Hex(await readLocalBuffer(localPath)),
98
+ first_source_id: sids[0],
99
+ last_source_id: sids.at(-1),
100
+ source,
101
+ };
102
+ }
103
+ /**
104
+ * Options for {@link assembleOverlayManifest}.
105
+ */
106
+ /**
107
+ * The filename {@linkcode assembleOverlayManifest} writes its plan under, beside `MANIFEST.json`.
108
+ */
109
+ export const OVERLAY_PLAN_FILE = "OVERLAY_PLAN.json";
110
+ /**
111
+ * How one overlay file's split was decided.
112
+ */
113
+ export const SplitRoute = {
114
+ /**
115
+ * `corpus split-slice` applied `splitForRow` to every row and wrote the split into the filename.
116
+ */
117
+ SplitSlice: "split-slice",
118
+ /**
119
+ * The caller supplied the split.
120
+ * The holdout policy never saw these rows.
121
+ */
122
+ Caller: "caller",
123
+ };
124
+ /**
125
+ * Which route decided one file's split.
126
+ */
127
+ export function splitRouteFor(parquet) {
128
+ return splitFromFilename(parquet) ? SplitRoute.SplitSlice : SplitRoute.Caller;
129
+ }
130
+ /**
131
+ * Rewrites a base manifest's file path to its location in the corpus tree mounted on Modal at `/data`.
132
+ */
133
+ export function rerootBaseFilePath(path, baseManifestPath) {
134
+ const versionedIndex = path.indexOf("/corpus/versioned/");
135
+ if (versionedIndex !== -1)
136
+ return "/data" + path.slice(versionedIndex);
137
+ if (/^\/data\/(?:train|val|test)\//u.test(path)) {
138
+ const localBaseDir = dirname(baseManifestPath);
139
+ const baseModalRoot = `/data/corpus/versioned/${basename(dirname(localBaseDir))}/${basename(localBaseDir)}`;
140
+ return baseModalRoot + path.slice("/data".length);
141
+ }
142
+ return path;
143
+ }
144
+ /**
145
+ * Maps a Modal path under `/data/` to the same file under the local data root
146
+ * and returns any other path unchanged.
147
+ */
148
+ export function localManifestFilePath(path) {
149
+ return path.startsWith("/data/") ? dataRootPath(path.slice("/data/".length)).toString() : path;
150
+ }
151
+ /**
152
+ * Writes a corpus manifest that keeps every file of `args.base` and appends `args.files`.
153
+ *
154
+ * A `val` or `test` file must use the `.<split>.parquet` suffix written by `splitOverlaySlice`.
155
+ * The suffix shows that the holdout policy chose its rows.
156
+ * A `train` file has no naming requirement.
157
+ *
158
+ * @throws When the base lists no files or a held-out file lacks the matching filename suffix.
159
+ */
160
+ export async function assembleOverlayManifest(args) {
161
+ if (!args.files.length)
162
+ throw new Error("an overlay must add at least one parquet file");
163
+ const base = await readLocalJSONFile(args.base);
164
+ const baseFiles = baseManifestFiles(base);
165
+ // An overlay on a base without files would contain only the added files.
166
+ if (!baseFiles.length) {
167
+ throw new Error(`${args.base} lists no files, so an overlay on it would carry only the ${args.files.length} file(s) added ` +
168
+ `here while naming itself an extension of that base.`);
169
+ }
170
+ for (const file of args.files) {
171
+ if (baseFiles.some((s) => s.source === file.source)) {
172
+ console.log(`WARN: base already contains source '${file.source}' — is this the right base?`);
173
+ }
174
+ }
175
+ // Held-out rows must come from the holdout policy.
176
+ // The filename records the split.
177
+ // A caller that picks held-out rows by hand can leak them into train.
178
+ for (const file of args.files) {
179
+ if (!file.split || file.split === "train")
180
+ continue;
181
+ const named = splitFromFilename(file.parquet);
182
+ if (named === file.split)
183
+ continue;
184
+ throw new Error(named
185
+ ? `${file.parquet} names split '${named}' and the caller asked for '${file.split}'. The filename is ` +
186
+ `what splitOverlaySlice wrote, so it decides; pass the split the name carries or route the file again.`
187
+ : `${file.parquet} was asked for split '${file.split}' and carries no '.${file.split}.parquet' suffix, ` +
188
+ `so it did not come from 'mailwoman corpus split-slice' and its held-out rows were chosen by the ` +
189
+ `caller rather than by the holdout policy. Route it through split-slice and add the per-split ` +
190
+ `outputs it writes. A file appended to 'train' needs none of this.`);
191
+ }
192
+ const kept = baseFiles.map((s) => ({ ...s, path: rerootBaseFilePath(s.path, args.base) }));
193
+ const added = [];
194
+ const newDir = PathBuilder.from(args.newDir);
195
+ for (const file of args.files) {
196
+ const split = file.split ?? "train";
197
+ added.push(await descriptor(
198
+ // DuckDB reads the path inside SQL text, so it must be a string.
199
+ newDir(split, file.parquet).toString(), `${args.modalRoot}/${split}/${file.parquet}`, split, file.source));
200
+ }
201
+ const addedRows = { train: 0, val: 0, test: 0 };
202
+ for (const file of added) {
203
+ addedRows[file.split] += file.rows;
204
+ }
205
+ const sources = args.files.map((file) => file.source).join(", ");
206
+ const manifest = {
207
+ corpus_version: args.version,
208
+ overlay_base: base.corpus_version ?? null,
209
+ note: args.note || `${base.corpus_version} files (all kept verbatim) + ${sources}. Pure overlay add.`,
210
+ schema: base.schema,
211
+ rows_per_slice: baseRowsPerFile(base),
212
+ row_group_size: base.row_group_size,
213
+ slices: [...kept, ...added],
214
+ counts: {
215
+ train: base.counts.train + addedRows.train,
216
+ val: base.counts.val + addedRows.val,
217
+ test: base.counts.test + addedRows.test,
218
+ },
219
+ total_rows: base.total_rows + addedRows.train + addedRows.val + addedRows.test,
220
+ // An overlay corpus's manifest reads no row's `license` column, so it states no license set.
221
+ // This value in the artifact prevents a reader from deriving an attribution table
222
+ // from `slices` and treating the base build's set as the whole.
223
+ // The overlay sources listed here have their own terms, recorded per source in
224
+ // `packages/corpus/lib/recipes/sources.ts` and per row in the `license` column itself.
225
+ licenses_cover: `no license set is measured here. The base build's MANIFEST.json covers the rows it aligned, and ` +
226
+ `these ${added.length} added file(s) carry their own: ${[...new Set(args.files.map((file) => file.source))].toSorted().join(", ")}.`,
227
+ };
228
+ const out = newDir("MANIFEST.json");
229
+ await writeLocalJSONFile(manifest, out);
230
+ // The plan is written beside the manifest it produced, because the previous overlay
231
+ // assembly's plan existed only as three scripts in a session scratch directory
232
+ // and was copied into the next assembly as a hand-written list.
233
+ // Commit this file under `packages/corpus/data/builds/<version>/`.
234
+ const plan = {
235
+ corpusVersion: args.version,
236
+ overlayBase: base.corpus_version ?? null,
237
+ baseManifest: args.base,
238
+ modalRoot: args.modalRoot,
239
+ note: manifest.note,
240
+ appliedAt: new Date().toISOString(),
241
+ files: args.files.map((file) => ({
242
+ parquet: file.parquet,
243
+ source: file.source,
244
+ split: file.split ?? "train",
245
+ route: splitRouteFor(file.parquet),
246
+ })),
247
+ };
248
+ const planPath = newDir(OVERLAY_PLAN_FILE);
249
+ await writeLocalJSONFile(plan, planPath);
250
+ console.log(`wrote ${planPath}`);
251
+ console.log(`wrote ${out}`);
252
+ console.log(` files: ${manifest.slices.length} (${kept.length} base kept, +${added.length} added)`);
253
+ console.log(` counts: ${stringifyJSON(manifest.counts)} total: ${manifest.total_rows}`);
254
+ // The route prints beside the split, because a hand-placed split is otherwise visible only in a filename.
255
+ for (const [index, file] of added.entries()) {
256
+ const route = plan.files[index]?.route ?? SplitRoute.Caller;
257
+ console.log(` ${file.source} ${file.split} via ${route}: ${file.rows} rows (${file.bytes} bytes)`);
258
+ }
259
+ }
260
+ //# sourceMappingURL=manifest.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"manifest.js","sourceRoot":"","sources":["../../../lib/tools/overlay/manifest.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,2BAA2B,CAAA;AACxD,OAAO,EAAE,eAAe,EAAE,iBAAiB,EAAE,QAAQ,EAAE,MAAM,4BAA4B,CAAA;AACzF,OAAO,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AAC/D,OAAO,EAAE,SAAS,EAAE,MAAM,sBAAsB,CAAA;AAChD,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACpD,OAAO,EAAE,QAAQ,EAAE,OAAO,EAAE,WAAW,EAAwB,MAAM,SAAS,CAAA;AAE9E,OAAO,EAAE,eAAe,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAA;AAkB7D;;;;;;GAMG;AACH,MAAM,oBAAoB,GAAG,KAAK,MAAM,EAAW,CAAA;AACnD,MAAM,4BAA4B,GAAG,cAAc,KAAK,EAAW,CAAA;AAcnE;;;;;;GAMG;AACH,MAAM,UAAU,iBAAiB,CAAC,QAAgE;IACjG,MAAM,MAAM,GAAG,QAAQ,CAAC,MAAM,CAAA;IAC9B,MAAM,SAAS,GAAG,QAAQ,CAAC,oBAAoB,CAAC,CAAA;IAEhD,IAAI,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,IAAI,MAAM,CAAC,MAAM;QAAE,OAAO,MAAwB,CAAA;IAE3E,IAAI,KAAK,CAAC,OAAO,CAAC,SAAS,CAAC;QAAE,OAAO,SAA2B,CAAA;IAEhE,IAAI,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC;QAAE,OAAO,MAAwB,CAAA;IAE1D,MAAM,IAAI,KAAK,CACd,0GAA0G;QACzG,4FAA4F,CAC7F,CAAA;AACF,CAAC;AAED;;GAEG;AACH,SAAS,eAAe,CAAC,IAAkB;IAC1C,OAAO,IAAI,CAAC,cAAc,IAAI,IAAI,CAAC,4BAA4B,CAAC,CAAA;AACjE,CAAC;AAkBD;;;;GAIG;AACH,MAAM,UAAU,iBAAiB,CAAC,OAAe;IAChD,MAAM,KAAK,GAAG,8BAA8B,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;IAE1D,OAAO,KAAK,CAAC,CAAC,CAAE,KAAK,CAAC,CAAC,CAAe,CAAC,CAAC,CAAC,IAAI,CAAA;AAC9C,CAAC;AAED;;;;;;;;;;;;;;;;;;GAkBG;AACH,KAAK,UAAU,UAAU,CACxB,SAAiB,EACjB,SAAiB,EACjB,KAAgB,EAChB,MAAc;IAEd,2DAA2D;IAC3D,MAAM,EAAE,GAAG,MAAM,UAAU,EAAE,CAAA;IAE7B,MAAM,MAAM,GAAG,MAAM,EAAE,CAAC,aAAa,CAAC,+CAA+C,eAAe,CAAC,SAAS,CAAC,IAAI,CAAC,CAAA;IAEpH,MAAM,IAAI,GAAG,MAAM,CAAC,aAAa,EAAE,CAAA;IACnC,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,SAAmB,CAAC,CAAA;IACnD,MAAM,MAAM,GAAG,IAAI,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAgB,CAAC,CAAC,CAAA;IAE3D,IAAI,MAAM,CAAC,IAAI,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,EAAE,CAAC;QAC9C,MAAM,KAAK,GAAG,CAAC,GAAG,MAAM,CAAC,CAAC,QAAQ,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAE/C,MAAM,IAAI,KAAK,CACd,GAAG,SAAS,iBAAiB,MAAM,CAAC,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,KAAK,GAAG,CAAC,CAAC,CAAC,GAAG,MAAM,CAAC,IAAI,YAAY,KAAK,GAAG,IAAI;YACrG,yCAAyC,MAAM,gDAAgD;YAC/F,sGAAsG,CACvG,CAAA;IACF,CAAC;IAED,OAAO;QACN,KAAK;QACL,IAAI,EAAE,SAAS;QACf,MAAM,EAAE,SAAS;QACjB,WAAW,EAAE,QAAQ;QACrB,IAAI,EAAE,IAAI,CAAC,MAAM;QACjB,KAAK,EAAE,CAAC,MAAM,QAAQ,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI;QACvC,MAAM,EAAE,SAAS,CAAC,MAAM,eAAe,CAAC,SAAS,CAAC,CAAC;QACnD,eAAe,EAAE,IAAI,CAAC,CAAC,CAAE;QACzB,cAAc,EAAE,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,CAAE;QAC5B,MAAM;KACN,CAAA;AACF,CAAC;AAED;;GAEG;AACH;;GAEG;AACH,MAAM,CAAC,MAAM,iBAAiB,GAAG,mBAAmB,CAAA;AA6BpD;;GAEG;AACH,MAAM,CAAC,MAAM,UAAU,GAAG;IACzB;;OAEG;IACH,UAAU,EAAE,aAAa;IAEzB;;;OAGG;IACH,MAAM,EAAE,QAAQ;CACP,CAAA;AAIV;;GAEG;AACH,MAAM,UAAU,aAAa,CAAC,OAAe;IAC5C,OAAO,iBAAiB,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,UAAU,CAAC,CAAC,CAAC,UAAU,CAAC,MAAM,CAAA;AAC9E,CAAC;AAcD;;GAEG;AACH,MAAM,UAAU,kBAAkB,CAAC,IAAY,EAAE,gBAAwB;IACxE,MAAM,cAAc,GAAG,IAAI,CAAC,OAAO,CAAC,oBAAoB,CAAC,CAAA;IAEzD,IAAI,cAAc,KAAK,CAAC,CAAC;QAAE,OAAO,OAAO,GAAG,IAAI,CAAC,KAAK,CAAC,cAAc,CAAC,CAAA;IAEtE,IAAI,gCAAgC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;QACjD,MAAM,YAAY,GAAG,OAAO,CAAC,gBAAgB,CAAC,CAAA;QAC9C,MAAM,aAAa,GAAG,0BAA0B,QAAQ,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC,IAAI,QAAQ,CAAC,YAAY,CAAC,EAAE,CAAA;QAE3G,OAAO,aAAa,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;IAClD,CAAC;IAED,OAAO,IAAI,CAAA;AACZ,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,qBAAqB,CAAC,IAAY;IACjD,OAAO,IAAI,CAAC,UAAU,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,IAAI,CAAA;AAC/F,CAAC;AAED;;;;;;;;GAQG;AACH,MAAM,CAAC,KAAK,UAAU,uBAAuB,CAAC,IAA4B;IACzE,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM;QAAE,MAAM,IAAI,KAAK,CAAC,+CAA+C,CAAC,CAAA;IAExF,MAAM,IAAI,GAAG,MAAM,iBAAiB,CAAe,IAAI,CAAC,IAAI,CAAC,CAAA;IAC7D,MAAM,SAAS,GAAG,iBAAiB,CAAC,IAAI,CAAC,CAAA;IAEzC,yEAAyE;IACzE,IAAI,CAAC,SAAS,CAAC,MAAM,EAAE,CAAC;QACvB,MAAM,IAAI,KAAK,CACd,GAAG,IAAI,CAAC,IAAI,6DAA6D,IAAI,CAAC,KAAK,CAAC,MAAM,iBAAiB;YAC1G,qDAAqD,CACtD,CAAA;IACF,CAAC;IAED,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,KAAK,EAAE,CAAC;QAC/B,IAAI,SAAS,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,IAAI,CAAC,MAAM,CAAC,EAAE,CAAC;YACrD,OAAO,CAAC,GAAG,CAAC,uCAAuC,IAAI,CAAC,MAAM,6BAA6B,CAAC,CAAA;QAC7F,CAAC;IACF,CAAC;IAED,mDAAmD;IACnD,kCAAkC;IAClC,sEAAsE;IACtE,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,KAAK,EAAE,CAAC;QAC/B,IAAI,CAAC,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,OAAO;YAAE,SAAQ;QAEnD,MAAM,KAAK,GAAG,iBAAiB,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;QAE7C,IAAI,KAAK,KAAK,IAAI,CAAC,KAAK;YAAE,SAAQ;QAElC,MAAM,IAAI,KAAK,CACd,KAAK;YACJ,CAAC,CAAC,GAAG,IAAI,CAAC,OAAO,iBAAiB,KAAK,+BAA+B,IAAI,CAAC,KAAK,qBAAqB;gBACnG,uGAAuG;YACzG,CAAC,CAAC,GAAG,IAAI,CAAC,OAAO,yBAAyB,IAAI,CAAC,KAAK,sBAAsB,IAAI,CAAC,KAAK,oBAAoB;gBACtG,kGAAkG;gBAClG,+FAA+F;gBAC/F,mEAAmE,CACtE,CAAA;IACF,CAAC;IAED,MAAM,IAAI,GAAG,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,IAAI,EAAE,kBAAkB,CAAC,CAAC,CAAC,IAAI,EAAE,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAA;IAC1F,MAAM,KAAK,GAA4B,EAAE,CAAA;IAEzC,MAAM,MAAM,GAAG,WAAW,CAAC,IAAI,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;IAE5C,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,KAAK,EAAE,CAAC;QAC/B,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,IAAI,OAAO,CAAA;QAEnC,KAAK,CAAC,IAAI,CACT,MAAM,UAAU;QACf,iEAAiE;QACjE,MAAM,CAAC,KAAK,EAAE,IAAI,CAAC,OAAO,CAAC,CAAC,QAAQ,EAAE,EACtC,GAAG,IAAI,CAAC,SAAS,IAAI,KAAK,IAAI,IAAI,CAAC,OAAO,EAAE,EAC5C,KAAK,EACL,IAAI,CAAC,MAAM,CACX,CACD,CAAA;IACF,CAAC;IAED,MAAM,SAAS,GAA8B,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,IAAI,EAAE,CAAC,EAAE,CAAA;IAE1E,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QAC1B,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,IAAI,CAAC,IAAI,CAAA;IACnC,CAAC;IAED,MAAM,OAAO,GAAG,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;IAEhE,MAAM,QAAQ,GAAG;QAChB,cAAc,EAAE,IAAI,CAAC,OAAO;QAC5B,YAAY,EAAE,IAAI,CAAC,cAAc,IAAI,IAAI;QACzC,IAAI,EAAE,IAAI,CAAC,IAAI,IAAI,GAAG,IAAI,CAAC,cAAc,gCAAgC,OAAO,qBAAqB;QACrG,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,cAAc,EAAE,eAAe,CAAC,IAAI,CAAC;QACrC,cAAc,EAAE,IAAI,CAAC,cAAc;QACnC,MAAM,EAAE,CAAC,GAAG,IAAI,EAAE,GAAG,KAAK,CAAC;QAC3B,MAAM,EAAE;YACP,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,GAAG,SAAS,CAAC,KAAK;YAC1C,GAAG,EAAE,IAAI,CAAC,MAAM,CAAC,GAAG,GAAG,SAAS,CAAC,GAAG;YACpC,IAAI,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI,GAAG,SAAS,CAAC,IAAI;SACvC;QACD,UAAU,EAAE,IAAI,CAAC,UAAU,GAAG,SAAS,CAAC,KAAK,GAAG,SAAS,CAAC,GAAG,GAAG,SAAS,CAAC,IAAI;QAC9E,6FAA6F;QAC7F,kFAAkF;QAClF,gEAAgE;QAChE,+EAA+E;QAC/E,uFAAuF;QACvF,cAAc,EACb,kGAAkG;YAClG,SAAS,KAAK,CAAC,MAAM,mCAAmC,CAAC,GAAG,IAAI,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,QAAQ,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG;KACrI,CAAA;IAED,MAAM,GAAG,GAAG,MAAM,CAAC,eAAe,CAAC,CAAA;IACnC,MAAM,kBAAkB,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAA;IAEvC,oFAAoF;IACpF,+EAA+E;IAC/E,gEAAgE;IAChE,mEAAmE;IACnE,MAAM,IAAI,GAAgB;QACzB,aAAa,EAAE,IAAI,CAAC,OAAO;QAC3B,WAAW,EAAE,IAAI,CAAC,cAAc,IAAI,IAAI;QACxC,YAAY,EAAE,IAAI,CAAC,IAAI;QACvB,SAAS,EAAE,IAAI,CAAC,SAAS;QACzB,IAAI,EAAE,QAAQ,CAAC,IAAI;QACnB,SAAS,EAAE,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE;QACnC,KAAK,EAAE,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC;YAChC,OAAO,EAAE,IAAI,CAAC,OAAO;YACrB,MAAM,EAAE,IAAI,CAAC,MAAM;YACnB,KAAK,EAAE,IAAI,CAAC,KAAK,IAAI,OAAO;YAC5B,KAAK,EAAE,aAAa,CAAC,IAAI,CAAC,OAAO,CAAC;SAClC,CAAC,CAAC;KACH,CAAA;IAED,MAAM,QAAQ,GAAG,MAAM,CAAC,iBAAiB,CAAC,CAAA;IAE1C,MAAM,kBAAkB,CAAC,IAAI,EAAE,QAAQ,CAAC,CAAA;IAExC,OAAO,CAAC,GAAG,CAAC,SAAS,QAAQ,EAAE,CAAC,CAAA;IAChC,OAAO,CAAC,GAAG,CAAC,SAAS,GAAG,EAAE,CAAC,CAAA;IAC3B,OAAO,CAAC,GAAG,CAAC,YAAY,QAAQ,CAAC,MAAM,CAAC,MAAM,KAAK,IAAI,CAAC,MAAM,gBAAgB,KAAK,CAAC,MAAM,SAAS,CAAC,CAAA;IACpG,OAAO,CAAC,GAAG,CAAC,aAAa,aAAa,CAAC,QAAQ,CAAC,MAAM,CAAC,YAAY,QAAQ,CAAC,UAAU,EAAE,CAAC,CAAA;IAEzF,0GAA0G;IAC1G,KAAK,MAAM,CAAC,KAAK,EAAE,IAAI,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,EAAE,CAAC;QAC7C,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,EAAE,KAAK,IAAI,UAAU,CAAC,MAAM,CAAA;QAE3D,OAAO,CAAC,GAAG,CAAC,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,KAAK,QAAQ,KAAK,KAAK,IAAI,CAAC,IAAI,UAAU,IAAI,CAAC,KAAK,SAAS,CAAC,CAAA;IACpG,CAAC;AACF,CAAC"}
@@ -0,0 +1,73 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ * @file Merges one source's overlay parquet files into shuffled output files.
6
+ *
7
+ * The training sampler can read a whole per-source draw from one row group,
8
+ * so a source split into per-country files would reach only one country per draw.
9
+ * A shuffle across the inputs mixes each source's countries within each row group.
10
+ */
11
+ import type { PathBuilderLike } from "path-ts";
12
+ /**
13
+ * Options for {@link mergeSourceFiles}.
14
+ */
15
+ export interface MergeSourceOptions {
16
+ /**
17
+ * The parquet files to merge, read in the order given.
18
+ */
19
+ inputs: readonly PathBuilderLike[];
20
+ /**
21
+ * The output path.
22
+ * Each written file receives a five-digit index in its stem.
23
+ */
24
+ output: PathBuilderLike;
25
+ /**
26
+ * The number of rows held in memory while shuffling.
27
+ * It defaults to the corpus writer's window.
28
+ */
29
+ windowSize?: number;
30
+ /**
31
+ * The shuffle seed.
32
+ *
33
+ * The default matches the corpus writer's seed.
34
+ * This makes rebuilds reproducible.
35
+ */
36
+ seed?: number;
37
+ /**
38
+ * The maximum rows per output file.
39
+ *
40
+ * `writeParquetFile` builds one Arrow table per file.
41
+ * Arrow's list builder overflows on very large tables.
42
+ */
43
+ rowsPerFile?: number;
44
+ }
45
+ /**
46
+ * The files that {@link mergeSourceFiles} wrote and the row counts they hold.
47
+ */
48
+ export interface MergeSourceResult {
49
+ inputs: readonly string[];
50
+ /**
51
+ * The files written, in order.
52
+ */
53
+ outputs: readonly string[];
54
+ rows: number;
55
+ /**
56
+ * Row counts per `country` value.
57
+ */
58
+ byCountry: Record<string, number>;
59
+ /**
60
+ * Row counts per `source` value.
61
+ * A successful merge has one entry.
62
+ */
63
+ bySource: Record<string, number>;
64
+ }
65
+ /**
66
+ * Merges the inputs through a windowed shuffle into one or more parquet files and reports their contents.
67
+ *
68
+ * @throws When the inputs contain more than one `source`, because a manifest
69
+ * entry records one source per file.
70
+ * The check runs after the stream ends, so files flushed before it remain on disk.
71
+ */
72
+ export declare function mergeSourceFiles(options: MergeSourceOptions): Promise<MergeSourceResult>;
73
+ //# sourceMappingURL=merge-source.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"merge-source.d.ts","sourceRoot":"","sources":["../../../lib/tools/overlay/merge-source.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAGH,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,SAAS,CAAA;AAO9C;;GAEG;AACH,MAAM,WAAW,kBAAkB;IAClC;;OAEG;IACH,MAAM,EAAE,SAAS,eAAe,EAAE,CAAA;IAClC;;;OAGG;IACH,MAAM,EAAE,eAAe,CAAA;IACvB;;;OAGG;IACH,UAAU,CAAC,EAAE,MAAM,CAAA;IACnB;;;;;OAKG;IACH,IAAI,CAAC,EAAE,MAAM,CAAA;IACb;;;;;OAKG;IACH,WAAW,CAAC,EAAE,MAAM,CAAA;CACpB;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAA;IACzB;;OAEG;IACH,OAAO,EAAE,SAAS,MAAM,EAAE,CAAA;IAC1B,IAAI,EAAE,MAAM,CAAA;IACZ;;OAEG;IACH,SAAS,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IACjC;;;OAGG;IACH,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;CAChC;AAQD;;;;;;GAMG;AACH,wBAAsB,gBAAgB,CAAC,OAAO,EAAE,kBAAkB,GAAG,OAAO,CAAC,iBAAiB,CAAC,CAkD9F"}
@@ -0,0 +1,63 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ * @file Merges one source's overlay parquet files into shuffled output files.
6
+ *
7
+ * The training sampler can read a whole per-source draw from one row group,
8
+ * so a source split into per-country files would reach only one country per draw.
9
+ * A shuffle across the inputs mixes each source's countries within each row group.
10
+ */
11
+ import { mulberry32 } from "@mailwoman/core/utils";
12
+ import { ROWS_PER_FILE } from "#parquet/schema";
13
+ import { DEFAULT_SHUFFLE_SEED, DEFAULT_SHUFFLE_WINDOW, shuffleWithinWindow } from "#parquet/shuffle";
14
+ import { openParquetRowStream } from "#parquet/streams";
15
+ import { writeParquetFile } from "#parquet/writers";
16
+ async function* readAll(inputs) {
17
+ for (const input of inputs) {
18
+ yield* openParquetRowStream(input);
19
+ }
20
+ }
21
+ /**
22
+ * Merges the inputs through a windowed shuffle into one or more parquet files and reports their contents.
23
+ *
24
+ * @throws When the inputs contain more than one `source`, because a manifest
25
+ * entry records one source per file.
26
+ * The check runs after the stream ends, so files flushed before it remain on disk.
27
+ */
28
+ export async function mergeSourceFiles(options) {
29
+ const inputs = options.inputs.map((input) => input.toString());
30
+ const random = mulberry32(options.seed ?? DEFAULT_SHUFFLE_SEED);
31
+ const rowsPerFile = options.rowsPerFile ?? ROWS_PER_FILE;
32
+ const stem = options.output.toString().replace(/\.parquet$/u, "");
33
+ const byCountry = {};
34
+ const bySource = {};
35
+ const outputs = [];
36
+ let pending = [];
37
+ let rows = 0;
38
+ const flush = async () => {
39
+ if (!pending.length)
40
+ return;
41
+ const output = `${stem}-${String(outputs.length).padStart(5, "0")}.parquet`;
42
+ await writeParquetFile(pending, output);
43
+ outputs.push(output);
44
+ pending = [];
45
+ };
46
+ for await (const row of shuffleWithinWindow(readAll(options.inputs), random, options.windowSize ?? DEFAULT_SHUFFLE_WINDOW)) {
47
+ pending.push(row);
48
+ rows++;
49
+ byCountry[row.country] = (byCountry[row.country] ?? 0) + 1;
50
+ bySource[row.source] = (bySource[row.source] ?? 0) + 1;
51
+ if (pending.length >= rowsPerFile) {
52
+ await flush();
53
+ }
54
+ }
55
+ const sources = Object.keys(bySource);
56
+ if (sources.length > 1) {
57
+ throw new Error(`the inputs carry ${sources.length} sources (${sources.join(", ")}), and a merged file's manifest entry names one. ` +
58
+ `Merge the files of one source at a time.`);
59
+ }
60
+ await flush();
61
+ return { inputs, outputs, rows, byCountry, bySource };
62
+ }
63
+ //# sourceMappingURL=merge-source.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"merge-source.js","sourceRoot":"","sources":["../../../lib/tools/overlay/merge-source.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,EAAE,UAAU,EAAE,MAAM,uBAAuB,CAAA;AAGlD,OAAO,EAAmB,aAAa,EAAE,MAAM,iBAAiB,CAAA;AAChE,OAAO,EAAE,oBAAoB,EAAE,sBAAsB,EAAE,mBAAmB,EAAE,MAAM,kBAAkB,CAAA;AACpG,OAAO,EAAE,oBAAoB,EAAE,MAAM,kBAAkB,CAAA;AACvD,OAAO,EAAE,gBAAgB,EAAE,MAAM,kBAAkB,CAAA;AAyDnD,KAAK,SAAS,CAAC,CAAC,OAAO,CAAC,MAAkC;IACzD,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;QAC5B,KAAK,CAAC,CAAC,oBAAoB,CAAa,KAAK,CAAC,CAAA;IAC/C,CAAC;AACF,CAAC;AAED;;;;;;GAMG;AACH,MAAM,CAAC,KAAK,UAAU,gBAAgB,CAAC,OAA2B;IACjE,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,QAAQ,EAAE,CAAC,CAAA;IAC9D,MAAM,MAAM,GAAG,UAAU,CAAC,OAAO,CAAC,IAAI,IAAI,oBAAoB,CAAC,CAAA;IAC/D,MAAM,WAAW,GAAG,OAAO,CAAC,WAAW,IAAI,aAAa,CAAA;IACxD,MAAM,IAAI,GAAG,OAAO,CAAC,MAAM,CAAC,QAAQ,EAAE,CAAC,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC,CAAA;IAEjE,MAAM,SAAS,GAA2B,EAAE,CAAA;IAC5C,MAAM,QAAQ,GAA2B,EAAE,CAAA;IAC3C,MAAM,OAAO,GAAa,EAAE,CAAA;IAC5B,IAAI,OAAO,GAAiB,EAAE,CAAA;IAC9B,IAAI,IAAI,GAAG,CAAC,CAAA;IAEZ,MAAM,KAAK,GAAG,KAAK,IAAmB,EAAE;QACvC,IAAI,CAAC,OAAO,CAAC,MAAM;YAAE,OAAM;QAE3B,MAAM,MAAM,GAAG,GAAG,IAAI,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,QAAQ,CAAC,CAAC,EAAE,GAAG,CAAC,UAAU,CAAA;QAE3E,MAAM,gBAAgB,CAAC,OAAO,EAAE,MAAM,CAAC,CAAA;QACvC,OAAO,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;QACpB,OAAO,GAAG,EAAE,CAAA;IACb,CAAC,CAAA;IAED,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,mBAAmB,CAC1C,OAAO,CAAC,OAAO,CAAC,MAAM,CAAC,EACvB,MAAM,EACN,OAAO,CAAC,UAAU,IAAI,sBAAsB,CAC5C,EAAE,CAAC;QACH,OAAO,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAEjB,IAAI,EAAE,CAAA;QACN,SAAS,CAAC,GAAG,CAAC,OAAO,CAAC,GAAG,CAAC,SAAS,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA;QAC1D,QAAQ,CAAC,GAAG,CAAC,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA;QAEtD,IAAI,OAAO,CAAC,MAAM,IAAI,WAAW,EAAE,CAAC;YACnC,MAAM,KAAK,EAAE,CAAA;QACd,CAAC;IACF,CAAC;IAED,MAAM,OAAO,GAAG,MAAM,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;IAErC,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACxB,MAAM,IAAI,KAAK,CACd,oBAAoB,OAAO,CAAC,MAAM,aAAa,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,mDAAmD;YACnH,0CAA0C,CAC3C,CAAA;IACF,CAAC;IAED,MAAM,KAAK,EAAE,CAAA;IAEb,OAAO,EAAE,MAAM,EAAE,OAAO,EAAE,IAAI,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAA;AACtD,CAAC"}
@@ -0,0 +1,66 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ * @file Route an already-written overlay parquet through the holdout policy.
6
+ *
7
+ * A parquet row has no `components` map, so each component is read back from `raw` over the `span_starts` /
8
+ * `span_ends` / `span_tags` triple that `alignRow` wrote.
9
+ */
10
+ import { type PathBuilderLike } from "path-ts";
11
+ import type { CanonicalRow } from "#types";
12
+ import { type CountryHoldout, type SplitName } from "#utils/split";
13
+ /**
14
+ * The fields {@linkcode holdoutComponents} reads, with the span triple optional
15
+ * so the absent-triple refusal is reachable from a test rather than only from a corrupt file.
16
+ */
17
+ export interface SpannedRow {
18
+ raw: string;
19
+ source_id: string;
20
+ span_starts?: readonly number[];
21
+ span_ends?: readonly number[];
22
+ span_tags?: readonly string[];
23
+ }
24
+ /**
25
+ * Read the components a holdout decision needs back out of one labeled row.
26
+ *
27
+ * Raises on an absent or non-parallel span triple, because treating an unreadable row
28
+ * as one with no held-out component would put it in the train split.
29
+ */
30
+ export declare function holdoutComponents(row: SpannedRow, index: number, input: string): CanonicalRow["components"];
31
+ export interface SplitSliceOptions {
32
+ /**
33
+ * The parquet to route.
34
+ * Its rows are read once and held per split in memory.
35
+ */
36
+ input: PathBuilderLike;
37
+ /**
38
+ * Where the per-split parquets are written.
39
+ * Each filename combines the input and its split.
40
+ */
41
+ outputDir: PathBuilderLike;
42
+ /**
43
+ * Defaults to `defaultHoldouts()`, which is the policy the base build applies.
44
+ */
45
+ holdouts?: Record<string, CountryHoldout>;
46
+ }
47
+ export interface SplitSliceResult {
48
+ input: string;
49
+ rows: number;
50
+ counts: Record<SplitName, number>;
51
+ /**
52
+ * The file written per split.
53
+ *
54
+ * A split with zero rows has no output file.
55
+ * An empty parquet would suggest that the writer lost its rows.
56
+ */
57
+ outputs: Partial<Record<SplitName, string>>;
58
+ }
59
+ /**
60
+ * Split one overlay parquet into up to three, by the same policy the base build applies.
61
+ *
62
+ * Each output filename combines the input name and split name.
63
+ * The function leaves the input file in place.
64
+ */
65
+ export declare function splitOverlaySlice(options: SplitSliceOptions): Promise<SplitSliceResult>;
66
+ //# sourceMappingURL=split-slice.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"split-slice.d.ts","sourceRoot":"","sources":["../../../lib/tools/overlay/split-slice.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AAGH,OAAO,EAAY,KAAK,eAAe,EAAe,MAAM,SAAS,CAAA;AAKrE,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,QAAQ,CAAA;AAC1C,OAAO,EAAE,KAAK,cAAc,EAAmB,KAAK,SAAS,EAAe,MAAM,cAAc,CAAA;AAUhG;;;GAGG;AACH,MAAM,WAAW,UAAU;IAC1B,GAAG,EAAE,MAAM,CAAA;IACX,SAAS,EAAE,MAAM,CAAA;IACjB,WAAW,CAAC,EAAE,SAAS,MAAM,EAAE,CAAA;IAC/B,SAAS,CAAC,EAAE,SAAS,MAAM,EAAE,CAAA;IAC7B,SAAS,CAAC,EAAE,SAAS,MAAM,EAAE,CAAA;CAC7B;AAED;;;;;GAKG;AACH,wBAAgB,iBAAiB,CAAC,GAAG,EAAE,UAAU,EAAE,KAAK,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,GAAG,YAAY,CAAC,YAAY,CAAC,CA0B3G;AAED,MAAM,WAAW,iBAAiB;IACjC;;;OAGG;IACH,KAAK,EAAE,eAAe,CAAA;IACtB;;;OAGG;IACH,SAAS,EAAE,eAAe,CAAA;IAC1B;;OAEG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,cAAc,CAAC,CAAA;CACzC;AAED,MAAM,WAAW,gBAAgB;IAChC,KAAK,EAAE,MAAM,CAAA;IACb,IAAI,EAAE,MAAM,CAAA;IACZ,MAAM,EAAE,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;IACjC;;;;;OAKG;IACH,OAAO,EAAE,OAAO,CAAC,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAA;CAC3C;AAED;;;;;GAKG;AACH,wBAAsB,iBAAiB,CAAC,OAAO,EAAE,iBAAiB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CAkC7F"}
@@ -0,0 +1,78 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ * @file Route an already-written overlay parquet through the holdout policy.
6
+ *
7
+ * A parquet row has no `components` map, so each component is read back from `raw` over the `span_starts` /
8
+ * `span_ends` / `span_tags` triple that `alignRow` wrote.
9
+ */
10
+ import { basename, resolvePath } from "path-ts";
11
+ import { openParquetRowStream } from "#parquet/streams";
12
+ import { writeParquetFile } from "#parquet/writers";
13
+ import { defaultHoldouts, splitForRow } from "#utils/split";
14
+ /**
15
+ * The component tags the holdout policy reads.
16
+ *
17
+ * `splitForRow` consults `region`, `postcode` and `locality` and no other field,
18
+ * so reconstructing the rest would cost a string slice per span for no decision.
19
+ */
20
+ const HOLDOUT_TAGS = new Set(["region", "postcode", "locality"]);
21
+ /**
22
+ * Read the components a holdout decision needs back out of one labeled row.
23
+ *
24
+ * Raises on an absent or non-parallel span triple, because treating an unreadable row
25
+ * as one with no held-out component would put it in the train split.
26
+ */
27
+ export function holdoutComponents(row, index, input) {
28
+ const { raw, span_starts: starts, span_ends: ends, span_tags: tags } = row;
29
+ if (!starts || !ends || !tags) {
30
+ throw new Error(`${input} row ${index} (${row.source_id}) carries no span triple, so its components cannot be read. ` +
31
+ `Re-emit the file through \`alignRow\` before splitting it.`);
32
+ }
33
+ if (starts.length !== ends.length || starts.length !== tags.length) {
34
+ throw new Error(`${input} row ${index} (${row.source_id}) has a span triple of ${starts.length}/${ends.length}/${tags.length}; ` +
35
+ `the three are parallel arrays and must agree in length.`);
36
+ }
37
+ const components = {};
38
+ for (const [span, tag] of tags.entries()) {
39
+ if (!HOLDOUT_TAGS.has(tag) || components[tag] !== undefined)
40
+ continue;
41
+ components[tag] = raw.slice(starts[span], ends[span]);
42
+ }
43
+ return components;
44
+ }
45
+ /**
46
+ * Split one overlay parquet into up to three, by the same policy the base build applies.
47
+ *
48
+ * Each output filename combines the input name and split name.
49
+ * The function leaves the input file in place.
50
+ */
51
+ export async function splitOverlaySlice(options) {
52
+ const input = options.input.toString();
53
+ const holdouts = options.holdouts ?? defaultHoldouts();
54
+ const stem = basename(input).replace(/\.parquet$/u, "");
55
+ const buckets = { train: [], val: [], test: [] };
56
+ let index = 0;
57
+ for await (const row of openParquetRowStream(input)) {
58
+ const components = holdoutComponents(row, index, input);
59
+ const split = splitForRow({ source_id: row.source_id, country: row.country, components }, holdouts);
60
+ buckets[split].push(row);
61
+ index++;
62
+ }
63
+ const outputs = {};
64
+ for (const split of ["train", "val", "test"]) {
65
+ if (!buckets[split].length)
66
+ continue;
67
+ const output = resolvePath(options.outputDir, `${stem}.${split}.parquet`);
68
+ await writeParquetFile(buckets[split], output);
69
+ outputs[split] = output;
70
+ }
71
+ return {
72
+ input,
73
+ rows: index,
74
+ counts: { train: buckets.train.length, val: buckets.val.length, test: buckets.test.length },
75
+ outputs,
76
+ };
77
+ }
78
+ //# sourceMappingURL=split-slice.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"split-slice.js","sourceRoot":"","sources":["../../../lib/tools/overlay/split-slice.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AAGH,OAAO,EAAE,QAAQ,EAAwB,WAAW,EAAE,MAAM,SAAS,CAAA;AAGrE,OAAO,EAAE,oBAAoB,EAAE,MAAM,kBAAkB,CAAA;AACvD,OAAO,EAAE,gBAAgB,EAAE,MAAM,kBAAkB,CAAA;AAEnD,OAAO,EAAuB,eAAe,EAAkB,WAAW,EAAE,MAAM,cAAc,CAAA;AAEhG;;;;;GAKG;AACH,MAAM,YAAY,GAAG,IAAI,GAAG,CAAS,CAAC,QAAQ,EAAE,UAAU,EAAE,UAAU,CAAC,CAAC,CAAA;AAcxE;;;;;GAKG;AACH,MAAM,UAAU,iBAAiB,CAAC,GAAe,EAAE,KAAa,EAAE,KAAa;IAC9E,MAAM,EAAE,GAAG,EAAE,WAAW,EAAE,MAAM,EAAE,SAAS,EAAE,IAAI,EAAE,SAAS,EAAE,IAAI,EAAE,GAAG,GAAG,CAAA;IAE1E,IAAI,CAAC,MAAM,IAAI,CAAC,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC;QAC/B,MAAM,IAAI,KAAK,CACd,GAAG,KAAK,QAAQ,KAAK,KAAK,GAAG,CAAC,SAAS,8DAA8D;YACpG,4DAA4D,CAC7D,CAAA;IACF,CAAC;IAED,IAAI,MAAM,CAAC,MAAM,KAAK,IAAI,CAAC,MAAM,IAAI,MAAM,CAAC,MAAM,KAAK,IAAI,CAAC,MAAM,EAAE,CAAC;QACpE,MAAM,IAAI,KAAK,CACd,GAAG,KAAK,QAAQ,KAAK,KAAK,GAAG,CAAC,SAAS,0BAA0B,MAAM,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,IAAI;YAC/G,yDAAyD,CAC1D,CAAA;IACF,CAAC;IAED,MAAM,UAAU,GAA2B,EAAE,CAAA;IAE7C,KAAK,MAAM,CAAC,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,CAAC,OAAO,EAAE,EAAE,CAAC;QAC1C,IAAI,CAAC,YAAY,CAAC,GAAG,CAAC,GAAG,CAAC,IAAI,UAAU,CAAC,GAAG,CAAC,KAAK,SAAS;YAAE,SAAQ;QAErE,UAAU,CAAC,GAAG,CAAC,GAAG,GAAG,CAAC,KAAK,CAAC,MAAM,CAAC,IAAI,CAAE,EAAE,IAAI,CAAC,IAAI,CAAE,CAAC,CAAA;IACxD,CAAC;IAED,OAAO,UAAmD,CAAA;AAC3D,CAAC;AAgCD;;;;;GAKG;AACH,MAAM,CAAC,KAAK,UAAU,iBAAiB,CAAC,OAA0B;IACjE,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,QAAQ,EAAE,CAAA;IACtC,MAAM,QAAQ,GAAG,OAAO,CAAC,QAAQ,IAAI,eAAe,EAAE,CAAA;IACtD,MAAM,IAAI,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAC,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC,CAAA;IAEvD,MAAM,OAAO,GAAoC,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,EAAE,EAAE,IAAI,EAAE,EAAE,EAAE,CAAA;IACjF,IAAI,KAAK,GAAG,CAAC,CAAA;IAEb,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,oBAAoB,CAAa,KAAK,CAAC,EAAE,CAAC;QACjE,MAAM,UAAU,GAAG,iBAAiB,CAAC,GAAG,EAAE,KAAK,EAAE,KAAK,CAAC,CAAA;QACvD,MAAM,KAAK,GAAG,WAAW,CAAC,EAAE,SAAS,EAAE,GAAG,CAAC,SAAS,EAAE,OAAO,EAAE,GAAG,CAAC,OAAO,EAAE,UAAU,EAAE,EAAE,QAAQ,CAAC,CAAA;QAEnG,OAAO,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAExB,KAAK,EAAE,CAAA;IACR,CAAC;IAED,MAAM,OAAO,GAAuC,EAAE,CAAA;IAEtD,KAAK,MAAM,KAAK,IAAI,CAAC,OAAO,EAAE,KAAK,EAAE,MAAM,CAAU,EAAE,CAAC;QACvD,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,MAAM;YAAE,SAAQ;QAEpC,MAAM,MAAM,GAAG,WAAW,CAAC,OAAO,CAAC,SAAS,EAAE,GAAG,IAAI,IAAI,KAAK,UAAU,CAAC,CAAA;QAEzE,MAAM,gBAAgB,CAAC,OAAO,CAAC,KAAK,CAAC,EAAE,MAAM,CAAC,CAAA;QAC9C,OAAO,CAAC,KAAK,CAAC,GAAG,MAAM,CAAA;IACxB,CAAC;IAED,OAAO;QACN,KAAK;QACL,IAAI,EAAE,KAAK;QACX,MAAM,EAAE,EAAE,KAAK,EAAE,OAAO,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,EAAE,OAAO,CAAC,GAAG,CAAC,MAAM,EAAE,IAAI,EAAE,OAAO,CAAC,IAAI,CAAC,MAAM,EAAE;QAC3F,OAAO;KACP,CAAA;AACF,CAAC"}
@@ -0,0 +1,54 @@
1
+ /**
2
+ * @copyright Sister Software
3
+ * @license AGPL-3.0
4
+ * @author Teffen Ellis, et al.
5
+ *
6
+ * Build per-script parquet files 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 files.
8
+ *
9
+ * This module sits beside `kryptonite.ts`. Both modules use the same composition pattern: take a
10
+ * base manifest, append new parquet files, 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 parquet file per script so `audit.ts` can attribute each file to its
15
+ * synthetic source without relying on filename-prefix inference.
16
+ * - Each file 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 file paths are
19
+ * rewritten from `$MAILWOMAN_DATA_ROOT/...` to `/data/...` in the combined manifest so
20
+ * all paths share one container-friendly form.
21
+ *
22
+ * See docs/records/engineering/CORPUS_V0_4_0_GENERATION.mdx for prompts, model and the
23
+ * reproducibility interface.
24
+ *
25
+ * Invoke via `mailwoman corpus slice translit \
26
+ * --jsonl /data/corpus/versioned/v0.4.0/transliteration/canonical-transliteration.jsonl \
27
+ * --base-manifest /data/corpus/versioned/v0.4.0/corpus-v0.4.0/manifest.json \
28
+ * --out-dir /data/corpus/versioned/v0.4.0`
29
+ */
30
+ import { type PathBuilderLike } from "path-ts";
31
+ export interface TranslitOverlayOptions {
32
+ jsonl: string;
33
+ baseManifest: string;
34
+ outDir: PathBuilderLike;
35
+ /**
36
+ * Default `"0.4.0"`.
37
+ */
38
+ corpusVersion?: string;
39
+ /**
40
+ * Default `"/data/"`.
41
+ */
42
+ canonicalPathPrefix?: string;
43
+ /**
44
+ * Prefix currently used in the base manifest's file paths, to be rewritten
45
+ * to {@link TranslitOverlayOptions.canonicalPathPrefix}.
46
+ *
47
+ * Defaults to `dataRootPath()` with a trailing slash — the root that wrote those paths.
48
+ * Pass it explicitly when translating a manifest generated under a different
49
+ * `$MAILWOMAN_DATA_ROOT` than the one you are running with.
50
+ */
51
+ legacyPathPrefix?: string;
52
+ }
53
+ export declare function buildTranslitOverlay(options: TranslitOverlayOptions, report?: (line: string) => void): Promise<void>;
54
+ //# sourceMappingURL=translit.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"translit.d.ts","sourceRoot":"","sources":["../../../lib/tools/overlay/translit.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAOH,OAAO,EAAe,KAAK,eAAe,EAAE,MAAM,SAAS,CAAA;AAU3D,MAAM,WAAW,sBAAsB;IACtC,KAAK,EAAE,MAAM,CAAA;IACb,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,eAAe,CAAA;IACvB;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB;;OAEG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAA;IAC5B;;;;;;;OAOG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAA;CACzB;AAkED,wBAAsB,oBAAoB,CACzC,OAAO,EAAE,sBAAsB,EAC/B,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAC7B,OAAO,CAAC,IAAI,CAAC,CAkGf"}