@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.
- package/README.md +134 -32
- package/data/PROVENANCE.md +623 -17
- package/data/address-source-register.json +14249 -0
- package/data/builds/v0.7.0-de-holdout/inputs.json +22 -0
- package/data/builds/v0.8.0-registers/inputs.json +92 -0
- package/data/license-decisions.json +1603 -0
- package/data/licenses/PROVENANCE.md +65 -0
- package/data/licenses/cdla-permissive-2.0.md +43 -0
- package/data/licenses/geonames-publication.md +34 -0
- package/data/licenses/licence-ouverte-2.0.md +36 -0
- package/data/licenses/national-address-database.md +46 -0
- package/data/licenses/nlod-2.0.md +90 -0
- package/data/licenses/sirene-publication.md +38 -0
- package/data/licenses/whosonfirst-licenses.md +65 -0
- package/data/source-resolutions.json +804 -0
- package/data/sub-venue-lexicon.json +1 -1
- package/data/training-manifests/v0.7.0-de-holdout.effective.json +477 -0
- package/data/training-manifests/v0.7.0-de-holdout.json +78 -0
- package/lib/adapters/cordis/adapter.ts +785 -0
- package/lib/adapters/dedup-key.ts +40 -0
- package/lib/adapters/errors.ts +21 -0
- package/lib/adapters/geonames/adapter.ts +43 -53
- package/lib/adapters/geonames/postal/adapter.ts +116 -0
- package/lib/adapters/gleif/adapter.ts +828 -0
- package/lib/adapters/libpostal-dictionary.ts +39 -0
- package/lib/adapters/openaddresses/adapter.ts +54 -61
- package/lib/adapters/osm/adapter.ts +356 -0
- package/lib/adapters/overture/adapter.ts +65 -41
- package/lib/adapters/po-box/adapter.ts +202 -0
- package/lib/adapters/registry.ts +79 -0
- package/lib/adapters/sha256-stream.ts +49 -0
- package/lib/adapters/source-id.ts +49 -0
- package/lib/adapters/street-line.ts +135 -0
- package/lib/adapters/wof/admin/json/adapter.ts +254 -0
- package/lib/adapters/wof/json-rows.ts +211 -0
- package/lib/adapters/wof/postalcode-json/adapter.ts +194 -0
- package/lib/adapters.ts +230 -0
- package/lib/au/adapters/gnaf/adapter.ts +144 -0
- package/lib/au/adapters/gnaf/assemble.ts +205 -0
- package/lib/be/adapters/brussels/adapter.ts +566 -0
- package/lib/be/adapters/vlaanderen/adapter.ts +670 -0
- package/lib/be/adapters/wallonie/adapter.ts +537 -0
- package/lib/be/tools/fetch/brussels.ts +239 -0
- package/lib/be/tools/fetch/wallonie.ts +274 -0
- package/lib/build/adapters.ts +110 -0
- package/lib/build/align.ts +364 -0
- package/lib/build/checkpoint.ts +131 -0
- package/lib/build/eligibility.ts +96 -0
- package/lib/build/heap.ts +53 -0
- package/lib/build/inputs.ts +81 -0
- package/lib/build/rows.ts +21 -0
- package/lib/build/types.ts +232 -0
- package/lib/build.ts +127 -315
- package/lib/cn/recipes/organizational-units.ts +178 -0
- package/lib/cz/adapters/cuzk/adapter.ts +317 -0
- package/lib/cz/recipes/pcfirst-preposition.ts +125 -0
- package/lib/cz/tools/fetch/cuzk.ts +699 -0
- package/lib/de/recipes/locale.ts +370 -0
- package/lib/directory.ts +31 -0
- package/lib/dk/adapters/inspire/adapter.ts +345 -0
- package/lib/dk/tools/fetch/inspire.ts +321 -0
- package/lib/ee/adapters/ads/adapter.ts +246 -0
- package/lib/ee/tools/fetch/ads.ts +270 -0
- package/lib/env.ts +6 -2
- package/lib/es/adapters/bizkaia/adapter.ts +388 -0
- package/lib/es/adapters/catastro/adapter.ts +343 -0
- package/lib/es/adapters/gipuzkoa/adapter.ts +263 -0
- package/lib/es/adapters/navarra/adapter.ts +289 -0
- package/lib/es/cadastre.ts +194 -0
- package/lib/es/tools/fetch/bizkaia.ts +748 -0
- package/lib/es/tools/fetch/catastro.ts +754 -0
- package/lib/es/tools/fetch/gipuzkoa.ts +336 -0
- package/lib/es/tools/fetch/navarra.ts +488 -0
- package/lib/fi/adapters/ryhti/adapter.ts +323 -0
- package/lib/fi/tools/fetch/ryhti.ts +476 -0
- package/lib/fingerprints.ts +154 -0
- package/lib/fr/adapters/annuaire-education/adapter.ts +598 -0
- package/lib/fr/adapters/ban/adapter.ts +187 -0
- package/lib/fr/adapters/finess/adapter.ts +618 -0
- package/lib/fr/adapters/finess/postal-lines.ts +157 -0
- package/lib/fr/adapters/finess/voie-type.ts +200 -0
- package/lib/fr/insee-country.ts +62 -0
- package/lib/fr/recipes/admin-split.ts +221 -0
- package/lib/fr/recipes/bare-street.ts +158 -0
- package/lib/fr/recipes/fragment.ts +251 -0
- package/lib/fr/recipes/lieudit.ts +310 -0
- package/lib/fr/recipes/order.ts +219 -0
- package/lib/fr/street-decompose.ts +67 -0
- package/lib/fr/tools/fetch/annuaire-education.ts +306 -0
- package/lib/fr/tools/fetch/ban.ts +298 -0
- package/lib/fr/tools/fetch/finess.ts +352 -0
- package/lib/gauntlet-inputs.ts +92 -0
- package/lib/index.ts +10 -4
- package/lib/inspire/address.ts +457 -0
- package/lib/inspire/archive.ts +44 -0
- package/lib/inspire/errors.ts +71 -0
- package/lib/inspire/stream.ts +262 -0
- package/lib/international/recipes/locale.ts +556 -0
- package/lib/it/adapters/anac/adapter.ts +522 -0
- package/lib/it/tools/fetch/anac.ts +394 -0
- package/lib/jp/adapters/wof-admin/adapter.ts +224 -0
- package/lib/jp/tools/fetch/houjin.ts +139 -0
- package/lib/kr/adapters/juso/label-rows.ts +239 -0
- package/lib/kr/adapters/localdata/align.ts +288 -0
- package/lib/kr/adapters/localdata/key-index.ts +177 -0
- package/lib/kr/adapters/localdata/permit-rows.ts +193 -0
- package/lib/kr/tools/fetch/juso.ts +170 -0
- package/lib/kr/tools/fetch/localdata.ts +205 -0
- package/lib/lu/adapters/bd-adresses/adapter.ts +240 -0
- package/lib/lu/tools/fetch/bd-adresses.ts +351 -0
- package/lib/lv/adapters/varis/adapter.ts +499 -0
- package/lib/lv/tools/fetch/varis.ts +214 -0
- package/lib/nl/adapters/kadaster/adapter.ts +386 -0
- package/lib/nl/recipes/postcode.ts +139 -0
- package/lib/nl/tools/fetch/kadaster.ts +298 -0
- package/lib/no/adapters/matrikkelen/adapter.ts +187 -0
- package/lib/no/recipes/fragment.ts +189 -0
- package/lib/no/recipes/street/led.ts +154 -0
- package/lib/no/recipes/street.ts +109 -0
- package/lib/no/tools/fetch/matrikkelen.ts +519 -0
- package/lib/parquet/duckdb.ts +144 -0
- package/lib/parquet/readers.ts +86 -0
- package/lib/parquet/schema.ts +214 -0
- package/lib/parquet/shuffle.ts +59 -0
- package/lib/parquet/span-sql.ts +54 -0
- package/lib/parquet/streams.ts +113 -0
- package/lib/parquet/writers.ts +451 -0
- package/lib/pl/adapters/emuia/adapter.ts +202 -0
- package/lib/pl/tools/fetch/emuia.ts +360 -0
- package/lib/recipes/anchor-absorption.ts +30 -19
- package/lib/recipes/bare/country.ts +87 -0
- package/lib/recipes/bare/postcode/eval.ts +99 -0
- package/lib/recipes/bare/postcode.ts +346 -0
- package/lib/recipes/boundary-stress.ts +36 -19
- package/lib/recipes/country-balanced.ts +67 -87
- package/lib/recipes/house-venue.ts +31 -13
- package/lib/recipes/intersection.ts +83 -61
- package/lib/recipes/po/box/cedex/au-nz.ts +82 -0
- package/lib/recipes/po/box/cedex/ca.ts +59 -0
- package/lib/recipes/po/box/cedex/fr.ts +77 -0
- package/lib/recipes/po/box/cedex/phrases.ts +109 -0
- package/lib/recipes/po/box/cedex/recipe.ts +215 -0
- package/lib/recipes/po/box/cedex/sources.ts +189 -0
- package/lib/recipes/po/box/cedex/types.ts +35 -0
- package/lib/recipes/po/box/cedex/us.ts +74 -0
- package/lib/recipes/po/box/cedex/vocabulary.ts +106 -0
- package/lib/recipes/po/box/cedex.ts +9 -0
- package/lib/recipes/po/box.ts +149 -0
- package/lib/recipes/scaffold.ts +258 -182
- package/lib/recipes/sources.ts +399 -0
- package/lib/recipes/street/affix.ts +707 -0
- package/lib/recipes/street/bare.ts +115 -0
- package/lib/recipes/street.ts +32 -8
- package/lib/recipes/sub/venue/context.ts +156 -0
- package/lib/recipes/sub/venue/render.ts +170 -0
- package/lib/recipes/sub/venue/sources.ts +588 -0
- package/lib/recipes/sub/venue.ts +801 -0
- package/lib/recipes/trailing-region.ts +70 -81
- package/lib/recipes/unit/sources.ts +44 -0
- package/lib/recipes/unit.ts +102 -90
- package/lib/recipes.ts +106 -0
- package/lib/registers.ts +278 -0
- package/lib/ro/adapters/retea-scolara/adapter.ts +264 -0
- package/lib/ro/tools/fetch/retea-scolara.ts +187 -0
- package/lib/runner.ts +220 -84
- package/lib/sg/recipes/register.ts +264 -0
- package/lib/sg/tools/fetch/acra.ts +174 -0
- package/lib/si/recipes/bare-village.ts +116 -0
- package/lib/sk/adapters/inspire/adapter.ts +395 -0
- package/lib/sk/tools/fetch/inspire.ts +402 -0
- package/lib/source-register/effective-manifest.ts +659 -0
- package/lib/source-register/register.ts +539 -0
- package/lib/source-register/training-manifest.ts +218 -0
- package/lib/source-register/types.ts +590 -0
- package/lib/source-register.ts +71 -0
- package/lib/south-asia/recipes/register.ts +233 -0
- package/lib/subvenue/harvest.ts +251 -0
- package/lib/subvenue/promotions.ts +293 -0
- package/lib/subvenue/surfaces.ts +94 -0
- package/lib/subvenue/table.ts +264 -0
- package/lib/surfaces/locale.ts +246 -0
- package/lib/synthesizers/anchor-absorption.ts +55 -91
- package/lib/synthesizers/boundary-stress.ts +49 -131
- package/lib/synthesizers/house-venue.ts +44 -102
- package/lib/synthesizers/intersection.ts +37 -26
- package/lib/synthesizers/no-street.ts +70 -119
- package/lib/synthesizers/po-box.ts +90 -105
- package/lib/synthesizers/street.ts +47 -35
- package/lib/synthesizers/utils.ts +65 -219
- package/lib/test-kit/corpus-recipe.ts +45 -28
- package/lib/test-kit.ts +116 -0
- package/lib/tools/align-canonical.ts +64 -0
- package/lib/tools/audit.ts +114 -118
- package/lib/tools/corpus-stats.ts +46 -32
- package/lib/tools/fetch/atom.ts +164 -0
- package/lib/tools/fetch/cordis.ts +318 -0
- package/lib/tools/fetch/download/manifest.ts +114 -0
- package/lib/tools/fetch/download/network.ts +250 -0
- package/lib/tools/fetch/download/retry.ts +174 -0
- package/lib/tools/fetch/download.ts +4 -471
- package/lib/tools/fetch/geonames/dump.ts +275 -0
- package/lib/tools/fetch/geonames/postal.ts +145 -0
- package/lib/tools/fetch/gleif.ts +360 -0
- package/lib/tools/fetch/header.ts +69 -0
- package/lib/tools/fetch/inspire-addresses.ts +495 -0
- package/lib/tools/fetch/openaddresses.ts +20 -109
- package/lib/tools/fetch/ourairports.ts +16 -49
- package/lib/tools/fetch/ppd.ts +10 -10
- package/lib/tools/fetch/vlaanderen-ad.ts +435 -0
- package/lib/tools/fetch/wfs-harvest.ts +943 -0
- package/lib/tools/fetch/wikidata-subvenue.ts +43 -143
- package/lib/tools/fetch/zip-archive.ts +100 -0
- package/lib/tools/fetch.ts +143 -0
- package/lib/tools/golden/expand.ts +605 -0
- package/lib/tools/golden/promote.ts +349 -0
- package/lib/tools/golden/relabel-street/directory.ts +367 -0
- package/lib/tools/golden/relabel-street/row.ts +283 -0
- package/lib/tools/golden/relabel-street.ts +9 -0
- package/lib/tools/ingest-csv.ts +45 -50
- package/lib/tools/jsonl-to-parquet.ts +59 -98
- package/lib/tools/lint/recipe-output/vocab.ts +366 -0
- package/lib/tools/lint/recipe-output.ts +482 -0
- package/lib/tools/migrate/overlay-parquet.ts +173 -0
- package/lib/tools/migrate/recipe-output.ts +153 -0
- package/lib/tools/migrate/source-names.ts +161 -0
- package/lib/tools/overlay/kryptonite.ts +144 -0
- package/lib/tools/overlay/manifest.ts +407 -0
- package/lib/tools/overlay/merge-source.ts +138 -0
- package/lib/tools/overlay/split-slice.ts +144 -0
- package/lib/tools/overlay/translit.ts +234 -0
- package/lib/tools/overture-subvenue.ts +18 -91
- package/lib/tools/postcode-triples.ts +362 -167
- package/lib/tools/source-register/build.ts +819 -0
- package/lib/tools/sub/venue/head-nouns.ts +208 -0
- package/lib/tools/sub/venue/lexicon.ts +394 -0
- package/lib/tools/sub/venue/wikidata.ts +88 -0
- package/lib/tools.ts +31 -0
- package/lib/tw/tools/fetch/gcis.ts +181 -0
- package/lib/types.ts +304 -101
- package/lib/us/adapters/fcc-bdc/adapter.ts +158 -0
- package/lib/us/adapters/state/hi-schools/adapter.ts +171 -0
- package/lib/us/adapters/state/hi-schools/workbook.ts +40 -0
- package/lib/us/adapters/state/ia-contractors/adapter.ts +148 -0
- package/lib/us/adapters/state/ny-notaries/adapter.ts +132 -0
- package/lib/us/adapters/state/tx-notaries/adapter.ts +155 -0
- package/lib/us/adapters/tiger/adapter.ts +225 -0
- package/lib/us/adapters/tiger/street-decompose.ts +73 -0
- package/lib/us/adapters/usgov/hrsa-fqhc/adapter.ts +171 -0
- package/lib/us/adapters/usgov/imls-pls/adapter.ts +140 -0
- package/lib/us/adapters/usgov/irs-bmf/adapter.ts +154 -0
- package/lib/us/adapters/usgov/nad/adapter.ts +263 -0
- package/lib/us/adapters/usgov/nppes/adapter.ts +180 -0
- package/lib/us/adapters/usgov/samhsa-treatment-locator/adapter.ts +195 -0
- package/lib/us/fips-state.ts +122 -0
- package/lib/us/tools/fetch/hrsa.ts +58 -0
- package/lib/us/tools/fetch/imls-pls.ts +131 -0
- package/lib/us/tools/fetch/nad.ts +359 -0
- package/lib/us/tools/fetch/nppes.ts +153 -0
- package/lib/us/tools/fetch/state/hi-schools.ts +167 -0
- package/lib/us/tools/fetch/state/sources.ts +146 -0
- package/lib/us/tools/fetch/tiger-full.ts +256 -0
- package/lib/utils/align.ts +43 -53
- package/lib/utils/cooccurrence-stats.ts +101 -0
- package/lib/utils/golden.ts +39 -50
- package/lib/utils/license.ts +238 -26
- package/lib/utils/split.ts +205 -88
- package/lib/utils/tokenize.ts +22 -34
- package/lib/utils/wof-json.ts +134 -87
- package/lib/utils.ts +13 -0
- package/lib/ve/recipes/reviewed-postcode-tail.ts +217 -0
- package/out/adapters/cordis/adapter.d.ts +324 -0
- package/out/adapters/cordis/adapter.d.ts.map +1 -0
- package/out/adapters/cordis/adapter.js +616 -0
- package/out/adapters/cordis/adapter.js.map +1 -0
- package/out/adapters/dedup-key.d.ts +30 -0
- package/out/adapters/dedup-key.d.ts.map +1 -0
- package/out/adapters/dedup-key.js +35 -0
- package/out/adapters/dedup-key.js.map +1 -0
- package/out/adapters/errors.d.ts +17 -0
- package/out/adapters/errors.d.ts.map +1 -0
- package/out/adapters/errors.js +20 -0
- package/out/adapters/errors.js.map +1 -0
- package/out/adapters/geonames/adapter.d.ts +26 -28
- package/out/adapters/geonames/adapter.d.ts.map +1 -1
- package/out/adapters/geonames/adapter.js +40 -50
- package/out/adapters/geonames/adapter.js.map +1 -1
- package/out/adapters/geonames/postal/adapter.d.ts +41 -0
- package/out/adapters/geonames/postal/adapter.d.ts.map +1 -0
- package/out/adapters/geonames/postal/adapter.js +102 -0
- package/out/adapters/geonames/postal/adapter.js.map +1 -0
- package/out/adapters/gleif/adapter.d.ts +293 -0
- package/out/adapters/gleif/adapter.d.ts.map +1 -0
- package/out/adapters/gleif/adapter.js +684 -0
- package/out/adapters/gleif/adapter.js.map +1 -0
- package/out/adapters/libpostal-dictionary.d.ts +22 -0
- package/out/adapters/libpostal-dictionary.d.ts.map +1 -0
- package/out/adapters/libpostal-dictionary.js +36 -0
- package/out/adapters/libpostal-dictionary.js.map +1 -0
- package/out/adapters/openaddresses/adapter.d.ts +28 -43
- package/out/adapters/openaddresses/adapter.d.ts.map +1 -1
- package/out/adapters/openaddresses/adapter.js +37 -51
- package/out/adapters/openaddresses/adapter.js.map +1 -1
- package/out/adapters/osm/adapter.d.ts +101 -0
- package/out/adapters/osm/adapter.d.ts.map +1 -0
- package/out/adapters/osm/adapter.js +260 -0
- package/out/adapters/osm/adapter.js.map +1 -0
- package/out/adapters/overture/adapter.d.ts +14 -30
- package/out/adapters/overture/adapter.d.ts.map +1 -1
- package/out/adapters/overture/adapter.js +57 -41
- package/out/adapters/overture/adapter.js.map +1 -1
- package/out/adapters/po-box/adapter.d.ts +63 -0
- package/out/adapters/po-box/adapter.d.ts.map +1 -0
- package/out/adapters/po-box/adapter.js +139 -0
- package/out/adapters/po-box/adapter.js.map +1 -0
- package/out/adapters/registry.d.ts +55 -0
- package/out/adapters/registry.d.ts.map +1 -0
- package/out/adapters/registry.js +43 -0
- package/out/adapters/registry.js.map +1 -0
- package/out/adapters/sha256-stream.d.ts +26 -0
- package/out/adapters/sha256-stream.d.ts.map +1 -0
- package/out/adapters/sha256-stream.js +34 -0
- package/out/adapters/sha256-stream.js.map +1 -0
- package/out/adapters/source-id.d.ts +33 -0
- package/out/adapters/source-id.d.ts.map +1 -0
- package/out/adapters/source-id.js +42 -0
- package/out/adapters/source-id.js.map +1 -0
- package/out/adapters/street-line.d.ts +102 -0
- package/out/adapters/street-line.d.ts.map +1 -0
- package/out/adapters/street-line.js +117 -0
- package/out/adapters/street-line.js.map +1 -0
- package/out/adapters/wof/admin/json/adapter.d.ts +88 -0
- package/out/adapters/wof/admin/json/adapter.d.ts.map +1 -0
- package/out/adapters/wof/admin/json/adapter.js +220 -0
- package/out/adapters/wof/admin/json/adapter.js.map +1 -0
- package/out/adapters/wof/json-rows.d.ts +89 -0
- package/out/adapters/wof/json-rows.d.ts.map +1 -0
- package/out/adapters/wof/json-rows.js +135 -0
- package/out/adapters/wof/json-rows.js.map +1 -0
- package/out/adapters/wof/postalcode-json/adapter.d.ts +49 -0
- package/out/adapters/wof/postalcode-json/adapter.d.ts.map +1 -0
- package/out/adapters/wof/postalcode-json/adapter.js +165 -0
- package/out/adapters/wof/postalcode-json/adapter.js.map +1 -0
- package/out/adapters.d.ts +75 -0
- package/out/adapters.d.ts.map +1 -0
- package/out/adapters.js +167 -0
- package/out/adapters.js.map +1 -0
- package/out/au/adapters/gnaf/adapter.d.ts +37 -0
- package/out/au/adapters/gnaf/adapter.d.ts.map +1 -0
- package/out/au/adapters/gnaf/adapter.js +119 -0
- package/out/au/adapters/gnaf/adapter.js.map +1 -0
- package/out/au/adapters/gnaf/assemble.d.ts +54 -0
- package/out/au/adapters/gnaf/assemble.d.ts.map +1 -0
- package/out/au/adapters/gnaf/assemble.js +134 -0
- package/out/au/adapters/gnaf/assemble.js.map +1 -0
- package/out/be/adapters/brussels/adapter.d.ts +212 -0
- package/out/be/adapters/brussels/adapter.d.ts.map +1 -0
- package/out/be/adapters/brussels/adapter.js +435 -0
- package/out/be/adapters/brussels/adapter.js.map +1 -0
- package/out/be/adapters/vlaanderen/adapter.d.ts +233 -0
- package/out/be/adapters/vlaanderen/adapter.d.ts.map +1 -0
- package/out/be/adapters/vlaanderen/adapter.js +416 -0
- package/out/be/adapters/vlaanderen/adapter.js.map +1 -0
- package/out/be/adapters/wallonie/adapter.d.ts +196 -0
- package/out/be/adapters/wallonie/adapter.d.ts.map +1 -0
- package/out/be/adapters/wallonie/adapter.js +437 -0
- package/out/be/adapters/wallonie/adapter.js.map +1 -0
- package/out/be/tools/fetch/brussels.d.ts +119 -0
- package/out/be/tools/fetch/brussels.d.ts.map +1 -0
- package/out/be/tools/fetch/brussels.js +146 -0
- package/out/be/tools/fetch/brussels.js.map +1 -0
- package/out/be/tools/fetch/wallonie.d.ts +119 -0
- package/out/be/tools/fetch/wallonie.d.ts.map +1 -0
- package/out/be/tools/fetch/wallonie.js +167 -0
- package/out/be/tools/fetch/wallonie.js.map +1 -0
- package/out/build/adapters.d.ts +33 -0
- package/out/build/adapters.d.ts.map +1 -0
- package/out/build/adapters.js +68 -0
- package/out/build/adapters.js.map +1 -0
- package/out/build/align.d.ts +73 -0
- package/out/build/align.d.ts.map +1 -0
- package/out/build/align.js +226 -0
- package/out/build/align.js.map +1 -0
- package/out/build/checkpoint.d.ts +94 -0
- package/out/build/checkpoint.d.ts.map +1 -0
- package/out/build/checkpoint.js +59 -0
- package/out/build/checkpoint.js.map +1 -0
- package/out/build/eligibility.d.ts +47 -0
- package/out/build/eligibility.d.ts.map +1 -0
- package/out/build/eligibility.js +80 -0
- package/out/build/eligibility.js.map +1 -0
- package/out/build/heap.d.ts +28 -0
- package/out/build/heap.d.ts.map +1 -0
- package/out/build/heap.js +47 -0
- package/out/build/heap.js.map +1 -0
- package/out/build/inputs.d.ts +52 -0
- package/out/build/inputs.d.ts.map +1 -0
- package/out/build/inputs.js +49 -0
- package/out/build/inputs.js.map +1 -0
- package/out/build/rows.d.ts +16 -0
- package/out/build/rows.d.ts.map +1 -0
- package/out/build/rows.js +19 -0
- package/out/build/rows.js.map +1 -0
- package/out/build/types.d.ts +220 -0
- package/out/build/types.d.ts.map +1 -0
- package/out/build/types.js +34 -0
- package/out/build/types.js.map +1 -0
- package/out/build.d.ts +6 -121
- package/out/build.d.ts.map +1 -1
- package/out/build.js +112 -207
- package/out/build.js.map +1 -1
- package/out/cn/recipes/organizational-units.d.ts +19 -0
- package/out/cn/recipes/organizational-units.d.ts.map +1 -0
- package/out/cn/recipes/organizational-units.js +138 -0
- package/out/cn/recipes/organizational-units.js.map +1 -0
- package/out/cz/adapters/cuzk/adapter.d.ts +52 -0
- package/out/cz/adapters/cuzk/adapter.d.ts.map +1 -0
- package/out/cz/adapters/cuzk/adapter.js +266 -0
- package/out/cz/adapters/cuzk/adapter.js.map +1 -0
- package/out/cz/recipes/pcfirst-preposition.d.ts +18 -0
- package/out/cz/recipes/pcfirst-preposition.d.ts.map +1 -0
- package/out/cz/recipes/pcfirst-preposition.js +110 -0
- package/out/cz/recipes/pcfirst-preposition.js.map +1 -0
- package/out/cz/tools/fetch/cuzk.d.ts +272 -0
- package/out/cz/tools/fetch/cuzk.d.ts.map +1 -0
- package/out/cz/tools/fetch/cuzk.js +446 -0
- package/out/cz/tools/fetch/cuzk.js.map +1 -0
- package/out/de/recipes/locale.d.ts +42 -0
- package/out/de/recipes/locale.d.ts.map +1 -0
- package/out/de/recipes/locale.js +299 -0
- package/out/de/recipes/locale.js.map +1 -0
- package/out/directory.d.ts +20 -0
- package/out/directory.d.ts.map +1 -0
- package/out/directory.js +27 -0
- package/out/directory.js.map +1 -0
- package/out/dk/adapters/inspire/adapter.d.ts +154 -0
- package/out/dk/adapters/inspire/adapter.d.ts.map +1 -0
- package/out/dk/adapters/inspire/adapter.js +258 -0
- package/out/dk/adapters/inspire/adapter.js.map +1 -0
- package/out/dk/tools/fetch/inspire.d.ts +100 -0
- package/out/dk/tools/fetch/inspire.d.ts.map +1 -0
- package/out/dk/tools/fetch/inspire.js +230 -0
- package/out/dk/tools/fetch/inspire.js.map +1 -0
- package/out/ee/adapters/ads/adapter.d.ts +109 -0
- package/out/ee/adapters/ads/adapter.d.ts.map +1 -0
- package/out/ee/adapters/ads/adapter.js +191 -0
- package/out/ee/adapters/ads/adapter.js.map +1 -0
- package/out/ee/tools/fetch/ads.d.ts +123 -0
- package/out/ee/tools/fetch/ads.d.ts.map +1 -0
- package/out/ee/tools/fetch/ads.js +195 -0
- package/out/ee/tools/fetch/ads.js.map +1 -0
- package/out/env.d.ts +7 -3
- package/out/env.d.ts.map +1 -1
- package/out/env.js +6 -2
- package/out/env.js.map +1 -1
- package/out/es/adapters/bizkaia/adapter.d.ts +143 -0
- package/out/es/adapters/bizkaia/adapter.d.ts.map +1 -0
- package/out/es/adapters/bizkaia/adapter.js +332 -0
- package/out/es/adapters/bizkaia/adapter.js.map +1 -0
- package/out/es/adapters/catastro/adapter.d.ts +104 -0
- package/out/es/adapters/catastro/adapter.d.ts.map +1 -0
- package/out/es/adapters/catastro/adapter.js +288 -0
- package/out/es/adapters/catastro/adapter.js.map +1 -0
- package/out/es/adapters/gipuzkoa/adapter.d.ts +91 -0
- package/out/es/adapters/gipuzkoa/adapter.d.ts.map +1 -0
- package/out/es/adapters/gipuzkoa/adapter.js +225 -0
- package/out/es/adapters/gipuzkoa/adapter.js.map +1 -0
- package/out/es/adapters/navarra/adapter.d.ts +105 -0
- package/out/es/adapters/navarra/adapter.d.ts.map +1 -0
- package/out/es/adapters/navarra/adapter.js +241 -0
- package/out/es/adapters/navarra/adapter.js.map +1 -0
- package/out/es/cadastre.d.ts +109 -0
- package/out/es/cadastre.d.ts.map +1 -0
- package/out/es/cadastre.js +130 -0
- package/out/es/cadastre.js.map +1 -0
- package/out/es/tools/fetch/bizkaia.d.ts +278 -0
- package/out/es/tools/fetch/bizkaia.d.ts.map +1 -0
- package/out/es/tools/fetch/bizkaia.js +477 -0
- package/out/es/tools/fetch/bizkaia.js.map +1 -0
- package/out/es/tools/fetch/catastro.d.ts +272 -0
- package/out/es/tools/fetch/catastro.d.ts.map +1 -0
- package/out/es/tools/fetch/catastro.js +486 -0
- package/out/es/tools/fetch/catastro.js.map +1 -0
- package/out/es/tools/fetch/gipuzkoa.d.ts +162 -0
- package/out/es/tools/fetch/gipuzkoa.d.ts.map +1 -0
- package/out/es/tools/fetch/gipuzkoa.js +209 -0
- package/out/es/tools/fetch/gipuzkoa.js.map +1 -0
- package/out/es/tools/fetch/navarra.d.ts +189 -0
- package/out/es/tools/fetch/navarra.d.ts.map +1 -0
- package/out/es/tools/fetch/navarra.js +316 -0
- package/out/es/tools/fetch/navarra.js.map +1 -0
- package/out/fi/adapters/ryhti/adapter.d.ts +115 -0
- package/out/fi/adapters/ryhti/adapter.d.ts.map +1 -0
- package/out/fi/adapters/ryhti/adapter.js +244 -0
- package/out/fi/adapters/ryhti/adapter.js.map +1 -0
- package/out/fi/tools/fetch/ryhti.d.ts +206 -0
- package/out/fi/tools/fetch/ryhti.d.ts.map +1 -0
- package/out/fi/tools/fetch/ryhti.js +327 -0
- package/out/fi/tools/fetch/ryhti.js.map +1 -0
- package/out/fingerprints.d.ts +68 -0
- package/out/fingerprints.d.ts.map +1 -0
- package/out/fingerprints.js +130 -0
- package/out/fingerprints.js.map +1 -0
- package/out/fr/adapters/annuaire-education/adapter.d.ts +236 -0
- package/out/fr/adapters/annuaire-education/adapter.d.ts.map +1 -0
- package/out/fr/adapters/annuaire-education/adapter.js +475 -0
- package/out/fr/adapters/annuaire-education/adapter.js.map +1 -0
- package/out/fr/adapters/ban/adapter.d.ts +38 -0
- package/out/fr/adapters/ban/adapter.d.ts.map +1 -0
- package/out/fr/adapters/ban/adapter.js +134 -0
- package/out/fr/adapters/ban/adapter.js.map +1 -0
- package/out/fr/adapters/finess/adapter.d.ts +276 -0
- package/out/fr/adapters/finess/adapter.d.ts.map +1 -0
- package/out/fr/adapters/finess/adapter.js +519 -0
- package/out/fr/adapters/finess/adapter.js.map +1 -0
- package/out/fr/adapters/finess/postal-lines.d.ts +61 -0
- package/out/fr/adapters/finess/postal-lines.d.ts.map +1 -0
- package/out/fr/adapters/finess/postal-lines.js +114 -0
- package/out/fr/adapters/finess/postal-lines.js.map +1 -0
- package/out/fr/adapters/finess/voie-type.d.ts +77 -0
- package/out/fr/adapters/finess/voie-type.d.ts.map +1 -0
- package/out/fr/adapters/finess/voie-type.js +172 -0
- package/out/fr/adapters/finess/voie-type.js.map +1 -0
- package/out/fr/insee-country.d.ts +45 -0
- package/out/fr/insee-country.d.ts.map +1 -0
- package/out/fr/insee-country.js +60 -0
- package/out/fr/insee-country.js.map +1 -0
- package/out/fr/recipes/admin-split.d.ts +15 -0
- package/out/fr/recipes/admin-split.d.ts.map +1 -0
- package/out/fr/recipes/admin-split.js +170 -0
- package/out/fr/recipes/admin-split.js.map +1 -0
- package/out/fr/recipes/bare-street.d.ts +13 -0
- package/out/fr/recipes/bare-street.d.ts.map +1 -0
- package/out/fr/recipes/bare-street.js +129 -0
- package/out/fr/recipes/bare-street.js.map +1 -0
- package/out/fr/recipes/fragment.d.ts +18 -0
- package/out/fr/recipes/fragment.d.ts.map +1 -0
- package/out/fr/recipes/fragment.js +197 -0
- package/out/fr/recipes/fragment.js.map +1 -0
- package/out/fr/recipes/lieudit.d.ts +30 -0
- package/out/fr/recipes/lieudit.d.ts.map +1 -0
- package/out/fr/recipes/lieudit.js +231 -0
- package/out/fr/recipes/lieudit.js.map +1 -0
- package/out/fr/recipes/order.d.ts +15 -0
- package/out/fr/recipes/order.d.ts.map +1 -0
- package/out/fr/recipes/order.js +171 -0
- package/out/fr/recipes/order.js.map +1 -0
- package/out/fr/street-decompose.d.ts +29 -0
- package/out/fr/street-decompose.d.ts.map +1 -0
- package/out/fr/street-decompose.js +52 -0
- package/out/fr/street-decompose.js.map +1 -0
- package/out/fr/tools/fetch/annuaire-education.d.ts +130 -0
- package/out/fr/tools/fetch/annuaire-education.d.ts.map +1 -0
- package/out/fr/tools/fetch/annuaire-education.js +196 -0
- package/out/fr/tools/fetch/annuaire-education.js.map +1 -0
- package/out/fr/tools/fetch/ban.d.ts +33 -0
- package/out/fr/tools/fetch/ban.d.ts.map +1 -0
- package/out/fr/tools/fetch/ban.js +248 -0
- package/out/fr/tools/fetch/ban.js.map +1 -0
- package/out/fr/tools/fetch/finess.d.ts +145 -0
- package/out/fr/tools/fetch/finess.d.ts.map +1 -0
- package/out/fr/tools/fetch/finess.js +228 -0
- package/out/fr/tools/fetch/finess.js.map +1 -0
- package/out/gauntlet-inputs.d.ts +42 -0
- package/out/gauntlet-inputs.d.ts.map +1 -0
- package/out/gauntlet-inputs.js +78 -0
- package/out/gauntlet-inputs.js.map +1 -0
- package/out/index.d.ts +10 -4
- package/out/index.d.ts.map +1 -1
- package/out/index.js +10 -4
- package/out/index.js.map +1 -1
- package/out/inspire/address.d.ts +251 -0
- package/out/inspire/address.d.ts.map +1 -0
- package/out/inspire/address.js +384 -0
- package/out/inspire/address.js.map +1 -0
- package/out/inspire/archive.d.ts +36 -0
- package/out/inspire/archive.d.ts.map +1 -0
- package/out/inspire/archive.js +42 -0
- package/out/inspire/archive.js.map +1 -0
- package/out/inspire/errors.d.ts +55 -0
- package/out/inspire/errors.d.ts.map +1 -0
- package/out/inspire/errors.js +64 -0
- package/out/inspire/errors.js.map +1 -0
- package/out/inspire/stream.d.ts +137 -0
- package/out/inspire/stream.d.ts.map +1 -0
- package/out/inspire/stream.js +125 -0
- package/out/inspire/stream.js.map +1 -0
- package/out/international/recipes/locale.d.ts +90 -0
- package/out/international/recipes/locale.d.ts.map +1 -0
- package/out/international/recipes/locale.js +419 -0
- package/out/international/recipes/locale.js.map +1 -0
- package/out/it/adapters/anac/adapter.d.ts +267 -0
- package/out/it/adapters/anac/adapter.d.ts.map +1 -0
- package/out/it/adapters/anac/adapter.js +429 -0
- package/out/it/adapters/anac/adapter.js.map +1 -0
- package/out/it/tools/fetch/anac.d.ts +174 -0
- package/out/it/tools/fetch/anac.d.ts.map +1 -0
- package/out/it/tools/fetch/anac.js +272 -0
- package/out/it/tools/fetch/anac.js.map +1 -0
- package/out/jp/adapters/wof-admin/adapter.d.ts +68 -0
- package/out/jp/adapters/wof-admin/adapter.d.ts.map +1 -0
- package/out/jp/adapters/wof-admin/adapter.js +172 -0
- package/out/jp/adapters/wof-admin/adapter.js.map +1 -0
- package/out/jp/tools/fetch/houjin.d.ts +27 -0
- package/out/jp/tools/fetch/houjin.d.ts.map +1 -0
- package/out/jp/tools/fetch/houjin.js +106 -0
- package/out/jp/tools/fetch/houjin.js.map +1 -0
- package/out/kr/adapters/juso/label-rows.d.ts +62 -0
- package/out/kr/adapters/juso/label-rows.d.ts.map +1 -0
- package/out/kr/adapters/juso/label-rows.js +172 -0
- package/out/kr/adapters/juso/label-rows.js.map +1 -0
- package/out/kr/adapters/localdata/align.d.ts +31 -0
- package/out/kr/adapters/localdata/align.d.ts.map +1 -0
- package/out/kr/adapters/localdata/align.js +212 -0
- package/out/kr/adapters/localdata/align.js.map +1 -0
- package/out/kr/adapters/localdata/key-index.d.ts +62 -0
- package/out/kr/adapters/localdata/key-index.d.ts.map +1 -0
- package/out/kr/adapters/localdata/key-index.js +127 -0
- package/out/kr/adapters/localdata/key-index.js.map +1 -0
- package/out/kr/adapters/localdata/permit-rows.d.ts +73 -0
- package/out/kr/adapters/localdata/permit-rows.d.ts.map +1 -0
- package/out/kr/adapters/localdata/permit-rows.js +136 -0
- package/out/kr/adapters/localdata/permit-rows.js.map +1 -0
- package/out/kr/tools/fetch/juso.d.ts +33 -0
- package/out/kr/tools/fetch/juso.d.ts.map +1 -0
- package/out/kr/tools/fetch/juso.js +121 -0
- package/out/kr/tools/fetch/juso.js.map +1 -0
- package/out/kr/tools/fetch/localdata.d.ts +37 -0
- package/out/kr/tools/fetch/localdata.d.ts.map +1 -0
- package/out/kr/tools/fetch/localdata.js +147 -0
- package/out/kr/tools/fetch/localdata.js.map +1 -0
- package/out/lu/adapters/bd-adresses/adapter.d.ts +109 -0
- package/out/lu/adapters/bd-adresses/adapter.d.ts.map +1 -0
- package/out/lu/adapters/bd-adresses/adapter.js +196 -0
- package/out/lu/adapters/bd-adresses/adapter.js.map +1 -0
- package/out/lu/tools/fetch/bd-adresses.d.ts +163 -0
- package/out/lu/tools/fetch/bd-adresses.d.ts.map +1 -0
- package/out/lu/tools/fetch/bd-adresses.js +223 -0
- package/out/lu/tools/fetch/bd-adresses.js.map +1 -0
- package/out/lv/adapters/varis/adapter.d.ts +283 -0
- package/out/lv/adapters/varis/adapter.d.ts.map +1 -0
- package/out/lv/adapters/varis/adapter.js +404 -0
- package/out/lv/adapters/varis/adapter.js.map +1 -0
- package/out/lv/tools/fetch/varis.d.ts +80 -0
- package/out/lv/tools/fetch/varis.d.ts.map +1 -0
- package/out/lv/tools/fetch/varis.js +133 -0
- package/out/lv/tools/fetch/varis.js.map +1 -0
- package/out/nl/adapters/kadaster/adapter.d.ts +125 -0
- package/out/nl/adapters/kadaster/adapter.d.ts.map +1 -0
- package/out/nl/adapters/kadaster/adapter.js +310 -0
- package/out/nl/adapters/kadaster/adapter.js.map +1 -0
- package/out/nl/recipes/postcode.d.ts +24 -0
- package/out/nl/recipes/postcode.d.ts.map +1 -0
- package/out/nl/recipes/postcode.js +121 -0
- package/out/nl/recipes/postcode.js.map +1 -0
- package/out/nl/tools/fetch/kadaster.d.ts +155 -0
- package/out/nl/tools/fetch/kadaster.d.ts.map +1 -0
- package/out/nl/tools/fetch/kadaster.js +179 -0
- package/out/nl/tools/fetch/kadaster.js.map +1 -0
- package/out/no/adapters/matrikkelen/adapter.d.ts +44 -0
- package/out/no/adapters/matrikkelen/adapter.d.ts.map +1 -0
- package/out/no/adapters/matrikkelen/adapter.js +143 -0
- package/out/no/adapters/matrikkelen/adapter.js.map +1 -0
- package/out/no/recipes/fragment.d.ts +11 -0
- package/out/no/recipes/fragment.d.ts.map +1 -0
- package/out/no/recipes/fragment.js +149 -0
- package/out/no/recipes/fragment.js.map +1 -0
- package/out/no/recipes/street/led.d.ts +21 -0
- package/out/no/recipes/street/led.d.ts.map +1 -0
- package/out/no/recipes/street/led.js +124 -0
- package/out/no/recipes/street/led.js.map +1 -0
- package/out/no/recipes/street.d.ts +19 -0
- package/out/no/recipes/street.d.ts.map +1 -0
- package/out/no/recipes/street.js +88 -0
- package/out/no/recipes/street.js.map +1 -0
- package/out/no/tools/fetch/matrikkelen.d.ts +251 -0
- package/out/no/tools/fetch/matrikkelen.d.ts.map +1 -0
- package/out/no/tools/fetch/matrikkelen.js +345 -0
- package/out/no/tools/fetch/matrikkelen.js.map +1 -0
- package/out/parquet/duckdb.d.ts +100 -0
- package/out/parquet/duckdb.d.ts.map +1 -0
- package/out/parquet/duckdb.js +92 -0
- package/out/parquet/duckdb.js.map +1 -0
- package/out/parquet/readers.d.ts +44 -0
- package/out/parquet/readers.d.ts.map +1 -0
- package/out/parquet/readers.js +69 -0
- package/out/parquet/readers.js.map +1 -0
- package/out/parquet/schema.d.ts +95 -0
- package/out/parquet/schema.d.ts.map +1 -0
- package/out/parquet/schema.js +157 -0
- package/out/parquet/schema.js.map +1 -0
- package/out/parquet/shuffle.d.ts +25 -0
- package/out/parquet/shuffle.d.ts.map +1 -0
- package/out/parquet/shuffle.js +44 -0
- package/out/parquet/shuffle.js.map +1 -0
- package/out/parquet/span-sql.d.ts +49 -0
- package/out/parquet/span-sql.d.ts.map +1 -0
- package/out/parquet/span-sql.js +53 -0
- package/out/parquet/span-sql.js.map +1 -0
- package/out/parquet/streams.d.ts +47 -0
- package/out/parquet/streams.d.ts.map +1 -0
- package/out/parquet/streams.js +78 -0
- package/out/parquet/streams.js.map +1 -0
- package/out/parquet/writers.d.ts +124 -0
- package/out/parquet/writers.d.ts.map +1 -0
- package/out/parquet/writers.js +251 -0
- package/out/parquet/writers.js.map +1 -0
- package/out/pl/adapters/emuia/adapter.d.ts +81 -0
- package/out/pl/adapters/emuia/adapter.d.ts.map +1 -0
- package/out/pl/adapters/emuia/adapter.js +178 -0
- package/out/pl/adapters/emuia/adapter.js.map +1 -0
- package/out/pl/tools/fetch/emuia.d.ts +165 -0
- package/out/pl/tools/fetch/emuia.d.ts.map +1 -0
- package/out/pl/tools/fetch/emuia.js +260 -0
- package/out/pl/tools/fetch/emuia.js.map +1 -0
- package/out/recipes/anchor-absorption.d.ts +9 -9
- package/out/recipes/anchor-absorption.d.ts.map +1 -1
- package/out/recipes/anchor-absorption.js +29 -19
- package/out/recipes/anchor-absorption.js.map +1 -1
- package/out/recipes/bare/country.d.ts +7 -0
- package/out/recipes/bare/country.d.ts.map +1 -0
- package/out/recipes/bare/country.js +77 -0
- package/out/recipes/bare/country.js.map +1 -0
- package/out/recipes/bare/postcode/eval.d.ts +29 -0
- package/out/recipes/bare/postcode/eval.d.ts.map +1 -0
- package/out/recipes/bare/postcode/eval.js +85 -0
- package/out/recipes/bare/postcode/eval.js.map +1 -0
- package/out/recipes/bare/postcode.d.ts +41 -0
- package/out/recipes/bare/postcode.d.ts.map +1 -0
- package/out/recipes/bare/postcode.js +279 -0
- package/out/recipes/bare/postcode.js.map +1 -0
- package/out/recipes/boundary-stress.d.ts +8 -8
- package/out/recipes/boundary-stress.d.ts.map +1 -1
- package/out/recipes/boundary-stress.js +36 -19
- package/out/recipes/boundary-stress.js.map +1 -1
- package/out/recipes/country-balanced.d.ts +3 -21
- package/out/recipes/country-balanced.d.ts.map +1 -1
- package/out/recipes/country-balanced.js +58 -76
- package/out/recipes/country-balanced.js.map +1 -1
- package/out/recipes/house-venue.d.ts +8 -6
- package/out/recipes/house-venue.d.ts.map +1 -1
- package/out/recipes/house-venue.js +29 -13
- package/out/recipes/house-venue.js.map +1 -1
- package/out/recipes/intersection.d.ts +20 -20
- package/out/recipes/intersection.d.ts.map +1 -1
- package/out/recipes/intersection.js +65 -53
- package/out/recipes/intersection.js.map +1 -1
- package/out/recipes/po/box/cedex/au-nz.d.ts +7 -0
- package/out/recipes/po/box/cedex/au-nz.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/au-nz.js +56 -0
- package/out/recipes/po/box/cedex/au-nz.js.map +1 -0
- package/out/recipes/po/box/cedex/ca.d.ts +7 -0
- package/out/recipes/po/box/cedex/ca.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/ca.js +38 -0
- package/out/recipes/po/box/cedex/ca.js.map +1 -0
- package/out/recipes/po/box/cedex/fr.d.ts +7 -0
- package/out/recipes/po/box/cedex/fr.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/fr.js +57 -0
- package/out/recipes/po/box/cedex/fr.js.map +1 -0
- package/out/recipes/po/box/cedex/phrases.d.ts +8 -0
- package/out/recipes/po/box/cedex/phrases.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/phrases.js +77 -0
- package/out/recipes/po/box/cedex/phrases.js.map +1 -0
- package/out/recipes/po/box/cedex/recipe.d.ts +9 -0
- package/out/recipes/po/box/cedex/recipe.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/recipe.js +163 -0
- package/out/recipes/po/box/cedex/recipe.js.map +1 -0
- package/out/recipes/po/box/cedex/sources.d.ts +86 -0
- package/out/recipes/po/box/cedex/sources.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/sources.js +143 -0
- package/out/recipes/po/box/cedex/sources.js.map +1 -0
- package/out/recipes/po/box/cedex/types.d.ts +31 -0
- package/out/recipes/po/box/cedex/types.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/types.js +5 -0
- package/out/recipes/po/box/cedex/types.js.map +1 -0
- package/out/recipes/po/box/cedex/us.d.ts +7 -0
- package/out/recipes/po/box/cedex/us.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/us.js +54 -0
- package/out/recipes/po/box/cedex/us.js.map +1 -0
- package/out/recipes/po/box/cedex/vocabulary.d.ts +68 -0
- package/out/recipes/po/box/cedex/vocabulary.d.ts.map +1 -0
- package/out/recipes/po/box/cedex/vocabulary.js +90 -0
- package/out/recipes/po/box/cedex/vocabulary.js.map +1 -0
- package/out/recipes/po/box/cedex.d.ts +9 -0
- package/out/recipes/po/box/cedex.d.ts.map +1 -0
- package/out/recipes/po/box/cedex.js +9 -0
- package/out/recipes/po/box/cedex.js.map +1 -0
- package/out/recipes/po/box.d.ts +18 -0
- package/out/recipes/po/box.d.ts.map +1 -0
- package/out/recipes/po/box.js +123 -0
- package/out/recipes/po/box.js.map +1 -0
- package/out/recipes/scaffold.d.ts +214 -121
- package/out/recipes/scaffold.d.ts.map +1 -1
- package/out/recipes/scaffold.js +78 -88
- package/out/recipes/scaffold.js.map +1 -1
- package/out/recipes/sources.d.ts +178 -0
- package/out/recipes/sources.d.ts.map +1 -0
- package/out/recipes/sources.js +348 -0
- package/out/recipes/sources.js.map +1 -0
- package/out/recipes/street/affix.d.ts +59 -0
- package/out/recipes/street/affix.d.ts.map +1 -0
- package/out/recipes/street/affix.js +477 -0
- package/out/recipes/street/affix.js.map +1 -0
- package/out/recipes/street/bare.d.ts +27 -0
- package/out/recipes/street/bare.d.ts.map +1 -0
- package/out/recipes/street/bare.js +98 -0
- package/out/recipes/street/bare.js.map +1 -0
- package/out/recipes/street.d.ts +6 -4
- package/out/recipes/street.d.ts.map +1 -1
- package/out/recipes/street.js +29 -8
- package/out/recipes/street.js.map +1 -1
- package/out/recipes/sub/venue/context.d.ts +32 -0
- package/out/recipes/sub/venue/context.d.ts.map +1 -0
- package/out/recipes/sub/venue/context.js +110 -0
- package/out/recipes/sub/venue/context.js.map +1 -0
- package/out/recipes/sub/venue/render.d.ts +58 -0
- package/out/recipes/sub/venue/render.d.ts.map +1 -0
- package/out/recipes/sub/venue/render.js +118 -0
- package/out/recipes/sub/venue/render.js.map +1 -0
- package/out/recipes/sub/venue/sources.d.ts +166 -0
- package/out/recipes/sub/venue/sources.d.ts.map +1 -0
- package/out/recipes/sub/venue/sources.js +419 -0
- package/out/recipes/sub/venue/sources.js.map +1 -0
- package/out/recipes/sub/venue.d.ts +176 -0
- package/out/recipes/sub/venue.d.ts.map +1 -0
- package/out/recipes/sub/venue.js +562 -0
- package/out/recipes/sub/venue.js.map +1 -0
- package/out/recipes/trailing-region.d.ts +9 -61
- package/out/recipes/trailing-region.d.ts.map +1 -1
- package/out/recipes/trailing-region.js +56 -81
- package/out/recipes/trailing-region.js.map +1 -1
- package/out/recipes/unit/sources.d.ts +28 -0
- package/out/recipes/unit/sources.d.ts.map +1 -0
- package/out/recipes/unit/sources.js +30 -0
- package/out/recipes/unit/sources.js.map +1 -0
- package/out/recipes/unit.d.ts +30 -17
- package/out/recipes/unit.d.ts.map +1 -1
- package/out/recipes/unit.js +90 -79
- package/out/recipes/unit.js.map +1 -1
- package/out/recipes.d.ts +26 -0
- package/out/recipes.d.ts.map +1 -0
- package/out/recipes.js +100 -0
- package/out/recipes.js.map +1 -0
- package/out/registers.d.ts +277 -0
- package/out/registers.d.ts.map +1 -0
- package/out/registers.js +273 -0
- package/out/registers.js.map +1 -0
- package/out/ro/adapters/retea-scolara/adapter.d.ts +114 -0
- package/out/ro/adapters/retea-scolara/adapter.d.ts.map +1 -0
- package/out/ro/adapters/retea-scolara/adapter.js +209 -0
- package/out/ro/adapters/retea-scolara/adapter.js.map +1 -0
- package/out/ro/tools/fetch/retea-scolara.d.ts +69 -0
- package/out/ro/tools/fetch/retea-scolara.d.ts.map +1 -0
- package/out/ro/tools/fetch/retea-scolara.js +117 -0
- package/out/ro/tools/fetch/retea-scolara.js.map +1 -0
- package/out/runner.d.ts +120 -60
- package/out/runner.d.ts.map +1 -1
- package/out/runner.js +130 -56
- package/out/runner.js.map +1 -1
- package/out/sg/recipes/register.d.ts +55 -0
- package/out/sg/recipes/register.d.ts.map +1 -0
- package/out/sg/recipes/register.js +206 -0
- package/out/sg/recipes/register.js.map +1 -0
- package/out/sg/tools/fetch/acra.d.ts +20 -0
- package/out/sg/tools/fetch/acra.d.ts.map +1 -0
- package/out/sg/tools/fetch/acra.js +123 -0
- package/out/sg/tools/fetch/acra.js.map +1 -0
- package/out/si/recipes/bare-village.d.ts +31 -0
- package/out/si/recipes/bare-village.d.ts.map +1 -0
- package/out/si/recipes/bare-village.js +103 -0
- package/out/si/recipes/bare-village.js.map +1 -0
- package/out/sk/adapters/inspire/adapter.d.ts +119 -0
- package/out/sk/adapters/inspire/adapter.d.ts.map +1 -0
- package/out/sk/adapters/inspire/adapter.js +318 -0
- package/out/sk/adapters/inspire/adapter.js.map +1 -0
- package/out/sk/tools/fetch/inspire.d.ts +195 -0
- package/out/sk/tools/fetch/inspire.d.ts.map +1 -0
- package/out/sk/tools/fetch/inspire.js +280 -0
- package/out/sk/tools/fetch/inspire.js.map +1 -0
- package/out/source-register/effective-manifest.d.ts +355 -0
- package/out/source-register/effective-manifest.d.ts.map +1 -0
- package/out/source-register/effective-manifest.js +355 -0
- package/out/source-register/effective-manifest.js.map +1 -0
- package/out/source-register/register.d.ts +79 -0
- package/out/source-register/register.d.ts.map +1 -0
- package/out/source-register/register.js +400 -0
- package/out/source-register/register.js.map +1 -0
- package/out/source-register/training-manifest.d.ts +111 -0
- package/out/source-register/training-manifest.d.ts.map +1 -0
- package/out/source-register/training-manifest.js +120 -0
- package/out/source-register/training-manifest.js.map +1 -0
- package/out/source-register/types.d.ts +539 -0
- package/out/source-register/types.d.ts.map +1 -0
- package/out/source-register/types.js +263 -0
- package/out/source-register/types.js.map +1 -0
- package/out/source-register.d.ts +19 -0
- package/out/source-register.d.ts.map +1 -0
- package/out/source-register.js +16 -0
- package/out/source-register.js.map +1 -0
- package/out/south-asia/recipes/register.d.ts +54 -0
- package/out/south-asia/recipes/register.d.ts.map +1 -0
- package/out/south-asia/recipes/register.js +174 -0
- package/out/south-asia/recipes/register.js.map +1 -0
- package/out/subvenue/harvest.d.ts +117 -0
- package/out/subvenue/harvest.d.ts.map +1 -0
- package/out/subvenue/harvest.js +172 -0
- package/out/subvenue/harvest.js.map +1 -0
- package/out/subvenue/promotions.d.ts +65 -0
- package/out/subvenue/promotions.d.ts.map +1 -0
- package/out/subvenue/promotions.js +226 -0
- package/out/subvenue/promotions.js.map +1 -0
- package/out/subvenue/surfaces.d.ts +39 -0
- package/out/subvenue/surfaces.d.ts.map +1 -0
- package/out/subvenue/surfaces.js +75 -0
- package/out/subvenue/surfaces.js.map +1 -0
- package/out/subvenue/table.d.ts +212 -0
- package/out/subvenue/table.d.ts.map +1 -0
- package/out/subvenue/table.js +94 -0
- package/out/subvenue/table.js.map +1 -0
- package/out/surfaces/locale.d.ts +113 -0
- package/out/surfaces/locale.d.ts.map +1 -0
- package/out/surfaces/locale.js +136 -0
- package/out/surfaces/locale.js.map +1 -0
- package/out/synthesizers/anchor-absorption.d.ts +22 -31
- package/out/synthesizers/anchor-absorption.d.ts.map +1 -1
- package/out/synthesizers/anchor-absorption.js +30 -85
- package/out/synthesizers/anchor-absorption.js.map +1 -1
- package/out/synthesizers/boundary-stress.d.ts +17 -48
- package/out/synthesizers/boundary-stress.d.ts.map +1 -1
- package/out/synthesizers/boundary-stress.js +27 -121
- package/out/synthesizers/boundary-stress.js.map +1 -1
- package/out/synthesizers/house-venue.d.ts +22 -25
- package/out/synthesizers/house-venue.d.ts.map +1 -1
- package/out/synthesizers/house-venue.js +28 -97
- package/out/synthesizers/house-venue.js.map +1 -1
- package/out/synthesizers/intersection.d.ts +10 -8
- package/out/synthesizers/intersection.d.ts.map +1 -1
- package/out/synthesizers/intersection.js +33 -26
- package/out/synthesizers/intersection.js.map +1 -1
- package/out/synthesizers/no-street.d.ts +7 -37
- package/out/synthesizers/no-street.d.ts.map +1 -1
- package/out/synthesizers/no-street.js +55 -111
- package/out/synthesizers/no-street.js.map +1 -1
- package/out/synthesizers/po-box.d.ts +46 -45
- package/out/synthesizers/po-box.d.ts.map +1 -1
- package/out/synthesizers/po-box.js +57 -97
- package/out/synthesizers/po-box.js.map +1 -1
- package/out/synthesizers/street.d.ts +18 -12
- package/out/synthesizers/street.d.ts.map +1 -1
- package/out/synthesizers/street.js +37 -30
- package/out/synthesizers/street.js.map +1 -1
- package/out/synthesizers/utils.d.ts +32 -129
- package/out/synthesizers/utils.d.ts.map +1 -1
- package/out/synthesizers/utils.js +54 -166
- package/out/synthesizers/utils.js.map +1 -1
- package/out/test-kit/corpus-recipe.d.ts +29 -23
- package/out/test-kit/corpus-recipe.d.ts.map +1 -1
- package/out/test-kit/corpus-recipe.js +28 -13
- package/out/test-kit/corpus-recipe.js.map +1 -1
- package/out/test-kit.d.ts +63 -0
- package/out/test-kit.d.ts.map +1 -0
- package/out/test-kit.js +88 -0
- package/out/test-kit.js.map +1 -0
- package/out/tools/align-canonical.d.ts +7 -0
- package/out/tools/align-canonical.d.ts.map +1 -0
- package/out/tools/align-canonical.js +52 -0
- package/out/tools/align-canonical.js.map +1 -0
- package/out/tools/audit.d.ts +18 -11
- package/out/tools/audit.d.ts.map +1 -1
- package/out/tools/audit.js +87 -95
- package/out/tools/audit.js.map +1 -1
- package/out/tools/corpus-stats.d.ts +24 -15
- package/out/tools/corpus-stats.d.ts.map +1 -1
- package/out/tools/corpus-stats.js +35 -31
- package/out/tools/corpus-stats.js.map +1 -1
- package/out/tools/fetch/atom.d.ts +115 -0
- package/out/tools/fetch/atom.d.ts.map +1 -0
- package/out/tools/fetch/atom.js +84 -0
- package/out/tools/fetch/atom.js.map +1 -0
- package/out/tools/fetch/cordis.d.ts +123 -0
- package/out/tools/fetch/cordis.d.ts.map +1 -0
- package/out/tools/fetch/cordis.js +208 -0
- package/out/tools/fetch/cordis.js.map +1 -0
- package/out/tools/fetch/download/manifest.d.ts +56 -0
- package/out/tools/fetch/download/manifest.d.ts.map +1 -0
- package/out/tools/fetch/download/manifest.js +81 -0
- package/out/tools/fetch/download/manifest.js.map +1 -0
- package/out/tools/fetch/download/network.d.ts +126 -0
- package/out/tools/fetch/download/network.d.ts.map +1 -0
- package/out/tools/fetch/download/network.js +140 -0
- package/out/tools/fetch/download/network.js.map +1 -0
- package/out/tools/fetch/download/retry.d.ts +46 -0
- package/out/tools/fetch/download/retry.d.ts.map +1 -0
- package/out/tools/fetch/download/retry.js +136 -0
- package/out/tools/fetch/download/retry.js.map +1 -0
- package/out/tools/fetch/download.d.ts +4 -176
- package/out/tools/fetch/download.d.ts.map +1 -1
- package/out/tools/fetch/download.js +4 -312
- package/out/tools/fetch/download.js.map +1 -1
- package/out/tools/fetch/geonames/dump.d.ts +109 -0
- package/out/tools/fetch/geonames/dump.d.ts.map +1 -0
- package/out/tools/fetch/geonames/dump.js +181 -0
- package/out/tools/fetch/geonames/dump.js.map +1 -0
- package/out/tools/fetch/geonames/postal.d.ts +41 -0
- package/out/tools/fetch/geonames/postal.d.ts.map +1 -0
- package/out/tools/fetch/geonames/postal.js +87 -0
- package/out/tools/fetch/geonames/postal.js.map +1 -0
- package/out/tools/fetch/gleif.d.ts +137 -0
- package/out/tools/fetch/gleif.d.ts.map +1 -0
- package/out/tools/fetch/gleif.js +220 -0
- package/out/tools/fetch/gleif.js.map +1 -0
- package/out/tools/fetch/header.d.ts +41 -0
- package/out/tools/fetch/header.d.ts.map +1 -0
- package/out/tools/fetch/header.js +59 -0
- package/out/tools/fetch/header.js.map +1 -0
- package/out/tools/fetch/inspire-addresses.d.ts +249 -0
- package/out/tools/fetch/inspire-addresses.d.ts.map +1 -0
- package/out/tools/fetch/inspire-addresses.js +276 -0
- package/out/tools/fetch/inspire-addresses.js.map +1 -0
- package/out/tools/fetch/openaddresses.d.ts +0 -3
- package/out/tools/fetch/openaddresses.d.ts.map +1 -1
- package/out/tools/fetch/openaddresses.js +18 -97
- package/out/tools/fetch/openaddresses.js.map +1 -1
- package/out/tools/fetch/ourairports.d.ts +5 -32
- package/out/tools/fetch/ourairports.d.ts.map +1 -1
- package/out/tools/fetch/ourairports.js +13 -46
- package/out/tools/fetch/ourairports.js.map +1 -1
- package/out/tools/fetch/ppd.d.ts +4 -3
- package/out/tools/fetch/ppd.d.ts.map +1 -1
- package/out/tools/fetch/ppd.js +10 -9
- package/out/tools/fetch/ppd.js.map +1 -1
- package/out/tools/fetch/vlaanderen-ad.d.ts +131 -0
- package/out/tools/fetch/vlaanderen-ad.d.ts.map +1 -0
- package/out/tools/fetch/vlaanderen-ad.js +293 -0
- package/out/tools/fetch/vlaanderen-ad.js.map +1 -0
- package/out/tools/fetch/wfs-harvest.d.ts +410 -0
- package/out/tools/fetch/wfs-harvest.d.ts.map +1 -0
- package/out/tools/fetch/wfs-harvest.js +522 -0
- package/out/tools/fetch/wfs-harvest.js.map +1 -0
- package/out/tools/fetch/wikidata-subvenue.d.ts +31 -99
- package/out/tools/fetch/wikidata-subvenue.d.ts.map +1 -1
- package/out/tools/fetch/wikidata-subvenue.js +26 -121
- package/out/tools/fetch/wikidata-subvenue.js.map +1 -1
- package/out/tools/fetch/zip-archive.d.ts +54 -0
- package/out/tools/fetch/zip-archive.d.ts.map +1 -0
- package/out/tools/fetch/zip-archive.js +58 -0
- package/out/tools/fetch/zip-archive.js.map +1 -0
- package/out/tools/fetch.d.ts +140 -0
- package/out/tools/fetch.d.ts.map +1 -0
- package/out/tools/fetch.js +136 -0
- package/out/tools/fetch.js.map +1 -0
- package/out/tools/golden/expand.d.ts +105 -0
- package/out/tools/golden/expand.d.ts.map +1 -0
- package/out/tools/golden/expand.js +420 -0
- package/out/tools/golden/expand.js.map +1 -0
- package/out/tools/golden/promote.d.ts +77 -0
- package/out/tools/golden/promote.d.ts.map +1 -0
- package/out/tools/golden/promote.js +248 -0
- package/out/tools/golden/promote.js.map +1 -0
- package/out/tools/golden/relabel-street/directory.d.ts +98 -0
- package/out/tools/golden/relabel-street/directory.d.ts.map +1 -0
- package/out/tools/golden/relabel-street/directory.js +227 -0
- package/out/tools/golden/relabel-street/directory.js.map +1 -0
- package/out/tools/golden/relabel-street/row.d.ts +69 -0
- package/out/tools/golden/relabel-street/row.d.ts.map +1 -0
- package/out/tools/golden/relabel-street/row.js +174 -0
- package/out/tools/golden/relabel-street/row.js.map +1 -0
- package/out/tools/golden/relabel-street.d.ts +9 -0
- package/out/tools/golden/relabel-street.d.ts.map +1 -0
- package/out/tools/golden/relabel-street.js +9 -0
- package/out/tools/golden/relabel-street.js.map +1 -0
- package/out/tools/ingest-csv.d.ts +19 -15
- package/out/tools/ingest-csv.d.ts.map +1 -1
- package/out/tools/ingest-csv.js +39 -45
- package/out/tools/ingest-csv.js.map +1 -1
- package/out/tools/jsonl-to-parquet.d.ts +15 -28
- package/out/tools/jsonl-to-parquet.d.ts.map +1 -1
- package/out/tools/jsonl-to-parquet.js +49 -91
- package/out/tools/jsonl-to-parquet.js.map +1 -1
- package/out/tools/lint/recipe-output/vocab.d.ts +77 -0
- package/out/tools/lint/recipe-output/vocab.d.ts.map +1 -0
- package/out/tools/lint/recipe-output/vocab.js +239 -0
- package/out/tools/lint/recipe-output/vocab.js.map +1 -0
- package/out/tools/lint/recipe-output.d.ts +73 -0
- package/out/tools/lint/recipe-output.d.ts.map +1 -0
- package/out/tools/lint/recipe-output.js +302 -0
- package/out/tools/lint/recipe-output.js.map +1 -0
- package/out/tools/migrate/overlay-parquet.d.ts +30 -0
- package/out/tools/migrate/overlay-parquet.d.ts.map +1 -0
- package/out/tools/migrate/overlay-parquet.js +141 -0
- package/out/tools/migrate/overlay-parquet.js.map +1 -0
- package/out/tools/migrate/recipe-output.d.ts +35 -0
- package/out/tools/migrate/recipe-output.d.ts.map +1 -0
- package/out/tools/migrate/recipe-output.js +124 -0
- package/out/tools/migrate/recipe-output.js.map +1 -0
- package/out/tools/migrate/source-names.d.ts +52 -0
- package/out/tools/migrate/source-names.d.ts.map +1 -0
- package/out/tools/migrate/source-names.js +105 -0
- package/out/tools/migrate/source-names.js.map +1 -0
- package/out/tools/overlay/kryptonite.d.ts +33 -0
- package/out/tools/overlay/kryptonite.d.ts.map +1 -0
- package/out/tools/overlay/kryptonite.js +108 -0
- package/out/tools/overlay/kryptonite.js.map +1 -0
- package/out/tools/overlay/manifest.d.ts +142 -0
- package/out/tools/overlay/manifest.d.ts.map +1 -0
- package/out/tools/overlay/manifest.js +260 -0
- package/out/tools/overlay/manifest.js.map +1 -0
- package/out/tools/overlay/merge-source.d.ts +73 -0
- package/out/tools/overlay/merge-source.d.ts.map +1 -0
- package/out/tools/overlay/merge-source.js +63 -0
- package/out/tools/overlay/merge-source.js.map +1 -0
- package/out/tools/overlay/split-slice.d.ts +66 -0
- package/out/tools/overlay/split-slice.d.ts.map +1 -0
- package/out/tools/overlay/split-slice.js +78 -0
- package/out/tools/overlay/split-slice.js.map +1 -0
- package/out/tools/overlay/translit.d.ts +54 -0
- package/out/tools/overlay/translit.d.ts.map +1 -0
- package/out/tools/overlay/translit.js +177 -0
- package/out/tools/overlay/translit.js.map +1 -0
- package/out/tools/overture-subvenue.d.ts +17 -76
- package/out/tools/overture-subvenue.d.ts.map +1 -1
- package/out/tools/overture-subvenue.js +7 -74
- package/out/tools/overture-subvenue.js.map +1 -1
- package/out/tools/postcode-triples.d.ts +96 -109
- package/out/tools/postcode-triples.d.ts.map +1 -1
- package/out/tools/postcode-triples.js +237 -152
- package/out/tools/postcode-triples.js.map +1 -1
- package/out/tools/source-register/build.d.ts +145 -0
- package/out/tools/source-register/build.d.ts.map +1 -0
- package/out/tools/source-register/build.js +515 -0
- package/out/tools/source-register/build.js.map +1 -0
- package/out/tools/sub/venue/head-nouns.d.ts +21 -0
- package/out/tools/sub/venue/head-nouns.d.ts.map +1 -0
- package/out/tools/sub/venue/head-nouns.js +176 -0
- package/out/tools/sub/venue/head-nouns.js.map +1 -0
- package/out/tools/sub/venue/lexicon.d.ts +129 -0
- package/out/tools/sub/venue/lexicon.d.ts.map +1 -0
- package/out/tools/sub/venue/lexicon.js +239 -0
- package/out/tools/sub/venue/lexicon.js.map +1 -0
- package/out/tools/sub/venue/wikidata.d.ts +28 -0
- package/out/tools/sub/venue/wikidata.d.ts.map +1 -0
- package/out/tools/sub/venue/wikidata.js +65 -0
- package/out/tools/sub/venue/wikidata.js.map +1 -0
- package/out/tools.d.ts +31 -0
- package/out/tools.d.ts.map +1 -0
- package/out/tools.js +31 -0
- package/out/tools.js.map +1 -0
- package/out/tw/tools/fetch/gcis.d.ts +29 -0
- package/out/tw/tools/fetch/gcis.d.ts.map +1 -0
- package/out/tw/tools/fetch/gcis.js +141 -0
- package/out/tw/tools/fetch/gcis.js.map +1 -0
- package/out/types.d.ts +270 -101
- package/out/types.d.ts.map +1 -1
- package/out/types.js +133 -9
- package/out/types.js.map +1 -1
- package/out/us/adapters/fcc-bdc/adapter.d.ts +50 -0
- package/out/us/adapters/fcc-bdc/adapter.d.ts.map +1 -0
- package/out/us/adapters/fcc-bdc/adapter.js +129 -0
- package/out/us/adapters/fcc-bdc/adapter.js.map +1 -0
- package/out/us/adapters/state/hi-schools/adapter.d.ts +47 -0
- package/out/us/adapters/state/hi-schools/adapter.d.ts.map +1 -0
- package/out/us/adapters/state/hi-schools/adapter.js +146 -0
- package/out/us/adapters/state/hi-schools/adapter.js.map +1 -0
- package/out/us/adapters/state/hi-schools/workbook.d.ts +34 -0
- package/out/us/adapters/state/hi-schools/workbook.d.ts.map +1 -0
- package/out/us/adapters/state/hi-schools/workbook.js +22 -0
- package/out/us/adapters/state/hi-schools/workbook.js.map +1 -0
- package/out/us/adapters/state/ia-contractors/adapter.d.ts +41 -0
- package/out/us/adapters/state/ia-contractors/adapter.d.ts.map +1 -0
- package/out/us/adapters/state/ia-contractors/adapter.js +117 -0
- package/out/us/adapters/state/ia-contractors/adapter.js.map +1 -0
- package/out/us/adapters/state/ny-notaries/adapter.d.ts +33 -0
- package/out/us/adapters/state/ny-notaries/adapter.d.ts.map +1 -0
- package/out/us/adapters/state/ny-notaries/adapter.js +114 -0
- package/out/us/adapters/state/ny-notaries/adapter.js.map +1 -0
- package/out/us/adapters/state/tx-notaries/adapter.d.ts +34 -0
- package/out/us/adapters/state/tx-notaries/adapter.d.ts.map +1 -0
- package/out/us/adapters/state/tx-notaries/adapter.js +125 -0
- package/out/us/adapters/state/tx-notaries/adapter.js.map +1 -0
- package/out/us/adapters/tiger/adapter.d.ts +31 -0
- package/out/us/adapters/tiger/adapter.d.ts.map +1 -0
- package/out/us/adapters/tiger/adapter.js +173 -0
- package/out/us/adapters/tiger/adapter.js.map +1 -0
- package/out/us/adapters/tiger/street-decompose.d.ts +23 -0
- package/out/us/adapters/tiger/street-decompose.d.ts.map +1 -0
- package/out/us/adapters/tiger/street-decompose.js +58 -0
- package/out/us/adapters/tiger/street-decompose.js.map +1 -0
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts +62 -0
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.js +133 -0
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.js.map +1 -0
- package/out/us/adapters/usgov/imls-pls/adapter.d.ts +37 -0
- package/out/us/adapters/usgov/imls-pls/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/imls-pls/adapter.js +113 -0
- package/out/us/adapters/usgov/imls-pls/adapter.js.map +1 -0
- package/out/us/adapters/usgov/irs-bmf/adapter.d.ts +38 -0
- package/out/us/adapters/usgov/irs-bmf/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/irs-bmf/adapter.js +126 -0
- package/out/us/adapters/usgov/irs-bmf/adapter.js.map +1 -0
- package/out/us/adapters/usgov/nad/adapter.d.ts +28 -0
- package/out/us/adapters/usgov/nad/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/nad/adapter.js +185 -0
- package/out/us/adapters/usgov/nad/adapter.js.map +1 -0
- package/out/us/adapters/usgov/nppes/adapter.d.ts +29 -0
- package/out/us/adapters/usgov/nppes/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/nppes/adapter.js +140 -0
- package/out/us/adapters/usgov/nppes/adapter.js.map +1 -0
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts +47 -0
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts.map +1 -0
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js +153 -0
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js.map +1 -0
- package/out/us/fips-state.d.ts +50 -0
- package/out/us/fips-state.d.ts.map +1 -0
- package/out/us/fips-state.js +109 -0
- package/out/us/fips-state.js.map +1 -0
- package/out/us/tools/fetch/hrsa.d.ts +18 -0
- package/out/us/tools/fetch/hrsa.d.ts.map +1 -0
- package/out/us/tools/fetch/hrsa.js +45 -0
- package/out/us/tools/fetch/hrsa.js.map +1 -0
- package/out/us/tools/fetch/imls-pls.d.ts +14 -0
- package/out/us/tools/fetch/imls-pls.d.ts.map +1 -0
- package/out/us/tools/fetch/imls-pls.js +91 -0
- package/out/us/tools/fetch/imls-pls.js.map +1 -0
- package/out/us/tools/fetch/nad.d.ts +82 -0
- package/out/us/tools/fetch/nad.d.ts.map +1 -0
- package/out/us/tools/fetch/nad.js +239 -0
- package/out/us/tools/fetch/nad.js.map +1 -0
- package/out/us/tools/fetch/nppes.d.ts +16 -0
- package/out/us/tools/fetch/nppes.d.ts.map +1 -0
- package/out/us/tools/fetch/nppes.js +114 -0
- package/out/us/tools/fetch/nppes.js.map +1 -0
- package/out/us/tools/fetch/state/hi-schools.d.ts +29 -0
- package/out/us/tools/fetch/state/hi-schools.d.ts.map +1 -0
- package/out/us/tools/fetch/state/hi-schools.js +114 -0
- package/out/us/tools/fetch/state/hi-schools.js.map +1 -0
- package/out/us/tools/fetch/state/sources.d.ts +27 -0
- package/out/us/tools/fetch/state/sources.d.ts.map +1 -0
- package/out/us/tools/fetch/state/sources.js +110 -0
- package/out/us/tools/fetch/state/sources.js.map +1 -0
- package/out/us/tools/fetch/tiger-full.d.ts +9 -0
- package/out/us/tools/fetch/tiger-full.d.ts.map +1 -0
- package/out/us/tools/fetch/tiger-full.js +178 -0
- package/out/us/tools/fetch/tiger-full.js.map +1 -0
- package/out/utils/align.d.ts +25 -20
- package/out/utils/align.d.ts.map +1 -1
- package/out/utils/align.js +27 -42
- package/out/utils/align.js.map +1 -1
- package/out/utils/cooccurrence-stats.d.ts +58 -0
- package/out/utils/cooccurrence-stats.d.ts.map +1 -0
- package/out/utils/cooccurrence-stats.js +62 -0
- package/out/utils/cooccurrence-stats.js.map +1 -0
- package/out/utils/golden.d.ts +23 -29
- package/out/utils/golden.d.ts.map +1 -1
- package/out/utils/golden.js +33 -45
- package/out/utils/golden.js.map +1 -1
- package/out/utils/license.d.ts +158 -21
- package/out/utils/license.d.ts.map +1 -1
- package/out/utils/license.js +156 -26
- package/out/utils/license.js.map +1 -1
- package/out/utils/split.d.ts +66 -70
- package/out/utils/split.d.ts.map +1 -1
- package/out/utils/split.js +158 -69
- package/out/utils/split.js.map +1 -1
- package/out/utils/tokenize.d.ts +16 -28
- package/out/utils/tokenize.d.ts.map +1 -1
- package/out/utils/tokenize.js +18 -31
- package/out/utils/tokenize.js.map +1 -1
- package/out/utils/wof-json.d.ts +88 -62
- package/out/utils/wof-json.d.ts.map +1 -1
- package/out/utils/wof-json.js +81 -65
- package/out/utils/wof-json.js.map +1 -1
- package/out/utils.d.ts +13 -0
- package/out/utils.d.ts.map +1 -0
- package/out/utils.js +13 -0
- package/out/utils.js.map +1 -0
- package/out/ve/recipes/reviewed-postcode-tail.d.ts +65 -0
- package/out/ve/recipes/reviewed-postcode-tail.d.ts.map +1 -0
- package/out/ve/recipes/reviewed-postcode-tail.js +146 -0
- package/out/ve/recipes/reviewed-postcode-tail.js.map +1 -0
- package/package.json +46 -330
- package/lib/adapters/ban/adapter.ts +0 -190
- package/lib/adapters/ban/street-decompose.ts +0 -66
- package/lib/adapters/fcc-bdc/adapter.ts +0 -174
- package/lib/adapters/geonames-postal/adapter.ts +0 -122
- package/lib/adapters/gnaf/adapter.ts +0 -149
- package/lib/adapters/gnaf/assemble.ts +0 -225
- package/lib/adapters/index.ts +0 -153
- package/lib/adapters/state-hi-schools/adapter.ts +0 -170
- package/lib/adapters/state-hi-schools/workbook.ts +0 -35
- package/lib/adapters/state-ia-contractors/adapter.ts +0 -144
- package/lib/adapters/state-ny-notaries/adapter.ts +0 -141
- package/lib/adapters/state-tx-notaries/adapter.ts +0 -157
- package/lib/adapters/synth-po-box/adapter.ts +0 -204
- package/lib/adapters/tiger/adapter.ts +0 -252
- package/lib/adapters/tiger/street-decompose.ts +0 -83
- package/lib/adapters/usgov-hrsa-fqhc/adapter.ts +0 -167
- package/lib/adapters/usgov-imls-pls/adapter.ts +0 -143
- package/lib/adapters/usgov-irs-bmf/adapter.ts +0 -151
- package/lib/adapters/usgov-nad/adapter.ts +0 -334
- package/lib/adapters/usgov-nppes/adapter.ts +0 -151
- package/lib/adapters/usgov-samhsa-treatment-locator/adapter.ts +0 -193
- package/lib/adapters/utils/index.ts +0 -307
- package/lib/adapters/wof-admin-jp/adapter.ts +0 -214
- package/lib/adapters/wof-admin-json/adapter.ts +0 -231
- package/lib/adapters/wof-json-rows.ts +0 -129
- package/lib/adapters/wof-postalcode-json/adapter.ts +0 -176
- package/lib/codex/us-fips-state.ts +0 -113
- package/lib/recipes/bare-country.ts +0 -95
- package/lib/recipes/cn-organizational-units.ts +0 -199
- package/lib/recipes/cz-pcfirst-preposition.ts +0 -112
- package/lib/recipes/fr-admin-split.ts +0 -243
- package/lib/recipes/fr-bare-street.ts +0 -171
- package/lib/recipes/fr-fragment.ts +0 -334
- package/lib/recipes/fr-lieudit.ts +0 -301
- package/lib/recipes/fr-order.ts +0 -246
- package/lib/recipes/german.ts +0 -360
- package/lib/recipes/index.ts +0 -97
- package/lib/recipes/locale.ts +0 -600
- package/lib/recipes/nl-postcode.ts +0 -117
- package/lib/recipes/no-fragment.ts +0 -213
- package/lib/recipes/no-street-led.ts +0 -135
- package/lib/recipes/no-street.ts +0 -91
- package/lib/recipes/po-box-cedex.ts +0 -976
- package/lib/recipes/po-box.ts +0 -124
- package/lib/recipes/reviewed-postcode-tail.ts +0 -190
- package/lib/recipes/si-bare-village.ts +0 -99
- package/lib/recipes/street-affix.ts +0 -805
- package/lib/recipes/street-bare.ts +0 -102
- package/lib/recipes/sub-venue-sources.ts +0 -878
- package/lib/recipes/sub-venue.ts +0 -998
- package/lib/synthesizers/german.ts +0 -260
- package/lib/test-kit/index.ts +0 -100
- package/lib/tools/align-slice.ts +0 -68
- package/lib/tools/fetch/acra-sg.ts +0 -174
- package/lib/tools/fetch/ban.ts +0 -289
- package/lib/tools/fetch/gcis-tw.ts +0 -175
- package/lib/tools/fetch/geonames-dump.ts +0 -257
- package/lib/tools/fetch/geonames-postal.ts +0 -180
- package/lib/tools/fetch/houjin-jp.ts +0 -135
- package/lib/tools/fetch/hrsa.ts +0 -59
- package/lib/tools/fetch/imls-pls.ts +0 -149
- package/lib/tools/fetch/index.ts +0 -174
- package/lib/tools/fetch/juso-kr.ts +0 -165
- package/lib/tools/fetch/localdata-kr.ts +0 -194
- package/lib/tools/fetch/nad.ts +0 -341
- package/lib/tools/fetch/nppes.ts +0 -165
- package/lib/tools/fetch/state-hi-schools.ts +0 -161
- package/lib/tools/fetch/state-sources.ts +0 -146
- package/lib/tools/fetch/tiger-full.ts +0 -308
- package/lib/tools/gb-title-case.ts +0 -69
- package/lib/tools/golden-expand.ts +0 -574
- package/lib/tools/golden-promote.ts +0 -337
- package/lib/tools/golden-relabel-street.ts +0 -686
- package/lib/tools/index.ts +0 -28
- package/lib/tools/lint-slice-vocab.ts +0 -388
- package/lib/tools/lint-slice.ts +0 -496
- package/lib/tools/overlay-manifest.ts +0 -148
- package/lib/tools/slice-kryptonite.ts +0 -137
- package/lib/tools/slice-translit.ts +0 -218
- package/lib/tools/sub-venue/harvest.ts +0 -234
- package/lib/tools/sub-venue/head-nouns.ts +0 -245
- package/lib/tools/sub-venue/surfaces.ts +0 -95
- package/lib/tools/sub-venue/table.ts +0 -281
- package/lib/tools/sub-venue/wikidata.ts +0 -85
- package/lib/tools/sub-venue-lexicon.ts +0 -429
- package/lib/tools/sub-venue-promotions.ts +0 -330
- package/lib/utils/index.ts +0 -14
- package/lib/utils/parquet.ts +0 -582
- package/lib/utils/slice-stats.ts +0 -95
- package/out/adapters/ban/adapter.d.ts +0 -44
- package/out/adapters/ban/adapter.d.ts.map +0 -1
- package/out/adapters/ban/adapter.js +0 -149
- package/out/adapters/ban/adapter.js.map +0 -1
- package/out/adapters/ban/street-decompose.d.ts +0 -28
- package/out/adapters/ban/street-decompose.d.ts.map +0 -1
- package/out/adapters/ban/street-decompose.js +0 -51
- package/out/adapters/ban/street-decompose.js.map +0 -1
- package/out/adapters/fcc-bdc/adapter.d.ts +0 -68
- package/out/adapters/fcc-bdc/adapter.d.ts.map +0 -1
- package/out/adapters/fcc-bdc/adapter.js +0 -144
- package/out/adapters/fcc-bdc/adapter.js.map +0 -1
- package/out/adapters/geonames-postal/adapter.d.ts +0 -54
- package/out/adapters/geonames-postal/adapter.d.ts.map +0 -1
- package/out/adapters/geonames-postal/adapter.js +0 -107
- package/out/adapters/geonames-postal/adapter.js.map +0 -1
- package/out/adapters/gnaf/adapter.d.ts +0 -49
- package/out/adapters/gnaf/adapter.d.ts.map +0 -1
- package/out/adapters/gnaf/adapter.js +0 -124
- package/out/adapters/gnaf/adapter.js.map +0 -1
- package/out/adapters/gnaf/assemble.d.ts +0 -65
- package/out/adapters/gnaf/assemble.d.ts.map +0 -1
- package/out/adapters/gnaf/assemble.js +0 -154
- package/out/adapters/gnaf/assemble.js.map +0 -1
- package/out/adapters/index.d.ts +0 -51
- package/out/adapters/index.d.ts.map +0 -1
- package/out/adapters/index.js +0 -95
- package/out/adapters/index.js.map +0 -1
- package/out/adapters/state-hi-schools/adapter.d.ts +0 -46
- package/out/adapters/state-hi-schools/adapter.d.ts.map +0 -1
- package/out/adapters/state-hi-schools/adapter.js +0 -145
- package/out/adapters/state-hi-schools/adapter.js.map +0 -1
- package/out/adapters/state-hi-schools/workbook.d.ts +0 -29
- package/out/adapters/state-hi-schools/workbook.d.ts.map +0 -1
- package/out/adapters/state-hi-schools/workbook.js +0 -20
- package/out/adapters/state-hi-schools/workbook.js.map +0 -1
- package/out/adapters/state-ia-contractors/adapter.d.ts +0 -34
- package/out/adapters/state-ia-contractors/adapter.d.ts.map +0 -1
- package/out/adapters/state-ia-contractors/adapter.js +0 -111
- package/out/adapters/state-ia-contractors/adapter.js.map +0 -1
- package/out/adapters/state-ny-notaries/adapter.d.ts +0 -32
- package/out/adapters/state-ny-notaries/adapter.d.ts.map +0 -1
- package/out/adapters/state-ny-notaries/adapter.js +0 -119
- package/out/adapters/state-ny-notaries/adapter.js.map +0 -1
- package/out/adapters/state-tx-notaries/adapter.d.ts +0 -33
- package/out/adapters/state-tx-notaries/adapter.d.ts.map +0 -1
- package/out/adapters/state-tx-notaries/adapter.js +0 -125
- package/out/adapters/state-tx-notaries/adapter.js.map +0 -1
- package/out/adapters/synth-po-box/adapter.d.ts +0 -67
- package/out/adapters/synth-po-box/adapter.d.ts.map +0 -1
- package/out/adapters/synth-po-box/adapter.js +0 -140
- package/out/adapters/synth-po-box/adapter.js.map +0 -1
- package/out/adapters/tiger/adapter.d.ts +0 -55
- package/out/adapters/tiger/adapter.d.ts.map +0 -1
- package/out/adapters/tiger/adapter.js +0 -201
- package/out/adapters/tiger/adapter.js.map +0 -1
- package/out/adapters/tiger/street-decompose.d.ts +0 -30
- package/out/adapters/tiger/street-decompose.d.ts.map +0 -1
- package/out/adapters/tiger/street-decompose.js +0 -68
- package/out/adapters/tiger/street-decompose.js.map +0 -1
- package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts +0 -56
- package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-hrsa-fqhc/adapter.js +0 -132
- package/out/adapters/usgov-hrsa-fqhc/adapter.js.map +0 -1
- package/out/adapters/usgov-imls-pls/adapter.d.ts +0 -36
- package/out/adapters/usgov-imls-pls/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-imls-pls/adapter.js +0 -114
- package/out/adapters/usgov-imls-pls/adapter.js.map +0 -1
- package/out/adapters/usgov-irs-bmf/adapter.d.ts +0 -37
- package/out/adapters/usgov-irs-bmf/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-irs-bmf/adapter.js +0 -120
- package/out/adapters/usgov-irs-bmf/adapter.js.map +0 -1
- package/out/adapters/usgov-nad/adapter.d.ts +0 -48
- package/out/adapters/usgov-nad/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-nad/adapter.js +0 -249
- package/out/adapters/usgov-nad/adapter.js.map +0 -1
- package/out/adapters/usgov-nppes/adapter.d.ts +0 -39
- package/out/adapters/usgov-nppes/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-nppes/adapter.js +0 -118
- package/out/adapters/usgov-nppes/adapter.js.map +0 -1
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts +0 -46
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts.map +0 -1
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.js +0 -154
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.js.map +0 -1
- package/out/adapters/utils/index.d.ts +0 -191
- package/out/adapters/utils/index.d.ts.map +0 -1
- package/out/adapters/utils/index.js +0 -206
- package/out/adapters/utils/index.js.map +0 -1
- package/out/adapters/wof-admin-jp/adapter.d.ts +0 -65
- package/out/adapters/wof-admin-jp/adapter.d.ts.map +0 -1
- package/out/adapters/wof-admin-jp/adapter.js +0 -161
- package/out/adapters/wof-admin-jp/adapter.js.map +0 -1
- package/out/adapters/wof-admin-json/adapter.d.ts +0 -79
- package/out/adapters/wof-admin-json/adapter.d.ts.map +0 -1
- package/out/adapters/wof-admin-json/adapter.js +0 -197
- package/out/adapters/wof-admin-json/adapter.js.map +0 -1
- package/out/adapters/wof-json-rows.d.ts +0 -62
- package/out/adapters/wof-json-rows.d.ts.map +0 -1
- package/out/adapters/wof-json-rows.js +0 -83
- package/out/adapters/wof-json-rows.js.map +0 -1
- package/out/adapters/wof-postalcode-json/adapter.d.ts +0 -65
- package/out/adapters/wof-postalcode-json/adapter.d.ts.map +0 -1
- package/out/adapters/wof-postalcode-json/adapter.js +0 -151
- package/out/adapters/wof-postalcode-json/adapter.js.map +0 -1
- package/out/codex/us-fips-state.d.ts +0 -41
- package/out/codex/us-fips-state.d.ts.map +0 -1
- package/out/codex/us-fips-state.js +0 -100
- package/out/codex/us-fips-state.js.map +0 -1
- package/out/recipes/bare-country.d.ts +0 -25
- package/out/recipes/bare-country.d.ts.map +0 -1
- package/out/recipes/bare-country.js +0 -88
- package/out/recipes/bare-country.js.map +0 -1
- package/out/recipes/cn-organizational-units.d.ts +0 -36
- package/out/recipes/cn-organizational-units.d.ts.map +0 -1
- package/out/recipes/cn-organizational-units.js +0 -159
- package/out/recipes/cn-organizational-units.js.map +0 -1
- package/out/recipes/cz-pcfirst-preposition.d.ts +0 -22
- package/out/recipes/cz-pcfirst-preposition.d.ts.map +0 -1
- package/out/recipes/cz-pcfirst-preposition.js +0 -101
- package/out/recipes/cz-pcfirst-preposition.js.map +0 -1
- package/out/recipes/fr-admin-split.d.ts +0 -38
- package/out/recipes/fr-admin-split.d.ts.map +0 -1
- package/out/recipes/fr-admin-split.js +0 -192
- package/out/recipes/fr-admin-split.js.map +0 -1
- package/out/recipes/fr-bare-street.d.ts +0 -28
- package/out/recipes/fr-bare-street.d.ts.map +0 -1
- package/out/recipes/fr-bare-street.js +0 -144
- package/out/recipes/fr-bare-street.js.map +0 -1
- package/out/recipes/fr-fragment.d.ts +0 -77
- package/out/recipes/fr-fragment.d.ts.map +0 -1
- package/out/recipes/fr-fragment.js +0 -277
- package/out/recipes/fr-fragment.js.map +0 -1
- package/out/recipes/fr-lieudit.d.ts +0 -32
- package/out/recipes/fr-lieudit.d.ts.map +0 -1
- package/out/recipes/fr-lieudit.js +0 -227
- package/out/recipes/fr-lieudit.js.map +0 -1
- package/out/recipes/fr-order.d.ts +0 -38
- package/out/recipes/fr-order.d.ts.map +0 -1
- package/out/recipes/fr-order.js +0 -198
- package/out/recipes/fr-order.js.map +0 -1
- package/out/recipes/german.d.ts +0 -44
- package/out/recipes/german.d.ts.map +0 -1
- package/out/recipes/german.js +0 -289
- package/out/recipes/german.js.map +0 -1
- package/out/recipes/index.d.ts +0 -24
- package/out/recipes/index.d.ts.map +0 -1
- package/out/recipes/index.js +0 -91
- package/out/recipes/index.js.map +0 -1
- package/out/recipes/locale.d.ts +0 -136
- package/out/recipes/locale.d.ts.map +0 -1
- package/out/recipes/locale.js +0 -460
- package/out/recipes/locale.js.map +0 -1
- package/out/recipes/nl-postcode.d.ts +0 -26
- package/out/recipes/nl-postcode.d.ts.map +0 -1
- package/out/recipes/nl-postcode.js +0 -103
- package/out/recipes/nl-postcode.js.map +0 -1
- package/out/recipes/no-fragment.d.ts +0 -43
- package/out/recipes/no-fragment.d.ts.map +0 -1
- package/out/recipes/no-fragment.js +0 -183
- package/out/recipes/no-fragment.js.map +0 -1
- package/out/recipes/no-street-led.d.ts +0 -24
- package/out/recipes/no-street-led.d.ts.map +0 -1
- package/out/recipes/no-street-led.js +0 -114
- package/out/recipes/no-street-led.js.map +0 -1
- package/out/recipes/no-street.d.ts +0 -18
- package/out/recipes/no-street.d.ts.map +0 -1
- package/out/recipes/no-street.js +0 -72
- package/out/recipes/no-street.js.map +0 -1
- package/out/recipes/po-box-cedex.d.ts +0 -38
- package/out/recipes/po-box-cedex.d.ts.map +0 -1
- package/out/recipes/po-box-cedex.js +0 -768
- package/out/recipes/po-box-cedex.js.map +0 -1
- package/out/recipes/po-box.d.ts +0 -17
- package/out/recipes/po-box.d.ts.map +0 -1
- package/out/recipes/po-box.js +0 -101
- package/out/recipes/po-box.js.map +0 -1
- package/out/recipes/reviewed-postcode-tail.d.ts +0 -47
- package/out/recipes/reviewed-postcode-tail.d.ts.map +0 -1
- package/out/recipes/reviewed-postcode-tail.js +0 -125
- package/out/recipes/reviewed-postcode-tail.js.map +0 -1
- package/out/recipes/si-bare-village.d.ts +0 -32
- package/out/recipes/si-bare-village.d.ts.map +0 -1
- package/out/recipes/si-bare-village.js +0 -89
- package/out/recipes/si-bare-village.js.map +0 -1
- package/out/recipes/street-affix.d.ts +0 -89
- package/out/recipes/street-affix.d.ts.map +0 -1
- package/out/recipes/street-affix.js +0 -563
- package/out/recipes/street-affix.js.map +0 -1
- package/out/recipes/street-bare.d.ts +0 -25
- package/out/recipes/street-bare.d.ts.map +0 -1
- package/out/recipes/street-bare.js +0 -87
- package/out/recipes/street-bare.js.map +0 -1
- package/out/recipes/sub-venue-sources.d.ts +0 -229
- package/out/recipes/sub-venue-sources.d.ts.map +0 -1
- package/out/recipes/sub-venue-sources.js +0 -654
- package/out/recipes/sub-venue-sources.js.map +0 -1
- package/out/recipes/sub-venue.d.ts +0 -269
- package/out/recipes/sub-venue.d.ts.map +0 -1
- package/out/recipes/sub-venue.js +0 -726
- package/out/recipes/sub-venue.js.map +0 -1
- package/out/synthesizers/german.d.ts +0 -115
- package/out/synthesizers/german.d.ts.map +0 -1
- package/out/synthesizers/german.js +0 -149
- package/out/synthesizers/german.js.map +0 -1
- package/out/test-kit/index.d.ts +0 -48
- package/out/test-kit/index.d.ts.map +0 -1
- package/out/test-kit/index.js +0 -78
- package/out/test-kit/index.js.map +0 -1
- package/out/tools/align-slice.d.ts +0 -7
- package/out/tools/align-slice.d.ts.map +0 -1
- package/out/tools/align-slice.js +0 -55
- package/out/tools/align-slice.js.map +0 -1
- package/out/tools/fetch/acra-sg.d.ts +0 -23
- package/out/tools/fetch/acra-sg.d.ts.map +0 -1
- package/out/tools/fetch/acra-sg.js +0 -126
- package/out/tools/fetch/acra-sg.js.map +0 -1
- package/out/tools/fetch/ban.d.ts +0 -31
- package/out/tools/fetch/ban.d.ts.map +0 -1
- package/out/tools/fetch/ban.js +0 -239
- package/out/tools/fetch/ban.js.map +0 -1
- package/out/tools/fetch/gcis-tw.d.ts +0 -23
- package/out/tools/fetch/gcis-tw.d.ts.map +0 -1
- package/out/tools/fetch/gcis-tw.js +0 -135
- package/out/tools/fetch/gcis-tw.js.map +0 -1
- package/out/tools/fetch/geonames-dump.d.ts +0 -91
- package/out/tools/fetch/geonames-dump.d.ts.map +0 -1
- package/out/tools/fetch/geonames-dump.js +0 -171
- package/out/tools/fetch/geonames-dump.js.map +0 -1
- package/out/tools/fetch/geonames-postal.d.ts +0 -76
- package/out/tools/fetch/geonames-postal.d.ts.map +0 -1
- package/out/tools/fetch/geonames-postal.js +0 -126
- package/out/tools/fetch/geonames-postal.js.map +0 -1
- package/out/tools/fetch/houjin-jp.d.ts +0 -24
- package/out/tools/fetch/houjin-jp.d.ts.map +0 -1
- package/out/tools/fetch/houjin-jp.js +0 -102
- package/out/tools/fetch/houjin-jp.js.map +0 -1
- package/out/tools/fetch/hrsa.d.ts +0 -18
- package/out/tools/fetch/hrsa.d.ts.map +0 -1
- package/out/tools/fetch/hrsa.js +0 -46
- package/out/tools/fetch/hrsa.js.map +0 -1
- package/out/tools/fetch/imls-pls.d.ts +0 -25
- package/out/tools/fetch/imls-pls.d.ts.map +0 -1
- package/out/tools/fetch/imls-pls.js +0 -108
- package/out/tools/fetch/imls-pls.js.map +0 -1
- package/out/tools/fetch/index.d.ts +0 -171
- package/out/tools/fetch/index.d.ts.map +0 -1
- package/out/tools/fetch/index.js +0 -170
- package/out/tools/fetch/index.js.map +0 -1
- package/out/tools/fetch/juso-kr.d.ts +0 -30
- package/out/tools/fetch/juso-kr.d.ts.map +0 -1
- package/out/tools/fetch/juso-kr.js +0 -120
- package/out/tools/fetch/juso-kr.js.map +0 -1
- package/out/tools/fetch/localdata-kr.d.ts +0 -30
- package/out/tools/fetch/localdata-kr.d.ts.map +0 -1
- package/out/tools/fetch/localdata-kr.js +0 -139
- package/out/tools/fetch/localdata-kr.js.map +0 -1
- package/out/tools/fetch/nad.d.ts +0 -70
- package/out/tools/fetch/nad.d.ts.map +0 -1
- package/out/tools/fetch/nad.js +0 -234
- package/out/tools/fetch/nad.js.map +0 -1
- package/out/tools/fetch/nppes.d.ts +0 -26
- package/out/tools/fetch/nppes.d.ts.map +0 -1
- package/out/tools/fetch/nppes.js +0 -126
- package/out/tools/fetch/nppes.js.map +0 -1
- package/out/tools/fetch/state-hi-schools.d.ts +0 -26
- package/out/tools/fetch/state-hi-schools.d.ts.map +0 -1
- package/out/tools/fetch/state-hi-schools.js +0 -112
- package/out/tools/fetch/state-hi-schools.js.map +0 -1
- package/out/tools/fetch/state-sources.d.ts +0 -26
- package/out/tools/fetch/state-sources.d.ts.map +0 -1
- package/out/tools/fetch/state-sources.js +0 -111
- package/out/tools/fetch/state-sources.js.map +0 -1
- package/out/tools/fetch/tiger-full.d.ts +0 -22
- package/out/tools/fetch/tiger-full.d.ts.map +0 -1
- package/out/tools/fetch/tiger-full.js +0 -217
- package/out/tools/fetch/tiger-full.js.map +0 -1
- package/out/tools/gb-title-case.d.ts +0 -2
- package/out/tools/gb-title-case.d.ts.map +0 -1
- package/out/tools/gb-title-case.js +0 -65
- package/out/tools/gb-title-case.js.map +0 -1
- package/out/tools/golden-expand.d.ts +0 -90
- package/out/tools/golden-expand.d.ts.map +0 -1
- package/out/tools/golden-expand.js +0 -406
- package/out/tools/golden-expand.js.map +0 -1
- package/out/tools/golden-promote.d.ts +0 -72
- package/out/tools/golden-promote.d.ts.map +0 -1
- package/out/tools/golden-promote.js +0 -240
- package/out/tools/golden-promote.js.map +0 -1
- package/out/tools/golden-relabel-street.d.ts +0 -196
- package/out/tools/golden-relabel-street.d.ts.map +0 -1
- package/out/tools/golden-relabel-street.js +0 -439
- package/out/tools/golden-relabel-street.js.map +0 -1
- package/out/tools/index.d.ts +0 -28
- package/out/tools/index.d.ts.map +0 -1
- package/out/tools/index.js +0 -28
- package/out/tools/index.js.map +0 -1
- package/out/tools/lint-slice-vocab.d.ts +0 -90
- package/out/tools/lint-slice-vocab.d.ts.map +0 -1
- package/out/tools/lint-slice-vocab.js +0 -267
- package/out/tools/lint-slice-vocab.js.map +0 -1
- package/out/tools/lint-slice.d.ts +0 -93
- package/out/tools/lint-slice.d.ts.map +0 -1
- package/out/tools/lint-slice.js +0 -322
- package/out/tools/lint-slice.js.map +0 -1
- package/out/tools/overlay-manifest.d.ts +0 -39
- package/out/tools/overlay-manifest.d.ts.map +0 -1
- package/out/tools/overlay-manifest.js +0 -90
- package/out/tools/overlay-manifest.js.map +0 -1
- package/out/tools/slice-kryptonite.d.ts +0 -32
- package/out/tools/slice-kryptonite.d.ts.map +0 -1
- package/out/tools/slice-kryptonite.js +0 -99
- package/out/tools/slice-kryptonite.js.map +0 -1
- package/out/tools/slice-translit.d.ts +0 -51
- package/out/tools/slice-translit.d.ts.map +0 -1
- package/out/tools/slice-translit.js +0 -166
- package/out/tools/slice-translit.js.map +0 -1
- package/out/tools/sub-venue/harvest.d.ts +0 -100
- package/out/tools/sub-venue/harvest.d.ts.map +0 -1
- package/out/tools/sub-venue/harvest.js +0 -168
- package/out/tools/sub-venue/harvest.js.map +0 -1
- package/out/tools/sub-venue/head-nouns.d.ts +0 -50
- package/out/tools/sub-venue/head-nouns.d.ts.map +0 -1
- package/out/tools/sub-venue/head-nouns.js +0 -213
- package/out/tools/sub-venue/head-nouns.js.map +0 -1
- package/out/tools/sub-venue/surfaces.d.ts +0 -45
- package/out/tools/sub-venue/surfaces.d.ts.map +0 -1
- package/out/tools/sub-venue/surfaces.js +0 -77
- package/out/tools/sub-venue/surfaces.js.map +0 -1
- package/out/tools/sub-venue/table.d.ts +0 -229
- package/out/tools/sub-venue/table.d.ts.map +0 -1
- package/out/tools/sub-venue/table.js +0 -108
- package/out/tools/sub-venue/table.js.map +0 -1
- package/out/tools/sub-venue/wikidata.d.ts +0 -26
- package/out/tools/sub-venue/wikidata.d.ts.map +0 -1
- package/out/tools/sub-venue/wikidata.js +0 -62
- package/out/tools/sub-venue/wikidata.js.map +0 -1
- package/out/tools/sub-venue-lexicon.d.ts +0 -156
- package/out/tools/sub-venue-lexicon.d.ts.map +0 -1
- package/out/tools/sub-venue-lexicon.js +0 -283
- package/out/tools/sub-venue-lexicon.js.map +0 -1
- package/out/tools/sub-venue-promotions.d.ts +0 -94
- package/out/tools/sub-venue-promotions.d.ts.map +0 -1
- package/out/tools/sub-venue-promotions.js +0 -266
- package/out/tools/sub-venue-promotions.js.map +0 -1
- package/out/utils/index.d.ts +0 -14
- package/out/utils/index.d.ts.map +0 -1
- package/out/utils/index.js +0 -14
- package/out/utils/index.js.map +0 -1
- package/out/utils/parquet.d.ts +0 -178
- package/out/utils/parquet.d.ts.map +0 -1
- package/out/utils/parquet.js +0 -371
- package/out/utils/parquet.js.map +0 -1
- package/out/utils/slice-stats.d.ts +0 -52
- package/out/utils/slice-stats.d.ts.map +0 -1
- package/out/utils/slice-stats.js +0 -58
- 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"}
|