@mailwoman/corpus 10.0.0 → 10.2.0
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 +791 -0
- package/lib/adapters/dedup-key.ts +40 -0
- package/lib/adapters/errors.ts +21 -0
- package/lib/adapters/geonames/adapter.ts +43 -53
- package/lib/adapters/geonames/postal/adapter.ts +31 -37
- package/lib/adapters/gleif/adapter.ts +829 -0
- package/lib/adapters/libpostal-dictionary.ts +37 -0
- package/lib/adapters/openaddresses/adapter.ts +52 -55
- package/lib/adapters/osm/adapter.ts +162 -83
- package/lib/adapters/overture/adapter.ts +52 -43
- package/lib/adapters/po-box/adapter.ts +207 -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 +136 -0
- package/lib/adapters/wof/admin/json/adapter.ts +50 -35
- package/lib/adapters/wof/json-rows.ts +57 -26
- package/lib/adapters/wof/postalcode-json/adapter.ts +63 -61
- package/lib/adapters.ts +230 -0
- package/lib/au/adapters/gnaf/adapter.ts +36 -44
- 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 +667 -0
- package/lib/be/adapters/wallonie/adapter.ts +537 -0
- package/lib/be/tools/fetch/brussels.ts +242 -0
- package/lib/be/tools/fetch/wallonie.ts +275 -0
- package/lib/build/adapters.ts +111 -0
- package/lib/build/align.ts +365 -0
- package/lib/build/checkpoint.ts +133 -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 +234 -0
- package/lib/build.ts +127 -315
- package/lib/cn/recipes/organizational-units.ts +20 -41
- package/lib/cz/adapters/cuzk/adapter.ts +320 -0
- package/lib/cz/recipes/pcfirst-preposition.ts +40 -27
- package/lib/cz/tools/fetch/cuzk.ts +710 -0
- package/lib/de/recipes/locale.ts +69 -59
- package/lib/directory.ts +31 -0
- package/lib/dk/adapters/inspire/adapter.ts +345 -0
- package/lib/dk/tools/fetch/inspire.ts +323 -0
- package/lib/ee/adapters/ads/adapter.ts +240 -0
- package/lib/ee/tools/fetch/ads.ts +272 -0
- package/lib/env.ts +6 -2
- package/lib/es/adapters/bizkaia/adapter.ts +388 -0
- package/lib/es/adapters/catastro/adapter.ts +345 -0
- package/lib/es/adapters/gipuzkoa/adapter.ts +256 -0
- package/lib/es/adapters/navarra/adapter.ts +290 -0
- package/lib/es/cadastre.ts +194 -0
- package/lib/es/tools/fetch/bizkaia.ts +752 -0
- package/lib/es/tools/fetch/catastro.ts +765 -0
- package/lib/es/tools/fetch/gipuzkoa.ts +338 -0
- package/lib/es/tools/fetch/navarra.ts +496 -0
- package/lib/fi/adapters/ryhti/adapter.ts +319 -0
- package/lib/fi/tools/fetch/ryhti.ts +476 -0
- package/lib/fingerprints.ts +154 -0
- package/lib/fr/adapters/annuaire-education/adapter.ts +600 -0
- package/lib/fr/adapters/ban/adapter.ts +87 -59
- package/lib/fr/adapters/finess/adapter.ts +619 -0
- package/lib/fr/adapters/finess/postal-lines.ts +157 -0
- package/lib/fr/adapters/finess/voie-type.ts +201 -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 +111 -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 +309 -0
- package/lib/fr/tools/fetch/ban.ts +30 -20
- package/lib/fr/tools/fetch/finess.ts +353 -0
- package/lib/gauntlet-inputs.ts +92 -0
- package/lib/index.ts +9 -3
- package/lib/inspire/address.ts +465 -0
- package/lib/inspire/archive.ts +44 -0
- package/lib/inspire/errors.ts +72 -0
- package/lib/inspire/stream.ts +265 -0
- package/lib/international/recipes/locale.ts +146 -184
- package/lib/it/adapters/anac/adapter.ts +527 -0
- package/lib/it/tools/fetch/anac.ts +391 -0
- package/lib/jp/adapters/wof-admin/adapter.ts +29 -23
- package/lib/jp/tools/fetch/houjin.ts +19 -21
- 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 +34 -27
- package/lib/kr/tools/fetch/juso.ts +19 -21
- package/lib/kr/tools/fetch/localdata.ts +29 -30
- package/lib/lu/adapters/bd-adresses/adapter.ts +240 -0
- package/lib/lu/tools/fetch/bd-adresses.ts +353 -0
- package/lib/lv/adapters/varis/adapter.ts +500 -0
- package/lib/lv/tools/fetch/varis.ts +215 -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 +302 -0
- package/lib/no/adapters/matrikkelen/adapter.ts +188 -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 +531 -0
- package/lib/parquet/duckdb.ts +147 -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 +452 -0
- package/lib/pl/adapters/emuia/adapter.ts +202 -0
- package/lib/pl/tools/fetch/emuia.ts +361 -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 +37 -20
- package/lib/recipes/country-balanced.ts +66 -86
- package/lib/recipes/house-venue.ts +36 -14
- package/lib/recipes/intersection.ts +84 -62
- 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 +31 -19
- package/lib/recipes/po/box/cedex/sources.ts +16 -4
- 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 +267 -158
- package/lib/recipes/sources.ts +412 -0
- package/lib/recipes/street/affix.ts +54 -154
- package/lib/recipes/street/bare.ts +30 -17
- package/lib/recipes/street.ts +35 -9
- package/lib/recipes/sub/venue/context.ts +32 -23
- package/lib/recipes/sub/venue/render.ts +170 -0
- package/lib/recipes/sub/venue/sources.ts +92 -248
- package/lib/recipes/sub/venue.ts +795 -0
- package/lib/recipes/trailing-region.ts +75 -87
- package/lib/recipes/unit/sources.ts +44 -0
- package/lib/recipes/unit.ts +101 -89
- package/lib/recipes/venue.ts +347 -0
- package/lib/{recipes/index.ts → recipes.ts} +15 -11
- package/lib/registers.ts +280 -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 +221 -84
- package/lib/sg/recipes/register.ts +47 -53
- package/lib/sg/tools/fetch/acra.ts +23 -28
- package/lib/si/recipes/bare-village.ts +34 -17
- package/lib/sk/adapters/inspire/adapter.ts +393 -0
- package/lib/sk/tools/fetch/inspire.ts +403 -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 +589 -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 +248 -0
- package/lib/synthesizers/anchor-absorption.ts +56 -91
- package/lib/synthesizers/boundary-stress.ts +67 -155
- package/lib/synthesizers/house-venue.ts +65 -127
- package/lib/synthesizers/no-street.ts +71 -119
- package/lib/synthesizers/po-box.ts +93 -105
- package/lib/synthesizers/tail.ts +55 -0
- package/lib/synthesizers/utils.ts +82 -219
- package/lib/test-kit/corpus-recipe.ts +45 -28
- package/lib/test-kit.ts +116 -0
- package/lib/tools/align-canonical.ts +64 -0
- package/lib/tools/audit.ts +117 -120
- package/lib/tools/corpus-stats.ts +45 -31
- package/lib/tools/fetch/atom.ts +164 -0
- package/lib/tools/fetch/cordis.ts +320 -0
- package/lib/tools/fetch/download/manifest.ts +36 -23
- package/lib/tools/fetch/download/network.ts +55 -31
- package/lib/tools/fetch/download/retry.ts +30 -29
- package/lib/tools/fetch/geonames/dump.ts +75 -55
- package/lib/tools/fetch/geonames/postal.ts +27 -62
- package/lib/tools/fetch/gleif.ts +362 -0
- package/lib/tools/fetch/header.ts +71 -0
- package/lib/tools/fetch/inspire-addresses.ts +500 -0
- package/lib/tools/fetch/openaddresses.ts +23 -112
- 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 +954 -0
- package/lib/tools/fetch/wikidata-subvenue.ts +45 -145
- package/lib/tools/fetch/zip-archive.ts +100 -0
- package/lib/tools/fetch.ts +101 -0
- package/lib/tools/golden/expand.ts +75 -44
- package/lib/tools/golden/promote.ts +47 -34
- package/lib/tools/golden/relabel-street/directory.ts +67 -39
- package/lib/tools/golden/relabel-street/row.ts +32 -109
- package/lib/tools/ingest-csv.ts +48 -51
- package/lib/tools/jsonl-to-parquet.ts +59 -98
- package/lib/tools/lint/recipe-output/vocab.ts +371 -0
- package/lib/tools/lint/{slice/index.ts → recipe-output.ts} +130 -126
- 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 +47 -39
- package/lib/tools/overlay/manifest.ts +409 -0
- package/lib/tools/overlay/merge-source.ts +139 -0
- package/lib/tools/overlay/split-slice.ts +144 -0
- package/lib/tools/overlay/translit.ts +236 -0
- package/lib/tools/overture-subvenue.ts +18 -91
- package/lib/tools/postcode-triples.ts +284 -212
- package/lib/tools/source-register/build.ts +822 -0
- package/lib/tools/sub/venue/head-nouns.ts +38 -75
- package/lib/tools/sub/venue/lexicon.ts +71 -110
- package/lib/tools/sub/venue/wikidata.ts +13 -10
- package/lib/tw/tools/fetch/gcis.ts +21 -20
- package/lib/types.ts +305 -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 +8 -3
- 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 +61 -122
- package/lib/us/adapters/usgov/nppes/adapter.ts +72 -35
- package/lib/us/adapters/usgov/samhsa-treatment-locator/adapter.ts +47 -34
- package/lib/us/fips-state.ts +19 -10
- package/lib/{synthesizers → us/synthesizers}/intersection.ts +45 -35
- package/lib/{synthesizers → us/synthesizers}/street.ts +56 -43
- package/lib/us/tools/fetch/hrsa.ts +7 -8
- package/lib/us/tools/fetch/imls-pls.ts +18 -36
- package/lib/us/tools/fetch/nad.ts +51 -33
- package/lib/us/tools/fetch/nppes.ts +27 -39
- package/lib/us/tools/fetch/state/hi-schools.ts +24 -17
- package/lib/us/tools/fetch/state/sources.ts +12 -12
- package/lib/us/tools/fetch/tiger-full.ts +22 -74
- package/lib/utils/align.ts +52 -59
- package/lib/utils/{slice-stats.ts → cooccurrence-stats.ts} +22 -16
- package/lib/utils/golden.ts +36 -51
- package/lib/utils/license.ts +239 -27
- package/lib/utils/split.ts +209 -88
- package/lib/utils/tokenize.ts +25 -56
- package/lib/utils/wof-json.ts +130 -77
- 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 +326 -0
- package/out/adapters/cordis/adapter.d.ts.map +1 -0
- package/out/adapters/cordis/adapter.js +622 -0
- package/out/adapters/cordis/adapter.js.map +1 -0
- package/out/adapters/dedup-key.d.ts +30 -0
- package/out/adapters/dedup-key.d.ts.map +1 -0
- package/out/adapters/dedup-key.js +35 -0
- package/out/adapters/dedup-key.js.map +1 -0
- package/out/adapters/errors.d.ts +17 -0
- package/out/adapters/errors.d.ts.map +1 -0
- package/out/adapters/errors.js +20 -0
- package/out/adapters/errors.js.map +1 -0
- package/out/adapters/geonames/adapter.d.ts +26 -28
- package/out/adapters/geonames/adapter.d.ts.map +1 -1
- package/out/adapters/geonames/adapter.js +40 -50
- package/out/adapters/geonames/adapter.js.map +1 -1
- package/out/adapters/geonames/postal/adapter.d.ts +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 +685 -0
- package/out/adapters/gleif/adapter.js.map +1 -0
- package/out/adapters/libpostal-dictionary.d.ts +20 -0
- package/out/adapters/libpostal-dictionary.d.ts.map +1 -0
- package/out/adapters/libpostal-dictionary.js +34 -0
- package/out/adapters/libpostal-dictionary.js.map +1 -0
- package/out/adapters/openaddresses/adapter.d.ts +29 -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 +59 -52
- package/out/adapters/osm/adapter.d.ts.map +1 -1
- package/out/adapters/osm/adapter.js +119 -69
- 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 +66 -0
- package/out/adapters/po-box/adapter.d.ts.map +1 -0
- package/out/adapters/{synth-po-box → po-box}/adapter.js +48 -47
- 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 +103 -0
- package/out/adapters/street-line.d.ts.map +1 -0
- package/out/adapters/street-line.js +118 -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 +48 -32
- package/out/adapters/wof/admin/json/adapter.js.map +1 -1
- package/out/adapters/wof/json-rows.d.ts +36 -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 +60 -58
- 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 +35 -43
- 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 +230 -0
- package/out/be/adapters/vlaanderen/adapter.d.ts.map +1 -0
- package/out/be/adapters/vlaanderen/adapter.js +418 -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 +122 -0
- package/out/be/tools/fetch/brussels.d.ts.map +1 -0
- package/out/be/tools/fetch/brussels.js +148 -0
- package/out/be/tools/fetch/brussels.js.map +1 -0
- package/out/be/tools/fetch/wallonie.d.ts +120 -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 +69 -0
- package/out/build/adapters.js.map +1 -0
- package/out/build/align.d.ts +74 -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 +61 -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 +222 -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 +269 -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 +281 -0
- package/out/cz/tools/fetch/cuzk.d.ts.map +1 -0
- package/out/cz/tools/fetch/cuzk.js +448 -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 +67 -57
- 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 +153 -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 +101 -0
- package/out/dk/tools/fetch/inspire.d.ts.map +1 -0
- package/out/dk/tools/fetch/inspire.js +232 -0
- package/out/dk/tools/fetch/inspire.js.map +1 -0
- package/out/ee/adapters/ads/adapter.d.ts +102 -0
- package/out/ee/adapters/ads/adapter.d.ts.map +1 -0
- package/out/ee/adapters/ads/adapter.js +185 -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 +197 -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 +103 -0
- package/out/es/adapters/catastro/adapter.d.ts.map +1 -0
- package/out/es/adapters/catastro/adapter.js +290 -0
- package/out/es/adapters/catastro/adapter.js.map +1 -0
- package/out/es/adapters/gipuzkoa/adapter.d.ts +84 -0
- package/out/es/adapters/gipuzkoa/adapter.d.ts.map +1 -0
- package/out/es/adapters/gipuzkoa/adapter.js +218 -0
- package/out/es/adapters/gipuzkoa/adapter.js.map +1 -0
- package/out/es/adapters/navarra/adapter.d.ts +106 -0
- package/out/es/adapters/navarra/adapter.d.ts.map +1 -0
- package/out/es/adapters/navarra/adapter.js +242 -0
- package/out/es/adapters/navarra/adapter.js.map +1 -0
- package/out/es/cadastre.d.ts +110 -0
- package/out/es/cadastre.d.ts.map +1 -0
- package/out/es/cadastre.js +128 -0
- package/out/es/cadastre.js.map +1 -0
- package/out/es/tools/fetch/bizkaia.d.ts +281 -0
- package/out/es/tools/fetch/bizkaia.d.ts.map +1 -0
- package/out/es/tools/fetch/bizkaia.js +478 -0
- package/out/es/tools/fetch/bizkaia.js.map +1 -0
- package/out/es/tools/fetch/catastro.d.ts +280 -0
- package/out/es/tools/fetch/catastro.d.ts.map +1 -0
- package/out/es/tools/fetch/catastro.js +491 -0
- package/out/es/tools/fetch/catastro.js.map +1 -0
- package/out/es/tools/fetch/gipuzkoa.d.ts +164 -0
- package/out/es/tools/fetch/gipuzkoa.d.ts.map +1 -0
- package/out/es/tools/fetch/gipuzkoa.js +210 -0
- package/out/es/tools/fetch/gipuzkoa.js.map +1 -0
- package/out/es/tools/fetch/navarra.d.ts +194 -0
- package/out/es/tools/fetch/navarra.d.ts.map +1 -0
- package/out/es/tools/fetch/navarra.js +321 -0
- package/out/es/tools/fetch/navarra.js.map +1 -0
- package/out/fi/adapters/ryhti/adapter.d.ts +111 -0
- package/out/fi/adapters/ryhti/adapter.d.ts.map +1 -0
- package/out/fi/adapters/ryhti/adapter.js +240 -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 +324 -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 +477 -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 +277 -0
- package/out/fr/adapters/finess/adapter.d.ts.map +1 -0
- package/out/fr/adapters/finess/adapter.js +520 -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 +173 -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 +89 -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 +133 -0
- package/out/fr/tools/fetch/annuaire-education.d.ts.map +1 -0
- package/out/fr/tools/fetch/annuaire-education.js +198 -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 +29 -19
- 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 +229 -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 +261 -0
- package/out/inspire/address.d.ts.map +1 -0
- package/out/inspire/address.js +395 -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 +56 -0
- package/out/inspire/errors.d.ts.map +1 -0
- package/out/inspire/errors.js +65 -0
- package/out/inspire/errors.js.map +1 -0
- package/out/inspire/stream.d.ts +140 -0
- package/out/inspire/stream.d.ts.map +1 -0
- package/out/inspire/stream.js +123 -0
- package/out/inspire/stream.js.map +1 -0
- package/out/international/recipes/locale.d.ts +40 -86
- package/out/international/recipes/locale.d.ts.map +1 -1
- package/out/international/recipes/locale.js +122 -157
- package/out/international/recipes/locale.js.map +1 -1
- package/out/it/adapters/anac/adapter.d.ts +269 -0
- package/out/it/adapters/anac/adapter.d.ts.map +1 -0
- package/out/it/adapters/anac/adapter.js +434 -0
- package/out/it/adapters/anac/adapter.js.map +1 -0
- package/out/it/tools/fetch/anac.d.ts +171 -0
- package/out/it/tools/fetch/anac.d.ts.map +1 -0
- package/out/it/tools/fetch/anac.js +269 -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 +17 -13
- 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 +31 -26
- 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 +24 -16
- 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 +165 -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 +284 -0
- package/out/lv/adapters/varis/adapter.d.ts.map +1 -0
- package/out/lv/adapters/varis/adapter.js +405 -0
- package/out/lv/adapters/varis/adapter.js.map +1 -0
- package/out/lv/tools/fetch/varis.d.ts +81 -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 +309 -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 +159 -0
- package/out/nl/tools/fetch/kadaster.d.ts.map +1 -0
- package/out/nl/tools/fetch/kadaster.js +182 -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 +263 -0
- package/out/no/tools/fetch/matrikkelen.d.ts.map +1 -0
- package/out/no/tools/fetch/matrikkelen.js +353 -0
- package/out/no/tools/fetch/matrikkelen.js.map +1 -0
- package/out/parquet/duckdb.d.ts +103 -0
- package/out/parquet/duckdb.d.ts.map +1 -0
- package/out/parquet/duckdb.js +93 -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 +252 -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 +166 -0
- package/out/pl/tools/fetch/emuia.d.ts.map +1 -0
- package/out/pl/tools/fetch/emuia.js +261 -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} +92 -80
- 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 +37 -20
- 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 +33 -14
- 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 +67 -55
- 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 +27 -16
- package/out/recipes/po/box/cedex/recipe.js.map +1 -1
- package/out/recipes/po/box/cedex/sources.d.ts +20 -14
- package/out/recipes/po/box/cedex/sources.d.ts.map +1 -1
- package/out/recipes/po/box/cedex/sources.js +2 -2
- 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 +223 -141
- package/out/recipes/scaffold.d.ts.map +1 -1
- package/out/recipes/scaffold.js +75 -65
- 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 +361 -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 +27 -16
- 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 +32 -9
- 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 +23 -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 +53 -93
- package/out/recipes/sub/venue/sources.d.ts.map +1 -1
- package/out/recipes/sub/venue/sources.js +57 -197
- package/out/recipes/sub/venue/sources.js.map +1 -1
- package/out/recipes/sub/venue.d.ts +173 -0
- package/out/recipes/sub/venue.d.ts.map +1 -0
- package/out/recipes/sub/venue.js +558 -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 +61 -86
- 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/venue.d.ts +119 -0
- package/out/recipes/venue.d.ts.map +1 -0
- package/out/recipes/venue.js +245 -0
- package/out/recipes/venue.js.map +1 -0
- package/out/recipes.d.ts +26 -0
- package/out/recipes.d.ts.map +1 -0
- package/out/{recipes/index.js → recipes.js} +16 -12
- package/out/recipes.js.map +1 -0
- package/out/registers.d.ts +279 -0
- package/out/registers.d.ts.map +1 -0
- package/out/registers.js +275 -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 +121 -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 +45 -50
- 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 +18 -21
- 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 +196 -0
- package/out/sk/tools/fetch/inspire.d.ts.map +1 -0
- package/out/sk/tools/fetch/inspire.js +281 -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 +538 -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 +138 -0
- package/out/surfaces/locale.js.map +1 -0
- package/out/synthesizers/anchor-absorption.d.ts +22 -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 +18 -49
- package/out/synthesizers/boundary-stress.d.ts.map +1 -1
- package/out/synthesizers/boundary-stress.js +45 -142
- package/out/synthesizers/boundary-stress.js.map +1 -1
- package/out/synthesizers/house-venue.d.ts +23 -25
- package/out/synthesizers/house-venue.d.ts.map +1 -1
- package/out/synthesizers/house-venue.js +47 -118
- package/out/synthesizers/house-venue.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 +48 -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/tail.d.ts +36 -0
- package/out/synthesizers/tail.d.ts.map +1 -0
- package/out/synthesizers/tail.js +32 -0
- package/out/synthesizers/tail.js.map +1 -0
- package/out/synthesizers/utils.d.ts +33 -129
- package/out/synthesizers/utils.d.ts.map +1 -1
- package/out/synthesizers/utils.js +71 -167
- 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 +19 -11
- package/out/tools/audit.d.ts.map +1 -1
- package/out/tools/audit.js +89 -97
- 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 +34 -30
- 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 +83 -0
- package/out/tools/fetch/atom.js.map +1 -0
- package/out/tools/fetch/cordis.d.ts +124 -0
- package/out/tools/fetch/cordis.d.ts.map +1 -0
- package/out/tools/fetch/cordis.js +210 -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 +22 -14
- 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 +28 -28
- 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 +51 -32
- package/out/tools/fetch/geonames/dump.d.ts.map +1 -1
- package/out/tools/fetch/geonames/dump.js +54 -43
- 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 +138 -0
- package/out/tools/fetch/gleif.d.ts.map +1 -0
- package/out/tools/fetch/gleif.js +222 -0
- package/out/tools/fetch/gleif.js.map +1 -0
- package/out/tools/fetch/header.d.ts +43 -0
- package/out/tools/fetch/header.d.ts.map +1 -0
- package/out/tools/fetch/header.js +61 -0
- package/out/tools/fetch/header.js.map +1 -0
- package/out/tools/fetch/inspire-addresses.d.ts +254 -0
- package/out/tools/fetch/inspire-addresses.d.ts.map +1 -0
- package/out/tools/fetch/inspire-addresses.js +277 -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 +20 -99
- 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 +420 -0
- package/out/tools/fetch/wfs-harvest.d.ts.map +1 -0
- package/out/tools/fetch/wfs-harvest.js +525 -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 +99 -0
- package/out/tools/fetch.d.ts.map +1 -0
- package/out/tools/fetch.js +95 -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 +51 -37
- 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 +38 -30
- package/out/tools/golden/promote.js.map +1 -1
- package/out/tools/golden/relabel-street/directory.d.ts +25 -10
- package/out/tools/golden/relabel-street/directory.d.ts.map +1 -1
- package/out/tools/golden/relabel-street/directory.js +44 -32
- package/out/tools/golden/relabel-street/directory.js.map +1 -1
- package/out/tools/golden/relabel-street/row.d.ts +18 -69
- package/out/tools/golden/relabel-street/row.d.ts.map +1 -1
- package/out/tools/golden/relabel-street/row.js +19 -77
- 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 +42 -46
- 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 +82 -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} +93 -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 +39 -29
- package/out/tools/overlay/kryptonite.js.map +1 -0
- package/out/tools/overlay/manifest.d.ts +143 -0
- package/out/tools/overlay/manifest.d.ts.map +1 -0
- package/out/tools/overlay/manifest.js +261 -0
- package/out/tools/overlay/manifest.js.map +1 -0
- package/out/tools/overlay/merge-source.d.ts +74 -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 +55 -0
- package/out/tools/overlay/translit.d.ts.map +1 -0
- package/out/tools/{slice → overlay}/translit.js +61 -49
- package/out/tools/overlay/translit.js.map +1 -0
- package/out/tools/overture-subvenue.d.ts +17 -76
- package/out/tools/overture-subvenue.d.ts.map +1 -1
- package/out/tools/overture-subvenue.js +7 -74
- package/out/tools/overture-subvenue.js.map +1 -1
- package/out/tools/postcode-triples.d.ts +78 -127
- package/out/tools/postcode-triples.d.ts.map +1 -1
- package/out/tools/postcode-triples.js +200 -193
- package/out/tools/postcode-triples.js.map +1 -1
- package/out/tools/source-register/build.d.ts +147 -0
- package/out/tools/source-register/build.d.ts.map +1 -0
- package/out/tools/source-register/build.js +517 -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 +53 -84
- package/out/tools/sub/venue/lexicon.d.ts.map +1 -1
- package/out/tools/sub/venue/lexicon.js +42 -90
- 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/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 +19 -13
- package/out/tw/tools/fetch/gcis.js.map +1 -1
- package/out/types.d.ts +271 -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 +8 -3
- 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 +57 -111
- 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 +38 -28
- package/out/us/adapters/usgov/samhsa-treatment-locator/adapter.js.map +1 -1
- package/out/us/fips-state.d.ts +19 -10
- 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/{synthesizers → us/synthesizers}/intersection.d.ts +10 -8
- package/out/us/synthesizers/intersection.d.ts.map +1 -0
- package/out/{synthesizers → us/synthesizers}/intersection.js +46 -35
- package/out/us/synthesizers/intersection.js.map +1 -0
- package/out/{synthesizers → us/synthesizers}/street.d.ts +19 -12
- package/out/us/synthesizers/street.d.ts.map +1 -0
- package/out/{synthesizers → us/synthesizers}/street.js +45 -38
- package/out/us/synthesizers/street.js.map +1 -0
- 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 +17 -34
- 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 +30 -25
- 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 +24 -36
- package/out/us/tools/fetch/nppes.js.map +1 -1
- package/out/us/tools/fetch/state/hi-schools.d.ts +9 -5
- package/out/us/tools/fetch/state/hi-schools.d.ts.map +1 -1
- package/out/us/tools/fetch/state/hi-schools.js +16 -14
- 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 +16 -55
- package/out/us/tools/fetch/tiger-full.js.map +1 -1
- package/out/utils/align.d.ts +26 -19
- package/out/utils/align.d.ts.map +1 -1
- package/out/utils/align.js +32 -47
- 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 +30 -46
- package/out/utils/golden.js.map +1 -1
- package/out/utils/license.d.ts +159 -22
- 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 +162 -69
- package/out/utils/split.js.map +1 -1
- package/out/utils/tokenize.d.ts +15 -44
- 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 +77 -56
- 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 +45 -488
- 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/slice/translit.ts +0 -218
- 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/intersection.d.ts.map +0 -1
- package/out/synthesizers/intersection.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/synthesizers/street.d.ts.map +0 -1
- package/out/synthesizers/street.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
|
@@ -0,0 +1,69 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file The adapter phase: drive each configured adapter to its own `canonical.jsonl`, or reuse the run a
|
|
6
|
+
* previous build finished.
|
|
7
|
+
*/
|
|
8
|
+
import { readLocalJSONFile } from "@mailwoman/core/fs/readers";
|
|
9
|
+
import { pathExists, statPath } from "@mailwoman/core/fs/readers/stat";
|
|
10
|
+
import { $public } from "#env";
|
|
11
|
+
import { runAdapter } from "#runner";
|
|
12
|
+
/**
|
|
13
|
+
* Reuse a finished adapter run, or return `null` when this build has to drive the adapter.
|
|
14
|
+
*
|
|
15
|
+
* @throws When the cached manifest and the file on disk disagree on the byte count.
|
|
16
|
+
* The writer flushes the canonical file before writing the manifest.
|
|
17
|
+
* Manifest presence establishes that the writer reached the end of its run.
|
|
18
|
+
* The manifest describes the file as of that moment.
|
|
19
|
+
* A truncated copy, an interrupted move or a partial transfer leaves the
|
|
20
|
+
* manifest intact beside a shorter file.
|
|
21
|
+
* The recorded byte count distinguishes those cases.
|
|
22
|
+
* A resumed build that reuses a short file can produce a corpus whose row count later readers accept.
|
|
23
|
+
*/
|
|
24
|
+
async function reusableRun(adapterID, adapterDir) {
|
|
25
|
+
if ($public.MAILWOMAN_RESUME !== "1")
|
|
26
|
+
return null;
|
|
27
|
+
const cachedManifest = adapterDir("MANIFEST.json");
|
|
28
|
+
const canonicalPath = adapterDir("canonical.jsonl");
|
|
29
|
+
if (!(await pathExists(cachedManifest)) || !(await pathExists(canonicalPath)))
|
|
30
|
+
return null;
|
|
31
|
+
const cached = await readLocalJSONFile(cachedManifest);
|
|
32
|
+
const onDisk = (await statPath(canonicalPath)).size;
|
|
33
|
+
if (onDisk !== cached.bytes) {
|
|
34
|
+
throw new Error(`resume refused for ${adapterID}: ${canonicalPath} is ${onDisk.toLocaleString()} bytes and its ` +
|
|
35
|
+
`manifest records ${cached.bytes.toLocaleString()}. Delete ${cachedManifest} to ` +
|
|
36
|
+
`re-run the adapter, or restore the file this manifest describes.`);
|
|
37
|
+
}
|
|
38
|
+
return cached;
|
|
39
|
+
}
|
|
40
|
+
/**
|
|
41
|
+
* Drive every configured adapter, reusing whatever a previous build already finished.
|
|
42
|
+
*/
|
|
43
|
+
export async function runAdapterPhase(opts) {
|
|
44
|
+
const runs = [];
|
|
45
|
+
const skipped = [];
|
|
46
|
+
for (const adapter of opts.adapters) {
|
|
47
|
+
const adapterOptions = opts.adapterInputs[adapter.id];
|
|
48
|
+
if (!adapterOptions) {
|
|
49
|
+
skipped.push(adapter.id);
|
|
50
|
+
opts.onProgress?.("adapter-run", `skipped ${adapter.id} (no input configured)`);
|
|
51
|
+
continue;
|
|
52
|
+
}
|
|
53
|
+
const cached = await reusableRun(adapter.id, opts.intermediateDir(adapter.id));
|
|
54
|
+
if (cached) {
|
|
55
|
+
opts.onProgress?.("adapter-run", `resumed ${adapter.id} (reused ${cached.yielded} canonical rows)`);
|
|
56
|
+
runs.push(cached);
|
|
57
|
+
continue;
|
|
58
|
+
}
|
|
59
|
+
opts.onProgress?.("adapter-run", `running ${adapter.id}`);
|
|
60
|
+
runs.push(await runAdapter({
|
|
61
|
+
adapter,
|
|
62
|
+
adapterOptions,
|
|
63
|
+
outputDir: opts.intermediateDir,
|
|
64
|
+
corpusVersion: opts.corpusVersion,
|
|
65
|
+
}));
|
|
66
|
+
}
|
|
67
|
+
return { runs, skipped };
|
|
68
|
+
}
|
|
69
|
+
//# sourceMappingURL=adapters.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"adapters.js","sourceRoot":"","sources":["../../lib/build/adapters.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAA;AAC9D,OAAO,EAAE,UAAU,EAAE,QAAQ,EAAE,MAAM,iCAAiC,CAAA;AAItE,OAAO,EAAE,OAAO,EAAE,MAAM,MAAM,CAAA;AAC9B,OAAO,EAAE,UAAU,EAA2B,MAAM,SAAS,CAAA;AAuB7D;;;;;;;;;;;GAWG;AACH,KAAK,UAAU,WAAW,CAAC,SAAiB,EAAE,UAAuB;IACpE,IAAI,OAAO,CAAC,gBAAgB,KAAK,GAAG;QAAE,OAAO,IAAI,CAAA;IAEjD,MAAM,cAAc,GAAG,UAAU,CAAC,eAAe,CAAC,CAAA;IAClD,MAAM,aAAa,GAAG,UAAU,CAAC,iBAAiB,CAAC,CAAA;IAEnD,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,cAAc,CAAC,CAAC,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,aAAa,CAAC,CAAC;QAAE,OAAO,IAAI,CAAA;IAE1F,MAAM,MAAM,GAAG,MAAM,iBAAiB,CAAqB,cAAc,CAAC,CAAA;IAC1E,MAAM,MAAM,GAAG,CAAC,MAAM,QAAQ,CAAC,aAAa,CAAC,CAAC,CAAC,IAAI,CAAA;IAEnD,IAAI,MAAM,KAAK,MAAM,CAAC,KAAK,EAAE,CAAC;QAC7B,MAAM,IAAI,KAAK,CACd,sBAAsB,SAAS,KAAK,aAAa,OAAO,MAAM,CAAC,cAAc,EAAE,iBAAiB;YAC/F,oBAAoB,MAAM,CAAC,KAAK,CAAC,cAAc,EAAE,YAAY,cAAc,MAAM;YACjF,kEAAkE,CACnE,CAAA;IACF,CAAC;IAED,OAAO,MAAM,CAAA;AACd,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,KAAK,UAAU,eAAe,CAAC,IAAyB;IAC9D,MAAM,IAAI,GAAyB,EAAE,CAAA;IACrC,MAAM,OAAO,GAAa,EAAE,CAAA;IAE5B,KAAK,MAAM,OAAO,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAC;QACrC,MAAM,cAAc,GAAG,IAAI,CAAC,aAAa,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;QAErD,IAAI,CAAC,cAAc,EAAE,CAAC;YACrB,OAAO,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;YACxB,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,wBAAwB,CAAC,CAAA;YAE/E,SAAQ;QACT,CAAC;QAED,MAAM,MAAM,GAAG,MAAM,WAAW,CAAC,OAAO,CAAC,EAAE,EAAE,IAAI,CAAC,eAAe,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC,CAAA;QAE9E,IAAI,MAAM,EAAE,CAAC;YACZ,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,YAAY,MAAM,CAAC,OAAO,kBAAkB,CAAC,CAAA;YACnG,IAAI,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;YAEjB,SAAQ;QACT,CAAC;QAED,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,EAAE,CAAC,CAAA;QAEzD,IAAI,CAAC,IAAI,CACR,MAAM,UAAU,CAAC;YAChB,OAAO;YACP,cAAc;YACd,SAAS,EAAE,IAAI,CAAC,eAAe;YAC/B,aAAa,EAAE,IAAI,CAAC,aAAa;SACjC,CAAC,CACF,CAAA;IACF,CAAC;IAED,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,CAAA;AACzB,CAAC"}
|
|
@@ -0,0 +1,74 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file The align phase reads each adapter's canonical rows and applies license and register refusals.
|
|
6
|
+
* It expands synthesis and aligns each row. It then routes the result to a split or quarantine.
|
|
7
|
+
*
|
|
8
|
+
* The phase appends to four files. It writes a checkpoint at each adapter boundary, allowing an interrupted
|
|
9
|
+
* build to resume at the next adapter without aligning earlier adapters again.
|
|
10
|
+
*/
|
|
11
|
+
import type { PathBuilder } from "path-ts";
|
|
12
|
+
import type { BuildProfile, BuildStage } from "#build/types";
|
|
13
|
+
import { type AdapterRunManifest } from "#runner";
|
|
14
|
+
import { type LicensePolicy, type LicenseRefusalKind, type LicenseVerdictCache } from "#utils/license";
|
|
15
|
+
import { type CountryHoldout, type SplitName } from "#utils/split";
|
|
16
|
+
export interface AlignPhaseOptions {
|
|
17
|
+
/**
|
|
18
|
+
* The adapter runs to align, in the order their rows enter the output files.
|
|
19
|
+
*/
|
|
20
|
+
adapterRuns: readonly AdapterRunManifest[];
|
|
21
|
+
corpusVersion: string;
|
|
22
|
+
synthesize: boolean;
|
|
23
|
+
profile: BuildProfile;
|
|
24
|
+
licensePolicy: LicensePolicy;
|
|
25
|
+
excludeLicenses: readonly RegExp[];
|
|
26
|
+
holdouts: Record<string, CountryHoldout>;
|
|
27
|
+
/**
|
|
28
|
+
* Each register source's ingest-eligibility reasons, or `null` under the exploratory profile.
|
|
29
|
+
*/
|
|
30
|
+
eligibility: ReadonlyMap<string, readonly string[]> | null;
|
|
31
|
+
labeledPaths: Record<SplitName, PathBuilder>;
|
|
32
|
+
quarantinePath: PathBuilder;
|
|
33
|
+
checkpointPath: PathBuilder;
|
|
34
|
+
onProgress?: (stage: BuildStage, message: string) => void;
|
|
35
|
+
}
|
|
36
|
+
/**
|
|
37
|
+
* The counts produced by the align phase.
|
|
38
|
+
*
|
|
39
|
+
* The build uses them in its manifest and frozen source record.
|
|
40
|
+
*/
|
|
41
|
+
export interface AlignPhaseResult {
|
|
42
|
+
aligned: number;
|
|
43
|
+
quarantined: number;
|
|
44
|
+
counts: Record<SplitName, number>;
|
|
45
|
+
licenseCounts: Map<string, number>;
|
|
46
|
+
refusedByKind: Map<LicenseRefusalKind, number>;
|
|
47
|
+
/**
|
|
48
|
+
* Rows the license policy refused, counted per `source`.
|
|
49
|
+
*
|
|
50
|
+
* A source here contributed zero rows to the corpus whatever its adapter reported writing.
|
|
51
|
+
* That is the distinction between a source present in corpus provenance
|
|
52
|
+
* and one present in model provenance.
|
|
53
|
+
*/
|
|
54
|
+
refusedBySource: Map<string, number>;
|
|
55
|
+
excludedByLicense: number;
|
|
56
|
+
admittedUnresolved: number;
|
|
57
|
+
excludedByEligibility: number;
|
|
58
|
+
rowsBySource: Map<string, {
|
|
59
|
+
rows: number;
|
|
60
|
+
license: string;
|
|
61
|
+
}>;
|
|
62
|
+
ineligibleSources: Map<string, readonly string[]>;
|
|
63
|
+
/**
|
|
64
|
+
* The license cache used by this phase.
|
|
65
|
+
*
|
|
66
|
+
* Its `refusedValues()` method returns every value it refused.
|
|
67
|
+
*/
|
|
68
|
+
licenses: LicenseVerdictCache;
|
|
69
|
+
}
|
|
70
|
+
/**
|
|
71
|
+
* Align every adapter run's rows into the four output files, writing a checkpoint at each adapter boundary.
|
|
72
|
+
*/
|
|
73
|
+
export declare function runAlignPhase(opts: AlignPhaseOptions): Promise<AlignPhaseResult>;
|
|
74
|
+
//# sourceMappingURL=align.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"align.d.ts","sourceRoot":"","sources":["../../lib/build/align.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAMH,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,SAAS,CAAA;AAY1C,OAAO,KAAK,EAAE,YAAY,EAAE,UAAU,EAAE,MAAM,cAAc,CAAA;AAE5D,OAAO,EAAqB,KAAK,kBAAkB,EAAE,MAAM,SAAS,CAAA;AAIpE,OAAO,EAEN,KAAK,aAAa,EAClB,KAAK,kBAAkB,EACvB,KAAK,mBAAmB,EACxB,MAAM,gBAAgB,CAAA;AACvB,OAAO,EAAe,KAAK,cAAc,EAAE,KAAK,SAAS,EAAE,MAAM,cAAc,CAAA;AAE/E,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,WAAW,EAAE,SAAS,kBAAkB,EAAE,CAAA;IAE1C,aAAa,EAAE,MAAM,CAAA;IACrB,UAAU,EAAE,OAAO,CAAA;IACnB,OAAO,EAAE,YAAY,CAAA;IACrB,aAAa,EAAE,aAAa,CAAA;IAC5B,eAAe,EAAE,SAAS,MAAM,EAAE,CAAA;IAClC,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,cAAc,CAAC,CAAA;IAExC;;OAEG;IACH,WAAW,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAC,GAAG,IAAI,CAAA;IAE1D,YAAY,EAAE,MAAM,CAAC,SAAS,EAAE,WAAW,CAAC,CAAA;IAC5C,cAAc,EAAE,WAAW,CAAA;IAC3B,cAAc,EAAE,WAAW,CAAA;IAE3B,UAAU,CAAC,EAAE,CAAC,KAAK,EAAE,UAAU,EAAE,OAAO,EAAE,MAAM,KAAK,IAAI,CAAA;CACzD;AAED;;;;GAIG;AACH,MAAM,WAAW,gBAAgB;IAChC,OAAO,EAAE,MAAM,CAAA;IACf,WAAW,EAAE,MAAM,CAAA;IACnB,MAAM,EAAE,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;IACjC,aAAa,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IAClC,aAAa,EAAE,GAAG,CAAC,kBAAkB,EAAE,MAAM,CAAC,CAAA;IAC9C;;;;;;OAMG;IACH,eAAe,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IACpC,iBAAiB,EAAE,MAAM,CAAA;IACzB,kBAAkB,EAAE,MAAM,CAAA;IAC1B,qBAAqB,EAAE,MAAM,CAAA;IAC7B,YAAY,EAAE,GAAG,CAAC,MAAM,EAAE;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAA;KAAE,CAAC,CAAA;IAC5D,iBAAiB,EAAE,GAAG,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAC,CAAA;IAEjD;;;;OAIG;IACH,QAAQ,EAAE,mBAAmB,CAAA;CAC7B;AA2DD;;GAEG;AACH,wBAAsB,aAAa,CAAC,IAAI,EAAE,iBAAiB,GAAG,OAAO,CAAC,gBAAgB,CAAC,CA6MtF"}
|
|
@@ -0,0 +1,226 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file The align phase reads each adapter's canonical rows and applies license and register refusals.
|
|
6
|
+
* It expands synthesis and aligns each row. It then routes the result to a split or quarantine.
|
|
7
|
+
*
|
|
8
|
+
* The phase appends to four files. It writes a checkpoint at each adapter boundary, allowing an interrupted
|
|
9
|
+
* build to resume at the next adapter without aligning earlier adapters again.
|
|
10
|
+
*/
|
|
11
|
+
import { openWriteStream } from "@mailwoman/core/fs/streams";
|
|
12
|
+
import { truncateFile } from "@mailwoman/core/fs/writers";
|
|
13
|
+
import { sha256Hex } from "@mailwoman/core/hash";
|
|
14
|
+
import { stringifyJSON } from "@mailwoman/core/json";
|
|
15
|
+
import { ALIGN_OUTPUTS, readAlignCheckpoint, writeAlignCheckpoint, } from "#build/checkpoint";
|
|
16
|
+
import { createIneligibilityReader } from "#build/eligibility";
|
|
17
|
+
import { streamJSONL } from "#build/rows";
|
|
18
|
+
import { $public } from "#env";
|
|
19
|
+
import { flushStream, once } from "#runner";
|
|
20
|
+
import { defaultAugmentationsForCountry, synthesizeRow } from "#synthesizers/utils";
|
|
21
|
+
import { alignRow } from "#utils/align";
|
|
22
|
+
import { createLicenseVerdictCache, } from "#utils/license";
|
|
23
|
+
import { splitForRow } from "#utils/split";
|
|
24
|
+
/**
|
|
25
|
+
* Return each output to the length the checkpoint recorded.
|
|
26
|
+
*
|
|
27
|
+
* Whatever the interrupted adapter wrote past the boundary is discarded, because the checkpoint's
|
|
28
|
+
* accumulators do not count those rows and the split manifests are computed from the accumulators.
|
|
29
|
+
*/
|
|
30
|
+
async function restoreOutputs(checkpoint, paths) {
|
|
31
|
+
for (const name of ALIGN_OUTPUTS) {
|
|
32
|
+
await truncateFile(paths[name], checkpoint.offsets[name]);
|
|
33
|
+
}
|
|
34
|
+
}
|
|
35
|
+
/**
|
|
36
|
+
* Restore the accumulators a checkpoint recorded into this run's containers.
|
|
37
|
+
*/
|
|
38
|
+
function restoreTally(checkpoint, result) {
|
|
39
|
+
for (const name of ["train", "val", "test"]) {
|
|
40
|
+
result.counts[name] = checkpoint.counts[name];
|
|
41
|
+
}
|
|
42
|
+
for (const [license, count] of Object.entries(checkpoint.license_counts)) {
|
|
43
|
+
result.licenseCounts.set(license, count);
|
|
44
|
+
}
|
|
45
|
+
for (const [kind, count] of Object.entries(checkpoint.refused_by_kind)) {
|
|
46
|
+
result.refusedByKind.set(kind, count);
|
|
47
|
+
}
|
|
48
|
+
for (const [source, count] of Object.entries(checkpoint.refused_by_source)) {
|
|
49
|
+
result.refusedBySource.set(source, count);
|
|
50
|
+
}
|
|
51
|
+
for (const [source, contributed] of Object.entries(checkpoint.rows_by_source)) {
|
|
52
|
+
result.rowsBySource.set(source, contributed);
|
|
53
|
+
}
|
|
54
|
+
return {
|
|
55
|
+
aligned: checkpoint.aligned,
|
|
56
|
+
quarantined: checkpoint.quarantined,
|
|
57
|
+
excludedByLicense: checkpoint.excluded_by_license,
|
|
58
|
+
admittedUnresolved: checkpoint.admitted_unresolved_license_rows,
|
|
59
|
+
excludedByEligibility: checkpoint.excluded_by_eligibility,
|
|
60
|
+
};
|
|
61
|
+
}
|
|
62
|
+
/**
|
|
63
|
+
* Align every adapter run's rows into the four output files, writing a checkpoint at each adapter boundary.
|
|
64
|
+
*/
|
|
65
|
+
export async function runAlignPhase(opts) {
|
|
66
|
+
const ineligibility = createIneligibilityReader(opts.eligibility);
|
|
67
|
+
const licenses = createLicenseVerdictCache(opts.licensePolicy, opts.excludeLicenses);
|
|
68
|
+
const result = {
|
|
69
|
+
aligned: 0,
|
|
70
|
+
quarantined: 0,
|
|
71
|
+
counts: { train: 0, val: 0, test: 0 },
|
|
72
|
+
licenseCounts: new Map(),
|
|
73
|
+
refusedByKind: new Map(),
|
|
74
|
+
excludedByLicense: 0,
|
|
75
|
+
admittedUnresolved: 0,
|
|
76
|
+
excludedByEligibility: 0,
|
|
77
|
+
refusedBySource: new Map(),
|
|
78
|
+
rowsBySource: new Map(),
|
|
79
|
+
ineligibleSources: ineligibility.refused,
|
|
80
|
+
licenses,
|
|
81
|
+
};
|
|
82
|
+
const settings = {
|
|
83
|
+
corpus_version: opts.corpusVersion,
|
|
84
|
+
synthesize: opts.synthesize,
|
|
85
|
+
license_policy: opts.licensePolicy,
|
|
86
|
+
exclude_licenses: opts.excludeLicenses.map((pattern) => pattern.source).toSorted(),
|
|
87
|
+
profile: opts.profile,
|
|
88
|
+
holdout_digest: sha256Hex(stringifyJSON(opts.holdouts)),
|
|
89
|
+
};
|
|
90
|
+
const paths = { ...opts.labeledPaths, quarantine: opts.quarantinePath };
|
|
91
|
+
const checkpoint = $public.MAILWOMAN_RESUME === "1"
|
|
92
|
+
? await readAlignCheckpoint(opts.checkpointPath, {
|
|
93
|
+
settings,
|
|
94
|
+
adapterOrder: opts.adapterRuns.map((run) => run.adapter_id),
|
|
95
|
+
})
|
|
96
|
+
: null;
|
|
97
|
+
let tally = {
|
|
98
|
+
aligned: 0,
|
|
99
|
+
quarantined: 0,
|
|
100
|
+
excludedByLicense: 0,
|
|
101
|
+
admittedUnresolved: 0,
|
|
102
|
+
excludedByEligibility: 0,
|
|
103
|
+
};
|
|
104
|
+
if (checkpoint) {
|
|
105
|
+
await restoreOutputs(checkpoint, paths);
|
|
106
|
+
tally = restoreTally(checkpoint, result);
|
|
107
|
+
opts.onProgress?.("align", `resumed align after ${checkpoint.completed_adapters.join(", ")} ` +
|
|
108
|
+
`(${tally.aligned.toLocaleString()} aligned rows)`);
|
|
109
|
+
}
|
|
110
|
+
// `a` rather than the default `w` on a resume, because the files hold the rows the checkpoint counted.
|
|
111
|
+
const flags = checkpoint ? "a" : "w";
|
|
112
|
+
const streams = {
|
|
113
|
+
train: openWriteStream(paths.train, { encoding: "utf8", flags }),
|
|
114
|
+
val: openWriteStream(paths.val, { encoding: "utf8", flags }),
|
|
115
|
+
test: openWriteStream(paths.test, { encoding: "utf8", flags }),
|
|
116
|
+
quarantine: openWriteStream(paths.quarantine, { encoding: "utf8", flags }),
|
|
117
|
+
};
|
|
118
|
+
const offsets = checkpoint
|
|
119
|
+
? { ...checkpoint.offsets }
|
|
120
|
+
: { train: 0, val: 0, test: 0, quarantine: 0 };
|
|
121
|
+
const writeLine = (name, line) => {
|
|
122
|
+
offsets[name] += Buffer.byteLength(line, "utf8");
|
|
123
|
+
streams[name].write(line);
|
|
124
|
+
};
|
|
125
|
+
const quarantine = (row, reason) => {
|
|
126
|
+
writeLine("quarantine", `${stringifyJSON({ row, reason })}\n`);
|
|
127
|
+
};
|
|
128
|
+
const completed = checkpoint ? [...checkpoint.completed_adapters] : [];
|
|
129
|
+
for (const adapterRun of opts.adapterRuns) {
|
|
130
|
+
if (completed.includes(adapterRun.adapter_id))
|
|
131
|
+
continue;
|
|
132
|
+
opts.onProgress?.("align", `aligning ${adapterRun.adapter_id}`);
|
|
133
|
+
for await (const row of streamJSONL(adapterRun.jsonl_path)) {
|
|
134
|
+
result.licenseCounts.set(row.license, (result.licenseCounts.get(row.license) ?? 0) + 1);
|
|
135
|
+
// The license count is incremented before the drop, so the manifest's license set
|
|
136
|
+
// reflects what the corpus contained and `excluded_by_license` what was removed.
|
|
137
|
+
const verdict = licenses.read(row.license);
|
|
138
|
+
if (verdict.refusal) {
|
|
139
|
+
tally.excludedByLicense++;
|
|
140
|
+
result.refusedByKind.set(verdict.refusal, (result.refusedByKind.get(verdict.refusal) ?? 0) + 1);
|
|
141
|
+
// Counted per source as well as per refusal kind, because an adapter's own
|
|
142
|
+
// `written` count records what it emitted rather than what entered the corpus.
|
|
143
|
+
// A source refused here contributed no row, and model provenance generated from model provenance
|
|
144
|
+
// generated from `adapters[].written` only would credit it with rows it did not supply.
|
|
145
|
+
result.refusedBySource.set(row.source, (result.refusedBySource.get(row.source) ?? 0) + 1);
|
|
146
|
+
continue;
|
|
147
|
+
}
|
|
148
|
+
// An admitted row whose license resolves to no expression entered with
|
|
149
|
+
// unknown obligations rather than with none.
|
|
150
|
+
// The manifest states the count so a consumer can tell the two apart.
|
|
151
|
+
if (!verdict.resolved) {
|
|
152
|
+
tally.admittedUnresolved++;
|
|
153
|
+
}
|
|
154
|
+
// Eligibility runs before augmentation: a synthetic row inherits its ancestor's `source`,
|
|
155
|
+
// so refusing the ancestor here refuses everything fanned from it.
|
|
156
|
+
const ineligible = ineligibility.read(row);
|
|
157
|
+
if (ineligible) {
|
|
158
|
+
tally.excludedByEligibility++;
|
|
159
|
+
quarantine(row, `source-ineligible:${ineligible.join("; ")}`);
|
|
160
|
+
continue;
|
|
161
|
+
}
|
|
162
|
+
// Counted after both refusals, so the frozen manifest records what a source
|
|
163
|
+
// contributed rather than what it offered.
|
|
164
|
+
const contributed = result.rowsBySource.get(row.source);
|
|
165
|
+
result.rowsBySource.set(row.source, { rows: (contributed?.rows ?? 0) + 1, license: row.license });
|
|
166
|
+
const fanned = [row];
|
|
167
|
+
if (opts.synthesize) {
|
|
168
|
+
for (const aug of synthesizeRow(row, defaultAugmentationsForCountry(row.country))) {
|
|
169
|
+
fanned.push(aug);
|
|
170
|
+
}
|
|
171
|
+
}
|
|
172
|
+
for (const r of fanned) {
|
|
173
|
+
let aligned;
|
|
174
|
+
try {
|
|
175
|
+
aligned = alignRow(r);
|
|
176
|
+
}
|
|
177
|
+
catch (error) {
|
|
178
|
+
// No single row may crash a multi-hour build.
|
|
179
|
+
// An unknown throw is quarantined as `align-threw`, and a rise in
|
|
180
|
+
// that reason's count is a finding.
|
|
181
|
+
quarantine(r, `align-threw:${error.message.slice(0, 160)}`);
|
|
182
|
+
tally.quarantined++;
|
|
183
|
+
continue;
|
|
184
|
+
}
|
|
185
|
+
if (aligned.kind === "labeled") {
|
|
186
|
+
const split = splitForRow(aligned.row, opts.holdouts);
|
|
187
|
+
writeLine(split, `${stringifyJSON(aligned.row)}\n`);
|
|
188
|
+
result.counts[split]++;
|
|
189
|
+
tally.aligned++;
|
|
190
|
+
}
|
|
191
|
+
else {
|
|
192
|
+
quarantine(r, aligned.row.reason);
|
|
193
|
+
tally.quarantined++;
|
|
194
|
+
}
|
|
195
|
+
}
|
|
196
|
+
}
|
|
197
|
+
// The checkpoint is written at the adapter boundary, after the bytes it counts have reached disk.
|
|
198
|
+
// Mid-adapter recovery is unavailable.
|
|
199
|
+
// The largest adapter takes about twenty minutes to align.
|
|
200
|
+
// A row offset inside one adapter's file would still require replaying that adapter's synthesis.
|
|
201
|
+
await Promise.all(ALIGN_OUTPUTS.map((name) => flushStream(streams[name])));
|
|
202
|
+
completed.push(adapterRun.adapter_id);
|
|
203
|
+
await writeAlignCheckpoint(opts.checkpointPath, {
|
|
204
|
+
settings,
|
|
205
|
+
completed_adapters: [...completed],
|
|
206
|
+
offsets: { ...offsets },
|
|
207
|
+
counts: { ...result.counts },
|
|
208
|
+
aligned: tally.aligned,
|
|
209
|
+
quarantined: tally.quarantined,
|
|
210
|
+
license_counts: Object.fromEntries(result.licenseCounts),
|
|
211
|
+
refused_by_kind: Object.fromEntries(result.refusedByKind),
|
|
212
|
+
refused_by_source: Object.fromEntries(result.refusedBySource),
|
|
213
|
+
excluded_by_license: tally.excludedByLicense,
|
|
214
|
+
admitted_unresolved_license_rows: tally.admittedUnresolved,
|
|
215
|
+
excluded_by_eligibility: tally.excludedByEligibility,
|
|
216
|
+
rows_by_source: Object.fromEntries(result.rowsBySource),
|
|
217
|
+
written_at: new Date().toISOString(),
|
|
218
|
+
});
|
|
219
|
+
}
|
|
220
|
+
for (const stream of Object.values(streams)) {
|
|
221
|
+
stream.end();
|
|
222
|
+
}
|
|
223
|
+
await Promise.all(Object.values(streams).map((stream) => once(stream, "close")));
|
|
224
|
+
return { ...result, ...tally };
|
|
225
|
+
}
|
|
226
|
+
//# sourceMappingURL=align.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"align.js","sourceRoot":"","sources":["../../lib/build/align.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,EAAE,eAAe,EAAoB,MAAM,4BAA4B,CAAA;AAC9E,OAAO,EAAE,YAAY,EAAE,MAAM,4BAA4B,CAAA;AACzD,OAAO,EAAE,SAAS,EAAE,MAAM,sBAAsB,CAAA;AAChD,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AAGpD,OAAO,EACN,aAAa,EACb,mBAAmB,EACnB,oBAAoB,GAIpB,MAAM,mBAAmB,CAAA;AAC1B,OAAO,EAAE,yBAAyB,EAAE,MAAM,oBAAoB,CAAA;AAC9D,OAAO,EAAE,WAAW,EAAE,MAAM,aAAa,CAAA;AAEzC,OAAO,EAAE,OAAO,EAAE,MAAM,MAAM,CAAA;AAC9B,OAAO,EAAE,WAAW,EAAE,IAAI,EAA2B,MAAM,SAAS,CAAA;AACpE,OAAO,EAAE,8BAA8B,EAAE,aAAa,EAAE,MAAM,qBAAqB,CAAA;AAEnF,OAAO,EAAE,QAAQ,EAAE,MAAM,cAAc,CAAA;AACvC,OAAO,EACN,yBAAyB,GAIzB,MAAM,gBAAgB,CAAA;AACvB,OAAO,EAAE,WAAW,EAAuC,MAAM,cAAc,CAAA;AAwE/E;;;;;GAKG;AACH,KAAK,UAAU,cAAc,CAAC,UAA2B,EAAE,KAA2C;IACrG,KAAK,MAAM,IAAI,IAAI,aAAa,EAAE,CAAC;QAClC,MAAM,YAAY,CAAC,KAAK,CAAC,IAAI,CAAC,EAAE,UAAU,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,CAAA;IAC1D,CAAC;AACF,CAAC;AAED;;GAEG;AACH,SAAS,YAAY,CAAC,UAA2B,EAAE,MAAwB;IAC1E,KAAK,MAAM,IAAI,IAAI,CAAC,OAAO,EAAE,KAAK,EAAE,MAAM,CAAU,EAAE,CAAC;QACtD,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;IAC9C,CAAC;IAED,KAAK,MAAM,CAAC,OAAO,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,UAAU,CAAC,cAAc,CAAC,EAAE,CAAC;QAC1E,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,OAAO,EAAE,KAAK,CAAC,CAAA;IACzC,CAAC;IAED,KAAK,MAAM,CAAC,IAAI,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,UAAU,CAAC,eAAe,CAAC,EAAE,CAAC;QACxE,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,IAA0B,EAAE,KAAK,CAAC,CAAA;IAC5D,CAAC;IAED,KAAK,MAAM,CAAC,MAAM,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,UAAU,CAAC,iBAAiB,CAAC,EAAE,CAAC;QAC5E,MAAM,CAAC,eAAe,CAAC,GAAG,CAAC,MAAM,EAAE,KAAK,CAAC,CAAA;IAC1C,CAAC;IAED,KAAK,MAAM,CAAC,MAAM,EAAE,WAAW,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,UAAU,CAAC,cAAc,CAAC,EAAE,CAAC;QAC/E,MAAM,CAAC,YAAY,CAAC,GAAG,CAAC,MAAM,EAAE,WAAW,CAAC,CAAA;IAC7C,CAAC;IAED,OAAO;QACN,OAAO,EAAE,UAAU,CAAC,OAAO;QAC3B,WAAW,EAAE,UAAU,CAAC,WAAW;QACnC,iBAAiB,EAAE,UAAU,CAAC,mBAAmB;QACjD,kBAAkB,EAAE,UAAU,CAAC,gCAAgC;QAC/D,qBAAqB,EAAE,UAAU,CAAC,uBAAuB;KACzD,CAAA;AACF,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,KAAK,UAAU,aAAa,CAAC,IAAuB;IAC1D,MAAM,aAAa,GAAG,yBAAyB,CAAC,IAAI,CAAC,WAAW,CAAC,CAAA;IACjE,MAAM,QAAQ,GAAG,yBAAyB,CAAC,IAAI,CAAC,aAAa,EAAE,IAAI,CAAC,eAAe,CAAC,CAAA;IAEpF,MAAM,MAAM,GAAqB;QAChC,OAAO,EAAE,CAAC;QACV,WAAW,EAAE,CAAC;QACd,MAAM,EAAE,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,IAAI,EAAE,CAAC,EAAE;QACrC,aAAa,EAAE,IAAI,GAAG,EAAE;QACxB,aAAa,EAAE,IAAI,GAAG,EAAE;QACxB,iBAAiB,EAAE,CAAC;QACpB,kBAAkB,EAAE,CAAC;QACrB,qBAAqB,EAAE,CAAC;QACxB,eAAe,EAAE,IAAI,GAAG,EAAE;QAC1B,YAAY,EAAE,IAAI,GAAG,EAAE;QACvB,iBAAiB,EAAE,aAAa,CAAC,OAAO;QACxC,QAAQ;KACR,CAAA;IAED,MAAM,QAAQ,GAAkB;QAC/B,cAAc,EAAE,IAAI,CAAC,aAAa;QAClC,UAAU,EAAE,IAAI,CAAC,UAAU;QAC3B,cAAc,EAAE,IAAI,CAAC,aAAa;QAClC,gBAAgB,EAAE,IAAI,CAAC,eAAe,CAAC,GAAG,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,QAAQ,EAAE;QAClF,OAAO,EAAE,IAAI,CAAC,OAAO;QACrB,cAAc,EAAE,SAAS,CAAC,aAAa,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;KACvD,CAAA;IAED,MAAM,KAAK,GAAyC,EAAE,GAAG,IAAI,CAAC,YAAY,EAAE,UAAU,EAAE,IAAI,CAAC,cAAc,EAAE,CAAA;IAE7G,MAAM,UAAU,GACf,OAAO,CAAC,gBAAgB,KAAK,GAAG;QAC/B,CAAC,CAAC,MAAM,mBAAmB,CAAC,IAAI,CAAC,cAAc,EAAE;YAC/C,QAAQ;YACR,YAAY,EAAE,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC;SAC3D,CAAC;QACH,CAAC,CAAC,IAAI,CAAA;IAER,IAAI,KAAK,GAAe;QACvB,OAAO,EAAE,CAAC;QACV,WAAW,EAAE,CAAC;QACd,iBAAiB,EAAE,CAAC;QACpB,kBAAkB,EAAE,CAAC;QACrB,qBAAqB,EAAE,CAAC;KACxB,CAAA;IAED,IAAI,UAAU,EAAE,CAAC;QAChB,MAAM,cAAc,CAAC,UAAU,EAAE,KAAK,CAAC,CAAA;QACvC,KAAK,GAAG,YAAY,CAAC,UAAU,EAAE,MAAM,CAAC,CAAA;QAExC,IAAI,CAAC,UAAU,EAAE,CAChB,OAAO,EACP,uBAAuB,UAAU,CAAC,kBAAkB,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG;YACjE,IAAI,KAAK,CAAC,OAAO,CAAC,cAAc,EAAE,gBAAgB,CACnD,CAAA;IACF,CAAC;IAED,uGAAuG;IACvG,MAAM,KAAK,GAAG,UAAU,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAA;IAEpC,MAAM,OAAO,GAAyC;QACrD,KAAK,EAAE,eAAe,CAAC,KAAK,CAAC,KAAK,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC;QAChE,GAAG,EAAE,eAAe,CAAC,KAAK,CAAC,GAAG,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC;QAC5D,IAAI,EAAE,eAAe,CAAC,KAAK,CAAC,IAAI,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC;QAC9D,UAAU,EAAE,eAAe,CAAC,KAAK,CAAC,UAAU,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC;KAC1E,CAAA;IAED,MAAM,OAAO,GAAoC,UAAU;QAC1D,CAAC,CAAC,EAAE,GAAG,UAAU,CAAC,OAAO,EAAE;QAC3B,CAAC,CAAC,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,IAAI,EAAE,CAAC,EAAE,UAAU,EAAE,CAAC,EAAE,CAAA;IAE/C,MAAM,SAAS,GAAG,CAAC,IAAqB,EAAE,IAAY,EAAQ,EAAE;QAC/D,OAAO,CAAC,IAAI,CAAC,IAAI,MAAM,CAAC,UAAU,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;QAChD,OAAO,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC1B,CAAC,CAAA;IAED,MAAM,UAAU,GAAG,CAAC,GAAiB,EAAE,MAAc,EAAQ,EAAE;QAC9D,SAAS,CAAC,YAAY,EAAE,GAAG,aAAa,CAAC,EAAE,GAAG,EAAE,MAAM,EAAE,CAAC,IAAI,CAAC,CAAA;IAC/D,CAAC,CAAA;IAED,MAAM,SAAS,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC,GAAG,UAAU,CAAC,kBAAkB,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;IAEtE,KAAK,MAAM,UAAU,IAAI,IAAI,CAAC,WAAW,EAAE,CAAC;QAC3C,IAAI,SAAS,CAAC,QAAQ,CAAC,UAAU,CAAC,UAAU,CAAC;YAAE,SAAQ;QAEvD,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,YAAY,UAAU,CAAC,UAAU,EAAE,CAAC,CAAA;QAE/D,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,WAAW,CAAe,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;YAC1E,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,GAAG,CAAC,OAAO,EAAE,CAAC,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;YAEvF,kFAAkF;YAClF,iFAAiF;YACjF,MAAM,OAAO,GAAG,QAAQ,CAAC,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,CAAA;YAE1C,IAAI,OAAO,CAAC,OAAO,EAAE,CAAC;gBACrB,KAAK,CAAC,iBAAiB,EAAE,CAAA;gBACzB,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,OAAO,CAAC,OAAO,EAAE,CAAC,MAAM,CAAC,aAAa,CAAC,GAAG,CAAC,OAAO,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;gBAE/F,2EAA2E;gBAC3E,+EAA+E;gBAC/E,iGAAiG;gBACjG,wFAAwF;gBACxF,MAAM,CAAC,eAAe,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,EAAE,CAAC,MAAM,CAAC,eAAe,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;gBAEzF,SAAQ;YACT,CAAC;YAED,uEAAuE;YACvE,6CAA6C;YAC7C,sEAAsE;YACtE,IAAI,CAAC,OAAO,CAAC,QAAQ,EAAE,CAAC;gBACvB,KAAK,CAAC,kBAAkB,EAAE,CAAA;YAC3B,CAAC;YAED,0FAA0F;YAC1F,mEAAmE;YACnE,MAAM,UAAU,GAAG,aAAa,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;YAE1C,IAAI,UAAU,EAAE,CAAC;gBAChB,KAAK,CAAC,qBAAqB,EAAE,CAAA;gBAE7B,UAAU,CAAC,GAAG,EAAE,qBAAqB,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;gBAE7D,SAAQ;YACT,CAAC;YAED,4EAA4E;YAC5E,2CAA2C;YAC3C,MAAM,WAAW,GAAG,MAAM,CAAC,YAAY,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;YAEvD,MAAM,CAAC,YAAY,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,EAAE,EAAE,IAAI,EAAE,CAAC,WAAW,EAAE,IAAI,IAAI,CAAC,CAAC,GAAG,CAAC,EAAE,OAAO,EAAE,GAAG,CAAC,OAAO,EAAE,CAAC,CAAA;YAEjG,MAAM,MAAM,GAAmB,CAAC,GAAG,CAAC,CAAA;YAEpC,IAAI,IAAI,CAAC,UAAU,EAAE,CAAC;gBACrB,KAAK,MAAM,GAAG,IAAI,aAAa,CAAC,GAAG,EAAE,8BAA8B,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC;oBACnF,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;gBACjB,CAAC;YACF,CAAC;YAED,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;gBACxB,IAAI,OAAoC,CAAA;gBAExC,IAAI,CAAC;oBACJ,OAAO,GAAG,QAAQ,CAAC,CAAC,CAAC,CAAA;gBACtB,CAAC;gBAAC,OAAO,KAAK,EAAE,CAAC;oBAChB,8CAA8C;oBAC9C,kEAAkE;oBAClE,oCAAoC;oBACpC,UAAU,CAAC,CAAC,EAAE,eAAgB,KAAe,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,CAAA;oBAEtE,KAAK,CAAC,WAAW,EAAE,CAAA;oBAEnB,SAAQ;gBACT,CAAC;gBAED,IAAI,OAAO,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;oBAChC,MAAM,KAAK,GAAG,WAAW,CAAC,OAAO,CAAC,GAAG,EAAE,IAAI,CAAC,QAAQ,CAAC,CAAA;oBAErD,SAAS,CAAC,KAAK,EAAE,GAAG,aAAa,CAAC,OAAO,CAAC,GAAG,CAAC,IAAI,CAAC,CAAA;oBAEnD,MAAM,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAA;oBAEtB,KAAK,CAAC,OAAO,EAAE,CAAA;gBAChB,CAAC;qBAAM,CAAC;oBACP,UAAU,CAAC,CAAC,EAAE,OAAO,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;oBAEjC,KAAK,CAAC,WAAW,EAAE,CAAA;gBACpB,CAAC;YACF,CAAC;QACF,CAAC;QAED,kGAAkG;QAClG,uCAAuC;QACvC,2DAA2D;QAC3D,iGAAiG;QACjG,MAAM,OAAO,CAAC,GAAG,CAAC,aAAa,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,WAAW,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAA;QAE1E,SAAS,CAAC,IAAI,CAAC,UAAU,CAAC,UAAU,CAAC,CAAA;QAErC,MAAM,oBAAoB,CAAC,IAAI,CAAC,cAAc,EAAE;YAC/C,QAAQ;YACR,kBAAkB,EAAE,CAAC,GAAG,SAAS,CAAC;YAClC,OAAO,EAAE,EAAE,GAAG,OAAO,EAAE;YACvB,MAAM,EAAE,EAAE,GAAG,MAAM,CAAC,MAAM,EAAE;YAC5B,OAAO,EAAE,KAAK,CAAC,OAAO;YACtB,WAAW,EAAE,KAAK,CAAC,WAAW;YAC9B,cAAc,EAAE,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,aAAa,CAAC;YACxD,eAAe,EAAE,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,aAAa,CAAC;YACzD,iBAAiB,EAAE,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,eAAe,CAAC;YAC7D,mBAAmB,EAAE,KAAK,CAAC,iBAAiB;YAC5C,gCAAgC,EAAE,KAAK,CAAC,kBAAkB;YAC1D,uBAAuB,EAAE,KAAK,CAAC,qBAAqB;YACpD,cAAc,EAAE,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,YAAY,CAAC;YACvD,UAAU,EAAE,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE;SACpC,CAAC,CAAA;IACH,CAAC;IAED,KAAK,MAAM,MAAM,IAAI,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,EAAE,CAAC;QAC7C,MAAM,CAAC,GAAG,EAAE,CAAA;IACb,CAAC;IAED,MAAM,OAAO,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,GAAG,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,IAAI,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC,CAAC,CAAA;IAEhF,OAAO,EAAE,GAAG,MAAM,EAAE,GAAG,KAAK,EAAE,CAAA;AAC/B,CAAC"}
|
|
@@ -0,0 +1,94 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file The align phase's resume point. It records aligned adapters and each output's length at that moment.
|
|
6
|
+
* It also records every accumulator used to compute the manifest.
|
|
7
|
+
*
|
|
8
|
+
* The adapter phase recovers from its per-adapter `MANIFEST.json`. Align, split and parquet ran 6 h 35 min
|
|
9
|
+
* of the 8 h 27 min `v0.7.0-de-holdout` build with no recovery point, because the four labeled streams are
|
|
10
|
+
* truncated at open and every accumulator is a local. This file is what a resumed align phase reads
|
|
11
|
+
* instead of starting over.
|
|
12
|
+
*
|
|
13
|
+
* The phase is replayable only when the same rows produce the same output, so the checkpoint records the
|
|
14
|
+
* settings that change alignment and refuses a resume under different ones. `align-determinism` in
|
|
15
|
+
* `build.test.ts` compares four output digests across two builds over one intermediate directory.
|
|
16
|
+
*/
|
|
17
|
+
import type { PathBuilderLike } from "path-ts";
|
|
18
|
+
/**
|
|
19
|
+
* The four append-only outputs of the align phase, by the name in each file.
|
|
20
|
+
*/
|
|
21
|
+
export declare const ALIGN_OUTPUTS: readonly ["train", "val", "test", "quarantine"];
|
|
22
|
+
export type AlignOutputName = (typeof ALIGN_OUTPUTS)[number];
|
|
23
|
+
/**
|
|
24
|
+
* The settings that change what the align phase writes for a given input row.
|
|
25
|
+
*
|
|
26
|
+
* A resume under different settings would mix two policies inside one output file,
|
|
27
|
+
* so the checkpoint records these and a mismatch refuses the resume rather than continuing.
|
|
28
|
+
*/
|
|
29
|
+
export interface AlignSettings {
|
|
30
|
+
corpus_version: string;
|
|
31
|
+
synthesize: boolean;
|
|
32
|
+
license_policy: string;
|
|
33
|
+
exclude_licenses: readonly string[];
|
|
34
|
+
profile: string;
|
|
35
|
+
holdout_digest: string;
|
|
36
|
+
}
|
|
37
|
+
/**
|
|
38
|
+
* The align phase's state at an adapter boundary.
|
|
39
|
+
*/
|
|
40
|
+
export interface AlignCheckpoint {
|
|
41
|
+
settings: AlignSettings;
|
|
42
|
+
/**
|
|
43
|
+
* Adapter ids whose rows are already in the four outputs, in the order they were aligned.
|
|
44
|
+
*
|
|
45
|
+
* A resume requires this to be a prefix of the current run's adapter order.
|
|
46
|
+
* Alignment appends, so a different order would place a later adapter's rows
|
|
47
|
+
* before an earlier one's and change every split file's contents.
|
|
48
|
+
*/
|
|
49
|
+
completed_adapters: string[];
|
|
50
|
+
/**
|
|
51
|
+
* Byte length of each output at the checkpoint, after its writes were flushed.
|
|
52
|
+
*
|
|
53
|
+
* A resumed build shortens each file to this length.
|
|
54
|
+
* The shortening discards whatever the interrupted adapter wrote past the boundary.
|
|
55
|
+
*/
|
|
56
|
+
offsets: Record<AlignOutputName, number>;
|
|
57
|
+
counts: Record<"train" | "val" | "test", number>;
|
|
58
|
+
aligned: number;
|
|
59
|
+
quarantined: number;
|
|
60
|
+
license_counts: Record<string, number>;
|
|
61
|
+
refused_by_kind: Record<string, number>;
|
|
62
|
+
/**
|
|
63
|
+
* Rows the license policy refused, counted per `source`, so a resumed build keeps the tally
|
|
64
|
+
* rather than restarting it at zero and understating the refusal in its manifest.
|
|
65
|
+
*/
|
|
66
|
+
refused_by_source: Record<string, number>;
|
|
67
|
+
excluded_by_license: number;
|
|
68
|
+
admitted_unresolved_license_rows: number;
|
|
69
|
+
excluded_by_eligibility: number;
|
|
70
|
+
rows_by_source: Record<string, {
|
|
71
|
+
rows: number;
|
|
72
|
+
license: string;
|
|
73
|
+
}>;
|
|
74
|
+
written_at: string;
|
|
75
|
+
}
|
|
76
|
+
/**
|
|
77
|
+
* Read the align checkpoint, or return `null` when the build has not reached one.
|
|
78
|
+
*
|
|
79
|
+
* @throws When the checkpoint describes different settings, or an adapter order
|
|
80
|
+
* the current run does not begin with.
|
|
81
|
+
* Both cases would append this run's rows to a file written under different rules.
|
|
82
|
+
* The resulting corpus could report a row count that later readers accept.
|
|
83
|
+
*/
|
|
84
|
+
export declare function readAlignCheckpoint(path: PathBuilderLike, expected: {
|
|
85
|
+
settings: AlignSettings;
|
|
86
|
+
adapterOrder: readonly string[];
|
|
87
|
+
}): Promise<AlignCheckpoint | null>;
|
|
88
|
+
/**
|
|
89
|
+
* Write the align checkpoint.
|
|
90
|
+
*
|
|
91
|
+
* The caller flushes the four outputs before calling this, so every byte the offsets count is on disk.
|
|
92
|
+
*/
|
|
93
|
+
export declare function writeAlignCheckpoint(path: PathBuilderLike, checkpoint: AlignCheckpoint): Promise<void>;
|
|
94
|
+
//# sourceMappingURL=checkpoint.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"checkpoint.d.ts","sourceRoot":"","sources":["../../lib/build/checkpoint.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AAOH,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,SAAS,CAAA;AAE9C;;GAEG;AACH,eAAO,MAAM,aAAa,YAAI,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,YAAY,CAAU,CAAA;AAE5E,MAAM,MAAM,eAAe,GAAG,CAAC,OAAO,aAAa,CAAC,CAAC,MAAM,CAAC,CAAA;AAE5D;;;;;GAKG;AACH,MAAM,WAAW,aAAa;IAC7B,cAAc,EAAE,MAAM,CAAA;IACtB,UAAU,EAAE,OAAO,CAAA;IACnB,cAAc,EAAE,MAAM,CAAA;IACtB,gBAAgB,EAAE,SAAS,MAAM,EAAE,CAAA;IACnC,OAAO,EAAE,MAAM,CAAA;IACf,cAAc,EAAE,MAAM,CAAA;CACtB;AAED;;GAEG;AACH,MAAM,WAAW,eAAe;IAC/B,QAAQ,EAAE,aAAa,CAAA;IAEvB;;;;;;OAMG;IACH,kBAAkB,EAAE,MAAM,EAAE,CAAA;IAE5B;;;;;OAKG;IACH,OAAO,EAAE,MAAM,CAAC,eAAe,EAAE,MAAM,CAAC,CAAA;IAExC,MAAM,EAAE,MAAM,CAAC,OAAO,GAAG,KAAK,GAAG,MAAM,EAAE,MAAM,CAAC,CAAA;IAChD,OAAO,EAAE,MAAM,CAAA;IACf,WAAW,EAAE,MAAM,CAAA;IACnB,cAAc,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IACtC,eAAe,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IACvC;;;OAGG;IACH,iBAAiB,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IACzC,mBAAmB,EAAE,MAAM,CAAA;IAC3B,gCAAgC,EAAE,MAAM,CAAA;IACxC,uBAAuB,EAAE,MAAM,CAAA;IAC/B,cAAc,EAAE,MAAM,CAAC,MAAM,EAAE;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAA;KAAE,CAAC,CAAA;IACjE,UAAU,EAAE,MAAM,CAAA;CAClB;AAED;;;;;;;GAOG;AACH,wBAAsB,mBAAmB,CACxC,IAAI,EAAE,eAAe,EACrB,QAAQ,EAAE;IAAE,QAAQ,EAAE,aAAa,CAAC;IAAC,YAAY,EAAE,SAAS,MAAM,EAAE,CAAA;CAAE,GACpE,OAAO,CAAC,eAAe,GAAG,IAAI,CAAC,CA0BjC;AAED;;;;GAIG;AACH,wBAAgB,oBAAoB,CAAC,IAAI,EAAE,eAAe,EAAE,UAAU,EAAE,eAAe,GAAG,OAAO,CAAC,IAAI,CAAC,CAEtG"}
|
|
@@ -0,0 +1,61 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file The align phase's resume point. It records aligned adapters and each output's length at that moment.
|
|
6
|
+
* It also records every accumulator used to compute the manifest.
|
|
7
|
+
*
|
|
8
|
+
* The adapter phase recovers from its per-adapter `MANIFEST.json`. Align, split and parquet ran 6 h 35 min
|
|
9
|
+
* of the 8 h 27 min `v0.7.0-de-holdout` build with no recovery point, because the four labeled streams are
|
|
10
|
+
* truncated at open and every accumulator is a local. This file is what a resumed align phase reads
|
|
11
|
+
* instead of starting over.
|
|
12
|
+
*
|
|
13
|
+
* The phase is replayable only when the same rows produce the same output, so the checkpoint records the
|
|
14
|
+
* settings that change alignment and refuses a resume under different ones. `align-determinism` in
|
|
15
|
+
* `build.test.ts` compares four output digests across two builds over one intermediate directory.
|
|
16
|
+
*/
|
|
17
|
+
import { readLocalJSONFile } from "@mailwoman/core/fs/readers";
|
|
18
|
+
import { pathExists } from "@mailwoman/core/fs/readers/stat";
|
|
19
|
+
import { writeLocalJSONFile } from "@mailwoman/core/fs/writers";
|
|
20
|
+
import { isIdentical } from "@mailwoman/core/identical";
|
|
21
|
+
import { stringifyJSON } from "@mailwoman/core/json";
|
|
22
|
+
/**
|
|
23
|
+
* The four append-only outputs of the align phase, by the name in each file.
|
|
24
|
+
*/
|
|
25
|
+
export const ALIGN_OUTPUTS = ["train", "val", "test", "quarantine"];
|
|
26
|
+
/**
|
|
27
|
+
* Read the align checkpoint, or return `null` when the build has not reached one.
|
|
28
|
+
*
|
|
29
|
+
* @throws When the checkpoint describes different settings, or an adapter order
|
|
30
|
+
* the current run does not begin with.
|
|
31
|
+
* Both cases would append this run's rows to a file written under different rules.
|
|
32
|
+
* The resulting corpus could report a row count that later readers accept.
|
|
33
|
+
*/
|
|
34
|
+
export async function readAlignCheckpoint(path, expected) {
|
|
35
|
+
if (!(await pathExists(path)))
|
|
36
|
+
return null;
|
|
37
|
+
const checkpoint = await readLocalJSONFile(path);
|
|
38
|
+
const found = stringifyJSON(checkpoint.settings);
|
|
39
|
+
const wanted = stringifyJSON(expected.settings);
|
|
40
|
+
if (found !== wanted) {
|
|
41
|
+
throw new Error(`resume refused: ${path.toString()} was written under ${found} and this build runs under ${wanted}. ` +
|
|
42
|
+
`Delete the intermediate directory's labeled and quarantine files together with this checkpoint to ` +
|
|
43
|
+
`align from the start.`);
|
|
44
|
+
}
|
|
45
|
+
const prefix = expected.adapterOrder.slice(0, checkpoint.completed_adapters.length);
|
|
46
|
+
if (!isIdentical(prefix, checkpoint.completed_adapters)) {
|
|
47
|
+
throw new Error(`resume refused: ${path.toString()} recorded ${stringifyJSON(checkpoint.completed_adapters)} as aligned ` +
|
|
48
|
+
`and this build's adapter order begins ${stringifyJSON(prefix)}. Alignment appends, so the rows would ` +
|
|
49
|
+
`land in a different order than the recorded offsets describe.`);
|
|
50
|
+
}
|
|
51
|
+
return checkpoint;
|
|
52
|
+
}
|
|
53
|
+
/**
|
|
54
|
+
* Write the align checkpoint.
|
|
55
|
+
*
|
|
56
|
+
* The caller flushes the four outputs before calling this, so every byte the offsets count is on disk.
|
|
57
|
+
*/
|
|
58
|
+
export function writeAlignCheckpoint(path, checkpoint) {
|
|
59
|
+
return writeLocalJSONFile(checkpoint, path).then(() => undefined);
|
|
60
|
+
}
|
|
61
|
+
//# sourceMappingURL=checkpoint.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"checkpoint.js","sourceRoot":"","sources":["../../lib/build/checkpoint.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAA;AAC9D,OAAO,EAAE,UAAU,EAAE,MAAM,iCAAiC,CAAA;AAC5D,OAAO,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AAC/D,OAAO,EAAE,WAAW,EAAE,MAAM,2BAA2B,CAAA;AACvD,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AAGpD;;GAEG;AACH,MAAM,CAAC,MAAM,aAAa,GAAG,CAAC,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,YAAY,CAAU,CAAA;AA2D5E;;;;;;;GAOG;AACH,MAAM,CAAC,KAAK,UAAU,mBAAmB,CACxC,IAAqB,EACrB,QAAsE;IAEtE,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,IAAI,CAAC,CAAC;QAAE,OAAO,IAAI,CAAA;IAE1C,MAAM,UAAU,GAAG,MAAM,iBAAiB,CAAkB,IAAI,CAAC,CAAA;IACjE,MAAM,KAAK,GAAG,aAAa,CAAC,UAAU,CAAC,QAAQ,CAAC,CAAA;IAChD,MAAM,MAAM,GAAG,aAAa,CAAC,QAAQ,CAAC,QAAQ,CAAC,CAAA;IAE/C,IAAI,KAAK,KAAK,MAAM,EAAE,CAAC;QACtB,MAAM,IAAI,KAAK,CACd,mBAAmB,IAAI,CAAC,QAAQ,EAAE,sBAAsB,KAAK,8BAA8B,MAAM,IAAI;YACpG,oGAAoG;YACpG,uBAAuB,CACxB,CAAA;IACF,CAAC;IAED,MAAM,MAAM,GAAG,QAAQ,CAAC,YAAY,CAAC,KAAK,CAAC,CAAC,EAAE,UAAU,CAAC,kBAAkB,CAAC,MAAM,CAAC,CAAA;IAEnF,IAAI,CAAC,WAAW,CAAC,MAAM,EAAE,UAAU,CAAC,kBAAkB,CAAC,EAAE,CAAC;QACzD,MAAM,IAAI,KAAK,CACd,mBAAmB,IAAI,CAAC,QAAQ,EAAE,aAAa,aAAa,CAAC,UAAU,CAAC,kBAAkB,CAAC,cAAc;YACxG,yCAAyC,aAAa,CAAC,MAAM,CAAC,yCAAyC;YACvG,+DAA+D,CAChE,CAAA;IACF,CAAC;IAED,OAAO,UAAU,CAAA;AAClB,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,oBAAoB,CAAC,IAAqB,EAAE,UAA2B;IACtF,OAAO,kBAAkB,CAAC,UAAU,EAAE,IAAI,CAAC,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,SAAS,CAAC,CAAA;AAClE,CAAC"}
|
|
@@ -0,0 +1,47 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* @copyright Sister Software
|
|
3
|
+
* @license AGPL-3.0
|
|
4
|
+
* @author Teffen Ellis, et al.
|
|
5
|
+
* @file What the address-source register records about a row's source, read once per build.
|
|
6
|
+
*/
|
|
7
|
+
import { type LicenseDecision } from "#source-register";
|
|
8
|
+
import type { CanonicalRow } from "#types";
|
|
9
|
+
/**
|
|
10
|
+
* The key a row is looked up under: the adapter that emitted it and the jurisdiction it describes.
|
|
11
|
+
*
|
|
12
|
+
* The register scopes a source to one publisher in one jurisdiction,
|
|
13
|
+
* so an adapter id by itself cannot address a record.
|
|
14
|
+
* One adapter serves several: `ban` emits eleven jurisdictions from one schema
|
|
15
|
+
* and each is its own source under its own license decision.
|
|
16
|
+
*/
|
|
17
|
+
export declare function sourceEligibilityKey(adapterID: string, iso2: string): string;
|
|
18
|
+
/**
|
|
19
|
+
* Every register source's ingest-eligibility reasons, keyed by the adapter
|
|
20
|
+
* and jurisdiction of the rows that recorded it.
|
|
21
|
+
*
|
|
22
|
+
* A source declaring no `adapterID` is absent from the map, because no adapter emits it
|
|
23
|
+
* and no row can arrive under it.
|
|
24
|
+
* An empty array is the only value a caller may read as permission.
|
|
25
|
+
* An absent key is refused rather than admitted.
|
|
26
|
+
*/
|
|
27
|
+
export declare function readSourceEligibility(): Promise<ReadonlyMap<string, readonly string[]>>;
|
|
28
|
+
/**
|
|
29
|
+
* The register's license decisions keyed by the label in a row.
|
|
30
|
+
*
|
|
31
|
+
* An unmatched adapter label records a `null` decision rather than inventing one.
|
|
32
|
+
*/
|
|
33
|
+
export declare function readRegisterDecisions(): Promise<ReadonlyMap<string, LicenseDecision>>;
|
|
34
|
+
/**
|
|
35
|
+
* Reads why a row's source may be refused from a release-eligible corpus.
|
|
36
|
+
* Records each refused source.
|
|
37
|
+
*
|
|
38
|
+
* `refused` accumulates as rows arrive.
|
|
39
|
+
* The build manifest reports the collected values.
|
|
40
|
+
*
|
|
41
|
+
* A build whose `eligibility` is `null` runs under the exploratory profile and every row reads eligible.
|
|
42
|
+
*/
|
|
43
|
+
export declare function createIneligibilityReader(eligibility: ReadonlyMap<string, readonly string[]> | null): {
|
|
44
|
+
refused: Map<string, readonly string[]>;
|
|
45
|
+
read: (row: CanonicalRow) => readonly string[] | null;
|
|
46
|
+
};
|
|
47
|
+
//# sourceMappingURL=eligibility.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"eligibility.d.ts","sourceRoot":"","sources":["../../lib/build/eligibility.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAIH,OAAO,EAAwD,KAAK,eAAe,EAAE,MAAM,kBAAkB,CAAA;AAC7G,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,QAAQ,CAAA;AAE1C;;;;;;;GAOG;AACH,wBAAgB,oBAAoB,CAAC,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,GAAG,MAAM,CAE5E;AAED;;;;;;;;GAQG;AACH,wBAAsB,qBAAqB,IAAI,OAAO,CAAC,WAAW,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAC,CAAC,CAW7F;AAED;;;;GAIG;AACH,wBAAsB,qBAAqB,IAAI,OAAO,CAAC,WAAW,CAAC,MAAM,EAAE,eAAe,CAAC,CAAC,CAI3F;AAED;;;;;;;;GAQG;AACH,wBAAgB,yBAAyB,CAAC,WAAW,EAAE,WAAW,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAC,GAAG,IAAI,GAAG;IACtG,OAAO,EAAE,GAAG,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,CAAC,CAAA;IACvC,IAAI,EAAE,CAAC,GAAG,EAAE,YAAY,KAAK,SAAS,MAAM,EAAE,GAAG,IAAI,CAAA;CACrD,CA0BA"}
|