@mailwoman/corpus 10.0.0 → 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 +120 -34
- 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 +42 -52
- package/lib/adapters/geonames/postal/adapter.ts +31 -37
- package/lib/adapters/gleif/adapter.ts +828 -0
- package/lib/adapters/libpostal-dictionary.ts +39 -0
- package/lib/adapters/openaddresses/adapter.ts +51 -55
- package/lib/adapters/osm/adapter.ts +148 -70
- package/lib/adapters/overture/adapter.ts +52 -43
- 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 +48 -33
- package/lib/adapters/wof/json-rows.ts +56 -26
- package/lib/adapters/wof/postalcode-json/adapter.ts +57 -59
- package/lib/adapters.ts +230 -0
- package/lib/au/adapters/gnaf/adapter.ts +35 -43
- package/lib/au/adapters/gnaf/assemble.ts +31 -51
- 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 +20 -41
- package/lib/cz/adapters/cuzk/adapter.ts +317 -0
- package/lib/cz/recipes/pcfirst-preposition.ts +40 -27
- package/lib/cz/tools/fetch/cuzk.ts +699 -0
- package/lib/de/recipes/locale.ts +68 -58
- 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 +87 -59
- 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 +38 -60
- package/lib/fr/recipes/bare-street.ts +37 -50
- package/lib/fr/recipes/fragment.ts +59 -142
- package/lib/fr/recipes/lieudit.ts +109 -73
- package/lib/fr/recipes/order.ts +32 -59
- package/lib/fr/{adapters/ban/street-decompose.ts → street-decompose.ts} +4 -3
- package/lib/fr/tools/fetch/annuaire-education.ts +306 -0
- package/lib/fr/tools/fetch/ban.ts +28 -19
- package/lib/fr/tools/fetch/finess.ts +352 -0
- package/lib/gauntlet-inputs.ts +92 -0
- package/lib/index.ts +9 -3
- 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 +134 -176
- 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 +27 -17
- package/lib/jp/tools/fetch/houjin.ts +16 -18
- package/lib/kr/adapters/juso/label-rows.ts +24 -44
- package/lib/kr/adapters/localdata/align.ts +20 -31
- package/lib/kr/adapters/localdata/key-index.ts +17 -25
- package/lib/kr/adapters/localdata/permit-rows.ts +33 -26
- package/lib/kr/tools/fetch/juso.ts +19 -21
- package/lib/kr/tools/fetch/localdata.ts +28 -29
- 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 +46 -24
- package/lib/nl/tools/fetch/kadaster.ts +298 -0
- package/lib/no/adapters/matrikkelen/adapter.ts +187 -0
- package/lib/no/recipes/fragment.ts +34 -58
- package/lib/no/recipes/street/led.ts +44 -25
- package/lib/no/recipes/{street/index.ts → street.ts} +29 -11
- 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 +24 -32
- package/lib/recipes/bare/postcode/eval.ts +1 -1
- package/lib/recipes/bare/postcode.ts +346 -0
- package/lib/recipes/boundary-stress.ts +36 -19
- package/lib/recipes/country-balanced.ts +66 -86
- package/lib/recipes/house-venue.ts +31 -13
- package/lib/recipes/intersection.ts +82 -60
- package/lib/recipes/po/box/cedex/au-nz.ts +3 -3
- package/lib/recipes/po/box/cedex/ca.ts +3 -2
- package/lib/recipes/po/box/cedex/fr.ts +4 -3
- package/lib/recipes/po/box/cedex/phrases.ts +7 -6
- package/lib/recipes/po/box/cedex/recipe.ts +28 -16
- package/lib/recipes/po/box/cedex/sources.ts +3 -3
- package/lib/recipes/po/box/cedex/us.ts +7 -2
- package/lib/recipes/po/{box/index.ts → box.ts} +39 -18
- package/lib/recipes/scaffold.ts +252 -153
- package/lib/recipes/sources.ts +399 -0
- package/lib/recipes/street/affix.ts +54 -154
- package/lib/recipes/street/bare.ts +28 -15
- package/lib/recipes/street.ts +32 -8
- package/lib/recipes/sub/venue/context.ts +30 -23
- package/lib/recipes/sub/venue/render.ts +170 -0
- package/lib/recipes/sub/venue/sources.ts +68 -219
- package/lib/recipes/sub/venue.ts +801 -0
- package/lib/recipes/trailing-region.ts +69 -85
- package/lib/recipes/unit/sources.ts +44 -0
- package/lib/recipes/unit.ts +101 -89
- package/lib/{recipes/index.ts → recipes.ts} +13 -11
- 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 +46 -52
- package/lib/sg/tools/fetch/acra.ts +20 -25
- package/lib/si/recipes/bare-village.ts +34 -17
- 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 +45 -26
- package/lib/subvenue/harvest.ts +251 -0
- package/lib/{tools/sub/venue → subvenue}/promotions.ts +33 -70
- 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 +46 -34
- package/lib/synthesizers/utils.ts +64 -218
- 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 +113 -118
- package/lib/tools/corpus-stats.ts +44 -30
- package/lib/tools/fetch/atom.ts +164 -0
- package/lib/tools/fetch/cordis.ts +318 -0
- package/lib/tools/fetch/download/manifest.ts +34 -22
- package/lib/tools/fetch/download/network.ts +55 -31
- package/lib/tools/fetch/download/retry.ts +26 -25
- package/lib/tools/fetch/geonames/dump.ts +72 -54
- package/lib/tools/fetch/geonames/postal.ts +27 -62
- 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 +22 -111
- package/lib/tools/fetch/ourairports.ts +18 -51
- package/lib/tools/fetch/ppd.ts +10 -8
- 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 +45 -145
- package/lib/tools/fetch/zip-archive.ts +100 -0
- package/lib/tools/fetch.ts +143 -0
- package/lib/tools/golden/expand.ts +74 -43
- package/lib/tools/golden/promote.ts +41 -32
- package/lib/tools/golden/relabel-street/directory.ts +63 -36
- package/lib/tools/golden/relabel-street/row.ts +28 -106
- package/lib/tools/ingest-csv.ts +45 -48
- package/lib/tools/jsonl-to-parquet.ts +59 -98
- package/lib/tools/lint/recipe-output/vocab.ts +366 -0
- package/lib/tools/lint/{slice/index.ts → recipe-output.ts} +107 -121
- 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/{slice → overlay}/kryptonite.ts +43 -36
- 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/{slice → overlay}/translit.ts +74 -58
- package/lib/tools/overture-subvenue.ts +17 -90
- package/lib/tools/postcode-triples.ts +281 -210
- package/lib/tools/source-register/build.ts +819 -0
- package/lib/tools/sub/venue/head-nouns.ts +38 -75
- package/lib/tools/sub/venue/lexicon.ts +71 -106
- package/lib/tools/sub/venue/wikidata.ts +13 -10
- package/lib/tools.ts +31 -0
- package/lib/tw/tools/fetch/gcis.ts +17 -16
- package/lib/types.ts +304 -101
- package/lib/us/adapters/fcc-bdc/adapter.ts +26 -41
- package/lib/us/adapters/state/hi-schools/adapter.ts +28 -17
- package/lib/us/adapters/state/hi-schools/workbook.ts +7 -2
- package/lib/us/adapters/state/ia-contractors/adapter.ts +27 -15
- package/lib/us/adapters/state/ny-notaries/adapter.ts +24 -25
- package/lib/us/adapters/state/tx-notaries/adapter.ts +21 -15
- package/lib/us/adapters/tiger/adapter.ts +19 -44
- package/lib/us/adapters/tiger/street-decompose.ts +7 -17
- package/lib/us/adapters/usgov/hrsa-fqhc/adapter.ts +49 -34
- package/lib/us/adapters/usgov/imls-pls/adapter.ts +25 -19
- package/lib/us/adapters/usgov/irs-bmf/adapter.ts +32 -23
- package/lib/us/adapters/usgov/nad/adapter.ts +51 -112
- package/lib/us/adapters/usgov/nppes/adapter.ts +72 -35
- package/lib/us/adapters/usgov/samhsa-treatment-locator/adapter.ts +43 -30
- package/lib/us/fips-state.ts +17 -8
- package/lib/us/tools/fetch/hrsa.ts +7 -8
- package/lib/us/tools/fetch/imls-pls.ts +17 -35
- package/lib/us/tools/fetch/nad.ts +50 -32
- package/lib/us/tools/fetch/nppes.ts +23 -35
- package/lib/us/tools/fetch/state/hi-schools.ts +22 -16
- package/lib/us/tools/fetch/state/sources.ts +12 -12
- package/lib/us/tools/fetch/tiger-full.ts +21 -73
- package/lib/utils/align.ts +42 -52
- package/lib/utils/{slice-stats.ts → cooccurrence-stats.ts} +22 -16
- package/lib/utils/golden.ts +34 -49
- 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 +129 -76
- package/lib/{utils/index.ts → utils.ts} +0 -1
- package/lib/ve/recipes/reviewed-postcode-tail.ts +40 -13
- 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 +39 -49
- package/out/adapters/geonames/adapter.js.map +1 -1
- package/out/adapters/geonames/postal/adapter.d.ts +14 -27
- package/out/adapters/geonames/postal/adapter.d.ts.map +1 -1
- package/out/adapters/geonames/postal/adapter.js +30 -35
- package/out/adapters/geonames/postal/adapter.js.map +1 -1
- 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 +34 -46
- package/out/adapters/openaddresses/adapter.js.map +1 -1
- package/out/adapters/osm/adapter.d.ts +48 -42
- package/out/adapters/osm/adapter.d.ts.map +1 -1
- package/out/adapters/osm/adapter.js +116 -67
- package/out/adapters/osm/adapter.js.map +1 -1
- package/out/adapters/overture/adapter.d.ts +11 -37
- package/out/adapters/overture/adapter.d.ts.map +1 -1
- package/out/adapters/overture/adapter.js +45 -42
- 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/{synth-po-box → po-box}/adapter.js +45 -46
- 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 +27 -18
- package/out/adapters/wof/admin/json/adapter.d.ts.map +1 -1
- package/out/adapters/wof/admin/json/adapter.js +47 -31
- package/out/adapters/wof/admin/json/adapter.js.map +1 -1
- package/out/adapters/wof/json-rows.d.ts +35 -19
- package/out/adapters/wof/json-rows.d.ts.map +1 -1
- package/out/adapters/wof/json-rows.js +35 -15
- package/out/adapters/wof/json-rows.js.map +1 -1
- package/out/adapters/wof/postalcode-json/adapter.d.ts +22 -38
- package/out/adapters/wof/postalcode-json/adapter.d.ts.map +1 -1
- package/out/adapters/wof/postalcode-json/adapter.js +55 -57
- package/out/adapters/wof/postalcode-json/adapter.js.map +1 -1
- 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 +15 -27
- package/out/au/adapters/gnaf/adapter.d.ts.map +1 -1
- package/out/au/adapters/gnaf/adapter.js +34 -42
- package/out/au/adapters/gnaf/adapter.js.map +1 -1
- package/out/au/adapters/gnaf/assemble.d.ts +15 -26
- package/out/au/adapters/gnaf/assemble.d.ts.map +1 -1
- package/out/au/adapters/gnaf/assemble.js +26 -46
- package/out/au/adapters/gnaf/assemble.js.map +1 -1
- 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 +5 -22
- package/out/cn/recipes/organizational-units.d.ts.map +1 -1
- package/out/cn/recipes/organizational-units.js +20 -41
- package/out/cn/recipes/organizational-units.js.map +1 -1
- 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 +7 -11
- package/out/cz/recipes/pcfirst-preposition.d.ts.map +1 -1
- package/out/cz/recipes/pcfirst-preposition.js +36 -27
- package/out/cz/recipes/pcfirst-preposition.js.map +1 -1
- 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 +24 -26
- package/out/de/recipes/locale.d.ts.map +1 -1
- package/out/de/recipes/locale.js +66 -56
- package/out/de/recipes/locale.js.map +1 -1
- 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 +6 -2
- 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 +19 -25
- package/out/fr/adapters/ban/adapter.d.ts.map +1 -1
- package/out/fr/adapters/ban/adapter.js +65 -54
- package/out/fr/adapters/ban/adapter.js.map +1 -1
- 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 +4 -27
- package/out/fr/recipes/admin-split.d.ts.map +1 -1
- package/out/fr/recipes/admin-split.js +36 -58
- package/out/fr/recipes/admin-split.js.map +1 -1
- package/out/fr/recipes/bare-street.d.ts +2 -17
- package/out/fr/recipes/bare-street.d.ts.map +1 -1
- package/out/fr/recipes/bare-street.js +35 -50
- package/out/fr/recipes/bare-street.js.map +1 -1
- package/out/fr/recipes/fragment.d.ts +6 -65
- package/out/fr/recipes/fragment.d.ts.map +1 -1
- package/out/fr/recipes/fragment.js +55 -135
- package/out/fr/recipes/fragment.js.map +1 -1
- package/out/fr/recipes/lieudit.d.ts +17 -19
- package/out/fr/recipes/lieudit.d.ts.map +1 -1
- package/out/fr/recipes/lieudit.js +88 -68
- package/out/fr/recipes/lieudit.js.map +1 -1
- package/out/fr/recipes/order.d.ts +4 -27
- package/out/fr/recipes/order.d.ts.map +1 -1
- package/out/fr/recipes/order.js +31 -58
- package/out/fr/recipes/order.js.map +1 -1
- package/out/fr/{adapters/ban/street-decompose.d.ts → street-decompose.d.ts} +3 -2
- package/out/fr/street-decompose.d.ts.map +1 -0
- package/out/fr/{adapters/ban/street-decompose.js → street-decompose.js} +4 -3
- 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 +9 -7
- package/out/fr/tools/fetch/ban.d.ts.map +1 -1
- package/out/fr/tools/fetch/ban.js +27 -18
- package/out/fr/tools/fetch/ban.js.map +1 -1
- 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 +9 -3
- package/out/index.d.ts.map +1 -1
- package/out/index.js +9 -3
- 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 +38 -84
- package/out/international/recipes/locale.d.ts.map +1 -1
- package/out/international/recipes/locale.js +112 -151
- package/out/international/recipes/locale.js.map +1 -1
- 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 +14 -11
- package/out/jp/adapters/wof-admin/adapter.d.ts.map +1 -1
- package/out/jp/adapters/wof-admin/adapter.js +27 -16
- package/out/jp/adapters/wof-admin/adapter.js.map +1 -1
- package/out/jp/tools/fetch/houjin.d.ts +7 -4
- package/out/jp/tools/fetch/houjin.d.ts.map +1 -1
- package/out/jp/tools/fetch/houjin.js +15 -11
- package/out/jp/tools/fetch/houjin.js.map +1 -1
- package/out/kr/adapters/juso/label-rows.d.ts +17 -31
- package/out/kr/adapters/juso/label-rows.d.ts.map +1 -1
- package/out/kr/adapters/juso/label-rows.js +19 -41
- package/out/kr/adapters/juso/label-rows.js.map +1 -1
- package/out/kr/adapters/localdata/align.d.ts +4 -16
- package/out/kr/adapters/localdata/align.d.ts.map +1 -1
- package/out/kr/adapters/localdata/align.js +19 -30
- package/out/kr/adapters/localdata/align.js.map +1 -1
- package/out/kr/adapters/localdata/key-index.d.ts +13 -22
- package/out/kr/adapters/localdata/key-index.d.ts.map +1 -1
- package/out/kr/adapters/localdata/key-index.js +15 -21
- package/out/kr/adapters/localdata/key-index.js.map +1 -1
- package/out/kr/adapters/localdata/permit-rows.d.ts +21 -14
- package/out/kr/adapters/localdata/permit-rows.d.ts.map +1 -1
- package/out/kr/adapters/localdata/permit-rows.js +30 -25
- package/out/kr/adapters/localdata/permit-rows.js.map +1 -1
- package/out/kr/tools/fetch/juso.d.ts +11 -8
- package/out/kr/tools/fetch/juso.d.ts.map +1 -1
- package/out/kr/tools/fetch/juso.js +15 -14
- package/out/kr/tools/fetch/juso.js.map +1 -1
- package/out/kr/tools/fetch/localdata.d.ts +16 -9
- package/out/kr/tools/fetch/localdata.d.ts.map +1 -1
- package/out/kr/tools/fetch/localdata.js +23 -15
- package/out/kr/tools/fetch/localdata.js.map +1 -1
- 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 +12 -14
- package/out/nl/recipes/postcode.d.ts.map +1 -1
- package/out/nl/recipes/postcode.js +42 -24
- package/out/nl/recipes/postcode.js.map +1 -1
- 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 +1 -33
- package/out/no/recipes/fragment.d.ts.map +1 -1
- package/out/no/recipes/fragment.js +23 -57
- package/out/no/recipes/fragment.js.map +1 -1
- package/out/no/recipes/street/led.d.ts +9 -12
- package/out/no/recipes/street/led.d.ts.map +1 -1
- package/out/no/recipes/street/led.js +34 -24
- package/out/no/recipes/street/led.js.map +1 -1
- package/out/no/recipes/street.d.ts +19 -0
- package/out/no/recipes/street.d.ts.map +1 -0
- package/out/no/recipes/{street/index.js → street.js} +28 -12
- 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 +2 -20
- package/out/recipes/bare/country.d.ts.map +1 -1
- package/out/recipes/bare/country.js +20 -31
- package/out/recipes/bare/country.js.map +1 -1
- package/out/recipes/bare/postcode/eval.d.ts +1 -1
- package/out/recipes/bare/postcode/eval.js +1 -1
- package/out/recipes/bare/postcode.d.ts +41 -0
- package/out/recipes/bare/postcode.d.ts.map +1 -0
- package/out/recipes/bare/{postcode/index.js → postcode.js} +91 -79
- 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.map +1 -1
- package/out/recipes/po/box/cedex/au-nz.js +3 -3
- package/out/recipes/po/box/cedex/au-nz.js.map +1 -1
- package/out/recipes/po/box/cedex/ca.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/ca.js +2 -2
- package/out/recipes/po/box/cedex/ca.js.map +1 -1
- package/out/recipes/po/box/cedex/fr.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/fr.js +3 -3
- package/out/recipes/po/box/cedex/fr.js.map +1 -1
- package/out/recipes/po/box/cedex/phrases.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/phrases.js +7 -6
- package/out/recipes/po/box/cedex/phrases.js.map +1 -1
- package/out/recipes/po/box/cedex/recipe.d.ts +1 -1
- package/out/recipes/po/box/cedex/recipe.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/recipe.js +25 -15
- package/out/recipes/po/box/cedex/recipe.js.map +1 -1
- package/out/recipes/po/box/cedex/sources.d.ts +12 -12
- package/out/recipes/po/box/cedex/sources.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/sources.js +1 -1
- package/out/recipes/po/box/cedex/sources.js.map +1 -1
- package/out/recipes/po/box/cedex/us.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/us.js +5 -2
- package/out/recipes/po/box/cedex/us.js.map +1 -1
- package/out/recipes/po/box.d.ts +18 -0
- package/out/recipes/po/box.d.ts.map +1 -0
- package/out/recipes/po/{box/index.js → box.js} +37 -19
- package/out/recipes/po/box.js.map +1 -0
- package/out/recipes/scaffold.d.ts +211 -138
- package/out/recipes/scaffold.d.ts.map +1 -1
- package/out/recipes/scaffold.js +72 -63
- 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 +20 -50
- package/out/recipes/street/affix.d.ts.map +1 -1
- package/out/recipes/street/affix.js +45 -133
- package/out/recipes/street/affix.js.map +1 -1
- package/out/recipes/street/bare.d.ts +9 -7
- package/out/recipes/street/bare.d.ts.map +1 -1
- package/out/recipes/street/bare.js +25 -14
- package/out/recipes/street/bare.js.map +1 -1
- 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 +12 -9
- package/out/recipes/sub/venue/context.d.ts.map +1 -1
- package/out/recipes/sub/venue/context.js +21 -17
- package/out/recipes/sub/venue/context.js.map +1 -1
- 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 +49 -92
- package/out/recipes/sub/venue/sources.d.ts.map +1 -1
- package/out/recipes/sub/venue/sources.js +43 -179
- package/out/recipes/sub/venue/sources.js.map +1 -1
- 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 +55 -85
- 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/index.js → recipes.js} +14 -12
- 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 +18 -32
- package/out/sg/recipes/register.d.ts.map +1 -1
- package/out/sg/recipes/register.js +44 -49
- package/out/sg/recipes/register.js.map +1 -1
- package/out/sg/tools/fetch/acra.d.ts +11 -14
- package/out/sg/tools/fetch/acra.d.ts.map +1 -1
- package/out/sg/tools/fetch/acra.js +16 -19
- package/out/sg/tools/fetch/acra.js.map +1 -1
- package/out/si/recipes/bare-village.d.ts +12 -13
- package/out/si/recipes/bare-village.d.ts.map +1 -1
- package/out/si/recipes/bare-village.js +31 -17
- package/out/si/recipes/bare-village.js.map +1 -1
- 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 +20 -17
- package/out/south-asia/recipes/register.d.ts.map +1 -1
- package/out/south-asia/recipes/register.js +44 -24
- package/out/south-asia/recipes/register.js.map +1 -1
- 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/{tools/sub/venue → subvenue}/promotions.js +15 -55
- 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 +21 -30
- 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 +36 -29
- 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 +86 -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 +33 -29
- 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 +27 -18
- package/out/tools/fetch/download/manifest.d.ts.map +1 -1
- package/out/tools/fetch/download/manifest.js +20 -13
- package/out/tools/fetch/download/manifest.js.map +1 -1
- package/out/tools/fetch/download/network.d.ts +48 -27
- package/out/tools/fetch/download/network.d.ts.map +1 -1
- package/out/tools/fetch/download/network.js +26 -16
- package/out/tools/fetch/download/network.js.map +1 -1
- package/out/tools/fetch/download/retry.d.ts +16 -15
- package/out/tools/fetch/download/retry.d.ts.map +1 -1
- package/out/tools/fetch/download/retry.js +24 -24
- package/out/tools/fetch/download/retry.js.map +1 -1
- package/out/tools/fetch/{download/index.d.ts → download.d.ts} +1 -1
- package/out/tools/fetch/download.d.ts.map +1 -0
- package/out/tools/fetch/{download/index.js → download.js} +1 -1
- package/out/tools/fetch/download.js.map +1 -0
- package/out/tools/fetch/geonames/dump.d.ts +50 -32
- package/out/tools/fetch/geonames/dump.d.ts.map +1 -1
- package/out/tools/fetch/geonames/dump.js +52 -42
- package/out/tools/fetch/geonames/dump.js.map +1 -1
- package/out/tools/fetch/geonames/postal.d.ts +14 -49
- package/out/tools/fetch/geonames/postal.d.ts.map +1 -1
- package/out/tools/fetch/geonames/postal.js +17 -56
- package/out/tools/fetch/geonames/postal.js.map +1 -1
- 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 +1 -4
- package/out/tools/fetch/openaddresses.d.ts.map +1 -1
- package/out/tools/fetch/openaddresses.js +19 -98
- package/out/tools/fetch/openaddresses.js.map +1 -1
- package/out/tools/fetch/ourairports.d.ts +6 -33
- package/out/tools/fetch/ourairports.d.ts.map +1 -1
- package/out/tools/fetch/ourairports.js +14 -47
- 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 -7
- 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 +32 -100
- package/out/tools/fetch/wikidata-subvenue.d.ts.map +1 -1
- package/out/tools/fetch/wikidata-subvenue.js +27 -122
- 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 +28 -13
- package/out/tools/golden/expand.d.ts.map +1 -1
- package/out/tools/golden/expand.js +50 -36
- package/out/tools/golden/expand.js.map +1 -1
- package/out/tools/golden/promote.d.ts +13 -8
- package/out/tools/golden/promote.d.ts.map +1 -1
- package/out/tools/golden/promote.js +33 -28
- package/out/tools/golden/promote.js.map +1 -1
- package/out/tools/golden/relabel-street/directory.d.ts +24 -9
- package/out/tools/golden/relabel-street/directory.d.ts.map +1 -1
- package/out/tools/golden/relabel-street/directory.js +41 -30
- package/out/tools/golden/relabel-street/directory.js.map +1 -1
- package/out/tools/golden/relabel-street/row.d.ts +17 -69
- package/out/tools/golden/relabel-street/row.d.ts.map +1 -1
- package/out/tools/golden/relabel-street/row.js +16 -74
- package/out/tools/golden/relabel-street/row.js.map +1 -1
- package/out/tools/golden/{relabel-street/index.d.ts → relabel-street.d.ts} +1 -1
- package/out/tools/golden/relabel-street.d.ts.map +1 -0
- package/out/tools/golden/{relabel-street/index.js → relabel-street.js} +1 -1
- 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 -43
- 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/{slice → recipe-output}/vocab.js +54 -69
- 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/{slice/index.js → recipe-output.js} +76 -96
- 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/{slice → overlay}/kryptonite.js +37 -28
- 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/{slice → overlay}/translit.js +59 -48
- package/out/tools/overlay/translit.js.map +1 -0
- package/out/tools/overture-subvenue.d.ts +16 -75
- 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 +77 -126
- package/out/tools/postcode-triples.d.ts.map +1 -1
- package/out/tools/postcode-triples.js +198 -192
- 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 +9 -38
- package/out/tools/sub/venue/head-nouns.d.ts.map +1 -1
- package/out/tools/sub/venue/head-nouns.js +37 -74
- package/out/tools/sub/venue/head-nouns.js.map +1 -1
- package/out/tools/sub/venue/lexicon.d.ts +55 -82
- package/out/tools/sub/venue/lexicon.d.ts.map +1 -1
- package/out/tools/sub/venue/lexicon.js +43 -87
- package/out/tools/sub/venue/lexicon.js.map +1 -1
- package/out/tools/sub/venue/wikidata.d.ts +8 -6
- package/out/tools/sub/venue/wikidata.d.ts.map +1 -1
- package/out/tools/sub/venue/wikidata.js +12 -9
- package/out/tools/sub/venue/wikidata.js.map +1 -1
- 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 +9 -3
- package/out/tw/tools/fetch/gcis.d.ts.map +1 -1
- package/out/tw/tools/fetch/gcis.js +16 -10
- package/out/tw/tools/fetch/gcis.js.map +1 -1
- 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 +18 -36
- package/out/us/adapters/fcc-bdc/adapter.d.ts.map +1 -1
- package/out/us/adapters/fcc-bdc/adapter.js +25 -38
- package/out/us/adapters/fcc-bdc/adapter.js.map +1 -1
- package/out/us/adapters/state/hi-schools/adapter.d.ts +12 -11
- package/out/us/adapters/state/hi-schools/adapter.d.ts.map +1 -1
- package/out/us/adapters/state/hi-schools/adapter.js +26 -16
- package/out/us/adapters/state/hi-schools/adapter.js.map +1 -1
- package/out/us/adapters/state/hi-schools/workbook.d.ts +7 -2
- package/out/us/adapters/state/hi-schools/workbook.d.ts.map +1 -1
- package/out/us/adapters/state/hi-schools/workbook.js +3 -1
- package/out/us/adapters/state/hi-schools/workbook.js.map +1 -1
- package/out/us/adapters/state/ia-contractors/adapter.d.ts +13 -6
- package/out/us/adapters/state/ia-contractors/adapter.d.ts.map +1 -1
- package/out/us/adapters/state/ia-contractors/adapter.js +27 -14
- package/out/us/adapters/state/ia-contractors/adapter.js.map +1 -1
- package/out/us/adapters/state/ny-notaries/adapter.d.ts +7 -6
- package/out/us/adapters/state/ny-notaries/adapter.d.ts.map +1 -1
- package/out/us/adapters/state/ny-notaries/adapter.js +24 -22
- package/out/us/adapters/state/ny-notaries/adapter.js.map +1 -1
- package/out/us/adapters/state/tx-notaries/adapter.d.ts +8 -7
- package/out/us/adapters/state/tx-notaries/adapter.d.ts.map +1 -1
- package/out/us/adapters/state/tx-notaries/adapter.js +21 -14
- package/out/us/adapters/state/tx-notaries/adapter.js.map +1 -1
- package/out/us/adapters/tiger/adapter.d.ts +10 -34
- package/out/us/adapters/tiger/adapter.d.ts.map +1 -1
- package/out/us/adapters/tiger/adapter.js +18 -42
- package/out/us/adapters/tiger/adapter.js.map +1 -1
- package/out/us/adapters/tiger/street-decompose.d.ts +6 -13
- package/out/us/adapters/tiger/street-decompose.d.ts.map +1 -1
- package/out/us/adapters/tiger/street-decompose.js +7 -17
- package/out/us/adapters/tiger/street-decompose.js.map +1 -1
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts +26 -20
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.js +41 -29
- package/out/us/adapters/usgov/hrsa-fqhc/adapter.js.map +1 -1
- package/out/us/adapters/usgov/imls-pls/adapter.d.ts +8 -7
- package/out/us/adapters/usgov/imls-pls/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/imls-pls/adapter.js +25 -18
- package/out/us/adapters/usgov/imls-pls/adapter.js.map +1 -1
- package/out/us/adapters/usgov/irs-bmf/adapter.d.ts +11 -10
- package/out/us/adapters/usgov/irs-bmf/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/irs-bmf/adapter.js +32 -22
- package/out/us/adapters/usgov/irs-bmf/adapter.js.map +1 -1
- package/out/us/adapters/usgov/nad/adapter.d.ts +11 -31
- package/out/us/adapters/usgov/nad/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/nad/adapter.js +51 -105
- package/out/us/adapters/usgov/nad/adapter.js.map +1 -1
- package/out/us/adapters/usgov/nppes/adapter.d.ts +12 -22
- package/out/us/adapters/usgov/nppes/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/nppes/adapter.js +62 -32
- package/out/us/adapters/usgov/nppes/adapter.js.map +1 -1
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts +16 -15
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.d.ts.map +1 -1
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js +36 -26
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js.map +1 -1
- package/out/us/fips-state.d.ts +17 -8
- package/out/us/fips-state.d.ts.map +1 -1
- package/out/us/fips-state.js +17 -8
- package/out/us/fips-state.js.map +1 -1
- package/out/us/tools/fetch/hrsa.d.ts +3 -3
- package/out/us/tools/fetch/hrsa.d.ts.map +1 -1
- package/out/us/tools/fetch/hrsa.js +6 -7
- package/out/us/tools/fetch/hrsa.js.map +1 -1
- package/out/us/tools/fetch/imls-pls.d.ts +5 -16
- package/out/us/tools/fetch/imls-pls.d.ts.map +1 -1
- package/out/us/tools/fetch/imls-pls.js +16 -33
- package/out/us/tools/fetch/imls-pls.js.map +1 -1
- package/out/us/tools/fetch/nad.d.ts +22 -10
- package/out/us/tools/fetch/nad.d.ts.map +1 -1
- package/out/us/tools/fetch/nad.js +29 -24
- package/out/us/tools/fetch/nad.js.map +1 -1
- package/out/us/tools/fetch/nppes.d.ts +6 -16
- package/out/us/tools/fetch/nppes.d.ts.map +1 -1
- package/out/us/tools/fetch/nppes.js +21 -33
- package/out/us/tools/fetch/nppes.js.map +1 -1
- package/out/us/tools/fetch/state/hi-schools.d.ts +8 -5
- package/out/us/tools/fetch/state/hi-schools.d.ts.map +1 -1
- package/out/us/tools/fetch/state/hi-schools.js +15 -13
- package/out/us/tools/fetch/state/hi-schools.js.map +1 -1
- package/out/us/tools/fetch/state/sources.d.ts +8 -7
- package/out/us/tools/fetch/state/sources.d.ts.map +1 -1
- package/out/us/tools/fetch/state/sources.js +8 -9
- package/out/us/tools/fetch/state/sources.js.map +1 -1
- package/out/us/tools/fetch/tiger-full.d.ts +1 -14
- package/out/us/tools/fetch/tiger-full.d.ts.map +1 -1
- package/out/us/tools/fetch/tiger-full.js +15 -54
- package/out/us/tools/fetch/tiger-full.js.map +1 -1
- package/out/utils/align.d.ts +24 -19
- 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 +21 -31
- package/out/utils/golden.d.ts.map +1 -1
- package/out/utils/golden.js +28 -44
- 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 +76 -55
- package/out/utils/wof-json.js.map +1 -1
- package/out/{utils/index.js → utils.d.ts} +1 -2
- package/out/utils.d.ts.map +1 -0
- package/out/{utils/index.d.ts → utils.js} +1 -2
- package/out/utils.js.map +1 -0
- package/out/ve/recipes/reviewed-postcode-tail.d.ts +25 -7
- package/out/ve/recipes/reviewed-postcode-tail.d.ts.map +1 -1
- package/out/ve/recipes/reviewed-postcode-tail.js +33 -12
- package/out/ve/recipes/reviewed-postcode-tail.js.map +1 -1
- package/package.json +46 -480
- package/lib/adapters/index.ts +0 -160
- package/lib/adapters/synth-po-box/adapter.ts +0 -204
- package/lib/adapters/utils/index.ts +0 -307
- package/lib/recipes/bare/postcode/index.ts +0 -327
- package/lib/recipes/sub/venue/index.ts +0 -997
- package/lib/synthesizers/locale.ts +0 -261
- package/lib/test-kit/index.ts +0 -100
- package/lib/tools/align-slice.ts +0 -68
- package/lib/tools/fetch/index.ts +0 -174
- package/lib/tools/gb-title-case.ts +0 -69
- package/lib/tools/index.ts +0 -28
- package/lib/tools/lint/slice/vocab.ts +0 -376
- package/lib/tools/overlay-manifest.ts +0 -215
- package/lib/tools/sub/venue/harvest.ts +0 -234
- package/lib/tools/sub/venue/surfaces.ts +0 -95
- package/lib/tools/sub/venue/table.ts +0 -281
- package/lib/utils/parquet.ts +0 -582
- package/out/adapters/index.d.ts +0 -51
- package/out/adapters/index.d.ts.map +0 -1
- package/out/adapters/index.js +0 -97
- package/out/adapters/index.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.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/fr/adapters/ban/street-decompose.d.ts.map +0 -1
- package/out/fr/adapters/ban/street-decompose.js.map +0 -1
- package/out/no/recipes/street/index.d.ts +0 -18
- package/out/no/recipes/street/index.d.ts.map +0 -1
- package/out/no/recipes/street/index.js.map +0 -1
- package/out/recipes/bare/postcode/index.d.ts +0 -62
- package/out/recipes/bare/postcode/index.d.ts.map +0 -1
- package/out/recipes/bare/postcode/index.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.map +0 -1
- package/out/recipes/po/box/index.d.ts +0 -17
- package/out/recipes/po/box/index.d.ts.map +0 -1
- package/out/recipes/po/box/index.js.map +0 -1
- package/out/recipes/sub/venue/index.d.ts +0 -270
- package/out/recipes/sub/venue/index.d.ts.map +0 -1
- package/out/recipes/sub/venue/index.js +0 -728
- package/out/recipes/sub/venue/index.js.map +0 -1
- package/out/synthesizers/locale.d.ts +0 -116
- package/out/synthesizers/locale.d.ts.map +0 -1
- package/out/synthesizers/locale.js +0 -150
- package/out/synthesizers/locale.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/download/index.d.ts.map +0 -1
- package/out/tools/fetch/download/index.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/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/relabel-street/index.d.ts.map +0 -1
- package/out/tools/golden/relabel-street/index.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/index.d.ts +0 -93
- package/out/tools/lint/slice/index.d.ts.map +0 -1
- package/out/tools/lint/slice/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.map +0 -1
- package/out/tools/overlay-manifest.d.ts +0 -89
- package/out/tools/overlay-manifest.d.ts.map +0 -1
- package/out/tools/overlay-manifest.js +0 -131
- 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.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.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/promotions.d.ts +0 -94
- package/out/tools/sub/venue/promotions.d.ts.map +0 -1
- package/out/tools/sub/venue/promotions.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/utils/index.d.ts.map +0 -1
- 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
- /package/lib/tools/fetch/{download/index.ts → download.ts} +0 -0
- /package/lib/tools/golden/{relabel-street/index.ts → relabel-street.ts} +0 -0
|
@@ -1,90 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* #511 base-consistency lint, GENERALIZED + COUNTRY-SCOPED (v2) — any synthetic slice vs the base.
|
|
7
|
-
*
|
|
8
|
-
* Ported from scripts/lint-slice-vocab.py (pyarrow → @duckdb/node-api); behavior preserved
|
|
9
|
-
* byte-for-byte (same flags, same stdout, same verdicts). The base-root default routes through
|
|
10
|
-
* `dataRootPath` so the lab `/mnt/playpen` literal stays in its one home
|
|
11
|
-
* (core/utils/data-root.ts) and `$MAILWOMAN_DATA_ROOT` is honored; with the env unset it equals
|
|
12
|
-
* the Python default.
|
|
13
|
-
*
|
|
14
|
-
* The #511 lesson: a synthetic slice must not label a token a tag the BASE dominantly labels
|
|
15
|
-
* something else, or training gets conflicting gradients on the same token and the minority (the
|
|
16
|
-
* slice) loses. This reads a slice's own (token -> tag) and checks each token against the base.
|
|
17
|
-
*
|
|
18
|
-
* WHY v2 IS COUNTRY-SCOPED + FULL-COUNT (the night-2026-06-18 lesson, learned the hard way over
|
|
19
|
-
* three tries): a token's correct tag is COUNTRY-specific — "Paris" is locality in FR data and
|
|
20
|
-
* street in US "Paris Ave"; "Marion" is a US town AND many US "Marion" streets. So:
|
|
21
|
-
*
|
|
22
|
-
* 1. A cross-COUNTRY aggregate mis-judges any country-specific token (v1 uniform AND a proportional
|
|
23
|
-
* retry both false-flagged FR cities as "street" from US street-contexts).
|
|
24
|
-
* 2. A SMALL sample is street-BIASED regardless, because the street sources (tiger 39 + nad 378 parts)
|
|
25
|
-
* dwarf the locality sources (a small US-scoped spot-check read Indianapolis 54% street vs
|
|
26
|
-
* its true 219700:29 LOCALITY). The fix: tally each slice token's base tag SCOPED to the
|
|
27
|
-
* country the slice uses it in (the base has a `country` column), over a LARGE/FULL scan
|
|
28
|
-
* (`fraction`, default 1.0). Pure-numeric tokens excluded (house_number/postcode are
|
|
29
|
-
* context-determined). An affix-split flag (slice street_suffix/_prefix vs base "street") is
|
|
30
|
-
* EXPECTED — the loader's affix-relabel handles it; weigh those separately.
|
|
31
|
-
*
|
|
32
|
-
* Usage: mailwoman dev lint slice-vocab --slice <slice.parquet>
|
|
33
|
-
* [--base-version v0.5.0] [--base-root <dir>] [--fraction 1.0] [--threshold 0.7] [--min-count
|
|
34
|
-
* 50]
|
|
35
|
-
*/
|
|
36
|
-
/**
|
|
37
|
-
* Options for {@linkcode lintSliceVocab}.
|
|
38
|
-
*/
|
|
39
|
-
export interface LintSliceVocabOptions {
|
|
40
|
-
/**
|
|
41
|
-
* The slice parquet to lint.
|
|
42
|
-
*/
|
|
43
|
-
slice: string;
|
|
44
|
-
/**
|
|
45
|
-
* Base corpus version. Default `v0.5.0`.
|
|
46
|
-
*/
|
|
47
|
-
baseVersion?: string;
|
|
48
|
-
/**
|
|
49
|
-
* Base corpus root. Default `$MAILWOMAN_DATA_ROOT/corpus/versioned`.
|
|
50
|
-
*/
|
|
51
|
-
baseRoot?: string;
|
|
52
|
-
/**
|
|
53
|
-
* Base-majority confidence floor for a contradiction. Default 0.7.
|
|
54
|
-
*/
|
|
55
|
-
threshold?: number;
|
|
56
|
-
/**
|
|
57
|
-
* Minimum base support to judge a token. Default 50.
|
|
58
|
-
*/
|
|
59
|
-
minCount?: number;
|
|
60
|
-
/**
|
|
61
|
-
* Fraction of base parts to scan (proportional per-source slice below 1.0). Default 1.0.
|
|
62
|
-
*/
|
|
63
|
-
fraction?: number;
|
|
64
|
-
}
|
|
65
|
-
/**
|
|
66
|
-
* One contradiction row: token, slice tag, base tag, base fraction, base total.
|
|
67
|
-
*/
|
|
68
|
-
export type SliceVocabRow = [token: string, sliceTag: string, baseTag: string, baseFrac: number, baseTotal: number];
|
|
69
|
-
/**
|
|
70
|
-
* Findings summary returned by {@linkcode lintSliceVocab}.
|
|
71
|
-
*/
|
|
72
|
-
export interface LintSliceVocabSummary {
|
|
73
|
-
/**
|
|
74
|
-
* Real contradictions — the command exits 1 when nonzero.
|
|
75
|
-
*/
|
|
76
|
-
errors: number;
|
|
77
|
-
/**
|
|
78
|
-
* Affix-split rows (EXPECTED — the loader's affix-relabel handles them).
|
|
79
|
-
*/
|
|
80
|
-
warnings: number;
|
|
81
|
-
findings: {
|
|
82
|
-
contradictions: SliceVocabRow[];
|
|
83
|
-
affixSplits: SliceVocabRow[];
|
|
84
|
-
};
|
|
85
|
-
}
|
|
86
|
-
/**
|
|
87
|
-
* Lint a synthetic slice's (token → tag) vocabulary against the base corpus, country-scoped.
|
|
88
|
-
*/
|
|
89
|
-
export declare function lintSliceVocab(options: LintSliceVocabOptions): Promise<LintSliceVocabSummary>;
|
|
90
|
-
//# sourceMappingURL=vocab.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"vocab.d.ts","sourceRoot":"","sources":["../../../../lib/tools/lint/slice/vocab.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AAwIH;;GAEG;AACH,MAAM,WAAW,qBAAqB;IACrC;;OAEG;IACH,KAAK,EAAE,MAAM,CAAA;IACb;;OAEG;IACH,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB;;OAEG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAA;IAClB;;OAEG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB;;OAEG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAA;CACjB;AAED;;GAEG;AACH,MAAM,MAAM,aAAa,GAAG,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,CAAC,CAAA;AAEnH;;GAEG;AACH,MAAM,WAAW,qBAAqB;IACrC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,QAAQ,EAAE,MAAM,CAAA;IAChB,QAAQ,EAAE;QAAE,cAAc,EAAE,aAAa,EAAE,CAAC;QAAC,WAAW,EAAE,aAAa,EAAE,CAAA;KAAE,CAAA;CAC3E;AAED;;GAEG;AACH,wBAAsB,cAAc,CAAC,OAAO,EAAE,qBAAqB,GAAG,OAAO,CAAC,qBAAqB,CAAC,CAwJnG"}
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"vocab.js","sourceRoot":"","sources":["../../../../lib/tools/lint/slice/vocab.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,2BAA2B,CAAA;AACxD,OAAO,EAAE,cAAc,EAAE,MAAM,sBAAsB,CAAA;AACrD,OAAO,EAAE,OAAO,EAAE,MAAM,yBAAyB,CAAA;AACjD,OAAO,EAAE,IAAI,EAAE,MAAM,SAAS,CAAA;AAC9B,OAAO,EAAE,UAAU,EAAE,MAAM,qBAAqB,CAAA;AAEhD,OAAO,EAAE,aAAa,EAAyB,MAAM,gBAAgB,CAAA;AAWrE;;GAEG;AACH,SAAS,QAAQ,CAAC,KAAa;IAC9B,MAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAA;IAE9B,OAAO,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,IAAI,CAAC,CAAC,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,KAAK,CAAA;AAC/D,CAAC;AAED;;;;GAIG;AACH,SAAS,OAAO,CAAC,KAAa;IAC7B,OAAO,KAAK,CAAC,MAAM,GAAG,CAAC,IAAI,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;AACpD,CAAC;AAED;;GAEG;AACH,SAAS,GAAG,CAAC,IAAY;IACxB,OAAO,GAAG,OAAO,CAAC,IAAI,GAAG,GAAG,CAAC,GAAG,CAAA;AACjC,CAAC;AAED;;;;GAIG;AACH,SAAS,aAAa,CAAC,CAAS;IAC/B,OAAO,MAAM,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,CAAA;AACtD,CAAC;AAED;;GAEG;AACH,SAAS,GAAG,CAAC,KAAa,EAAE,KAAa;IACxC,OAAO,KAAK,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;AAC3B,CAAC;AAED;;;GAGG;AACH,SAAS,QAAQ,CAAC,OAA4B;IAC7C,IAAI,KAAK,GAAG,CAAC,CAAA;IACb,IAAI,OAAO,GAAG,EAAE,CAAA;IAChB,IAAI,SAAS,GAAG,CAAC,CAAC,CAAA;IAElB,KAAK,MAAM,CAAC,GAAG,EAAE,KAAK,CAAC,IAAI,OAAO,EAAE,CAAC;QACpC,KAAK,IAAI,KAAK,CAAA;QAEd,IAAI,KAAK,GAAG,SAAS,EAAE,CAAC;YACvB,SAAS,GAAG,KAAK,CAAA;YACjB,OAAO,GAAG,GAAG,CAAA;QACd,CAAC;IACF,CAAC;IAED,IAAI,KAAK,KAAK,CAAC;QAAE,OAAO,CAAC,EAAE,EAAE,CAAC,EAAE,CAAC,CAAC,CAAA;IAElC,OAAO,CAAC,OAAO,EAAE,KAAK,EAAE,SAAS,GAAG,KAAK,CAAC,CAAA;AAC3C,CAAC;AAED;;GAEG;AACH,SAAS,IAAI,CAAC,KAAuC,EAAE,GAAW,EAAE,GAAW;IAC9E,IAAI,OAAO,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAA;IAE5B,IAAI,CAAC,OAAO,EAAE,CAAC;QACd,OAAO,GAAG,IAAI,GAAG,EAAE,CAAA;QACnB,KAAK,CAAC,GAAG,CAAC,GAAG,EAAE,OAAO,CAAC,CAAA;IACxB,CAAC;IAED,OAAO,CAAC,GAAG,CAAC,GAAG,EAAE,CAAC,OAAO,CAAC,GAAG,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;AAC9C,CAAC;AAED;;;;GAIG;AACH,KAAK,UAAU,QAAQ,CAAC,GAAqB,EAAE,IAAY;IAC1D,MAAM,MAAM,GAAG,MAAM,GAAG,CAAC,aAAa,CACrC,2FAA2F,IAAI,IAAI,CACnG,CAAA;IAED,MAAM,GAAG,GAAG,MAAM,CAAC,aAAa,EAAmE,CAAA;IACnG,MAAM,IAAI,GAAgB,EAAE,CAAA;IAE5B,KAAK,MAAM,CAAC,IAAI,GAAG,EAAE,CAAC;QACrB,MAAM,MAAM,GAAG,cAAc,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAA;QAC/C,MAAM,MAAM,GAAG,cAAc,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAA;QAE/C,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC;YAAE,SAAQ;QAE9D,IAAI,CAAC,IAAI,CAAC;YACT,MAAM,EAAE,MAAkB;YAC1B,MAAM,EAAE,MAAkB;YAC1B,OAAO,EAAE,CAAC,CAAC,OAAO,IAAI,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC;SACrD,CAAC,CAAA;IACH,CAAC;IAED,OAAO,IAAI,CAAA;AACZ,CAAC;AAED;;GAEG;AACH,KAAK,UAAU,UAAU,CAAC,GAAqB,EAAE,IAAY;IAC5D,MAAM,MAAM,GAAG,MAAM,GAAG,CAAC,aAAa,CAAC,oCAAoC,IAAI,YAAY,CAAC,CAAA;IAC5F,MAAM,IAAI,GAAG,MAAM,CAAC,aAAa,EAAgC,CAAA;IAEjE,OAAO,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,CAAE,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;AAClD,CAAC;AAoDD;;GAEG;AACH,MAAM,CAAC,KAAK,UAAU,cAAc,CAAC,OAA8B;IAClE,MAAM,WAAW,GAAG,OAAO,CAAC,WAAW,IAAI,QAAQ,CAAA;IACnD,MAAM,QAAQ,GAAG,OAAO,CAAC,QAAQ,IAAI,YAAY,CAAC,QAAQ,EAAE,WAAW,CAAC,CAAA;IACxE,MAAM,SAAS,GAAG,OAAO,CAAC,SAAS,IAAI,GAAG,CAAA;IAC1C,MAAM,QAAQ,GAAG,OAAO,CAAC,QAAQ,IAAI,EAAE,CAAA;IACvC,MAAM,QAAQ,GAAG,OAAO,CAAC,QAAQ,IAAI,CAAC,CAAA;IAEtC,MAAM,GAAG,GAAG,MAAM,aAAa,EAAE,CAAA;IAEjC,mFAAmF;IACnF,MAAM,SAAS,GAAG,MAAM,QAAQ,CAAC,GAAG,EAAE,OAAO,CAAC,KAAK,CAAC,CAAA;IACpD,MAAM,SAAS,GAAG,IAAI,GAAG,EAA+B,CAAA;IACxD,MAAM,cAAc,GAAG,IAAI,GAAG,EAA8B,CAAA;IAE5D,KAAK,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,OAAO,EAAE,IAAI,SAAS,EAAE,CAAC;QACrD,MAAM,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,MAAM,CAAC,CAAA;QAEhD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC5B,MAAM,CAAC,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;YACpB,MAAM,CAAC,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;YAEpB,IAAI,OAAO,CAAC,CAAC,CAAC;gBAAE,SAAQ,CAAC,4EAA4E;YACrG,IAAI,CAAC,SAAS,EAAE,CAAC,EAAE,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAA;YAC/B,IAAI,GAAG,GAAG,cAAc,CAAC,GAAG,CAAC,CAAC,CAAC,CAAA;YAE/B,IAAI,CAAC,GAAG,EAAE,CAAC;gBACV,GAAG,GAAG,IAAI,GAAG,EAAE,CAAA;gBACf,cAAc,CAAC,GAAG,CAAC,CAAC,EAAE,GAAG,CAAC,CAAA;YAC3B,CAAC;YAED,GAAG,CAAC,GAAG,CAAC,OAAO,CAAC,CAAA;QACjB,CAAC;IACF,CAAC;IAED,MAAM,UAAU,GAAG,IAAI,GAAG,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC,CAAA;IAE5C,OAAO,CAAC,GAAG,CAAC,UAAU,SAAS,CAAC,MAAM,UAAU,UAAU,CAAC,IAAI,gBAAgB,CAAC,CAAA;IAEhF,gGAAgG;IAChG,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,EAAE,WAAW,EAAE,UAAU,WAAW,EAAE,EAAE,OAAO,CAAC,CAAA;IAE9E,IAAI,KAAK,GAAG,CACX,MAAM,UAAU,CAAC,IAAI,CAAC,WAAW,EAAE;QAClC,GAAG,EAAE,QAAQ;QACb,QAAQ,EAAE,IAAI;KACd,CAAC,CAAC,OAAO,EAAE,CACZ,CAAC,QAAQ,EAAE,CAAA;IAEZ,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,CAAC;QACnB,MAAM,IAAI,KAAK,CAAC,qBAAqB,CAAC,CAAA;IACvC,CAAC;IAED,IAAI,QAAQ,GAAG,CAAC,EAAE,CAAC;QAClB,MAAM,KAAK,GAAG,IAAI,GAAG,EAAoB,CAAA;QAEzC,KAAK,MAAM,CAAC,IAAI,KAAK,EAAE,CAAC;YACvB,MAAM,GAAG,GAAG,MAAM,UAAU,CAAC,GAAG,EAAE,CAAC,CAAC,CAAA;YACpC,IAAI,IAAI,GAAG,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAA;YAEzB,IAAI,CAAC,IAAI,EAAE,CAAC;gBACX,IAAI,GAAG,EAAE,CAAA;gBACT,KAAK,CAAC,GAAG,CAAC,GAAG,EAAE,IAAI,CAAC,CAAA;YACrB,CAAC;YAED,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;QACb,CAAC;QAED,MAAM,MAAM,GAAa,EAAE,CAAA;QAE3B,KAAK,MAAM,EAAE,IAAI,KAAK,CAAC,MAAM,EAAE,EAAE,CAAC;YACjC,MAAM,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,OAAO,CAAC,EAAE,CAAC,MAAM,GAAG,QAAQ,CAAC,CAAC,CAAA;YAEvD,KAAK,MAAM,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,EAAE,CAAC;gBACnC,MAAM,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YACf,CAAC;QACF,CAAC;QAED,KAAK,GAAG,MAAM,CAAA;IACf,CAAC;IAED,OAAO,CAAC,GAAG,CACV,QAAQ,WAAW,cAAc,KAAK,CAAC,MAAM,oBAAoB,aAAa,CAAC,QAAQ,CAAC,mBAAmB,CAC3G,CAAA;IAED,mFAAmF;IACnF,MAAM,QAAQ,GAAG,IAAI,GAAG,EAA+B,CAAA;IAEvD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC,CAAE,CAAC,CAAA;QAE3C,KAAK,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,OAAO,EAAE,IAAI,IAAI,EAAE,CAAC;YAChD,MAAM,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,MAAM,CAAC,CAAA;YAEhD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC5B,MAAM,CAAC,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;gBAEpB,IAAI,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,cAAc,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,GAAG,CAAC,OAAO,CAAC,EAAE,CAAC;oBAC9D,IAAI,CAAC,QAAQ,EAAE,CAAC,EAAE,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,CAAC,CAAA;gBACxC,CAAC;YACF,CAAC;QACF,CAAC;QAED,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,GAAG,KAAK,CAAC,EAAE,CAAC;YACzB,OAAO,CAAC,GAAG,CAAC,QAAQ,CAAC,GAAG,CAAC,IAAI,KAAK,CAAC,MAAM,QAAQ,CAAC,CAAA;QACnD,CAAC;IACF,CAAC;IAED,kFAAkF;IAClF,MAAM,OAAO,GAAoB,EAAE,CAAA;IACnC,MAAM,KAAK,GAAoB,EAAE,CAAA;IAEjC,KAAK,MAAM,CAAC,IAAI,UAAU,EAAE,CAAC;QAC5B,MAAM,CAAC,IAAI,CAAC,GAAG,QAAQ,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,CAAA;QAC1C,MAAM,CAAC,IAAI,EAAE,MAAM,EAAE,KAAK,CAAC,GAAG,QAAQ,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,IAAI,GAAG,EAAE,CAAC,CAAA;QAEpE,IAAI,MAAM,GAAG,QAAQ,IAAI,CAAC,IAAI,IAAI,IAAI,KAAK,IAAI,IAAI,KAAK,GAAG,SAAS;YAAE,SAAQ;QAC9E,MAAM,GAAG,GAAkB,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,MAAM,CAAC,CAAA;QAEzD,IAAI,CAAC,IAAI,KAAK,eAAe,IAAI,IAAI,KAAK,eAAe,CAAC,IAAI,IAAI,KAAK,QAAQ,EAAE,CAAC;YACjF,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAChB,CAAC;aAAM,CAAC;YACP,OAAO,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAClB,CAAC;IACF,CAAC;IAED,MAAM,QAAQ,GAAqC;QAClD,CAAC,eAAe,EAAE,OAAO,CAAC;QAC1B,CAAC,gDAAgD,EAAE,KAAK,CAAC;KACzD,CAAA;IAED,KAAK,MAAM,CAAC,KAAK,EAAE,IAAI,CAAC,IAAI,QAAQ,EAAE,CAAC;QACtC,IAAI,CAAC,IAAI,CAAC,MAAM;YAAE,SAAQ;QAC1B,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;QAE/C,OAAO,CAAC,GAAG,CAAC,KAAK,KAAK,CAAC,UAAU,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,GAAG,IAAI,CAAC,MAAM,IAAI,KAAK,GAAG,CAAC,CAAA;QAErF,KAAK,MAAM,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,MAAM,CAAC,IAAI,IAAI,EAAE,CAAC;YACnD,OAAO,CAAC,GAAG,CAAC,KAAK,GAAG,CAAC,CAAC,EAAE,EAAE,CAAC,UAAU,GAAG,CAAC,IAAI,EAAE,EAAE,CAAC,SAAS,IAAI,KAAK,GAAG,CAAC,KAAK,CAAC,OAAO,MAAM,GAAG,CAAC,CAAA;QAChG,CAAC;IACF,CAAC;IAED,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC;QACrB,OAAO,CAAC,GAAG,CACV,yDAAyD,GAAG,CAAC,SAAS,CAAC,aAAa,QAAQ,2BAA2B,CACvH,CAAA;IACF,CAAC;IAED,OAAO;QACN,MAAM,EAAE,OAAO,CAAC,MAAM;QACtB,QAAQ,EAAE,KAAK,CAAC,MAAM;QACtB,QAAQ,EAAE,EAAE,cAAc,EAAE,OAAO,EAAE,WAAW,EAAE,KAAK,EAAE;KACzD,CAAA;AACF,CAAC"}
|
|
@@ -1,89 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* Assemble a corpus OVERLAY MANIFEST — generalized from assemble-fr-admin-split-overlay-manifest.
|
|
7
|
-
* ADDS slice parquets to a base corpus, keeping every base slice VERBATIM (pure overlay ADD), and
|
|
8
|
-
* re-roots base paths to /data (the Modal volume). Parameterized by --slice-parquet + --source, one
|
|
9
|
-
* label per parquet, so it works for any overlay (the fr-admin-split one is the original; #148's
|
|
10
|
-
* overture-multilocale is the second user; v0.29.0's eight target-family slices are why it takes a
|
|
11
|
-
* set rather than one — chaining eight single-slice overlays would leave seven dead directories and
|
|
12
|
-
* an eight-deep base chain for what is one version).
|
|
13
|
-
*
|
|
14
|
-
* Ported faithfully from scripts/assemble-overlay-manifest.py. The new slice's source_id column is
|
|
15
|
-
* read through DuckDB (`@duckdb/node-api`) instead of PyArrow; everything else is pure JSON.
|
|
16
|
-
*
|
|
17
|
-
* Pipeline (the recipe rides the result): node scripts/build-overture-multilocale-canonical.mjs
|
|
18
|
-
* --cap 150000 --out /tmp/ovl/overture-ml.canonical.jsonl node scripts/align-canonical-slice.ts
|
|
19
|
-
* --input <canonical> --output <labeled> --corpus-version 0.5.0 mailwoman dev jsonl-to-parquet
|
|
20
|
-
* --input <labeled> --output <NEW>/train/<slice-parquet> node
|
|
21
|
-
* scripts/assemble-overlay-manifest.ts --base <BASE>/MANIFEST.json --new-dir <NEW>\
|
|
22
|
-
* --modal-root /data/corpus/versioned/<ver>/<dir> --version <ver>\
|
|
23
|
-
* --slice-parquet <slice-parquet> --source <source> --note "..."
|
|
24
|
-
*
|
|
25
|
-
* # then push the overlay to R2 + sync + `modal run -d ... --config <recipe>.yaml --resume none`.
|
|
26
|
-
*/
|
|
27
|
-
type ManifestSlice = Record<string, unknown> & {
|
|
28
|
-
path: string;
|
|
29
|
-
source?: string;
|
|
30
|
-
};
|
|
31
|
-
/**
|
|
32
|
-
* The two keys a manifest written before the 2026-09-01 vocabulary rename uses for its slice list and its size. Every
|
|
33
|
-
* corpus built before that date carries them, here and on the Modal volume, and a built corpus is an immutable artifact
|
|
34
|
-
* — so a reader accepts either spelling and a writer emits only the current one. Spelled by concatenation because the
|
|
35
|
-
* word is banned in this tree and the ratchet's baseline is zero.
|
|
36
|
-
*/
|
|
37
|
-
declare const LEGACY_SLICES_KEY: "shards";
|
|
38
|
-
declare const LEGACY_ROWS_PER_SLICE_KEY: "rows_per_shard";
|
|
39
|
-
interface BaseManifest {
|
|
40
|
-
corpus_version?: string;
|
|
41
|
-
schema: unknown;
|
|
42
|
-
rows_per_slice?: unknown;
|
|
43
|
-
row_group_size: unknown;
|
|
44
|
-
slices?: ManifestSlice[];
|
|
45
|
-
counts: {
|
|
46
|
-
train: number;
|
|
47
|
-
val: number;
|
|
48
|
-
test: number;
|
|
49
|
-
};
|
|
50
|
-
total_rows: number;
|
|
51
|
-
/**
|
|
52
|
-
* The pre-rename spellings, present on every corpus built before 2026-09-01.
|
|
53
|
-
*/
|
|
54
|
-
[LEGACY_SLICES_KEY]?: ManifestSlice[];
|
|
55
|
-
[LEGACY_ROWS_PER_SLICE_KEY]?: unknown;
|
|
56
|
-
}
|
|
57
|
-
/**
|
|
58
|
-
* A base manifest's slice list under either key.
|
|
59
|
-
*
|
|
60
|
-
* The rename moved this reader and the trainer's to the new key without migrating the manifests, and the trainer
|
|
61
|
-
* measured the cost on 2026-09-09: `v0.28.0-reviewed-postcode-tail` declares 706 train slices under the old key and the
|
|
62
|
-
* loader resolved ONE. An overlay assembled from a base read as empty would carry no base slices at all.
|
|
63
|
-
*/
|
|
64
|
-
export declare function baseManifestSlices(base: BaseManifest): ManifestSlice[];
|
|
65
|
-
/**
|
|
66
|
-
* One parquet to add, with the source label its rows carry.
|
|
67
|
-
*/
|
|
68
|
-
export interface OverlaySlice {
|
|
69
|
-
parquet: string;
|
|
70
|
-
source: string;
|
|
71
|
-
}
|
|
72
|
-
export interface OverlayManifestOptions {
|
|
73
|
-
base: string;
|
|
74
|
-
newDir: string;
|
|
75
|
-
modalRoot: string;
|
|
76
|
-
version: string;
|
|
77
|
-
/**
|
|
78
|
-
* The parquets this overlay adds, in the order they should appear after the base's own.
|
|
79
|
-
*/
|
|
80
|
-
slices: readonly OverlaySlice[];
|
|
81
|
-
note: string;
|
|
82
|
-
}
|
|
83
|
-
/**
|
|
84
|
-
* Resolve a base manifest's slice path to the mounted corpus tree used by Modal.
|
|
85
|
-
*/
|
|
86
|
-
export declare function rerootBaseSlicePath(path: string, baseManifestPath: string): string;
|
|
87
|
-
export declare function assembleOverlayManifest(args: OverlayManifestOptions): Promise<void>;
|
|
88
|
-
export {};
|
|
89
|
-
//# sourceMappingURL=overlay-manifest.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"overlay-manifest.d.ts","sourceRoot":"","sources":["../../lib/tools/overlay-manifest.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;GAyBG;AAsBH,KAAK,aAAa,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG;IAAE,IAAI,EAAE,MAAM,CAAC;IAAC,MAAM,CAAC,EAAE,MAAM,CAAA;CAAE,CAAA;AAEhF;;;;;GAKG;AACH,QAAA,MAAM,iBAAiB,UAAyB,CAAA;AAChD,QAAA,MAAM,yBAAyB,kBAAiC,CAAA;AAEhE,UAAU,YAAY;IACrB,cAAc,CAAC,EAAE,MAAM,CAAA;IACvB,MAAM,EAAE,OAAO,CAAA;IACf,cAAc,CAAC,EAAE,OAAO,CAAA;IACxB,cAAc,EAAE,OAAO,CAAA;IACvB,MAAM,CAAC,EAAE,aAAa,EAAE,CAAA;IACxB,MAAM,EAAE;QAAE,KAAK,EAAE,MAAM,CAAC;QAAC,GAAG,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAA;IACpD,UAAU,EAAE,MAAM,CAAA;IAClB;;OAEG;IACH,CAAC,iBAAiB,CAAC,CAAC,EAAE,aAAa,EAAE,CAAA;IACrC,CAAC,yBAAyB,CAAC,CAAC,EAAE,OAAO,CAAA;CACrC;AAED;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,IAAI,EAAE,YAAY,GAAG,aAAa,EAAE,CAQtE;AASD;;GAEG;AACH,MAAM,WAAW,YAAY;IAC5B,OAAO,EAAE,MAAM,CAAA;IACf,MAAM,EAAE,MAAM,CAAA;CACd;AA0BD,MAAM,WAAW,sBAAsB;IACtC,IAAI,EAAE,MAAM,CAAA;IACZ,MAAM,EAAE,MAAM,CAAA;IACd,SAAS,EAAE,MAAM,CAAA;IACjB,OAAO,EAAE,MAAM,CAAA;IACf;;OAEG;IACH,MAAM,EAAE,SAAS,YAAY,EAAE,CAAA;IAC/B,IAAI,EAAE,MAAM,CAAA;CACZ;AAED;;GAEG;AACH,wBAAgB,mBAAmB,CAAC,IAAI,EAAE,MAAM,EAAE,gBAAgB,EAAE,MAAM,GAAG,MAAM,CAalF;AAED,wBAAsB,uBAAuB,CAAC,IAAI,EAAE,sBAAsB,GAAG,OAAO,CAAC,IAAI,CAAC,CAuDzF"}
|
|
@@ -1,131 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* Assemble a corpus OVERLAY MANIFEST — generalized from assemble-fr-admin-split-overlay-manifest.
|
|
7
|
-
* ADDS slice parquets to a base corpus, keeping every base slice VERBATIM (pure overlay ADD), and
|
|
8
|
-
* re-roots base paths to /data (the Modal volume). Parameterized by --slice-parquet + --source, one
|
|
9
|
-
* label per parquet, so it works for any overlay (the fr-admin-split one is the original; #148's
|
|
10
|
-
* overture-multilocale is the second user; v0.29.0's eight target-family slices are why it takes a
|
|
11
|
-
* set rather than one — chaining eight single-slice overlays would leave seven dead directories and
|
|
12
|
-
* an eight-deep base chain for what is one version).
|
|
13
|
-
*
|
|
14
|
-
* Ported faithfully from scripts/assemble-overlay-manifest.py. The new slice's source_id column is
|
|
15
|
-
* read through DuckDB (`@duckdb/node-api`) instead of PyArrow; everything else is pure JSON.
|
|
16
|
-
*
|
|
17
|
-
* Pipeline (the recipe rides the result): node scripts/build-overture-multilocale-canonical.mjs
|
|
18
|
-
* --cap 150000 --out /tmp/ovl/overture-ml.canonical.jsonl node scripts/align-canonical-slice.ts
|
|
19
|
-
* --input <canonical> --output <labeled> --corpus-version 0.5.0 mailwoman dev jsonl-to-parquet
|
|
20
|
-
* --input <labeled> --output <NEW>/train/<slice-parquet> node
|
|
21
|
-
* scripts/assemble-overlay-manifest.ts --base <BASE>/MANIFEST.json --new-dir <NEW>\
|
|
22
|
-
* --modal-root /data/corpus/versioned/<ver>/<dir> --version <ver>\
|
|
23
|
-
* --slice-parquet <slice-parquet> --source <source> --note "..."
|
|
24
|
-
*
|
|
25
|
-
* # then push the overlay to R2 + sync + `modal run -d ... --config <recipe>.yaml --resume none`.
|
|
26
|
-
*/
|
|
27
|
-
import { readLocalBuffer, readLocalJSONFile, statPath } from "@mailwoman/core/fs/readers";
|
|
28
|
-
import { writeLocalJSONFile } from "@mailwoman/core/fs/writers";
|
|
29
|
-
import { sha256Hex } from "@mailwoman/core/hash";
|
|
30
|
-
import { basename, dirname, join } from "path-ts";
|
|
31
|
-
import { connectDuckDB, escapeSQLString } from "#utils/parquet";
|
|
32
|
-
/**
|
|
33
|
-
* The two keys a manifest written before the 2026-09-01 vocabulary rename uses for its slice list and its size. Every
|
|
34
|
-
* corpus built before that date carries them, here and on the Modal volume, and a built corpus is an immutable artifact
|
|
35
|
-
* — so a reader accepts either spelling and a writer emits only the current one. Spelled by concatenation because the
|
|
36
|
-
* word is banned in this tree and the ratchet's baseline is zero.
|
|
37
|
-
*/
|
|
38
|
-
const LEGACY_SLICES_KEY = `sh${"ards"}`;
|
|
39
|
-
const LEGACY_ROWS_PER_SLICE_KEY = `rows_per_sh${"ard"}`;
|
|
40
|
-
/**
|
|
41
|
-
* A base manifest's slice list under either key.
|
|
42
|
-
*
|
|
43
|
-
* The rename moved this reader and the trainer's to the new key without migrating the manifests, and the trainer
|
|
44
|
-
* measured the cost on 2026-09-09: `v0.28.0-reviewed-postcode-tail` declares 706 train slices under the old key and the
|
|
45
|
-
* loader resolved ONE. An overlay assembled from a base read as empty would carry no base slices at all.
|
|
46
|
-
*/
|
|
47
|
-
export function baseManifestSlices(base) {
|
|
48
|
-
const slices = base.slices?.length ? base.slices : base[LEGACY_SLICES_KEY];
|
|
49
|
-
if (slices?.length)
|
|
50
|
-
return slices;
|
|
51
|
-
throw new Error(`base manifest lists no slices under "slices" or the pre-rename key — refusing to assemble an overlay whose base would be empty`);
|
|
52
|
-
}
|
|
53
|
-
/**
|
|
54
|
-
* The base's rows-per-slice under either key, carried through to the new manifest verbatim.
|
|
55
|
-
*/
|
|
56
|
-
function baseRowsPerSlice(base) {
|
|
57
|
-
return base.rows_per_slice ?? base[LEGACY_ROWS_PER_SLICE_KEY];
|
|
58
|
-
}
|
|
59
|
-
async function descriptor(localPath, modalPath, split, source) {
|
|
60
|
-
const db = await connectDuckDB();
|
|
61
|
-
const result = await db.runAndReadAll(`SELECT source_id FROM read_parquet('${escapeSQLString(localPath)}')`);
|
|
62
|
-
const sids = result.getRowObjects().map((r) => r.source_id);
|
|
63
|
-
return {
|
|
64
|
-
split,
|
|
65
|
-
path: modalPath,
|
|
66
|
-
format: "parquet",
|
|
67
|
-
compression: "SNAPPY",
|
|
68
|
-
rows: sids.length,
|
|
69
|
-
bytes: (await statPath(localPath)).size,
|
|
70
|
-
sha256: sha256Hex(await readLocalBuffer(localPath)),
|
|
71
|
-
first_source_id: sids[0],
|
|
72
|
-
last_source_id: sids.at(-1),
|
|
73
|
-
source,
|
|
74
|
-
};
|
|
75
|
-
}
|
|
76
|
-
/**
|
|
77
|
-
* Resolve a base manifest's slice path to the mounted corpus tree used by Modal.
|
|
78
|
-
*/
|
|
79
|
-
export function rerootBaseSlicePath(path, baseManifestPath) {
|
|
80
|
-
const versionedIndex = path.indexOf("/corpus/versioned/");
|
|
81
|
-
if (versionedIndex !== -1)
|
|
82
|
-
return "/data" + path.slice(versionedIndex);
|
|
83
|
-
if (/^\/data\/(?:train|val|test)\//u.test(path)) {
|
|
84
|
-
const localBaseDir = dirname(baseManifestPath);
|
|
85
|
-
const baseModalRoot = `/data/corpus/versioned/${basename(dirname(localBaseDir))}/${basename(localBaseDir)}`;
|
|
86
|
-
return baseModalRoot + path.slice("/data".length);
|
|
87
|
-
}
|
|
88
|
-
return path;
|
|
89
|
-
}
|
|
90
|
-
export async function assembleOverlayManifest(args) {
|
|
91
|
-
if (!args.slices.length)
|
|
92
|
-
throw new Error("an overlay must add at least one slice");
|
|
93
|
-
const base = await readLocalJSONFile(args.base);
|
|
94
|
-
const baseSlices = baseManifestSlices(base);
|
|
95
|
-
for (const slice of args.slices) {
|
|
96
|
-
if (baseSlices.some((s) => s.source === slice.source)) {
|
|
97
|
-
console.log(`WARN: base already contains source '${slice.source}' — is this the right base?`);
|
|
98
|
-
}
|
|
99
|
-
}
|
|
100
|
-
const kept = baseSlices.map((s) => ({ ...s, path: rerootBaseSlicePath(s.path, args.base) }));
|
|
101
|
-
const added = [];
|
|
102
|
-
for (const slice of args.slices) {
|
|
103
|
-
added.push(await descriptor(join(args.newDir, "train", slice.parquet), `${args.modalRoot}/train/${slice.parquet}`, "train", slice.source));
|
|
104
|
-
}
|
|
105
|
-
const addedRows = added.reduce((total, slice) => total + slice.rows, 0);
|
|
106
|
-
const sources = args.slices.map((slice) => slice.source).join(", ");
|
|
107
|
-
const manifest = {
|
|
108
|
-
corpus_version: args.version,
|
|
109
|
-
overlay_base: base.corpus_version ?? null,
|
|
110
|
-
note: args.note || `${base.corpus_version} slices (all kept verbatim) + ${sources}. Pure overlay add.`,
|
|
111
|
-
schema: base.schema,
|
|
112
|
-
rows_per_slice: baseRowsPerSlice(base),
|
|
113
|
-
row_group_size: base.row_group_size,
|
|
114
|
-
slices: [...kept, ...added],
|
|
115
|
-
counts: {
|
|
116
|
-
train: base.counts.train + addedRows,
|
|
117
|
-
val: base.counts.val,
|
|
118
|
-
test: base.counts.test,
|
|
119
|
-
},
|
|
120
|
-
total_rows: base.total_rows + addedRows,
|
|
121
|
-
};
|
|
122
|
-
const out = join(args.newDir, "MANIFEST.json");
|
|
123
|
-
await writeLocalJSONFile(manifest, out);
|
|
124
|
-
console.log(`wrote ${out}`);
|
|
125
|
-
console.log(` slices: ${manifest.slices.length} (${kept.length} base kept, +${added.length} added)`);
|
|
126
|
-
console.log(` counts: ${JSON.stringify(manifest.counts)} total: ${manifest.total_rows}`);
|
|
127
|
-
for (const slice of added) {
|
|
128
|
-
console.log(` ${slice.source} train: ${slice.rows} rows (${slice.bytes} bytes)`);
|
|
129
|
-
}
|
|
130
|
-
}
|
|
131
|
-
//# sourceMappingURL=overlay-manifest.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"overlay-manifest.js","sourceRoot":"","sources":["../../lib/tools/overlay-manifest.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;GAyBG;AAEH,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,QAAQ,EAAE,OAAO,EAAE,IAAI,EAAE,MAAM,SAAS,CAAA;AAEjD,OAAO,EAAE,aAAa,EAAE,eAAe,EAAE,MAAM,gBAAgB,CAAA;AAiB/D;;;;;GAKG;AACH,MAAM,iBAAiB,GAAG,KAAK,MAAM,EAAW,CAAA;AAChD,MAAM,yBAAyB,GAAG,cAAc,KAAK,EAAW,CAAA;AAiBhE;;;;;;GAMG;AACH,MAAM,UAAU,kBAAkB,CAAC,IAAkB;IACpD,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAA;IAE1E,IAAI,MAAM,EAAE,MAAM;QAAE,OAAO,MAAM,CAAA;IAEjC,MAAM,IAAI,KAAK,CACd,gIAAgI,CAChI,CAAA;AACF,CAAC;AAED;;GAEG;AACH,SAAS,gBAAgB,CAAC,IAAkB;IAC3C,OAAO,IAAI,CAAC,cAAc,IAAI,IAAI,CAAC,yBAAyB,CAAC,CAAA;AAC9D,CAAC;AAUD,KAAK,UAAU,UAAU,CACxB,SAAiB,EACjB,SAAiB,EACjB,KAAa,EACb,MAAc;IAEd,MAAM,EAAE,GAAG,MAAM,aAAa,EAAE,CAAA;IAChC,MAAM,MAAM,GAAG,MAAM,EAAE,CAAC,aAAa,CAAC,uCAAuC,eAAe,CAAC,SAAS,CAAC,IAAI,CAAC,CAAA;IAC5G,MAAM,IAAI,GAAG,MAAM,CAAC,aAAa,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,SAAmB,CAAC,CAAA;IAErE,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;AAcD;;GAEG;AACH,MAAM,UAAU,mBAAmB,CAAC,IAAY,EAAE,gBAAwB;IACzE,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,MAAM,CAAC,KAAK,UAAU,uBAAuB,CAAC,IAA4B;IACzE,IAAI,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM;QAAE,MAAM,IAAI,KAAK,CAAC,wCAAwC,CAAC,CAAA;IAElF,MAAM,IAAI,GAAG,MAAM,iBAAiB,CAAe,IAAI,CAAC,IAAI,CAAC,CAAA;IAC7D,MAAM,UAAU,GAAG,kBAAkB,CAAC,IAAI,CAAC,CAAA;IAE3C,KAAK,MAAM,KAAK,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC;QACjC,IAAI,UAAU,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,KAAK,CAAC,MAAM,CAAC,EAAE,CAAC;YACvD,OAAO,CAAC,GAAG,CAAC,uCAAuC,KAAK,CAAC,MAAM,6BAA6B,CAAC,CAAA;QAC9F,CAAC;IACF,CAAC;IAED,MAAM,IAAI,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,IAAI,EAAE,mBAAmB,CAAC,CAAC,CAAC,IAAI,EAAE,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAA;IAC5F,MAAM,KAAK,GAAsB,EAAE,CAAA;IAEnC,KAAK,MAAM,KAAK,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC;QACjC,KAAK,CAAC,IAAI,CACT,MAAM,UAAU,CACf,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE,OAAO,EAAE,KAAK,CAAC,OAAO,CAAC,EACzC,GAAG,IAAI,CAAC,SAAS,UAAU,KAAK,CAAC,OAAO,EAAE,EAC1C,OAAO,EACP,KAAK,CAAC,MAAM,CACZ,CACD,CAAA;IACF,CAAC;IAED,MAAM,SAAS,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,EAAE,CAAC,KAAK,GAAG,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,CAAA;IACvE,MAAM,OAAO,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;IAEnE,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,iCAAiC,OAAO,qBAAqB;QACtG,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,cAAc,EAAE,gBAAgB,CAAC,IAAI,CAAC;QACtC,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;YACpC,GAAG,EAAE,IAAI,CAAC,MAAM,CAAC,GAAG;YACpB,IAAI,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;SACtB;QACD,UAAU,EAAE,IAAI,CAAC,UAAU,GAAG,SAAS;KACvC,CAAA;IAED,MAAM,GAAG,GAAG,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE,eAAe,CAAC,CAAA;IAC9C,MAAM,kBAAkB,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAA;IAEvC,OAAO,CAAC,GAAG,CAAC,SAAS,GAAG,EAAE,CAAC,CAAA;IAC3B,OAAO,CAAC,GAAG,CAAC,aAAa,QAAQ,CAAC,MAAM,CAAC,MAAM,KAAK,IAAI,CAAC,MAAM,gBAAgB,KAAK,CAAC,MAAM,SAAS,CAAC,CAAA;IACrG,OAAO,CAAC,GAAG,CAAC,aAAa,IAAI,CAAC,SAAS,CAAC,QAAQ,CAAC,MAAM,CAAC,YAAY,QAAQ,CAAC,UAAU,EAAE,CAAC,CAAA;IAE1F,KAAK,MAAM,KAAK,IAAI,KAAK,EAAE,CAAC;QAC3B,OAAO,CAAC,GAAG,CAAC,KAAK,KAAK,CAAC,MAAM,WAAW,KAAK,CAAC,IAAI,UAAU,KAAK,CAAC,KAAK,SAAS,CAAC,CAAA;IAClF,CAAC;AACF,CAAC"}
|
|
@@ -1,32 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* Build a parquet slice from the DeepSeek-generated kryptonite JSONL and emit the corpus-v0.4.0
|
|
7
|
-
* MANIFEST. corpus-v0.4.0 is a pure adapter-addition revision: it points at every slice from
|
|
8
|
-
* v0.3.0 plus the new kryptonite slice(s). No v0.3.0 bytes are touched or re-shuffled.
|
|
9
|
-
*
|
|
10
|
-
* See docs/engineering/reference/CORPUS_V0_4_0_GENERATION.md for the why; that doc also pins the
|
|
11
|
-
* DeepSeek model version + prompt versions used to produce the JSONL.
|
|
12
|
-
*
|
|
13
|
-
* Invoke via `mailwoman corpus slice kryptonite \
|
|
14
|
-
* --jsonl /data/corpus/versioned/v0.4.0/kryptonite/canonical-kryptonite.jsonl \
|
|
15
|
-
* --base-manifest /data/corpus/versioned/v0.3.0/corpus-v0.3.0/MANIFEST.json \
|
|
16
|
-
* --out-dir /data/corpus/versioned/v0.4.0`
|
|
17
|
-
*/
|
|
18
|
-
export interface SliceKryptoniteOptions {
|
|
19
|
-
jsonl: string;
|
|
20
|
-
baseManifest: string;
|
|
21
|
-
outDir: string;
|
|
22
|
-
/**
|
|
23
|
-
* Default `"0.4.0"`.
|
|
24
|
-
*/
|
|
25
|
-
corpusVersion?: string;
|
|
26
|
-
/**
|
|
27
|
-
* Default `"deepseek-kryptonite"`.
|
|
28
|
-
*/
|
|
29
|
-
source?: string;
|
|
30
|
-
}
|
|
31
|
-
export declare function buildKryptoniteSlice(options: SliceKryptoniteOptions, report?: (line: string) => void): Promise<void>;
|
|
32
|
-
//# sourceMappingURL=kryptonite.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"kryptonite.d.ts","sourceRoot":"","sources":["../../../lib/tools/slice/kryptonite.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AAWH,MAAM,WAAW,sBAAsB;IACtC,KAAK,EAAE,MAAM,CAAA;IACb,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB;;OAEG;IACH,MAAM,CAAC,EAAE,MAAM,CAAA;CACf;AAiCD,wBAAsB,oBAAoB,CACzC,OAAO,EAAE,sBAAsB,EAC/B,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAC7B,OAAO,CAAC,IAAI,CAAC,CA6Df"}
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"kryptonite.js","sourceRoot":"","sources":["../../../lib/tools/slice/kryptonite.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AAEH,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAA;AAC1E,OAAO,EAAE,kBAAkB,EAAE,kBAAkB,EAAE,eAAe,EAAE,MAAM,4BAA4B,CAAA;AACpG,OAAO,EAAE,IAAI,EAAE,MAAM,SAAS,CAAA;AAC9B,OAAO,EAAE,eAAe,EAAE,MAAM,aAAa,CAAA;AAG7C,OAAO,EAAE,QAAQ,EAAE,eAAe,EAAE,cAAc,EAAE,iBAAiB,EAAE,WAAW,EAAE,MAAM,QAAQ,CAAA;AAiBlG,KAAK,SAAS,CAAC,CAAC,aAAa,CAAC,KAAa,EAAE,aAAqB;IACjE,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,eAAe,CAAC,SAAS,CAA0B,KAAK,CAAC,EAAE,CAAC;QACnF,oFAAoF;QACpF,MAAM,UAAU,GAAG,GAAG,CAAC,YAAY,CAA2B,CAAA;QAE9D,MAAM;YACL,GAAG,EAAE,GAAG,CAAC,KAAK,CAAW;YACzB,UAAU;YACV,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,IAAI;YAC3C,MAAM,EAAG,GAAG,CAAC,QAAQ,CAAY,IAAI,SAAS;YAC9C,MAAM,EAAG,GAAG,CAAC,QAAQ,CAAY,IAAI,qBAAqB;YAC1D,SAAS,EAAE,GAAG,CAAC,WAAW,CAAW;YACrC,cAAc,EAAE,aAAa;YAC7B,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,gDAAgD;YACvF,KAAK,EAAE,GAAG,CAAC,OAAO,CAA0B;SAC5C,CAAA;IACF,CAAC;AACF,CAAC;AAED,KAAK,SAAS,CAAC,CAAC,WAAW,CAAC,KAAa,EAAE,aAAqB,EAAE,aAAuB;IACxF,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,aAAa,CAAC,KAAK,EAAE,aAAa,CAAC,EAAE,CAAC;QAC7D,MAAM,MAAM,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAA;QAE5B,IAAI,MAAM,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;YAC/B,MAAM,MAAM,CAAC,GAAG,CAAA;QACjB,CAAC;aAAM,CAAC;YACP,aAAa,CAAC,IAAI,CAAC,GAAG,GAAG,CAAC,SAAS,KAAK,MAAM,CAAC,GAAG,CAAC,MAAM,EAAE,CAAC,CAAA;QAC7D,CAAC;IACF,CAAC;AACF,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,oBAAoB,CACzC,OAA+B,EAC/B,MAA+B;IAE/B,MAAM,aAAa,GAAG,OAAO,CAAC,aAAa,IAAI,OAAO,CAAA;IACtD,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,IAAI,qBAAqB,CAAA;IAEtD,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,oBAAoB,OAAO,CAAC,KAAK,EAAE,CAAC,CAAA;IAE5F,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,4BAA4B,OAAO,CAAC,YAAY,EAAE,CAAC,CAAA;IAElH,MAAM,eAAe,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;IAErC,MAAM,UAAU,GAAa,EAAE,CAAA;IAE/B,MAAM,WAAW,GAAG,MAAM,WAAW,CACpC,EAAE,KAAK,EAAE,WAAW,CAAC,OAAO,CAAC,KAAK,EAAE,aAAa,EAAE,UAAU,CAAC,EAAE,EAChE,EAAE,SAAS,EAAE,OAAO,CAAC,MAAM,EAAE,aAAa,EAAE,CAC5C,CAAA;IAED,MAAM,EAAE,CACP,SAAS,WAAW,CAAC,UAAU,cAAc,WAAW,CAAC,MAAM,CAAC,MAAM,aAAa;QAClF,eAAe,UAAU,CAAC,MAAM,EAAE,CACnC,CAAA;IAED,IAAI,UAAU,CAAC,MAAM,EAAE,CAAC;QACvB,MAAM,KAAK,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,WAAW,aAAa,EAAE,EAAE,2BAA2B,CAAC,CAAA;QAC3F,MAAM,kBAAkB,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,IAAI,EAAE,KAAK,CAAC,CAAA;QAC7D,MAAM,EAAE,CAAC,oBAAoB,KAAK,EAAE,CAAC,CAAA;IACtC,CAAC;IAED,mFAAmF;IACnF,sFAAsF;IACtF,yFAAyF;IACzF,KAAK,MAAM,EAAE,IAAI,WAAW,CAAC,MAAM,EAAE,CAAC;QACrC,EAAE,CAAC,MAAM,GAAG,MAAM,CAAA;IACnB,CAAC;IAED,sFAAsF;IACtF,MAAM,IAAI,GAAG,MAAM,iBAAiB,CAAgB,OAAO,CAAC,YAAY,CAAC,CAAA;IAEzE,MAAM,QAAQ,GAAkB;QAC/B,cAAc,EAAE,aAAa;QAC7B,MAAM,EAAE,eAAe;QACvB,cAAc,EAAE,IAAI,CAAC,cAAc;QACnC,cAAc,EAAE,IAAI,CAAC,cAAc,IAAI,cAAc;QACrD,MAAM,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,GAAG,WAAW,CAAC,MAAM,CAAC;QAC/C,MAAM,EAAE;YACP,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,GAAG,CAAC,WAAW,CAAC,MAAM,CAAC,KAAK,IAAI,CAAC,CAAC;YAC1D,GAAG,EAAE,IAAI,CAAC,MAAM,CAAC,GAAG;YACpB,IAAI,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;SACtB;QACD,UAAU,EAAE,IAAI,CAAC,UAAU,GAAG,WAAW,CAAC,UAAU;KACpD,CAAA;IAED,oFAAoF;IACpF,kFAAkF;IAClF,uFAAuF;IACvF,MAAM,YAAY,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,WAAW,aAAa,EAAE,EAAE,eAAe,CAAC,CAAA;IACtF,MAAM,kBAAkB,CAAC,QAAQ,EAAE,YAAY,CAAC,CAAA;IAChD,MAAM,EAAE,CAAC,6BAA6B,YAAY,EAAE,CAAC,CAAA;IACrD,MAAM,EAAE,CAAC,gBAAgB,QAAQ,CAAC,UAAU,UAAU,IAAI,CAAC,UAAU,WAAW,WAAW,CAAC,UAAU,GAAG,CAAC,CAAA;IAC1G,MAAM,EAAE,CAAC,YAAY,QAAQ,CAAC,MAAM,CAAC,MAAM,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,WAAW,WAAW,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAA;IAC/G,MAAM,EAAE,CAAC,iBAAiB,iBAAiB,EAAE,CAAC,CAAA;AAC/C,CAAC"}
|
|
@@ -1,51 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* Build per-script parquet slices from the DeepSeek-generated transliteration JSONL and emit the
|
|
7
|
-
* corpus-v0.4.0 MANIFEST that combines them with the existing kryptonite + v0.3.0 slices.
|
|
8
|
-
*
|
|
9
|
-
* Sibling to `slice-kryptonite.ts`. The two modules share the same composition pattern: take a
|
|
10
|
-
* base MANIFEST, append new slices, write a combined MANIFEST. Differences specific to
|
|
11
|
-
* transliteration:
|
|
12
|
-
*
|
|
13
|
-
* - One JSONL contains rows from N target scripts (source = `deepseek-translit-<slug>`). We bucket by
|
|
14
|
-
* `source` and write one slice per script so `audit.ts` can attribute each slice to its
|
|
15
|
-
* synthetic source without relying on filename-prefix inference.
|
|
16
|
-
* - Each slice is written to `train/part-translit-<slug>.parquet` (distinct from kryptonite's
|
|
17
|
-
* `part-0000.parquet`, which v0.4.0's first builder already produced).
|
|
18
|
-
* - Inherits the path-canonicalization fix flagged in Thread B's postmortem: v0.3.0 slice paths are
|
|
19
|
-
* rewritten from `$MAILWOMAN_DATA_ROOT/...` to `/data/...` in the combined MANIFEST so
|
|
20
|
-
* all paths share one container-friendly form.
|
|
21
|
-
*
|
|
22
|
-
* See docs/engineering/reference/CORPUS_V0_4_0_GENERATION.md for prompts, model, and the
|
|
23
|
-
* reproducibility contract.
|
|
24
|
-
*
|
|
25
|
-
* Invoke via `mailwoman corpus slice translit \
|
|
26
|
-
* --jsonl /data/corpus/versioned/v0.4.0/transliteration/canonical-transliteration.jsonl \
|
|
27
|
-
* --base-manifest /data/corpus/versioned/v0.4.0/corpus-v0.4.0/MANIFEST.json \
|
|
28
|
-
* --out-dir /data/corpus/versioned/v0.4.0`
|
|
29
|
-
*/
|
|
30
|
-
export interface SliceTranslitOptions {
|
|
31
|
-
jsonl: string;
|
|
32
|
-
baseManifest: string;
|
|
33
|
-
outDir: string;
|
|
34
|
-
/**
|
|
35
|
-
* Default `"0.4.0"`.
|
|
36
|
-
*/
|
|
37
|
-
corpusVersion?: string;
|
|
38
|
-
/**
|
|
39
|
-
* Default `"/data/"`.
|
|
40
|
-
*/
|
|
41
|
-
canonicalPathPrefix?: string;
|
|
42
|
-
/**
|
|
43
|
-
* Prefix the base manifest's slice paths currently carry, to be rewritten to
|
|
44
|
-
* {@link SliceTranslitOptions.canonicalPathPrefix}. Defaults to `mailwomanDataRoot()` with a trailing slash — the
|
|
45
|
-
* root that WROTE those paths. Pass it explicitly when translating a manifest generated under a different
|
|
46
|
-
* `$MAILWOMAN_DATA_ROOT` than the one you are running with.
|
|
47
|
-
*/
|
|
48
|
-
legacyPathPrefix?: string;
|
|
49
|
-
}
|
|
50
|
-
export declare function buildTranslitSlice(options: SliceTranslitOptions, report?: (line: string) => void): Promise<void>;
|
|
51
|
-
//# sourceMappingURL=translit.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"translit.d.ts","sourceRoot":"","sources":["../../../lib/tools/slice/translit.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAaH,MAAM,WAAW,oBAAoB;IACpC,KAAK,EAAE,MAAM,CAAA;IACb,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB;;OAEG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAA;IAC5B;;;;;OAKG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAA;CACzB;AA0DD,wBAAsB,kBAAkB,CACvC,OAAO,EAAE,oBAAoB,EAC7B,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAC7B,OAAO,CAAC,IAAI,CAAC,CAgGf"}
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"translit.js","sourceRoot":"","sources":["../../../lib/tools/slice/translit.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,2BAA2B,CAAA;AAC7D,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,OAAO,EAAE,MAAM,4BAA4B,CAAA;AACnF,OAAO,EAAE,kBAAkB,EAAE,kBAAkB,EAAE,eAAe,EAAE,MAAM,4BAA4B,CAAA;AACpG,OAAO,EAAE,UAAU,EAAE,MAAM,sBAAsB,CAAA;AACjD,OAAO,EAAE,IAAI,EAAE,MAAM,SAAS,CAAA;AAC9B,OAAO,EAAE,eAAe,EAAE,MAAM,aAAa,CAAA;AAG7C,OAAO,EAAE,QAAQ,EAAE,eAAe,EAAE,cAAc,EAAE,YAAY,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,QAAQ,CAAA;AAwBrH,SAAS,cAAc,CAAC,GAA4B,EAAE,aAAqB;IAC1E,OAAO;QACN,GAAG,EAAE,GAAG,CAAC,KAAK,CAAW;QACzB,UAAU,EAAE,GAAG,CAAC,YAAY,CAA2B;QACvD,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,IAAI;QAC3C,MAAM,EAAG,GAAG,CAAC,QAAQ,CAAY,IAAI,SAAS;QAC9C,MAAM,EAAE,GAAG,CAAC,QAAQ,CAAW;QAC/B,SAAS,EAAE,GAAG,CAAC,WAAW,CAAW;QACrC,cAAc,EAAE,aAAa;QAC7B,OAAO,EAAG,GAAG,CAAC,SAAS,CAAY,IAAI,gDAAgD;QACvF,KAAK,EAAE,GAAG,CAAC,OAAO,CAA0B;KAC5C,CAAA;AACF,CAAC;AAED;;;GAGG;AACH,KAAK,UAAU,aAAa,CAAC,IAA2B,EAAE,OAAe,EAAE,MAAc;IACxF,IAAI,aAAa,GAAG,EAAE,CAAA;IACtB,IAAI,YAAY,GAAG,EAAE,CAAA;IAErB,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QACxB,IAAI,aAAa,KAAK,EAAE,EAAE,CAAC;YAC1B,aAAa,GAAG,GAAG,CAAC,SAAS,CAAA;QAC9B,CAAC;QAED,YAAY,GAAG,GAAG,CAAC,SAAS,CAAA;IAC7B,CAAC;IAED,MAAM,gBAAgB,CAAC,IAAI,CAAC,GAAG,CAAC,YAAY,CAAC,EAAE,OAAO,CAAC,CAAA;IAEvD,MAAM,QAAQ,GAAG,MAAM,OAAO,CAAC,OAAO,CAAC,CAAA;IACvC,MAAM,MAAM,GAAG,MAAM,UAAU,CAAC,OAAO,CAAC,CAAA;IAExC,OAAO;QACN,KAAK,EAAE,OAAO;QACd,IAAI,EAAE,OAAO;QACb,MAAM,EAAE,SAAS;QACjB,WAAW,EAAE,iBAAiB;QAC9B,IAAI,EAAE,IAAI,CAAC,MAAM;QACjB,KAAK,EAAE,QAAQ,EAAE,IAAI,IAAI,CAAC;QAC1B,MAAM;QACN,eAAe,EAAE,aAAa;QAC9B,cAAc,EAAE,YAAY;QAC5B,mGAAmG;QACnG,MAAM;KACN,CAAA;AACF,CAAC;AAED,SAAS,qBAAqB,CAAC,IAAY,EAAE,YAAoB,EAAE,eAAuB;IACzF,IAAI,IAAI,CAAC,UAAU,CAAC,YAAY,CAAC;QAAE,OAAO,eAAe,GAAG,IAAI,CAAC,KAAK,CAAC,YAAY,CAAC,MAAM,CAAC,CAAA;IAE3F,OAAO,IAAI,CAAA;AACZ,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,kBAAkB,CACvC,OAA6B,EAC7B,MAA+B;IAE/B,MAAM,aAAa,GAAG,OAAO,CAAC,aAAa,IAAI,OAAO,CAAA;IACtD,wEAAwE;IACxE,MAAM,mBAAmB,GAAG,OAAO,CAAC,mBAAmB,IAAI,QAAQ,CAAA;IACnE,MAAM,gBAAgB,GAAG,OAAO,CAAC,gBAAgB,IAAI,GAAG,iBAAiB,EAAE,GAAG,CAAA;IAE9E,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,oBAAoB,OAAO,CAAC,KAAK,EAAE,CAAC,CAAA;IAE5F,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC;QAAE,MAAM,IAAI,KAAK,CAAC,4BAA4B,OAAO,CAAC,YAAY,EAAE,CAAC,CAAA;IAElH,MAAM,SAAS,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,WAAW,aAAa,EAAE,CAAC,CAAA;IAClE,MAAM,QAAQ,GAAG,IAAI,CAAC,SAAS,EAAE,OAAO,CAAC,CAAA;IACzC,MAAM,eAAe,CAAC,QAAQ,CAAC,CAAA;IAE/B,gEAAgE;IAChE,MAAM,OAAO,GAAG,IAAI,GAAG,EAAwB,CAAA;IAC/C,MAAM,UAAU,GAAa,EAAE,CAAA;IAC/B,IAAI,OAAO,GAAG,CAAC,CAAA;IAEf,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,eAAe,CAAC,SAAS,CAA0B,OAAO,CAAC,KAAK,CAAC,EAAE,CAAC;QAC3F,OAAO,EAAE,CAAA;QACT,MAAM,KAAK,GAAG,cAAc,CAAC,GAAG,EAAE,aAAa,CAAC,CAAA;QAChD,MAAM,MAAM,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAA;QAE9B,IAAI,MAAM,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;YAC/B,UAAU,CAAC,IAAI,CAAC,GAAG,KAAK,CAAC,SAAS,KAAK,MAAM,CAAC,GAAG,CAAC,MAAM,EAAE,CAAC,CAAA;YAE3D,SAAQ;QACT,CAAC;QAED,MAAM,MAAM,GAAG,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,MAAM,CAAC,CAAA;QAExC,IAAI,MAAM,EAAE,CAAC;YACZ,MAAM,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAA;QACxB,CAAC;aAAM,CAAC;YACP,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,MAAM,EAAE,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;QACxC,CAAC;IACF,CAAC;IAED,MAAM,EAAE,CAAC,QAAQ,OAAO,UAAU,UAAU,CAAC,MAAM,iBAAiB,OAAO,CAAC,IAAI,iBAAiB,CAAC,CAAA;IAElG,MAAM,SAAS,GAAsB,EAAE,CAAA;IACvC,MAAM,UAAU,GAAG,CAAC,GAAG,OAAO,CAAC,IAAI,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAA;IAEjD,KAAK,MAAM,MAAM,IAAI,UAAU,EAAE,CAAC;QACjC,MAAM,IAAI,GAAG,OAAO,CAAC,GAAG,CAAC,MAAM,CAAE,CAAA;QACjC,MAAM,IAAI,GAAG,MAAM,CAAC,UAAU,CAAC,oBAAoB,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,oBAAoB,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;QACzG,MAAM,OAAO,GAAG,IAAI,CAAC,QAAQ,EAAE,iBAAiB,IAAI,UAAU,CAAC,CAAA;QAC/D,MAAM,UAAU,GAAG,MAAM,aAAa,CAAC,IAAI,EAAE,OAAO,EAAE,MAAM,CAAC,CAAA;QAC7D,SAAS,CAAC,IAAI,CAAC,UAAU,CAAC,CAAA;QAC1B,MAAM,EAAE,CAAC,KAAK,MAAM,KAAK,UAAU,CAAC,IAAI,WAAW,OAAO,KAAK,UAAU,CAAC,KAAK,SAAS,CAAC,CAAA;IAC1F,CAAC;IAED,IAAI,UAAU,CAAC,MAAM,EAAE,CAAC;QACvB,MAAM,KAAK,GAAG,IAAI,CAAC,SAAS,EAAE,gCAAgC,CAAC,CAAA;QAC/D,MAAM,kBAAkB,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,IAAI,EAAE,KAAK,CAAC,CAAA;QAC7D,MAAM,EAAE,CAAC,oBAAoB,KAAK,KAAK,UAAU,CAAC,MAAM,QAAQ,CAAC,CAAA;IAClE,CAAC;IAED,iGAAiG;IACjG,+FAA+F;IAC/F,sBAAsB;IACtB,MAAM,IAAI,GAAG,MAAM,iBAAiB,CAAgB,OAAO,CAAC,YAAY,CAAC,CAAA;IAEzE,MAAM,aAAa,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,EAAE,EAAE,EAAE,CAAC,CAAC;QAC9C,GAAG,EAAE;QACL,IAAI,EAAE,qBAAqB,CAAC,EAAE,CAAC,IAAI,EAAE,gBAAgB,EAAE,mBAAmB,CAAC;KAC3E,CAAC,CAAC,CAAA;IAEH,MAAM,YAAY,GAAG,SAAS,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,EAAE,EAAE,EAAE,CAAC,GAAG,GAAG,EAAE,CAAC,IAAI,EAAE,CAAC,CAAC,CAAA;IAEpE,MAAM,QAAQ,GAAkB;QAC/B,cAAc,EAAE,aAAa;QAC7B,MAAM,EAAE,eAAe;QACvB,cAAc,EAAE,IAAI,CAAC,cAAc;QACnC,cAAc,EAAE,IAAI,CAAC,cAAc,IAAI,cAAc;QACrD,MAAM,EAAE,CAAC,GAAG,aAAa,EAAE,GAAG,SAAS,CAAC;QACxC,MAAM,EAAE;YACP,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,GAAG,YAAY;YACvC,GAAG,EAAE,IAAI,CAAC,MAAM,CAAC,GAAG;YACpB,IAAI,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;SACtB;QACD,UAAU,EAAE,IAAI,CAAC,UAAU,GAAG,YAAY;KAC1C,CAAA;IAED,MAAM,YAAY,GAAG,IAAI,CAAC,SAAS,EAAE,eAAe,CAAC,CAAA;IACrD,MAAM,kBAAkB,CAAC,QAAQ,EAAE,YAAY,CAAC,CAAA;IAChD,MAAM,EAAE,CAAC,6BAA6B,YAAY,EAAE,CAAC,CAAA;IACrD,MAAM,EAAE,CAAC,gBAAgB,QAAQ,CAAC,UAAU,UAAU,IAAI,CAAC,UAAU,WAAW,YAAY,GAAG,CAAC,CAAA;IAChG,MAAM,EAAE,CAAC,YAAY,QAAQ,CAAC,MAAM,CAAC,MAAM,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,WAAW,SAAS,CAAC,MAAM,GAAG,CAAC,CAAA;IACtG,MAAM,EAAE,CAAC,iBAAiB,iBAAiB,EAAE,CAAC,CAAA;IAC9C,MAAM,OAAO,GAAG,aAAa,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,IAAI,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,IAAI,CAAC,CAAC,MAAM,CAAA;IAEtF,IAAI,OAAO,GAAG,CAAC,EAAE,CAAC;QACjB,MAAM,EAAE,CAAC,qCAAqC,OAAO,aAAa,gBAAgB,QAAQ,mBAAmB,IAAI,CAAC,CAAA;IACnH,CAAC;AACF,CAAC"}
|
|
@@ -1,100 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* @file The harvest side of the sub-venue lexicon: what one row of a harvestable source looks like,
|
|
7
|
-
* how a file of them is read back, and what a pass over them extracts — attested surfaces plus the
|
|
8
|
-
* measured shape of each designator's identifier half.
|
|
9
|
-
*
|
|
10
|
-
* {@link SubVenueHarvestRow} is SOURCE-NEUTRAL, and verified so: an Overture Places row from the
|
|
11
|
-
* `airport_terminal` category is `{ designatorID, name }` and fits unchanged. It is declared here
|
|
12
|
-
* rather than imported so the builder does not take a dependency on `@mailwoman/osm` (which
|
|
13
|
-
* `@mailwoman/corpus` does not depend on) just to name a shape it reads from a file.
|
|
14
|
-
*
|
|
15
|
-
* Two measurements shape {@link extractAttestedPhrases}, and both are the reason it is safe to run
|
|
16
|
-
* over raw OSM at all:
|
|
17
|
-
*
|
|
18
|
-
* **A feature's `name` is usually the VENUE's name, not a sub-venue phrase.** On the Berlin extract
|
|
19
|
-
* (2,060 matched features) a `railway=platform` is named `Stendaler Straße` and an `amenity=university`
|
|
20
|
-
* is named `Hertie School`; across 250,116 named Great Britain features only 6,003 (2.40%) contain a
|
|
21
|
-
* designator token at all. So names are not harvested wholesale — a name contributes only when it
|
|
22
|
-
* CONTAINS a phrase already in the surface index, which is what makes `Terminal E (Untere Ebene)`
|
|
23
|
-
* evidence and `Otto Lilienthal Flughafen Berlin Tegel` not.
|
|
24
|
-
*
|
|
25
|
-
* **A matched phrase belongs to the record the PHRASE names, not the record the ROW carries.**
|
|
26
|
-
* Attributing every hit to `row.designatorID` — the rule that matched the FEATURE — produced `west →
|
|
27
|
-
* platform`, `hall → platform` and `biggin → platform` on the GB extract, because a bus stop tagged
|
|
28
|
-
* `public_transport=platform` is named "Village Hall" or "West Kensington"; 108 of 133 OSM-derived
|
|
29
|
-
* surfaces named a different record than the one they pointed at. Attribution therefore runs through
|
|
30
|
-
* the index, and the row's own designator is kept as `context` — exactly the axis a confound board
|
|
31
|
-
* needs, since a `hall` seen on a platform is a confound and a `hall` seen on a terminal is evidence.
|
|
32
|
-
*/
|
|
33
|
-
import { type SurfaceIndex } from "#tools/sub/venue/surfaces";
|
|
34
|
-
import type { IdentifierShape, SubVenueSurface } from "#tools/sub/venue/table";
|
|
35
|
-
/**
|
|
36
|
-
* One row of a harvestable source, as read back off JSONL or out of a layer database.
|
|
37
|
-
*
|
|
38
|
-
* SOURCE-NEUTRAL by design, and verified so in wave 2: an Overture Places row from the `airport_terminal` category is
|
|
39
|
-
* `{ designatorID, name }` and fits unchanged. What did NOT fit was the harvest function's hardcoded `osm:name` source
|
|
40
|
-
* stamp — see `overture-subvenue.ts`'s docstring. Declared locally so the builder does not import `@mailwoman/osm`
|
|
41
|
-
* (which `@mailwoman/corpus` does not depend on) just to name a shape it reads from a file.
|
|
42
|
-
*/
|
|
43
|
-
export interface SubVenueHarvestRow {
|
|
44
|
-
/**
|
|
45
|
-
* The designator the SOURCE's rule assigned to the FEATURE. Not necessarily the record a matched phrase names — see
|
|
46
|
-
* this file's header on phrase attribution. Carried into {@link SubVenueSurface.context}.
|
|
47
|
-
*/
|
|
48
|
-
designatorID: string;
|
|
49
|
-
/**
|
|
50
|
-
* `venue` (station, airport, campus) or `sub_venue`, when the source's rule assigns one. The harvest itself does not
|
|
51
|
-
* read it; the sub-venue slice recipe fills its venue slot from it.
|
|
52
|
-
*/
|
|
53
|
-
tier?: string;
|
|
54
|
-
name?: string | null;
|
|
55
|
-
ref?: string | null;
|
|
56
|
-
localizedNames?: Record<string, string>;
|
|
57
|
-
}
|
|
58
|
-
/**
|
|
59
|
-
* Classify an OSM `ref` into an {@link IdentifierShape} class.
|
|
60
|
-
*
|
|
61
|
-
* The classes are the ones Berlin's gates actually produced, plus the two aviation forms the corpus task names
|
|
62
|
-
* (`Terminal 2F` is digit-letter, `Concourse B` is letter). `range` covers both separators OSM uses for a gate serving
|
|
63
|
-
* more than one stand: `16-18` and `0/1`.
|
|
64
|
-
*/
|
|
65
|
-
export declare function classifyIdentifier(ref: string): string;
|
|
66
|
-
/**
|
|
67
|
-
* Options for one harvest pass — which source stamp its surfaces carry and which region they were attested in.
|
|
68
|
-
*
|
|
69
|
-
* Both default to the OSM/unknown-region values wave 1 hardcoded, so an existing caller is unchanged.
|
|
70
|
-
*/
|
|
71
|
-
export interface HarvestOptions {
|
|
72
|
-
/**
|
|
73
|
-
* Source family: `osm` yields `osm:name` / `osm:name:<lang>`, `overture` yields `overture:name`.
|
|
74
|
-
*/
|
|
75
|
-
source?: string;
|
|
76
|
-
/**
|
|
77
|
-
* ISO 3166-1 alpha-2 of the extract or partition. `""` when unknown.
|
|
78
|
-
*/
|
|
79
|
-
region?: string;
|
|
80
|
-
}
|
|
81
|
-
/**
|
|
82
|
-
* Harvest attested phrases and identifier shapes out of a source's rows.
|
|
83
|
-
*
|
|
84
|
-
* `index` gates the name harvest — a name contributes only when it CONTAINS a phrase already in the table. That filter
|
|
85
|
-
* is the whole reason this function is safe to run over raw OSM: see this file's header for the Berlin measurement that
|
|
86
|
-
* motivated it. The index also decides ATTRIBUTION: a hit is a surface of the record the PHRASE names, and the row's
|
|
87
|
-
* own designator is recorded as `context`.
|
|
88
|
-
*
|
|
89
|
-
* Returns surfaces with real `observations` counts, so the lexicon can rank `terminal` above a phrase attested once.
|
|
90
|
-
*/
|
|
91
|
-
export declare function extractAttestedPhrases(rows: Iterable<SubVenueHarvestRow>, index: SurfaceIndex, options?: HarvestOptions): {
|
|
92
|
-
surfaces: SubVenueSurface[];
|
|
93
|
-
identifierShapes: IdentifierShape[];
|
|
94
|
-
};
|
|
95
|
-
/**
|
|
96
|
-
* Read a JSONL file of {@link SubVenueHarvestRow}s. Blank lines and unparseable rows are skipped rather than fatal — an
|
|
97
|
-
* extract is a build output, and one malformed line should not cost the whole lexicon.
|
|
98
|
-
*/
|
|
99
|
-
export declare function readSubVenueJSONL(path: string): Promise<SubVenueHarvestRow[]>;
|
|
100
|
-
//# sourceMappingURL=harvest.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"harvest.d.ts","sourceRoot":"","sources":["../../../../lib/tools/sub/venue/harvest.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA+BG;AAMH,OAAO,EAAwB,KAAK,YAAY,EAAE,MAAM,2BAA2B,CAAA;AACnF,OAAO,KAAK,EAAE,eAAe,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AAE9E;;;;;;;GAOG;AACH,MAAM,WAAW,kBAAkB;IAClC;;;OAGG;IACH,YAAY,EAAE,MAAM,CAAA;IACpB;;;OAGG;IACH,IAAI,CAAC,EAAE,MAAM,CAAA;IACb,IAAI,CAAC,EAAE,MAAM,GAAG,IAAI,CAAA;IACpB,GAAG,CAAC,EAAE,MAAM,GAAG,IAAI,CAAA;IACnB,cAAc,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;CACvC;AAED;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CActD;AAYD;;;;GAIG;AACH,MAAM,WAAW,cAAc;IAC9B;;OAEG;IACH,MAAM,CAAC,EAAE,MAAM,CAAA;IACf;;OAEG;IACH,MAAM,CAAC,EAAE,MAAM,CAAA;CACf;AAED;;;;;;;;;GASG;AACH,wBAAgB,sBAAsB,CACrC,IAAI,EAAE,QAAQ,CAAC,kBAAkB,CAAC,EAClC,KAAK,EAAE,YAAY,EACnB,OAAO,GAAE,cAAmB,GAC1B;IAAE,QAAQ,EAAE,eAAe,EAAE,CAAC;IAAC,gBAAgB,EAAE,eAAe,EAAE,CAAA;CAAE,CAiFtE;AAED;;;GAGG;AACH,wBAAsB,iBAAiB,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,kBAAkB,EAAE,CAAC,CAmBnF"}
|