@mailwoman/corpus 9.2.0 → 9.3.0
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 +22 -22
- package/data/PROVENANCE.md +7 -7
- package/lib/adapters/ban/adapter.ts +190 -0
- package/lib/adapters/ban/street-decompose.ts +66 -0
- package/lib/adapters/fcc-bdc/adapter.ts +174 -0
- package/lib/adapters/geonames/adapter.ts +199 -0
- package/lib/adapters/geonames-postal/adapter.ts +122 -0
- package/lib/adapters/gnaf/adapter.ts +149 -0
- package/lib/adapters/gnaf/assemble.ts +225 -0
- package/lib/adapters/index.ts +153 -0
- package/lib/adapters/openaddresses/adapter.ts +251 -0
- package/lib/adapters/overture/adapter.ts +162 -0
- package/lib/adapters/state-hi-schools/adapter.ts +170 -0
- package/lib/adapters/state-hi-schools/workbook.ts +35 -0
- package/lib/adapters/synth-po-box/adapter.ts +204 -0
- package/lib/adapters/tiger/adapter.ts +252 -0
- package/lib/adapters/tiger/street-decompose.ts +83 -0
- package/lib/adapters/usgov-hrsa-fqhc/adapter.ts +167 -0
- package/lib/adapters/usgov-imls-pls/adapter.ts +143 -0
- package/lib/adapters/usgov-irs-bmf/adapter.ts +151 -0
- package/lib/adapters/usgov-nad/adapter.ts +334 -0
- package/lib/adapters/usgov-nppes/adapter.ts +151 -0
- package/lib/adapters/usgov-samhsa-treatment-locator/adapter.ts +193 -0
- package/lib/adapters/utils/index.ts +307 -0
- package/lib/adapters/wof-admin-jp/adapter.ts +214 -0
- package/lib/adapters/wof-admin-json/adapter.ts +231 -0
- package/lib/adapters/wof-json-rows.ts +129 -0
- package/lib/adapters/wof-postalcode-json/adapter.ts +176 -0
- package/lib/build.ts +378 -0
- package/lib/env.ts +50 -0
- package/lib/index.ts +13 -0
- package/lib/name-prone-us-suffixes.ts +12 -0
- package/lib/recipes/anchor-absorption.ts +101 -0
- package/lib/recipes/bare-country.ts +95 -0
- package/lib/recipes/boundary-stress.ts +100 -0
- package/lib/recipes/cn-organizational-units.ts +199 -0
- package/lib/recipes/country-balanced.ts +509 -0
- package/lib/recipes/cz-pcfirst-preposition.ts +111 -0
- package/lib/recipes/fr-admin-split.ts +242 -0
- package/lib/recipes/fr-bare-street.ts +170 -0
- package/lib/recipes/fr-fragment.ts +333 -0
- package/lib/recipes/fr-lieudit.ts +301 -0
- package/lib/recipes/fr-order.ts +245 -0
- package/lib/recipes/german.ts +359 -0
- package/lib/recipes/house-venue.ts +92 -0
- package/lib/recipes/index.ts +97 -0
- package/lib/recipes/intersection.ts +618 -0
- package/lib/recipes/locale.ts +599 -0
- package/lib/recipes/nl-postcode.ts +115 -0
- package/lib/recipes/no-fragment.ts +219 -0
- package/lib/recipes/no-street-led.ts +141 -0
- package/lib/recipes/no-street.ts +90 -0
- package/lib/recipes/po-box-cedex.ts +976 -0
- package/lib/recipes/po-box.ts +123 -0
- package/lib/recipes/reviewed-postcode-tail.ts +190 -0
- package/lib/recipes/scaffold.ts +469 -0
- package/lib/recipes/si-bare-village.ts +97 -0
- package/lib/recipes/street-affix.ts +804 -0
- package/lib/recipes/street-bare.ts +100 -0
- package/lib/recipes/street.ts +81 -0
- package/lib/recipes/sub-venue-sources.ts +877 -0
- package/lib/recipes/sub-venue.ts +997 -0
- package/lib/recipes/trailing-region.ts +183 -0
- package/lib/recipes/unit.ts +304 -0
- package/lib/runner.ts +312 -0
- package/lib/synthesizers/anchor-absorption.ts +288 -0
- package/lib/synthesizers/boundary-stress.ts +532 -0
- package/lib/synthesizers/german.ts +260 -0
- package/lib/synthesizers/house-venue.ts +308 -0
- package/lib/synthesizers/intersection.ts +233 -0
- package/lib/synthesizers/no-street.ts +327 -0
- package/lib/synthesizers/po-box.ts +323 -0
- package/lib/synthesizers/street.ts +269 -0
- package/lib/synthesizers/utils.ts +993 -0
- package/lib/test-kit/corpus-recipe.ts +86 -0
- package/lib/test-kit/index.ts +100 -0
- package/lib/tools/align-slice.ts +68 -0
- package/lib/tools/audit.ts +403 -0
- package/lib/tools/corpus-stats.ts +136 -0
- package/lib/tools/fetch/ban.ts +289 -0
- package/lib/tools/fetch/download.ts +262 -0
- package/lib/tools/fetch/geonames-dump.ts +257 -0
- package/lib/tools/fetch/geonames-postal.ts +180 -0
- package/lib/tools/fetch/hrsa.ts +59 -0
- package/lib/tools/fetch/imls-pls.ts +149 -0
- package/lib/tools/fetch/index.ts +147 -0
- package/lib/tools/fetch/nad.ts +341 -0
- package/lib/tools/fetch/nppes.ts +165 -0
- package/lib/tools/fetch/openaddresses.ts +347 -0
- package/lib/tools/fetch/ourairports.ts +170 -0
- package/lib/tools/fetch/ppd.ts +121 -0
- package/lib/tools/fetch/state-hi-schools.ts +161 -0
- package/lib/tools/fetch/state-sources.ts +146 -0
- package/lib/tools/fetch/tiger-full.ts +308 -0
- package/lib/tools/fetch/wikidata-subvenue.ts +384 -0
- package/lib/tools/golden-expand.ts +574 -0
- package/lib/tools/golden-promote.ts +337 -0
- package/lib/tools/golden-relabel-street.ts +683 -0
- package/lib/tools/index.ts +28 -0
- package/lib/tools/ingest-csv.ts +378 -0
- package/lib/tools/jsonl-to-parquet.ts +207 -0
- package/lib/tools/lint-slice-vocab.ts +388 -0
- package/lib/tools/lint-slice.ts +496 -0
- package/lib/tools/overlay-manifest.ts +148 -0
- package/lib/tools/overture-subvenue.ts +209 -0
- package/lib/tools/postcode-triples.ts +363 -0
- package/lib/tools/slice-kryptonite.ts +137 -0
- package/lib/tools/slice-translit.ts +218 -0
- package/lib/tools/sub-venue/harvest.ts +234 -0
- package/lib/tools/sub-venue/head-nouns.ts +245 -0
- package/lib/tools/sub-venue/surfaces.ts +95 -0
- package/lib/tools/sub-venue/table.ts +281 -0
- package/lib/tools/sub-venue/wikidata.ts +85 -0
- package/lib/tools/sub-venue-lexicon.ts +429 -0
- package/lib/tools/sub-venue-promotions.ts +330 -0
- package/lib/types.ts +247 -0
- package/lib/utils/align.ts +330 -0
- package/lib/utils/golden.ts +189 -0
- package/lib/utils/index.ts +14 -0
- package/lib/utils/parquet.ts +582 -0
- package/lib/utils/slice-stats.ts +95 -0
- package/lib/utils/split.ts +279 -0
- package/lib/utils/tokenize.ts +120 -0
- package/lib/utils/wof-json.ts +247 -0
- package/out/adapters/ban/adapter.d.ts.map +1 -0
- package/out/adapters/ban/adapter.js +149 -0
- package/out/adapters/ban/adapter.js.map +1 -0
- package/out/adapters/ban/street-decompose.d.ts.map +1 -0
- package/out/adapters/ban/street-decompose.js +51 -0
- package/out/adapters/ban/street-decompose.js.map +1 -0
- package/out/adapters/fcc-bdc/adapter.d.ts.map +1 -0
- package/out/adapters/fcc-bdc/adapter.js +144 -0
- package/out/adapters/fcc-bdc/adapter.js.map +1 -0
- package/out/adapters/geonames/adapter.d.ts.map +1 -0
- package/out/adapters/geonames/adapter.js +176 -0
- package/out/adapters/geonames/adapter.js.map +1 -0
- package/out/adapters/geonames-postal/adapter.d.ts +54 -0
- package/out/adapters/geonames-postal/adapter.d.ts.map +1 -0
- package/out/adapters/geonames-postal/adapter.js +107 -0
- package/out/adapters/geonames-postal/adapter.js.map +1 -0
- package/out/adapters/gnaf/adapter.d.ts +49 -0
- package/out/adapters/gnaf/adapter.d.ts.map +1 -0
- package/out/adapters/gnaf/adapter.js +124 -0
- package/out/adapters/gnaf/adapter.js.map +1 -0
- package/out/adapters/gnaf/assemble.d.ts +65 -0
- package/out/adapters/gnaf/assemble.d.ts.map +1 -0
- package/out/adapters/gnaf/assemble.js +154 -0
- package/out/adapters/gnaf/assemble.js.map +1 -0
- package/out/adapters/index.d.ts +51 -0
- package/out/adapters/index.d.ts.map +1 -0
- package/out/adapters/index.js +95 -0
- package/out/adapters/index.js.map +1 -0
- package/out/adapters/openaddresses/adapter.d.ts.map +1 -0
- package/out/adapters/openaddresses/adapter.js +186 -0
- package/out/adapters/openaddresses/adapter.js.map +1 -0
- package/out/adapters/overture/adapter.d.ts +48 -0
- package/out/adapters/overture/adapter.d.ts.map +1 -0
- package/out/adapters/overture/adapter.js +125 -0
- package/out/adapters/overture/adapter.js.map +1 -0
- package/out/adapters/state-hi-schools/adapter.d.ts +46 -0
- package/out/adapters/state-hi-schools/adapter.d.ts.map +1 -0
- package/out/adapters/state-hi-schools/adapter.js +145 -0
- package/out/adapters/state-hi-schools/adapter.js.map +1 -0
- package/out/adapters/state-hi-schools/workbook.d.ts +29 -0
- package/out/adapters/state-hi-schools/workbook.d.ts.map +1 -0
- package/out/adapters/state-hi-schools/workbook.js +20 -0
- package/out/adapters/state-hi-schools/workbook.js.map +1 -0
- package/out/adapters/state-ia-contractors/adapter.d.ts.map +1 -0
- package/out/adapters/state-ia-contractors/adapter.js.map +1 -0
- package/out/adapters/state-ny-notaries/adapter.d.ts.map +1 -0
- package/out/adapters/state-ny-notaries/adapter.js.map +1 -0
- package/out/adapters/state-tx-notaries/adapter.d.ts.map +1 -0
- package/out/adapters/state-tx-notaries/adapter.js.map +1 -0
- package/out/adapters/synth-po-box/adapter.d.ts.map +1 -0
- package/out/adapters/synth-po-box/adapter.js +140 -0
- package/out/adapters/synth-po-box/adapter.js.map +1 -0
- package/out/adapters/tiger/adapter.d.ts +55 -0
- package/out/adapters/tiger/adapter.d.ts.map +1 -0
- package/out/adapters/tiger/adapter.js +201 -0
- package/out/adapters/tiger/adapter.js.map +1 -0
- package/out/adapters/tiger/street-decompose.d.ts.map +1 -0
- package/out/adapters/tiger/street-decompose.js +68 -0
- package/out/adapters/tiger/street-decompose.js.map +1 -0
- package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts +56 -0
- package/out/adapters/usgov-hrsa-fqhc/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-hrsa-fqhc/adapter.js +132 -0
- package/out/adapters/usgov-hrsa-fqhc/adapter.js.map +1 -0
- package/out/adapters/usgov-imls-pls/adapter.d.ts +36 -0
- package/out/adapters/usgov-imls-pls/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-imls-pls/adapter.js +114 -0
- package/out/adapters/usgov-imls-pls/adapter.js.map +1 -0
- package/out/adapters/usgov-irs-bmf/adapter.d.ts +37 -0
- package/out/adapters/usgov-irs-bmf/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-irs-bmf/adapter.js +120 -0
- package/out/adapters/usgov-irs-bmf/adapter.js.map +1 -0
- package/out/adapters/usgov-nad/adapter.d.ts +48 -0
- package/out/adapters/usgov-nad/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-nad/adapter.js +249 -0
- package/out/adapters/usgov-nad/adapter.js.map +1 -0
- package/out/adapters/usgov-nppes/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-nppes/adapter.js +118 -0
- package/out/adapters/usgov-nppes/adapter.js.map +1 -0
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.d.ts.map +1 -0
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.js +154 -0
- package/out/adapters/usgov-samhsa-treatment-locator/adapter.js.map +1 -0
- package/out/adapters/utils/index.d.ts +191 -0
- package/out/adapters/utils/index.d.ts.map +1 -0
- package/out/adapters/utils/index.js +206 -0
- package/out/adapters/utils/index.js.map +1 -0
- package/out/adapters/wof-admin-jp/adapter.d.ts.map +1 -0
- package/out/adapters/wof-admin-jp/adapter.js +161 -0
- package/out/adapters/wof-admin-jp/adapter.js.map +1 -0
- package/out/adapters/wof-admin-json/adapter.d.ts +79 -0
- package/out/adapters/wof-admin-json/adapter.d.ts.map +1 -0
- package/out/adapters/wof-admin-json/adapter.js +197 -0
- package/out/adapters/wof-admin-json/adapter.js.map +1 -0
- package/out/adapters/wof-json-rows.d.ts +62 -0
- package/out/adapters/wof-json-rows.d.ts.map +1 -0
- package/out/adapters/wof-json-rows.js +83 -0
- package/out/adapters/wof-json-rows.js.map +1 -0
- package/out/adapters/wof-postalcode-json/adapter.d.ts +65 -0
- package/out/adapters/wof-postalcode-json/adapter.d.ts.map +1 -0
- package/out/adapters/wof-postalcode-json/adapter.js +151 -0
- package/out/adapters/wof-postalcode-json/adapter.js.map +1 -0
- package/out/build.d.ts +132 -0
- package/out/build.d.ts.map +1 -0
- package/out/build.js +245 -0
- package/out/build.js.map +1 -0
- package/out/codex/us-fips-state.d.ts.map +1 -0
- package/out/codex/us-fips-state.js.map +1 -0
- package/out/env.d.ts +46 -0
- package/out/env.d.ts.map +1 -0
- package/out/env.js +46 -0
- package/out/env.js.map +1 -0
- package/out/index.d.ts +13 -0
- package/out/index.d.ts.map +1 -0
- package/out/index.js +13 -0
- package/out/index.js.map +1 -0
- package/out/name-prone-us-suffixes.d.ts +12 -0
- package/out/name-prone-us-suffixes.d.ts.map +1 -0
- package/out/name-prone-us-suffixes.js +12 -0
- package/out/name-prone-us-suffixes.js.map +1 -0
- package/out/recipes/anchor-absorption.d.ts +20 -0
- package/out/recipes/anchor-absorption.d.ts.map +1 -0
- package/out/recipes/anchor-absorption.js +81 -0
- package/out/recipes/anchor-absorption.js.map +1 -0
- package/out/recipes/bare-country.d.ts +25 -0
- package/out/recipes/bare-country.d.ts.map +1 -0
- package/out/recipes/bare-country.js +87 -0
- package/out/recipes/bare-country.js.map +1 -0
- package/out/recipes/boundary-stress.d.ts +20 -0
- package/out/recipes/boundary-stress.d.ts.map +1 -0
- package/out/recipes/boundary-stress.js +83 -0
- package/out/recipes/boundary-stress.js.map +1 -0
- package/out/recipes/cn-organizational-units.d.ts +36 -0
- package/out/recipes/cn-organizational-units.d.ts.map +1 -0
- package/out/recipes/cn-organizational-units.js +159 -0
- package/out/recipes/cn-organizational-units.js.map +1 -0
- package/out/recipes/country-balanced.d.ts +32 -0
- package/out/recipes/country-balanced.d.ts.map +1 -0
- package/out/recipes/country-balanced.js +393 -0
- package/out/recipes/country-balanced.js.map +1 -0
- package/out/recipes/cz-pcfirst-preposition.d.ts +22 -0
- package/out/recipes/cz-pcfirst-preposition.d.ts.map +1 -0
- package/out/recipes/cz-pcfirst-preposition.js +100 -0
- package/out/recipes/cz-pcfirst-preposition.js.map +1 -0
- package/out/recipes/fr-admin-split.d.ts +38 -0
- package/out/recipes/fr-admin-split.d.ts.map +1 -0
- package/out/recipes/fr-admin-split.js +192 -0
- package/out/recipes/fr-admin-split.js.map +1 -0
- package/out/recipes/fr-bare-street.d.ts +28 -0
- package/out/recipes/fr-bare-street.d.ts.map +1 -0
- package/out/recipes/fr-bare-street.js +143 -0
- package/out/recipes/fr-bare-street.js.map +1 -0
- package/out/recipes/fr-fragment.d.ts +77 -0
- package/out/recipes/fr-fragment.d.ts.map +1 -0
- package/out/recipes/fr-fragment.js +276 -0
- package/out/recipes/fr-fragment.js.map +1 -0
- package/out/recipes/fr-lieudit.d.ts +32 -0
- package/out/recipes/fr-lieudit.d.ts.map +1 -0
- package/out/recipes/fr-lieudit.js +227 -0
- package/out/recipes/fr-lieudit.js.map +1 -0
- package/out/recipes/fr-order.d.ts +38 -0
- package/out/recipes/fr-order.d.ts.map +1 -0
- package/out/recipes/fr-order.js +197 -0
- package/out/recipes/fr-order.js.map +1 -0
- package/out/recipes/german.d.ts +44 -0
- package/out/recipes/german.d.ts.map +1 -0
- package/out/recipes/german.js +288 -0
- package/out/recipes/german.js.map +1 -0
- package/out/recipes/house-venue.d.ts +18 -0
- package/out/recipes/house-venue.d.ts.map +1 -0
- package/out/recipes/house-venue.js +73 -0
- package/out/recipes/house-venue.js.map +1 -0
- package/out/recipes/index.d.ts +24 -0
- package/out/recipes/index.d.ts.map +1 -0
- package/out/recipes/index.js +91 -0
- package/out/recipes/index.js.map +1 -0
- package/out/recipes/intersection.d.ts +42 -0
- package/out/recipes/intersection.d.ts.map +1 -0
- package/out/recipes/intersection.js +452 -0
- package/out/recipes/intersection.js.map +1 -0
- package/out/recipes/locale.d.ts +136 -0
- package/out/recipes/locale.d.ts.map +1 -0
- package/out/recipes/locale.js +460 -0
- package/out/recipes/locale.js.map +1 -0
- package/out/recipes/nl-postcode.d.ts +26 -0
- package/out/recipes/nl-postcode.d.ts.map +1 -0
- package/out/recipes/nl-postcode.js +102 -0
- package/out/recipes/nl-postcode.js.map +1 -0
- package/out/recipes/no-fragment.d.ts +43 -0
- package/out/recipes/no-fragment.d.ts.map +1 -0
- package/out/recipes/no-fragment.js +182 -0
- package/out/recipes/no-fragment.js.map +1 -0
- package/out/recipes/no-street-led.d.ts +24 -0
- package/out/recipes/no-street-led.d.ts.map +1 -0
- package/out/recipes/no-street-led.js +113 -0
- package/out/recipes/no-street-led.js.map +1 -0
- package/out/recipes/no-street.d.ts +18 -0
- package/out/recipes/no-street.d.ts.map +1 -0
- package/out/recipes/no-street.js +71 -0
- package/out/recipes/no-street.js.map +1 -0
- package/out/recipes/po-box-cedex.d.ts +38 -0
- package/out/recipes/po-box-cedex.d.ts.map +1 -0
- package/out/recipes/po-box-cedex.js +767 -0
- package/out/recipes/po-box-cedex.js.map +1 -0
- package/out/recipes/po-box.d.ts +17 -0
- package/out/recipes/po-box.d.ts.map +1 -0
- package/out/recipes/po-box.js +100 -0
- package/out/recipes/po-box.js.map +1 -0
- package/out/recipes/reviewed-postcode-tail.d.ts +47 -0
- package/out/recipes/reviewed-postcode-tail.d.ts.map +1 -0
- package/out/recipes/reviewed-postcode-tail.js +125 -0
- package/out/recipes/reviewed-postcode-tail.js.map +1 -0
- package/out/recipes/scaffold.d.ts +324 -0
- package/out/recipes/scaffold.d.ts.map +1 -0
- package/out/recipes/scaffold.js +186 -0
- package/out/recipes/scaffold.js.map +1 -0
- package/out/recipes/si-bare-village.d.ts +32 -0
- package/out/recipes/si-bare-village.d.ts.map +1 -0
- package/out/recipes/si-bare-village.js +88 -0
- package/out/recipes/si-bare-village.js.map +1 -0
- package/out/recipes/street-affix.d.ts +89 -0
- package/out/recipes/street-affix.d.ts.map +1 -0
- package/out/recipes/street-affix.js +563 -0
- package/out/recipes/street-affix.js.map +1 -0
- package/out/recipes/street-bare.d.ts +25 -0
- package/out/recipes/street-bare.d.ts.map +1 -0
- package/out/recipes/street-bare.js +86 -0
- package/out/recipes/street-bare.js.map +1 -0
- package/out/recipes/street.d.ts +15 -0
- package/out/recipes/street.d.ts.map +1 -0
- package/out/recipes/street.js +69 -0
- package/out/recipes/street.js.map +1 -0
- package/out/recipes/sub-venue-sources.d.ts +229 -0
- package/out/recipes/sub-venue-sources.d.ts.map +1 -0
- package/out/recipes/sub-venue-sources.js +653 -0
- package/out/recipes/sub-venue-sources.js.map +1 -0
- package/out/recipes/sub-venue.d.ts +269 -0
- package/out/recipes/sub-venue.d.ts.map +1 -0
- package/out/recipes/sub-venue.js +725 -0
- package/out/recipes/sub-venue.js.map +1 -0
- package/out/recipes/trailing-region.d.ts +74 -0
- package/out/recipes/trailing-region.d.ts.map +1 -0
- package/out/recipes/trailing-region.js +157 -0
- package/out/recipes/trailing-region.js.map +1 -0
- package/out/recipes/unit.d.ts +29 -0
- package/out/recipes/unit.d.ts.map +1 -0
- package/out/recipes/unit.js +230 -0
- package/out/recipes/unit.js.map +1 -0
- package/out/runner.d.ts +130 -0
- package/out/runner.d.ts.map +1 -0
- package/out/runner.js +185 -0
- package/out/runner.js.map +1 -0
- package/out/synthesizers/anchor-absorption.d.ts +63 -0
- package/out/synthesizers/anchor-absorption.d.ts.map +1 -0
- package/out/synthesizers/anchor-absorption.js +221 -0
- package/out/synthesizers/anchor-absorption.js.map +1 -0
- package/out/synthesizers/boundary-stress.d.ts +78 -0
- package/out/synthesizers/boundary-stress.d.ts.map +1 -0
- package/out/synthesizers/boundary-stress.js +455 -0
- package/out/synthesizers/boundary-stress.js.map +1 -0
- package/out/synthesizers/german.d.ts +115 -0
- package/out/synthesizers/german.d.ts.map +1 -0
- package/out/synthesizers/german.js +149 -0
- package/out/synthesizers/german.js.map +1 -0
- package/out/synthesizers/house-venue.d.ts +56 -0
- package/out/synthesizers/house-venue.d.ts.map +1 -0
- package/out/synthesizers/house-venue.js +247 -0
- package/out/synthesizers/house-venue.js.map +1 -0
- package/out/synthesizers/intersection.d.ts +52 -0
- package/out/synthesizers/intersection.d.ts.map +1 -0
- package/out/synthesizers/intersection.js +175 -0
- package/out/synthesizers/intersection.js.map +1 -0
- package/out/synthesizers/no-street.d.ts +72 -0
- package/out/synthesizers/no-street.d.ts.map +1 -0
- package/out/synthesizers/no-street.js +262 -0
- package/out/synthesizers/no-street.js.map +1 -0
- package/out/synthesizers/po-box.d.ts +108 -0
- package/out/synthesizers/po-box.d.ts.map +1 -0
- package/out/synthesizers/po-box.js +250 -0
- package/out/synthesizers/po-box.js.map +1 -0
- package/out/synthesizers/street.d.ts +54 -0
- package/out/synthesizers/street.d.ts.map +1 -0
- package/out/synthesizers/street.js +218 -0
- package/out/synthesizers/street.js.map +1 -0
- package/out/synthesizers/utils.d.ts +241 -0
- package/out/synthesizers/utils.d.ts.map +1 -0
- package/out/synthesizers/utils.js +776 -0
- package/out/synthesizers/utils.js.map +1 -0
- package/out/test-kit/corpus-recipe.d.ts +59 -0
- package/out/test-kit/corpus-recipe.d.ts.map +1 -0
- package/out/test-kit/corpus-recipe.js +39 -0
- package/out/test-kit/corpus-recipe.js.map +1 -0
- package/out/test-kit/index.d.ts +48 -0
- package/out/test-kit/index.d.ts.map +1 -0
- package/out/test-kit/index.js +78 -0
- package/out/test-kit/index.js.map +1 -0
- package/out/tools/align-slice.d.ts +7 -0
- package/out/tools/align-slice.d.ts.map +1 -0
- package/out/tools/align-slice.js +55 -0
- package/out/tools/align-slice.js.map +1 -0
- package/out/tools/audit.d.ts +29 -0
- package/out/tools/audit.d.ts.map +1 -0
- package/out/tools/audit.js +282 -0
- package/out/tools/audit.js.map +1 -0
- package/out/tools/corpus-stats.d.ts +44 -0
- package/out/tools/corpus-stats.d.ts.map +1 -0
- package/out/tools/corpus-stats.js +102 -0
- package/out/tools/corpus-stats.js.map +1 -0
- package/out/tools/fetch/ban.d.ts +31 -0
- package/out/tools/fetch/ban.d.ts.map +1 -0
- package/out/tools/fetch/ban.js +239 -0
- package/out/tools/fetch/ban.js.map +1 -0
- package/out/tools/fetch/download.d.ts +117 -0
- package/out/tools/fetch/download.d.ts.map +1 -0
- package/out/tools/fetch/download.js +160 -0
- package/out/tools/fetch/download.js.map +1 -0
- package/out/tools/fetch/geonames-dump.d.ts +91 -0
- package/out/tools/fetch/geonames-dump.d.ts.map +1 -0
- package/out/tools/fetch/geonames-dump.js +171 -0
- package/out/tools/fetch/geonames-dump.js.map +1 -0
- package/out/tools/fetch/geonames-postal.d.ts +76 -0
- package/out/tools/fetch/geonames-postal.d.ts.map +1 -0
- package/out/tools/fetch/geonames-postal.js +126 -0
- package/out/tools/fetch/geonames-postal.js.map +1 -0
- package/out/tools/fetch/hrsa.d.ts +18 -0
- package/out/tools/fetch/hrsa.d.ts.map +1 -0
- package/out/tools/fetch/hrsa.js +46 -0
- package/out/tools/fetch/hrsa.js.map +1 -0
- package/out/tools/fetch/imls-pls.d.ts +25 -0
- package/out/tools/fetch/imls-pls.d.ts.map +1 -0
- package/out/tools/fetch/imls-pls.js +108 -0
- package/out/tools/fetch/imls-pls.js.map +1 -0
- package/out/tools/fetch/index.d.ts +144 -0
- package/out/tools/fetch/index.d.ts.map +1 -0
- package/out/tools/fetch/index.js +143 -0
- package/out/tools/fetch/index.js.map +1 -0
- package/out/tools/fetch/nad.d.ts +70 -0
- package/out/tools/fetch/nad.d.ts.map +1 -0
- package/out/tools/fetch/nad.js +234 -0
- package/out/tools/fetch/nad.js.map +1 -0
- package/out/tools/fetch/nppes.d.ts +26 -0
- package/out/tools/fetch/nppes.d.ts.map +1 -0
- package/out/tools/fetch/nppes.js +126 -0
- package/out/tools/fetch/nppes.js.map +1 -0
- package/out/tools/fetch/openaddresses.d.ts +9 -0
- package/out/tools/fetch/openaddresses.d.ts.map +1 -0
- package/out/tools/fetch/openaddresses.js +279 -0
- package/out/tools/fetch/openaddresses.js.map +1 -0
- package/out/tools/fetch/ourairports.d.ts +45 -0
- package/out/tools/fetch/ourairports.d.ts.map +1 -0
- package/out/tools/fetch/ourairports.js +128 -0
- package/out/tools/fetch/ourairports.js.map +1 -0
- package/out/tools/fetch/ppd.d.ts.map +1 -0
- package/out/tools/fetch/ppd.js +85 -0
- package/out/tools/fetch/ppd.js.map +1 -0
- package/out/tools/fetch/state-hi-schools.d.ts +26 -0
- package/out/tools/fetch/state-hi-schools.d.ts.map +1 -0
- package/out/tools/fetch/state-hi-schools.js +112 -0
- package/out/tools/fetch/state-hi-schools.js.map +1 -0
- package/out/tools/fetch/state-sources.d.ts +26 -0
- package/out/tools/fetch/state-sources.d.ts.map +1 -0
- package/out/tools/fetch/state-sources.js +111 -0
- package/out/tools/fetch/state-sources.js.map +1 -0
- package/out/tools/fetch/tiger-full.d.ts +22 -0
- package/out/tools/fetch/tiger-full.d.ts.map +1 -0
- package/out/tools/fetch/tiger-full.js +217 -0
- package/out/tools/fetch/tiger-full.js.map +1 -0
- package/out/tools/fetch/wikidata-subvenue.d.ts +171 -0
- package/out/tools/fetch/wikidata-subvenue.d.ts.map +1 -0
- package/out/tools/fetch/wikidata-subvenue.js +274 -0
- package/out/tools/fetch/wikidata-subvenue.js.map +1 -0
- package/out/tools/gb-title-case.d.ts.map +1 -0
- package/out/tools/gb-title-case.js.map +1 -0
- package/out/tools/golden-expand.d.ts +90 -0
- package/out/tools/golden-expand.d.ts.map +1 -0
- package/out/tools/golden-expand.js +406 -0
- package/out/tools/golden-expand.js.map +1 -0
- package/out/tools/golden-promote.d.ts.map +1 -0
- package/out/tools/golden-promote.js +240 -0
- package/out/tools/golden-promote.js.map +1 -0
- package/out/tools/golden-relabel-street.d.ts +196 -0
- package/out/tools/golden-relabel-street.d.ts.map +1 -0
- package/out/tools/golden-relabel-street.js +440 -0
- package/out/tools/golden-relabel-street.js.map +1 -0
- package/out/tools/index.d.ts +28 -0
- package/out/tools/index.d.ts.map +1 -0
- package/out/tools/index.js +28 -0
- package/out/tools/index.js.map +1 -0
- package/out/tools/ingest-csv.d.ts.map +1 -0
- package/out/tools/ingest-csv.js +265 -0
- package/out/tools/ingest-csv.js.map +1 -0
- package/out/tools/jsonl-to-parquet.d.ts +61 -0
- package/out/tools/jsonl-to-parquet.d.ts.map +1 -0
- package/out/tools/jsonl-to-parquet.js +150 -0
- package/out/tools/jsonl-to-parquet.js.map +1 -0
- package/out/tools/lint-slice-vocab.d.ts +90 -0
- package/out/tools/lint-slice-vocab.d.ts.map +1 -0
- package/out/tools/lint-slice-vocab.js +267 -0
- package/out/tools/lint-slice-vocab.js.map +1 -0
- package/out/tools/lint-slice.d.ts +93 -0
- package/out/tools/lint-slice.d.ts.map +1 -0
- package/out/tools/lint-slice.js +322 -0
- package/out/tools/lint-slice.js.map +1 -0
- package/out/tools/overlay-manifest.d.ts +39 -0
- package/out/tools/overlay-manifest.d.ts.map +1 -0
- package/out/tools/overlay-manifest.js +90 -0
- package/out/tools/overlay-manifest.js.map +1 -0
- package/out/tools/overture-subvenue.d.ts +112 -0
- package/out/tools/overture-subvenue.d.ts.map +1 -0
- package/out/tools/overture-subvenue.js +139 -0
- package/out/tools/overture-subvenue.js.map +1 -0
- package/out/tools/postcode-triples.d.ts +174 -0
- package/out/tools/postcode-triples.d.ts.map +1 -0
- package/out/tools/postcode-triples.js +292 -0
- package/out/tools/postcode-triples.js.map +1 -0
- package/out/tools/slice-kryptonite.d.ts +32 -0
- package/out/tools/slice-kryptonite.d.ts.map +1 -0
- package/out/tools/slice-kryptonite.js +99 -0
- package/out/tools/slice-kryptonite.js.map +1 -0
- package/out/tools/slice-translit.d.ts +51 -0
- package/out/tools/slice-translit.d.ts.map +1 -0
- package/out/tools/slice-translit.js +166 -0
- package/out/tools/slice-translit.js.map +1 -0
- package/out/tools/sub-venue/harvest.d.ts +100 -0
- package/out/tools/sub-venue/harvest.d.ts.map +1 -0
- package/out/tools/sub-venue/harvest.js +168 -0
- package/out/tools/sub-venue/harvest.js.map +1 -0
- package/out/tools/sub-venue/head-nouns.d.ts +50 -0
- package/out/tools/sub-venue/head-nouns.d.ts.map +1 -0
- package/out/tools/sub-venue/head-nouns.js +213 -0
- package/out/tools/sub-venue/head-nouns.js.map +1 -0
- package/out/tools/sub-venue/surfaces.d.ts +45 -0
- package/out/tools/sub-venue/surfaces.d.ts.map +1 -0
- package/out/tools/sub-venue/surfaces.js +77 -0
- package/out/tools/sub-venue/surfaces.js.map +1 -0
- package/out/tools/sub-venue/table.d.ts +229 -0
- package/out/tools/sub-venue/table.d.ts.map +1 -0
- package/out/tools/sub-venue/table.js +108 -0
- package/out/tools/sub-venue/table.js.map +1 -0
- package/out/tools/sub-venue/wikidata.d.ts +26 -0
- package/out/tools/sub-venue/wikidata.d.ts.map +1 -0
- package/out/tools/sub-venue/wikidata.js +62 -0
- package/out/tools/sub-venue/wikidata.js.map +1 -0
- package/out/tools/sub-venue-lexicon.d.ts +156 -0
- package/out/tools/sub-venue-lexicon.d.ts.map +1 -0
- package/out/tools/sub-venue-lexicon.js +283 -0
- package/out/tools/sub-venue-lexicon.js.map +1 -0
- package/out/tools/sub-venue-promotions.d.ts +94 -0
- package/out/tools/sub-venue-promotions.d.ts.map +1 -0
- package/out/tools/sub-venue-promotions.js +266 -0
- package/out/tools/sub-venue-promotions.js.map +1 -0
- package/out/types.d.ts +221 -0
- package/out/types.d.ts.map +1 -0
- package/out/types.js.map +1 -0
- package/out/utils/align.d.ts +93 -0
- package/out/utils/align.d.ts.map +1 -0
- package/out/utils/align.js +234 -0
- package/out/utils/align.js.map +1 -0
- package/out/utils/golden.d.ts +85 -0
- package/out/utils/golden.d.ts.map +1 -0
- package/out/utils/golden.js +127 -0
- package/out/utils/golden.js.map +1 -0
- package/out/utils/index.d.ts +14 -0
- package/out/utils/index.d.ts.map +1 -0
- package/out/utils/index.js +14 -0
- package/out/utils/index.js.map +1 -0
- package/out/utils/license.d.ts.map +1 -0
- package/out/utils/license.js.map +1 -0
- package/out/utils/parquet.d.ts +178 -0
- package/out/utils/parquet.d.ts.map +1 -0
- package/out/utils/parquet.js +371 -0
- package/out/utils/parquet.js.map +1 -0
- package/out/utils/slice-stats.d.ts +52 -0
- package/out/utils/slice-stats.d.ts.map +1 -0
- package/out/utils/slice-stats.js +58 -0
- package/out/utils/slice-stats.js.map +1 -0
- package/out/utils/split.d.ts +128 -0
- package/out/utils/split.d.ts.map +1 -0
- package/out/utils/split.js +204 -0
- package/out/utils/split.js.map +1 -0
- package/out/utils/tokenize.d.ts +64 -0
- package/out/utils/tokenize.d.ts.map +1 -0
- package/out/utils/tokenize.js +82 -0
- package/out/utils/tokenize.js.map +1 -0
- package/out/utils/wof-json.d.ts +113 -0
- package/out/utils/wof-json.d.ts.map +1 -0
- package/out/utils/wof-json.js +173 -0
- package/out/utils/wof-json.js.map +1 -0
- package/package.json +259 -194
- package/out/src/adapters/ban/adapter.d.ts.map +0 -1
- package/out/src/adapters/ban/adapter.js +0 -149
- package/out/src/adapters/ban/adapter.js.map +0 -1
- package/out/src/adapters/ban/street-decompose.d.ts.map +0 -1
- package/out/src/adapters/ban/street-decompose.js +0 -75
- package/out/src/adapters/ban/street-decompose.js.map +0 -1
- package/out/src/adapters/fcc-bdc/adapter.d.ts.map +0 -1
- package/out/src/adapters/fcc-bdc/adapter.js +0 -144
- package/out/src/adapters/fcc-bdc/adapter.js.map +0 -1
- package/out/src/adapters/geonames/adapter.d.ts.map +0 -1
- package/out/src/adapters/geonames/adapter.js +0 -176
- package/out/src/adapters/geonames/adapter.js.map +0 -1
- package/out/src/adapters/geonames-postal/adapter.d.ts +0 -41
- package/out/src/adapters/geonames-postal/adapter.d.ts.map +0 -1
- package/out/src/adapters/geonames-postal/adapter.js +0 -104
- package/out/src/adapters/geonames-postal/adapter.js.map +0 -1
- package/out/src/adapters/gnaf/adapter.d.ts +0 -49
- package/out/src/adapters/gnaf/adapter.d.ts.map +0 -1
- package/out/src/adapters/gnaf/adapter.js +0 -124
- package/out/src/adapters/gnaf/adapter.js.map +0 -1
- package/out/src/adapters/gnaf/assemble.d.ts +0 -65
- package/out/src/adapters/gnaf/assemble.d.ts.map +0 -1
- package/out/src/adapters/gnaf/assemble.js +0 -154
- package/out/src/adapters/gnaf/assemble.js.map +0 -1
- package/out/src/adapters/index.d.ts +0 -51
- package/out/src/adapters/index.d.ts.map +0 -1
- package/out/src/adapters/index.js +0 -95
- package/out/src/adapters/index.js.map +0 -1
- package/out/src/adapters/openaddresses/adapter.d.ts.map +0 -1
- package/out/src/adapters/openaddresses/adapter.js +0 -186
- package/out/src/adapters/openaddresses/adapter.js.map +0 -1
- package/out/src/adapters/overture/adapter.d.ts +0 -48
- package/out/src/adapters/overture/adapter.d.ts.map +0 -1
- package/out/src/adapters/overture/adapter.js +0 -125
- package/out/src/adapters/overture/adapter.js.map +0 -1
- package/out/src/adapters/state-hi-schools/adapter.d.ts +0 -47
- package/out/src/adapters/state-hi-schools/adapter.d.ts.map +0 -1
- package/out/src/adapters/state-hi-schools/adapter.js +0 -133
- package/out/src/adapters/state-hi-schools/adapter.js.map +0 -1
- package/out/src/adapters/state-ia-contractors/adapter.d.ts.map +0 -1
- package/out/src/adapters/state-ia-contractors/adapter.js.map +0 -1
- package/out/src/adapters/state-ny-notaries/adapter.d.ts.map +0 -1
- package/out/src/adapters/state-ny-notaries/adapter.js.map +0 -1
- package/out/src/adapters/state-tx-notaries/adapter.d.ts.map +0 -1
- package/out/src/adapters/state-tx-notaries/adapter.js.map +0 -1
- package/out/src/adapters/synth-po-box/adapter.d.ts.map +0 -1
- package/out/src/adapters/synth-po-box/adapter.js +0 -140
- package/out/src/adapters/synth-po-box/adapter.js.map +0 -1
- package/out/src/adapters/tiger/adapter.d.ts +0 -55
- package/out/src/adapters/tiger/adapter.d.ts.map +0 -1
- package/out/src/adapters/tiger/adapter.js +0 -201
- package/out/src/adapters/tiger/adapter.js.map +0 -1
- package/out/src/adapters/tiger/street-decompose.d.ts.map +0 -1
- package/out/src/adapters/tiger/street-decompose.js +0 -90
- package/out/src/adapters/tiger/street-decompose.js.map +0 -1
- package/out/src/adapters/usgov-hrsa-fqhc/adapter.d.ts +0 -47
- package/out/src/adapters/usgov-hrsa-fqhc/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-hrsa-fqhc/adapter.js +0 -139
- package/out/src/adapters/usgov-hrsa-fqhc/adapter.js.map +0 -1
- package/out/src/adapters/usgov-imls-pls/adapter.d.ts +0 -36
- package/out/src/adapters/usgov-imls-pls/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-imls-pls/adapter.js +0 -114
- package/out/src/adapters/usgov-imls-pls/adapter.js.map +0 -1
- package/out/src/adapters/usgov-irs-bmf/adapter.d.ts +0 -37
- package/out/src/adapters/usgov-irs-bmf/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-irs-bmf/adapter.js +0 -124
- package/out/src/adapters/usgov-irs-bmf/adapter.js.map +0 -1
- package/out/src/adapters/usgov-nad/adapter.d.ts +0 -48
- package/out/src/adapters/usgov-nad/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-nad/adapter.js +0 -252
- package/out/src/adapters/usgov-nad/adapter.js.map +0 -1
- package/out/src/adapters/usgov-nppes/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-nppes/adapter.js +0 -116
- package/out/src/adapters/usgov-nppes/adapter.js.map +0 -1
- package/out/src/adapters/usgov-samhsa-treatment-locator/adapter.d.ts.map +0 -1
- package/out/src/adapters/usgov-samhsa-treatment-locator/adapter.js +0 -156
- package/out/src/adapters/usgov-samhsa-treatment-locator/adapter.js.map +0 -1
- package/out/src/adapters/utils/index.d.ts +0 -150
- package/out/src/adapters/utils/index.d.ts.map +0 -1
- package/out/src/adapters/utils/index.js +0 -159
- package/out/src/adapters/utils/index.js.map +0 -1
- package/out/src/adapters/wof-admin-jp/adapter.d.ts.map +0 -1
- package/out/src/adapters/wof-admin-jp/adapter.js +0 -164
- package/out/src/adapters/wof-admin-jp/adapter.js.map +0 -1
- package/out/src/adapters/wof-admin-json/adapter.d.ts +0 -82
- package/out/src/adapters/wof-admin-json/adapter.d.ts.map +0 -1
- package/out/src/adapters/wof-admin-json/adapter.js +0 -224
- package/out/src/adapters/wof-admin-json/adapter.js.map +0 -1
- package/out/src/adapters/wof-json-rows.d.ts +0 -32
- package/out/src/adapters/wof-json-rows.d.ts.map +0 -1
- package/out/src/adapters/wof-json-rows.js +0 -45
- package/out/src/adapters/wof-json-rows.js.map +0 -1
- package/out/src/adapters/wof-postalcode-json/adapter.d.ts +0 -65
- package/out/src/adapters/wof-postalcode-json/adapter.d.ts.map +0 -1
- package/out/src/adapters/wof-postalcode-json/adapter.js +0 -160
- package/out/src/adapters/wof-postalcode-json/adapter.js.map +0 -1
- package/out/src/build.d.ts +0 -132
- package/out/src/build.d.ts.map +0 -1
- package/out/src/build.js +0 -251
- package/out/src/build.js.map +0 -1
- package/out/src/codex/us-fips-state.d.ts.map +0 -1
- package/out/src/codex/us-fips-state.js.map +0 -1
- package/out/src/index.d.ts +0 -14
- package/out/src/index.d.ts.map +0 -1
- package/out/src/index.js +0 -14
- package/out/src/index.js.map +0 -1
- package/out/src/name-prone-us-suffixes.d.ts +0 -12
- package/out/src/name-prone-us-suffixes.d.ts.map +0 -1
- package/out/src/name-prone-us-suffixes.js +0 -12
- package/out/src/name-prone-us-suffixes.js.map +0 -1
- package/out/src/parquet-wrapper/index.d.ts +0 -9
- package/out/src/parquet-wrapper/index.d.ts.map +0 -1
- package/out/src/parquet-wrapper/index.js +0 -9
- package/out/src/parquet-wrapper/index.js.map +0 -1
- package/out/src/parquet-wrapper/reader.d.ts +0 -34
- package/out/src/parquet-wrapper/reader.d.ts.map +0 -1
- package/out/src/parquet-wrapper/reader.js +0 -69
- package/out/src/parquet-wrapper/reader.js.map +0 -1
- package/out/src/parquet-wrapper/schema.d.ts +0 -49
- package/out/src/parquet-wrapper/schema.d.ts.map +0 -1
- package/out/src/parquet-wrapper/schema.js +0 -55
- package/out/src/parquet-wrapper/schema.js.map +0 -1
- package/out/src/parquet-wrapper/writer.d.ts +0 -46
- package/out/src/parquet-wrapper/writer.d.ts.map +0 -1
- package/out/src/parquet-wrapper/writer.js +0 -76
- package/out/src/parquet-wrapper/writer.js.map +0 -1
- package/out/src/runner.d.ts +0 -126
- package/out/src/runner.d.ts.map +0 -1
- package/out/src/runner.js +0 -189
- package/out/src/runner.js.map +0 -1
- package/out/src/shard-recipes/anchor-absorption.d.ts +0 -20
- package/out/src/shard-recipes/anchor-absorption.d.ts.map +0 -1
- package/out/src/shard-recipes/anchor-absorption.js +0 -82
- package/out/src/shard-recipes/anchor-absorption.js.map +0 -1
- package/out/src/shard-recipes/bare-country.d.ts +0 -25
- package/out/src/shard-recipes/bare-country.d.ts.map +0 -1
- package/out/src/shard-recipes/bare-country.js +0 -87
- package/out/src/shard-recipes/bare-country.js.map +0 -1
- package/out/src/shard-recipes/boundary-stress.d.ts +0 -20
- package/out/src/shard-recipes/boundary-stress.d.ts.map +0 -1
- package/out/src/shard-recipes/boundary-stress.js +0 -83
- package/out/src/shard-recipes/boundary-stress.js.map +0 -1
- package/out/src/shard-recipes/country-balanced.d.ts +0 -32
- package/out/src/shard-recipes/country-balanced.d.ts.map +0 -1
- package/out/src/shard-recipes/country-balanced.js +0 -406
- package/out/src/shard-recipes/country-balanced.js.map +0 -1
- package/out/src/shard-recipes/cz-pcfirst-preposition.d.ts +0 -22
- package/out/src/shard-recipes/cz-pcfirst-preposition.d.ts.map +0 -1
- package/out/src/shard-recipes/cz-pcfirst-preposition.js +0 -100
- package/out/src/shard-recipes/cz-pcfirst-preposition.js.map +0 -1
- package/out/src/shard-recipes/fr-admin-split.d.ts +0 -38
- package/out/src/shard-recipes/fr-admin-split.d.ts.map +0 -1
- package/out/src/shard-recipes/fr-admin-split.js +0 -188
- package/out/src/shard-recipes/fr-admin-split.js.map +0 -1
- package/out/src/shard-recipes/fr-bare-street.d.ts +0 -28
- package/out/src/shard-recipes/fr-bare-street.d.ts.map +0 -1
- package/out/src/shard-recipes/fr-bare-street.js +0 -143
- package/out/src/shard-recipes/fr-bare-street.js.map +0 -1
- package/out/src/shard-recipes/fr-fragment.d.ts +0 -77
- package/out/src/shard-recipes/fr-fragment.d.ts.map +0 -1
- package/out/src/shard-recipes/fr-fragment.js +0 -276
- package/out/src/shard-recipes/fr-fragment.js.map +0 -1
- package/out/src/shard-recipes/fr-lieudit.d.ts +0 -32
- package/out/src/shard-recipes/fr-lieudit.d.ts.map +0 -1
- package/out/src/shard-recipes/fr-lieudit.js +0 -226
- package/out/src/shard-recipes/fr-lieudit.js.map +0 -1
- package/out/src/shard-recipes/fr-order.d.ts +0 -38
- package/out/src/shard-recipes/fr-order.d.ts.map +0 -1
- package/out/src/shard-recipes/fr-order.js +0 -208
- package/out/src/shard-recipes/fr-order.js.map +0 -1
- package/out/src/shard-recipes/german.d.ts +0 -23
- package/out/src/shard-recipes/german.d.ts.map +0 -1
- package/out/src/shard-recipes/german.js +0 -132
- package/out/src/shard-recipes/german.js.map +0 -1
- package/out/src/shard-recipes/house-venue.d.ts +0 -18
- package/out/src/shard-recipes/house-venue.d.ts.map +0 -1
- package/out/src/shard-recipes/house-venue.js +0 -74
- package/out/src/shard-recipes/house-venue.js.map +0 -1
- package/out/src/shard-recipes/index.d.ts +0 -24
- package/out/src/shard-recipes/index.d.ts.map +0 -1
- package/out/src/shard-recipes/index.js +0 -89
- package/out/src/shard-recipes/index.js.map +0 -1
- package/out/src/shard-recipes/intersection.d.ts +0 -42
- package/out/src/shard-recipes/intersection.d.ts.map +0 -1
- package/out/src/shard-recipes/intersection.js +0 -460
- package/out/src/shard-recipes/intersection.js.map +0 -1
- package/out/src/shard-recipes/locale.d.ts +0 -135
- package/out/src/shard-recipes/locale.d.ts.map +0 -1
- package/out/src/shard-recipes/locale.js +0 -459
- package/out/src/shard-recipes/locale.js.map +0 -1
- package/out/src/shard-recipes/nl-postcode.d.ts +0 -26
- package/out/src/shard-recipes/nl-postcode.d.ts.map +0 -1
- package/out/src/shard-recipes/nl-postcode.js +0 -102
- package/out/src/shard-recipes/nl-postcode.js.map +0 -1
- package/out/src/shard-recipes/no-fragment.d.ts +0 -43
- package/out/src/shard-recipes/no-fragment.d.ts.map +0 -1
- package/out/src/shard-recipes/no-fragment.js +0 -188
- package/out/src/shard-recipes/no-fragment.js.map +0 -1
- package/out/src/shard-recipes/no-street-led.d.ts +0 -24
- package/out/src/shard-recipes/no-street-led.d.ts.map +0 -1
- package/out/src/shard-recipes/no-street-led.js +0 -121
- package/out/src/shard-recipes/no-street-led.js.map +0 -1
- package/out/src/shard-recipes/no-street.d.ts +0 -18
- package/out/src/shard-recipes/no-street.d.ts.map +0 -1
- package/out/src/shard-recipes/no-street.js +0 -72
- package/out/src/shard-recipes/no-street.js.map +0 -1
- package/out/src/shard-recipes/po-box-cedex.d.ts +0 -38
- package/out/src/shard-recipes/po-box-cedex.d.ts.map +0 -1
- package/out/src/shard-recipes/po-box-cedex.js +0 -726
- package/out/src/shard-recipes/po-box-cedex.js.map +0 -1
- package/out/src/shard-recipes/po-box.d.ts +0 -17
- package/out/src/shard-recipes/po-box.d.ts.map +0 -1
- package/out/src/shard-recipes/po-box.js +0 -101
- package/out/src/shard-recipes/po-box.js.map +0 -1
- package/out/src/shard-recipes/reviewed-postcode-tail.d.ts +0 -47
- package/out/src/shard-recipes/reviewed-postcode-tail.d.ts.map +0 -1
- package/out/src/shard-recipes/reviewed-postcode-tail.js +0 -125
- package/out/src/shard-recipes/reviewed-postcode-tail.js.map +0 -1
- package/out/src/shard-recipes/scaffold.d.ts +0 -246
- package/out/src/shard-recipes/scaffold.d.ts.map +0 -1
- package/out/src/shard-recipes/scaffold.js +0 -137
- package/out/src/shard-recipes/scaffold.js.map +0 -1
- package/out/src/shard-recipes/si-bare-village.d.ts +0 -32
- package/out/src/shard-recipes/si-bare-village.d.ts.map +0 -1
- package/out/src/shard-recipes/si-bare-village.js +0 -88
- package/out/src/shard-recipes/si-bare-village.js.map +0 -1
- package/out/src/shard-recipes/street-affix.d.ts +0 -89
- package/out/src/shard-recipes/street-affix.d.ts.map +0 -1
- package/out/src/shard-recipes/street-affix.js +0 -578
- package/out/src/shard-recipes/street-affix.js.map +0 -1
- package/out/src/shard-recipes/street-bare.d.ts +0 -25
- package/out/src/shard-recipes/street-bare.d.ts.map +0 -1
- package/out/src/shard-recipes/street-bare.js +0 -86
- package/out/src/shard-recipes/street-bare.js.map +0 -1
- package/out/src/shard-recipes/street.d.ts +0 -15
- package/out/src/shard-recipes/street.d.ts.map +0 -1
- package/out/src/shard-recipes/street.js +0 -69
- package/out/src/shard-recipes/street.js.map +0 -1
- package/out/src/shard-recipes/sub-venue-sources.d.ts +0 -228
- package/out/src/shard-recipes/sub-venue-sources.d.ts.map +0 -1
- package/out/src/shard-recipes/sub-venue-sources.js +0 -664
- package/out/src/shard-recipes/sub-venue-sources.js.map +0 -1
- package/out/src/shard-recipes/sub-venue.d.ts +0 -269
- package/out/src/shard-recipes/sub-venue.d.ts.map +0 -1
- package/out/src/shard-recipes/sub-venue.js +0 -710
- package/out/src/shard-recipes/sub-venue.js.map +0 -1
- package/out/src/shard-recipes/trailing-region.d.ts +0 -74
- package/out/src/shard-recipes/trailing-region.d.ts.map +0 -1
- package/out/src/shard-recipes/trailing-region.js +0 -157
- package/out/src/shard-recipes/trailing-region.js.map +0 -1
- package/out/src/shard-recipes/unit.d.ts +0 -29
- package/out/src/shard-recipes/unit.d.ts.map +0 -1
- package/out/src/shard-recipes/unit.js +0 -243
- package/out/src/shard-recipes/unit.js.map +0 -1
- package/out/src/synthesizers/anchor-absorption.d.ts +0 -63
- package/out/src/synthesizers/anchor-absorption.d.ts.map +0 -1
- package/out/src/synthesizers/anchor-absorption.js +0 -223
- package/out/src/synthesizers/anchor-absorption.js.map +0 -1
- package/out/src/synthesizers/boundary-stress.d.ts +0 -78
- package/out/src/synthesizers/boundary-stress.d.ts.map +0 -1
- package/out/src/synthesizers/boundary-stress.js +0 -455
- package/out/src/synthesizers/boundary-stress.js.map +0 -1
- package/out/src/synthesizers/german.d.ts +0 -107
- package/out/src/synthesizers/german.d.ts.map +0 -1
- package/out/src/synthesizers/german.js +0 -148
- package/out/src/synthesizers/german.js.map +0 -1
- package/out/src/synthesizers/house-venue.d.ts +0 -56
- package/out/src/synthesizers/house-venue.d.ts.map +0 -1
- package/out/src/synthesizers/house-venue.js +0 -267
- package/out/src/synthesizers/house-venue.js.map +0 -1
- package/out/src/synthesizers/intersection.d.ts +0 -52
- package/out/src/synthesizers/intersection.d.ts.map +0 -1
- package/out/src/synthesizers/intersection.js +0 -175
- package/out/src/synthesizers/intersection.js.map +0 -1
- package/out/src/synthesizers/no-street.d.ts +0 -72
- package/out/src/synthesizers/no-street.d.ts.map +0 -1
- package/out/src/synthesizers/no-street.js +0 -282
- package/out/src/synthesizers/no-street.js.map +0 -1
- package/out/src/synthesizers/po-box.d.ts +0 -104
- package/out/src/synthesizers/po-box.d.ts.map +0 -1
- package/out/src/synthesizers/po-box.js +0 -266
- package/out/src/synthesizers/po-box.js.map +0 -1
- package/out/src/synthesizers/street.d.ts +0 -54
- package/out/src/synthesizers/street.d.ts.map +0 -1
- package/out/src/synthesizers/street.js +0 -221
- package/out/src/synthesizers/street.js.map +0 -1
- package/out/src/synthesizers/utils.d.ts +0 -203
- package/out/src/synthesizers/utils.d.ts.map +0 -1
- package/out/src/synthesizers/utils.js +0 -708
- package/out/src/synthesizers/utils.js.map +0 -1
- package/out/src/tools/align-shard.d.ts +0 -8
- package/out/src/tools/align-shard.d.ts.map +0 -1
- package/out/src/tools/align-shard.js +0 -56
- package/out/src/tools/align-shard.js.map +0 -1
- package/out/src/tools/audit.d.ts +0 -28
- package/out/src/tools/audit.d.ts.map +0 -1
- package/out/src/tools/audit.js +0 -285
- package/out/src/tools/audit.js.map +0 -1
- package/out/src/tools/corpus-stats.d.ts +0 -44
- package/out/src/tools/corpus-stats.d.ts.map +0 -1
- package/out/src/tools/corpus-stats.js +0 -138
- package/out/src/tools/corpus-stats.js.map +0 -1
- package/out/src/tools/fetch/ban.d.ts +0 -31
- package/out/src/tools/fetch/ban.d.ts.map +0 -1
- package/out/src/tools/fetch/ban.js +0 -239
- package/out/src/tools/fetch/ban.js.map +0 -1
- package/out/src/tools/fetch/download.d.ts +0 -78
- package/out/src/tools/fetch/download.d.ts.map +0 -1
- package/out/src/tools/fetch/download.js +0 -116
- package/out/src/tools/fetch/download.js.map +0 -1
- package/out/src/tools/fetch/geonames-dump.d.ts +0 -87
- package/out/src/tools/fetch/geonames-dump.d.ts.map +0 -1
- package/out/src/tools/fetch/geonames-dump.js +0 -178
- package/out/src/tools/fetch/geonames-dump.js.map +0 -1
- package/out/src/tools/fetch/geonames-postal.d.ts +0 -76
- package/out/src/tools/fetch/geonames-postal.d.ts.map +0 -1
- package/out/src/tools/fetch/geonames-postal.js +0 -125
- package/out/src/tools/fetch/geonames-postal.js.map +0 -1
- package/out/src/tools/fetch/hrsa.d.ts +0 -18
- package/out/src/tools/fetch/hrsa.d.ts.map +0 -1
- package/out/src/tools/fetch/hrsa.js +0 -46
- package/out/src/tools/fetch/hrsa.js.map +0 -1
- package/out/src/tools/fetch/imls-pls.d.ts +0 -25
- package/out/src/tools/fetch/imls-pls.d.ts.map +0 -1
- package/out/src/tools/fetch/imls-pls.js +0 -108
- package/out/src/tools/fetch/imls-pls.js.map +0 -1
- package/out/src/tools/fetch/index.d.ts +0 -144
- package/out/src/tools/fetch/index.d.ts.map +0 -1
- package/out/src/tools/fetch/index.js +0 -143
- package/out/src/tools/fetch/index.js.map +0 -1
- package/out/src/tools/fetch/nad.d.ts +0 -70
- package/out/src/tools/fetch/nad.d.ts.map +0 -1
- package/out/src/tools/fetch/nad.js +0 -234
- package/out/src/tools/fetch/nad.js.map +0 -1
- package/out/src/tools/fetch/nppes.d.ts +0 -26
- package/out/src/tools/fetch/nppes.d.ts.map +0 -1
- package/out/src/tools/fetch/nppes.js +0 -126
- package/out/src/tools/fetch/nppes.js.map +0 -1
- package/out/src/tools/fetch/openaddresses.d.ts +0 -61
- package/out/src/tools/fetch/openaddresses.d.ts.map +0 -1
- package/out/src/tools/fetch/openaddresses.js +0 -317
- package/out/src/tools/fetch/openaddresses.js.map +0 -1
- package/out/src/tools/fetch/ourairports.d.ts +0 -45
- package/out/src/tools/fetch/ourairports.d.ts.map +0 -1
- package/out/src/tools/fetch/ourairports.js +0 -128
- package/out/src/tools/fetch/ourairports.js.map +0 -1
- package/out/src/tools/fetch/ppd.d.ts.map +0 -1
- package/out/src/tools/fetch/ppd.js +0 -100
- package/out/src/tools/fetch/ppd.js.map +0 -1
- package/out/src/tools/fetch/state-hi-schools.d.ts +0 -25
- package/out/src/tools/fetch/state-hi-schools.d.ts.map +0 -1
- package/out/src/tools/fetch/state-hi-schools.js +0 -162
- package/out/src/tools/fetch/state-hi-schools.js.map +0 -1
- package/out/src/tools/fetch/state-sources.d.ts +0 -26
- package/out/src/tools/fetch/state-sources.d.ts.map +0 -1
- package/out/src/tools/fetch/state-sources.js +0 -111
- package/out/src/tools/fetch/state-sources.js.map +0 -1
- package/out/src/tools/fetch/tiger-full.d.ts +0 -42
- package/out/src/tools/fetch/tiger-full.d.ts.map +0 -1
- package/out/src/tools/fetch/tiger-full.js +0 -248
- package/out/src/tools/fetch/tiger-full.js.map +0 -1
- package/out/src/tools/fetch/wikidata-subvenue.d.ts +0 -171
- package/out/src/tools/fetch/wikidata-subvenue.d.ts.map +0 -1
- package/out/src/tools/fetch/wikidata-subvenue.js +0 -275
- package/out/src/tools/fetch/wikidata-subvenue.js.map +0 -1
- package/out/src/tools/gb-title-case.d.ts.map +0 -1
- package/out/src/tools/gb-title-case.js.map +0 -1
- package/out/src/tools/golden-expand.d.ts +0 -90
- package/out/src/tools/golden-expand.d.ts.map +0 -1
- package/out/src/tools/golden-expand.js +0 -413
- package/out/src/tools/golden-expand.js.map +0 -1
- package/out/src/tools/golden-promote.d.ts.map +0 -1
- package/out/src/tools/golden-promote.js +0 -244
- package/out/src/tools/golden-promote.js.map +0 -1
- package/out/src/tools/golden-relabel-street.d.ts +0 -196
- package/out/src/tools/golden-relabel-street.d.ts.map +0 -1
- package/out/src/tools/golden-relabel-street.js +0 -440
- package/out/src/tools/golden-relabel-street.js.map +0 -1
- package/out/src/tools/index.d.ts +0 -28
- package/out/src/tools/index.d.ts.map +0 -1
- package/out/src/tools/index.js +0 -28
- package/out/src/tools/index.js.map +0 -1
- package/out/src/tools/ingest-csv.d.ts.map +0 -1
- package/out/src/tools/ingest-csv.js +0 -296
- package/out/src/tools/ingest-csv.js.map +0 -1
- package/out/src/tools/jsonl-to-parquet.d.ts +0 -61
- package/out/src/tools/jsonl-to-parquet.d.ts.map +0 -1
- package/out/src/tools/jsonl-to-parquet.js +0 -164
- package/out/src/tools/jsonl-to-parquet.js.map +0 -1
- package/out/src/tools/lint-shard-vocab.d.ts +0 -90
- package/out/src/tools/lint-shard-vocab.d.ts.map +0 -1
- package/out/src/tools/lint-shard-vocab.js +0 -281
- package/out/src/tools/lint-shard-vocab.js.map +0 -1
- package/out/src/tools/lint-shard.d.ts +0 -93
- package/out/src/tools/lint-shard.d.ts.map +0 -1
- package/out/src/tools/lint-shard.js +0 -354
- package/out/src/tools/lint-shard.js.map +0 -1
- package/out/src/tools/overlay-manifest.d.ts +0 -39
- package/out/src/tools/overlay-manifest.d.ts.map +0 -1
- package/out/src/tools/overlay-manifest.js +0 -98
- package/out/src/tools/overlay-manifest.js.map +0 -1
- package/out/src/tools/overture-subvenue.d.ts +0 -112
- package/out/src/tools/overture-subvenue.d.ts.map +0 -1
- package/out/src/tools/overture-subvenue.js +0 -144
- package/out/src/tools/overture-subvenue.js.map +0 -1
- package/out/src/tools/postcode-triples.d.ts +0 -175
- package/out/src/tools/postcode-triples.d.ts.map +0 -1
- package/out/src/tools/postcode-triples.js +0 -304
- package/out/src/tools/postcode-triples.js.map +0 -1
- package/out/src/tools/shard-kryptonite.d.ts +0 -32
- package/out/src/tools/shard-kryptonite.d.ts.map +0 -1
- package/out/src/tools/shard-kryptonite.js +0 -100
- package/out/src/tools/shard-kryptonite.js.map +0 -1
- package/out/src/tools/shard-translit.d.ts +0 -51
- package/out/src/tools/shard-translit.d.ts.map +0 -1
- package/out/src/tools/shard-translit.js +0 -175
- package/out/src/tools/shard-translit.js.map +0 -1
- package/out/src/tools/sub-venue/harvest.d.ts +0 -100
- package/out/src/tools/sub-venue/harvest.d.ts.map +0 -1
- package/out/src/tools/sub-venue/harvest.js +0 -168
- package/out/src/tools/sub-venue/harvest.js.map +0 -1
- package/out/src/tools/sub-venue/head-nouns.d.ts +0 -50
- package/out/src/tools/sub-venue/head-nouns.d.ts.map +0 -1
- package/out/src/tools/sub-venue/head-nouns.js +0 -210
- package/out/src/tools/sub-venue/head-nouns.js.map +0 -1
- package/out/src/tools/sub-venue/surfaces.d.ts +0 -45
- package/out/src/tools/sub-venue/surfaces.d.ts.map +0 -1
- package/out/src/tools/sub-venue/surfaces.js +0 -77
- package/out/src/tools/sub-venue/surfaces.js.map +0 -1
- package/out/src/tools/sub-venue/table.d.ts +0 -229
- package/out/src/tools/sub-venue/table.d.ts.map +0 -1
- package/out/src/tools/sub-venue/table.js +0 -108
- package/out/src/tools/sub-venue/table.js.map +0 -1
- package/out/src/tools/sub-venue/wikidata.d.ts +0 -26
- package/out/src/tools/sub-venue/wikidata.d.ts.map +0 -1
- package/out/src/tools/sub-venue/wikidata.js +0 -62
- package/out/src/tools/sub-venue/wikidata.js.map +0 -1
- package/out/src/tools/sub-venue-lexicon.d.ts +0 -156
- package/out/src/tools/sub-venue-lexicon.d.ts.map +0 -1
- package/out/src/tools/sub-venue-lexicon.js +0 -282
- package/out/src/tools/sub-venue-lexicon.js.map +0 -1
- package/out/src/tools/sub-venue-promotions.d.ts +0 -94
- package/out/src/tools/sub-venue-promotions.d.ts.map +0 -1
- package/out/src/tools/sub-venue-promotions.js +0 -266
- package/out/src/tools/sub-venue-promotions.js.map +0 -1
- package/out/src/types.d.ts +0 -221
- package/out/src/types.d.ts.map +0 -1
- package/out/src/types.js.map +0 -1
- package/out/src/utils/align.d.ts +0 -93
- package/out/src/utils/align.d.ts.map +0 -1
- package/out/src/utils/align.js +0 -234
- package/out/src/utils/align.js.map +0 -1
- package/out/src/utils/golden.d.ts +0 -69
- package/out/src/utils/golden.d.ts.map +0 -1
- package/out/src/utils/golden.js +0 -127
- package/out/src/utils/golden.js.map +0 -1
- package/out/src/utils/index.d.ts +0 -14
- package/out/src/utils/index.d.ts.map +0 -1
- package/out/src/utils/index.js +0 -14
- package/out/src/utils/index.js.map +0 -1
- package/out/src/utils/license.d.ts.map +0 -1
- package/out/src/utils/license.js.map +0 -1
- package/out/src/utils/parquet.d.ts +0 -162
- package/out/src/utils/parquet.d.ts.map +0 -1
- package/out/src/utils/parquet.js +0 -262
- package/out/src/utils/parquet.js.map +0 -1
- package/out/src/utils/split.d.ts +0 -127
- package/out/src/utils/split.d.ts.map +0 -1
- package/out/src/utils/split.js +0 -204
- package/out/src/utils/split.js.map +0 -1
- package/out/src/utils/tokenize.d.ts +0 -56
- package/out/src/utils/tokenize.d.ts.map +0 -1
- package/out/src/utils/tokenize.js +0 -48
- package/out/src/utils/tokenize.js.map +0 -1
- package/out/src/utils/wof-json.d.ts +0 -109
- package/out/src/utils/wof-json.d.ts.map +0 -1
- package/out/src/utils/wof-json.js +0 -173
- package/out/src/utils/wof-json.js.map +0 -1
- package/src/adapters/ban/adapter.ts +0 -191
- package/src/adapters/ban/street-decompose.ts +0 -96
- package/src/adapters/fcc-bdc/adapter.ts +0 -174
- package/src/adapters/geonames/adapter.ts +0 -200
- package/src/adapters/geonames-postal/adapter.ts +0 -119
- package/src/adapters/gnaf/adapter.ts +0 -149
- package/src/adapters/gnaf/assemble.ts +0 -226
- package/src/adapters/index.ts +0 -149
- package/src/adapters/openaddresses/adapter.ts +0 -251
- package/src/adapters/overture/adapter.ts +0 -162
- package/src/adapters/state-hi-schools/adapter.ts +0 -164
- package/src/adapters/synth-po-box/adapter.ts +0 -204
- package/src/adapters/tiger/adapter.ts +0 -253
- package/src/adapters/tiger/street-decompose.ts +0 -111
- package/src/adapters/usgov-hrsa-fqhc/adapter.ts +0 -183
- package/src/adapters/usgov-imls-pls/adapter.ts +0 -143
- package/src/adapters/usgov-irs-bmf/adapter.ts +0 -163
- package/src/adapters/usgov-nad/adapter.ts +0 -348
- package/src/adapters/usgov-nppes/adapter.ts +0 -159
- package/src/adapters/usgov-samhsa-treatment-locator/adapter.ts +0 -204
- package/src/adapters/utils/index.ts +0 -233
- package/src/adapters/wof-admin-jp/adapter.ts +0 -216
- package/src/adapters/wof-admin-json/adapter.ts +0 -256
- package/src/adapters/wof-json-rows.ts +0 -79
- package/src/adapters/wof-postalcode-json/adapter.ts +0 -183
- package/src/build.ts +0 -386
- package/src/index.ts +0 -14
- package/src/name-prone-us-suffixes.ts +0 -12
- package/src/parquet-wrapper/index.ts +0 -9
- package/src/parquet-wrapper/reader.ts +0 -97
- package/src/parquet-wrapper/schema.ts +0 -88
- package/src/parquet-wrapper/writer.ts +0 -110
- package/src/runner.ts +0 -317
- package/src/shard-recipes/anchor-absorption.ts +0 -103
- package/src/shard-recipes/bare-country.ts +0 -95
- package/src/shard-recipes/boundary-stress.ts +0 -101
- package/src/shard-recipes/country-balanced.ts +0 -524
- package/src/shard-recipes/cz-pcfirst-preposition.ts +0 -111
- package/src/shard-recipes/fr-admin-split.ts +0 -239
- package/src/shard-recipes/fr-bare-street.ts +0 -170
- package/src/shard-recipes/fr-fragment.ts +0 -332
- package/src/shard-recipes/fr-lieudit.ts +0 -302
- package/src/shard-recipes/fr-order.ts +0 -258
- package/src/shard-recipes/german.ts +0 -170
- package/src/shard-recipes/house-venue.ts +0 -88
- package/src/shard-recipes/index.ts +0 -95
- package/src/shard-recipes/intersection.ts +0 -630
- package/src/shard-recipes/locale.ts +0 -599
- package/src/shard-recipes/nl-postcode.ts +0 -115
- package/src/shard-recipes/no-fragment.ts +0 -219
- package/src/shard-recipes/no-street-led.ts +0 -143
- package/src/shard-recipes/no-street.ts +0 -86
- package/src/shard-recipes/po-box-cedex.ts +0 -920
- package/src/shard-recipes/po-box.ts +0 -119
- package/src/shard-recipes/reviewed-postcode-tail.ts +0 -189
- package/src/shard-recipes/scaffold.ts +0 -360
- package/src/shard-recipes/si-bare-village.ts +0 -97
- package/src/shard-recipes/street-affix.ts +0 -826
- package/src/shard-recipes/street-bare.ts +0 -101
- package/src/shard-recipes/street.ts +0 -82
- package/src/shard-recipes/sub-venue-sources.ts +0 -885
- package/src/shard-recipes/sub-venue.ts +0 -984
- package/src/shard-recipes/trailing-region.ts +0 -183
- package/src/shard-recipes/unit.ts +0 -321
- package/src/synthesizers/anchor-absorption.ts +0 -290
- package/src/synthesizers/boundary-stress.ts +0 -531
- package/src/synthesizers/german.ts +0 -251
- package/src/synthesizers/house-venue.ts +0 -330
- package/src/synthesizers/intersection.ts +0 -232
- package/src/synthesizers/no-street.ts +0 -349
- package/src/synthesizers/po-box.ts +0 -339
- package/src/synthesizers/street.ts +0 -274
- package/src/synthesizers/utils.ts +0 -903
- package/src/tools/align-shard.ts +0 -70
- package/src/tools/audit.ts +0 -399
- package/src/tools/corpus-stats.ts +0 -184
- package/src/tools/fetch/ban.ts +0 -290
- package/src/tools/fetch/download.ts +0 -176
- package/src/tools/fetch/geonames-dump.ts +0 -267
- package/src/tools/fetch/geonames-postal.ts +0 -180
- package/src/tools/fetch/hrsa.ts +0 -68
- package/src/tools/fetch/imls-pls.ts +0 -158
- package/src/tools/fetch/index.ts +0 -147
- package/src/tools/fetch/nad.ts +0 -342
- package/src/tools/fetch/nppes.ts +0 -174
- package/src/tools/fetch/openaddresses.ts +0 -404
- package/src/tools/fetch/ourairports.ts +0 -171
- package/src/tools/fetch/ppd.ts +0 -137
- package/src/tools/fetch/state-hi-schools.ts +0 -208
- package/src/tools/fetch/state-sources.ts +0 -155
- package/src/tools/fetch/tiger-full.ts +0 -354
- package/src/tools/fetch/wikidata-subvenue.ts +0 -386
- package/src/tools/golden-expand.ts +0 -584
- package/src/tools/golden-promote.ts +0 -353
- package/src/tools/golden-relabel-street.ts +0 -685
- package/src/tools/index.ts +0 -28
- package/src/tools/ingest-csv.ts +0 -422
- package/src/tools/jsonl-to-parquet.ts +0 -221
- package/src/tools/lint-shard-vocab.ts +0 -409
- package/src/tools/lint-shard.ts +0 -543
- package/src/tools/overlay-manifest.ts +0 -157
- package/src/tools/overture-subvenue.ts +0 -215
- package/src/tools/postcode-triples.ts +0 -375
- package/src/tools/shard-kryptonite.ts +0 -139
- package/src/tools/shard-translit.ts +0 -245
- package/src/tools/sub-venue/harvest.ts +0 -235
- package/src/tools/sub-venue/head-nouns.ts +0 -242
- package/src/tools/sub-venue/surfaces.ts +0 -95
- package/src/tools/sub-venue/table.ts +0 -281
- package/src/tools/sub-venue/wikidata.ts +0 -85
- package/src/tools/sub-venue-lexicon.ts +0 -431
- package/src/tools/sub-venue-promotions.ts +0 -330
- package/src/types.ts +0 -247
- package/src/utils/align.ts +0 -331
- package/src/utils/golden.ts +0 -176
- package/src/utils/index.ts +0 -14
- package/src/utils/parquet.ts +0 -391
- package/src/utils/split.ts +0 -279
- package/src/utils/tokenize.ts +0 -78
- package/src/utils/wof-json.ts +0 -244
- /package/{src → lib}/adapters/state-ia-contractors/adapter.ts +0 -0
- /package/{src → lib}/adapters/state-ny-notaries/adapter.ts +0 -0
- /package/{src → lib}/adapters/state-tx-notaries/adapter.ts +0 -0
- /package/{src → lib}/codex/us-fips-state.ts +0 -0
- /package/{src → lib}/tools/gb-title-case.ts +0 -0
- /package/{src → lib}/utils/license.ts +0 -0
- /package/out/{src/adapters → adapters}/ban/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/ban/street-decompose.d.ts +0 -0
- /package/out/{src/adapters → adapters}/fcc-bdc/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/geonames/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/openaddresses/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/state-ia-contractors/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/state-ia-contractors/adapter.js +0 -0
- /package/out/{src/adapters → adapters}/state-ny-notaries/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/state-ny-notaries/adapter.js +0 -0
- /package/out/{src/adapters → adapters}/state-tx-notaries/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/state-tx-notaries/adapter.js +0 -0
- /package/out/{src/adapters → adapters}/synth-po-box/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/tiger/street-decompose.d.ts +0 -0
- /package/out/{src/adapters → adapters}/usgov-nppes/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/usgov-samhsa-treatment-locator/adapter.d.ts +0 -0
- /package/out/{src/adapters → adapters}/wof-admin-jp/adapter.d.ts +0 -0
- /package/out/{src/codex → codex}/us-fips-state.d.ts +0 -0
- /package/out/{src/codex → codex}/us-fips-state.js +0 -0
- /package/out/{src/tools → tools}/fetch/ppd.d.ts +0 -0
- /package/out/{src/tools → tools}/gb-title-case.d.ts +0 -0
- /package/out/{src/tools → tools}/gb-title-case.js +0 -0
- /package/out/{src/tools → tools}/golden-promote.d.ts +0 -0
- /package/out/{src/tools → tools}/ingest-csv.d.ts +0 -0
- /package/out/{src/types.js → types.js} +0 -0
- /package/out/{src/utils → utils}/license.d.ts +0 -0
- /package/out/{src/utils → utils}/license.js +0 -0
|
@@ -1,188 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* `fr-admin-split` shard recipe — the FR admin-split coverage shard (night 2026-06-19,
|
|
7
|
-
* surpass-v1.5.0). Teaches the model to SPLIT the département out of the locality on
|
|
8
|
-
* bare/space/comma-delimited French place rows — the admin-deciding failure class the pre-GPU
|
|
9
|
-
* self-validation proved moves the resolved coordinate (collision communes −61%; see
|
|
10
|
-
* docs/articles/evals/experiments/2026-06-19-fr-admin-split-prevalidation.md). Ported from
|
|
11
|
-
* scripts/build-fr-admin-split-shard.mjs.
|
|
12
|
-
*
|
|
13
|
-
* Failure shapes (the model currently mis-handles all of these):
|
|
14
|
-
*
|
|
15
|
-
* - `Thauron, Creuse` → région dropped to null (the comma+full-name miss)
|
|
16
|
-
* - `Montredon, Lozère` → région = "ère" (the diacritic subword split, #727)
|
|
17
|
-
* - (AU analog) `CANBERRA ACT` → the space-delimited admin fuse
|
|
18
|
-
*
|
|
19
|
-
* The département is the essential admin unit for FR postal geography and maps to the `region`
|
|
20
|
-
* component tag in our schema. We derive it DETERMINISTICALLY from the real postcode via codex
|
|
21
|
-
* `departementForCodePostal` (first two digits = département) — salvage-first, no re-derived
|
|
22
|
-
* table.
|
|
23
|
-
*
|
|
24
|
-
* Data (`--communes`, opts.communes): REAL BAN (Base Adresse Nationale) commune+postcode+coord
|
|
25
|
-
* tuples, one per line, TAB-separated `commune <TAB> postcode <TAB> lon <TAB> lat`. Build the
|
|
26
|
-
* input TSV once from the BAN staging CSV (see the legacy script header). Anchor-ON by
|
|
27
|
-
* construction: rows carry a REAL postcode token in `raw` + a `postcode` component, so the
|
|
28
|
-
* training loader paints the anchor feature onto that span automatically. The trailing-postcode
|
|
29
|
-
* anchor REINFORCES the FR split (FR postcode is trailing, unlike German PLZ-leading — the v0.9.2
|
|
30
|
-
* scar is positional, not universal).
|
|
31
|
-
*/
|
|
32
|
-
import { departementForCodePostal } from "@mailwoman/codex/fr";
|
|
33
|
-
import { tempRootPath } from "@mailwoman/core/utils";
|
|
34
|
-
import { CSVSpliterator, Delimiters } from "spliterator";
|
|
35
|
-
import { stableSourceID } from "#adapters/utils";
|
|
36
|
-
import { alignRow } from "#utils";
|
|
37
|
-
import { makeMulberry32 } from "./scaffold.js";
|
|
38
|
-
/* oxlint-disable sister-software/no-unnamed-threshold -- the bare decimals below are weighted-sampler
|
|
39
|
-
cutoffs, not thresholds: `const r = random()` followed by a cascade of `r < 0.4` branches IS the
|
|
40
|
-
output distribution, and reading the cascade top-to-bottom is how you see it. Naming each cutoff
|
|
41
|
-
would hide the distribution behind a wall of identifiers. Genuine thresholds in these files are
|
|
42
|
-
extracted as named constants above. */
|
|
43
|
-
const DEFAULT_COMMUNES = tempRootPath("reg", "fr-communes.tsv");
|
|
44
|
-
const LICENSE = "BAN (Base Adresse Nationale) commune+postcode tuples, rendered admin-split — see ingest SOURCE";
|
|
45
|
-
/**
|
|
46
|
-
* Read the distinct commune TSV (commune, postcode, lon, lat); derive the département name.
|
|
47
|
-
*/
|
|
48
|
-
async function readCommunes(path) {
|
|
49
|
-
const rows = [];
|
|
50
|
-
// The TSV is headerless — every line is a commune tuple — so `header: false` keeps row 1 instead of
|
|
51
|
-
// spending it on column names. Source is repo-generated (LF); even under CRLF only the trailing `lat`
|
|
52
|
-
// column carries a CR, and it's consumed via `Number()` (whitespace-trimming), so it stays harmless.
|
|
53
|
-
for await (const [commune, postcode, lon, lat] of CSVSpliterator.fromAsync(path, {
|
|
54
|
-
columnDelimiter: Delimiters.Tab,
|
|
55
|
-
header: false,
|
|
56
|
-
})) {
|
|
57
|
-
if (!commune || !postcode)
|
|
58
|
-
continue;
|
|
59
|
-
const dep = departementForCodePostal(postcode);
|
|
60
|
-
if (!dep)
|
|
61
|
-
continue; // bad/unmappable postcode — skip (CEDEX, etc.)
|
|
62
|
-
// Substring invariant: a département whose name isn't a clean token (none are) or a commune
|
|
63
|
-
// containing the département name would confuse alignment — both are vanishingly rare here.
|
|
64
|
-
rows.push({ commune, postcode, departement: dep.name, lon, lat });
|
|
65
|
-
}
|
|
66
|
-
return rows;
|
|
67
|
-
}
|
|
68
|
-
/**
|
|
69
|
-
* Render one admin-split variant. The CORE teaching signal: the département, even as a full word after a comma or a
|
|
70
|
-
* space, is `region` — never folded into `locality`. Variants 1-3 are the failure class; 4-5 are canonical-FR
|
|
71
|
-
* preservation so the model doesn't over-fire region on every trailing token (and the bare commune still resolves).
|
|
72
|
-
*/
|
|
73
|
-
function render(random, c) {
|
|
74
|
-
const r = random();
|
|
75
|
-
const loc = random() < 0.1 ? c.commune.toUpperCase() : c.commune;
|
|
76
|
-
const dep = c.departement;
|
|
77
|
-
const pc = c.postcode;
|
|
78
|
-
let out;
|
|
79
|
-
if (r < 0.25) {
|
|
80
|
-
// 1. bare comma, NO postcode — the Thauron/#727 shape (anchor off)
|
|
81
|
-
out = { raw: `${loc}, ${dep}`, components: { locality: loc, region: dep }, order: "bare-comma" };
|
|
82
|
-
}
|
|
83
|
-
else if (r < 0.5) {
|
|
84
|
-
// 2. bare comma + postcode — anchor ON
|
|
85
|
-
out = {
|
|
86
|
-
raw: `${loc}, ${dep} ${pc}`,
|
|
87
|
-
components: { locality: loc, region: dep, postcode: pc },
|
|
88
|
-
order: "bare-comma-pc",
|
|
89
|
-
};
|
|
90
|
-
}
|
|
91
|
-
else if (r < 0.7) {
|
|
92
|
-
// 3. space-delimited admin (the AU `CANBERRA ACT` fuse applied to FR) — anchor ON
|
|
93
|
-
out = { raw: `${loc} ${dep} ${pc}`, components: { locality: loc, region: dep, postcode: pc }, order: "space-pc" };
|
|
94
|
-
}
|
|
95
|
-
else if (r < 0.85) {
|
|
96
|
-
// 4. canonical FR postcode-first (NO département) — preservation, anchor ON
|
|
97
|
-
out = { raw: `${pc} ${loc}`, components: { postcode: pc, locality: loc }, order: "canonical-pc-first" };
|
|
98
|
-
}
|
|
99
|
-
else {
|
|
100
|
-
// 5. commune + postcode (NO département) — preservation, anchor ON
|
|
101
|
-
out = { raw: `${loc} ${pc}`, components: { locality: loc, postcode: pc }, order: "commune-pc" };
|
|
102
|
-
}
|
|
103
|
-
// fr.country preservation (the v1.8.0 #728 finding): the v1.8.0 shard's bare rows carried NO country
|
|
104
|
-
// token, so the model under-emitted country on FR (fr.country −3.5pp). ~20% of rows now append an
|
|
105
|
-
// explicit "France" + a `country` component — the model relearns to emit country WHEN the token is
|
|
106
|
-
// present without over-firing it on the (still-majority) country-less rows. Substring invariant holds.
|
|
107
|
-
if (random() < 0.2) {
|
|
108
|
-
out = {
|
|
109
|
-
raw: `${out.raw}, France`,
|
|
110
|
-
components: { ...out.components, country: "France" },
|
|
111
|
-
order: `${out.order}+fr`,
|
|
112
|
-
};
|
|
113
|
-
}
|
|
114
|
-
return out;
|
|
115
|
-
}
|
|
116
|
-
/**
|
|
117
|
-
* Shard recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
|
|
118
|
-
* and `description` below for the surface form it generates.
|
|
119
|
-
*/
|
|
120
|
-
export const frAdminSplitRecipe = {
|
|
121
|
-
name: "fr-admin-split",
|
|
122
|
-
description: "FR admin-split rows: BAN communes → split département into `region` (+ canonical-FR preservation)",
|
|
123
|
-
mode: "generate",
|
|
124
|
-
options: [
|
|
125
|
-
{ flag: "--communes <tsv>", description: "BAN commune+postcode+coord TSV. Default /tmp/reg/fr-communes.tsv" },
|
|
126
|
-
],
|
|
127
|
-
async run(opts, write) {
|
|
128
|
-
// Legacy build-fr-admin-split-shard.mjs seeded `mulberry32(opts.seed)`.
|
|
129
|
-
const random = makeMulberry32(opts.seed);
|
|
130
|
-
const count = opts.count ?? 60_000;
|
|
131
|
-
const source = opts.sourceName ?? "synth-fr-admin-split";
|
|
132
|
-
const communesPath = opts.communes ?? DEFAULT_COMMUNES;
|
|
133
|
-
const pool = await readCommunes(communesPath);
|
|
134
|
-
console.error(` ${communesPath}: ${pool.length} communes with derived département`);
|
|
135
|
-
if (!pool.length) {
|
|
136
|
-
throw new Error("No communes — build the TSV from BAN first (see the recipe header).");
|
|
137
|
-
}
|
|
138
|
-
let emitted = 0;
|
|
139
|
-
let skipped = 0;
|
|
140
|
-
let guard = 0;
|
|
141
|
-
const orderCounts = {};
|
|
142
|
-
const N = pool.length;
|
|
143
|
-
while (emitted < count && guard++ < count * 12) {
|
|
144
|
-
const base = pool[Math.floor(random() * N)];
|
|
145
|
-
const { raw, components, order } = render(random, base);
|
|
146
|
-
// Alignment precondition: every component surface appears verbatim in raw.
|
|
147
|
-
const values = Object.values(components).filter((v) => Boolean(v));
|
|
148
|
-
if (!values.every((v) => raw.includes(v))) {
|
|
149
|
-
skipped++;
|
|
150
|
-
continue;
|
|
151
|
-
}
|
|
152
|
-
if (opts.golden) {
|
|
153
|
-
// Held-out eval slice for the centroid gate — carries the truth coordinate.
|
|
154
|
-
write(JSON.stringify({ raw, components, country: "FR", lat: Number(base.lat), lon: Number(base.lon) }) + "\n");
|
|
155
|
-
emitted++;
|
|
156
|
-
orderCounts[order] = (orderCounts[order] ?? 0) + 1;
|
|
157
|
-
continue;
|
|
158
|
-
}
|
|
159
|
-
const sourceID = stableSourceID(source, {
|
|
160
|
-
locality: components.locality,
|
|
161
|
-
region: components.region,
|
|
162
|
-
postcode: components.postcode,
|
|
163
|
-
});
|
|
164
|
-
const canonical = {
|
|
165
|
-
raw,
|
|
166
|
-
components,
|
|
167
|
-
country: "FR",
|
|
168
|
-
locale: "fr-FR",
|
|
169
|
-
source,
|
|
170
|
-
source_id: sourceID,
|
|
171
|
-
corpus_version: "0.5.0",
|
|
172
|
-
license: LICENSE,
|
|
173
|
-
};
|
|
174
|
-
const aligned = alignRow(canonical);
|
|
175
|
-
if (aligned.kind !== "labeled" || !aligned.row) {
|
|
176
|
-
skipped++;
|
|
177
|
-
continue;
|
|
178
|
-
}
|
|
179
|
-
write(JSON.stringify({ ...aligned.row, synth_method: "fr-admin-split", synth_order: order, synth_base_id: null }) +
|
|
180
|
-
"\n");
|
|
181
|
-
emitted++;
|
|
182
|
-
orderCounts[order] = (orderCounts[order] ?? 0) + 1;
|
|
183
|
-
}
|
|
184
|
-
console.error(` emitted=${emitted} skipped=${skipped} order-mix=${JSON.stringify(orderCounts)}`);
|
|
185
|
-
return { emitted, skipped };
|
|
186
|
-
},
|
|
187
|
-
};
|
|
188
|
-
//# sourceMappingURL=fr-admin-split.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"fr-admin-split.js","sourceRoot":"","sources":["../../../src/shard-recipes/fr-admin-split.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AAEH,OAAO,EAAE,wBAAwB,EAAE,MAAM,qBAAqB,CAAA;AAE9D,OAAO,EAAE,YAAY,EAAE,MAAM,uBAAuB,CAAA;AACpD,OAAO,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,aAAa,CAAA;AAExD,OAAO,EAAE,cAAc,EAAE,MAAM,iBAAiB,CAAA;AAEhD,OAAO,EAAE,QAAQ,EAAE,MAAM,QAAQ,CAAA;AAEjC,OAAO,EAAE,cAAc,EAAoB,MAAM,eAAe,CAAA;AAEhE;;;;yCAIyC;AAEzC,MAAM,gBAAgB,GAAG,YAAY,CAAC,KAAK,EAAE,iBAAiB,CAAC,CAAA;AAC/D,MAAM,OAAO,GAAG,gGAAgG,CAAA;AAsBhH;;GAEG;AACH,KAAK,UAAU,YAAY,CAAC,IAAY;IACvC,MAAM,IAAI,GAAiB,EAAE,CAAA;IAE7B,oGAAoG;IACpG,sGAAsG;IACtG,qGAAqG;IACrG,IAAI,KAAK,EAAE,MAAM,CAAC,OAAO,EAAE,QAAQ,EAAE,GAAG,EAAE,GAAG,CAAC,IAAI,cAAc,CAAC,SAAS,CAAC,IAAI,EAAE;QAChF,eAAe,EAAE,UAAU,CAAC,GAAG;QAC/B,MAAM,EAAE,KAAK;KACb,CAAC,EAAE,CAAC;QACJ,IAAI,CAAC,OAAO,IAAI,CAAC,QAAQ;YAAE,SAAQ;QACnC,MAAM,GAAG,GAAG,wBAAwB,CAAC,QAAQ,CAAC,CAAA;QAE9C,IAAI,CAAC,GAAG;YAAE,SAAQ,CAAC,+CAA+C;QAClE,4FAA4F;QAC5F,4FAA4F;QAC5F,IAAI,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,QAAQ,EAAE,WAAW,EAAE,GAAG,CAAC,IAAI,EAAE,GAAG,EAAE,GAAG,EAAE,CAAC,CAAA;IAClE,CAAC;IAED,OAAO,IAAI,CAAA;AACZ,CAAC;AAED;;;;GAIG;AACH,SAAS,MAAM,CAAC,MAAoB,EAAE,CAAa;IAClD,MAAM,CAAC,GAAG,MAAM,EAAE,CAAA;IAClB,MAAM,GAAG,GAAG,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,OAAO,CAAA;IAChE,MAAM,GAAG,GAAG,CAAC,CAAC,WAAW,CAAA;IACzB,MAAM,EAAE,GAAG,CAAC,CAAC,QAAQ,CAAA;IACrB,IAAI,GAAsB,CAAA;IAE1B,IAAI,CAAC,GAAG,IAAI,EAAE,CAAC;QACd,mEAAmE;QACnE,GAAG,GAAG,EAAE,GAAG,EAAE,GAAG,GAAG,KAAK,GAAG,EAAE,EAAE,UAAU,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,EAAE,KAAK,EAAE,YAAY,EAAE,CAAA;IACjG,CAAC;SAAM,IAAI,CAAC,GAAG,GAAG,EAAE,CAAC;QACpB,uCAAuC;QACvC,GAAG,GAAG;YACL,GAAG,EAAE,GAAG,GAAG,KAAK,GAAG,IAAI,EAAE,EAAE;YAC3B,UAAU,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,QAAQ,EAAE,EAAE,EAAE;YACxD,KAAK,EAAE,eAAe;SACtB,CAAA;IACF,CAAC;SAAM,IAAI,CAAC,GAAG,GAAG,EAAE,CAAC;QACpB,kFAAkF;QAClF,GAAG,GAAG,EAAE,GAAG,EAAE,GAAG,GAAG,IAAI,GAAG,IAAI,EAAE,EAAE,EAAE,UAAU,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,QAAQ,EAAE,EAAE,EAAE,EAAE,KAAK,EAAE,UAAU,EAAE,CAAA;IAClH,CAAC;SAAM,IAAI,CAAC,GAAG,IAAI,EAAE,CAAC;QACrB,4EAA4E;QAC5E,GAAG,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,IAAI,GAAG,EAAE,EAAE,UAAU,EAAE,EAAE,QAAQ,EAAE,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,EAAE,KAAK,EAAE,oBAAoB,EAAE,CAAA;IACxG,CAAC;SAAM,CAAC;QACP,mEAAmE;QACnE,GAAG,GAAG,EAAE,GAAG,EAAE,GAAG,GAAG,IAAI,EAAE,EAAE,EAAE,UAAU,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,QAAQ,EAAE,EAAE,EAAE,EAAE,KAAK,EAAE,YAAY,EAAE,CAAA;IAChG,CAAC;IAED,qGAAqG;IACrG,kGAAkG;IAClG,mGAAmG;IACnG,uGAAuG;IACvG,IAAI,MAAM,EAAE,GAAG,GAAG,EAAE,CAAC;QACpB,GAAG,GAAG;YACL,GAAG,EAAE,GAAG,GAAG,CAAC,GAAG,UAAU;YACzB,UAAU,EAAE,EAAE,GAAG,GAAG,CAAC,UAAU,EAAE,OAAO,EAAE,QAAQ,EAAE;YACpD,KAAK,EAAE,GAAG,GAAG,CAAC,KAAK,KAAK;SACxB,CAAA;IACF,CAAC;IAED,OAAO,GAAG,CAAA;AACX,CAAC;AAED;;;GAGG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAgB;IAC9C,IAAI,EAAE,gBAAgB;IACtB,WAAW,EAAE,mGAAmG;IAChH,IAAI,EAAE,UAAU;IAChB,OAAO,EAAE;QACR,EAAE,IAAI,EAAE,kBAAkB,EAAE,WAAW,EAAE,kEAAkE,EAAE;KAC7G;IACD,KAAK,CAAC,GAAG,CAAC,IAAI,EAAE,KAAK;QACpB,wEAAwE;QACxE,MAAM,MAAM,GAAG,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACxC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,IAAI,MAAM,CAAA;QAClC,MAAM,MAAM,GAAG,IAAI,CAAC,UAAU,IAAI,sBAAsB,CAAA;QACxD,MAAM,YAAY,GAAG,IAAI,CAAC,QAAQ,IAAI,gBAAgB,CAAA;QAEtD,MAAM,IAAI,GAAG,MAAM,YAAY,CAAC,YAAY,CAAC,CAAA;QAE7C,OAAO,CAAC,KAAK,CAAC,KAAK,YAAY,KAAK,IAAI,CAAC,MAAM,oCAAoC,CAAC,CAAA;QAEpF,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE,CAAC;YAClB,MAAM,IAAI,KAAK,CAAC,qEAAqE,CAAC,CAAA;QACvF,CAAC;QAED,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,KAAK,GAAG,CAAC,CAAA;QACb,MAAM,WAAW,GAA2B,EAAE,CAAA;QAC9C,MAAM,CAAC,GAAG,IAAI,CAAC,MAAM,CAAA;QAErB,OAAO,OAAO,GAAG,KAAK,IAAI,KAAK,EAAE,GAAG,KAAK,GAAG,EAAE,EAAE,CAAC;YAChD,MAAM,IAAI,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,CAAE,CAAA;YAC5C,MAAM,EAAE,GAAG,EAAE,UAAU,EAAE,KAAK,EAAE,GAAG,MAAM,CAAC,MAAM,EAAE,IAAI,CAAC,CAAA;YAEvD,2EAA2E;YAC3E,MAAM,MAAM,GAAG,MAAM,CAAC,MAAM,CAAC,UAAU,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,EAAe,EAAE,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAA;YAE/E,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC;gBAC3C,OAAO,EAAE,CAAA;gBAET,SAAQ;YACT,CAAC;YAED,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC;gBACjB,4EAA4E;gBAC5E,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,EAAE,GAAG,EAAE,UAAU,EAAE,OAAO,EAAE,IAAI,EAAE,GAAG,EAAE,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,EAAE,GAAG,EAAE,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,CAAA;gBAE9G,OAAO,EAAE,CAAA;gBACT,WAAW,CAAC,KAAK,CAAC,GAAG,CAAC,WAAW,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA;gBAElD,SAAQ;YACT,CAAC;YAED,MAAM,QAAQ,GAAG,cAAc,CAAC,MAAM,EAAE;gBACvC,QAAQ,EAAE,UAAU,CAAC,QAAQ;gBAC7B,MAAM,EAAE,UAAU,CAAC,MAAM;gBACzB,QAAQ,EAAE,UAAU,CAAC,QAAQ;aAC7B,CAAC,CAAA;YAEF,MAAM,SAAS,GAAiB;gBAC/B,GAAG;gBACH,UAAU;gBACV,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,OAAO;gBACf,MAAM;gBACN,SAAS,EAAE,QAAQ;gBACnB,cAAc,EAAE,OAAO;gBACvB,OAAO,EAAE,OAAO;aAChB,CAAA;YAED,MAAM,OAAO,GAAG,QAAQ,CAAC,SAAS,CAAC,CAAA;YAEnC,IAAI,OAAO,CAAC,IAAI,KAAK,SAAS,IAAI,CAAC,OAAO,CAAC,GAAG,EAAE,CAAC;gBAChD,OAAO,EAAE,CAAA;gBAET,SAAQ;YACT,CAAC;YAED,KAAK,CACJ,IAAI,CAAC,SAAS,CAAC,EAAE,GAAG,OAAO,CAAC,GAAG,EAAE,YAAY,EAAE,gBAAgB,EAAE,WAAW,EAAE,KAAK,EAAE,aAAa,EAAE,IAAI,EAAE,CAAC;gBAC1G,IAAI,CACL,CAAA;YAED,OAAO,EAAE,CAAA;YACT,WAAW,CAAC,KAAK,CAAC,GAAG,CAAC,WAAW,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA;QACnD,CAAC;QAED,OAAO,CAAC,KAAK,CAAC,aAAa,OAAO,YAAY,OAAO,cAAc,IAAI,CAAC,SAAS,CAAC,WAAW,CAAC,EAAE,CAAC,CAAA;QAEjG,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,CAAA;IAC5B,CAAC;CACD,CAAA"}
|
|
@@ -1,28 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* `fr-bare-street` shard recipe (#251) — the postcode-anchoring-imbalance lever. BAN (and every
|
|
7
|
-
* other comprehensive FR source) is postcode-COMPLETE, so the model learned the French
|
|
8
|
-
* street→locality boundary as "the token after the 5-digit postcode," never as "comma + city." Strip
|
|
9
|
-
* the postcode and it leaks the street's proper-noun tokens into the following locality ("Rue René
|
|
10
|
-
* Cassin, Paris" → street="Rue Ren", locality="Cassin"). This recipe mints the MISSING distribution:
|
|
11
|
-
* the BARE comma form, NO postcode, real `(street, number, city)` tuples from BAN (Licence Ouverte —
|
|
12
|
-
* permissive; the model stays clean of ODbL, unlike the opt-in OSM rooftop shards).
|
|
13
|
-
*
|
|
14
|
-
* Each tuple → `<n> <Rue/Avenue/…> <proper-noun name>, <City>` with the FR prefix split
|
|
15
|
-
* ({@link decomposeFrStreet}: "Rue" → street_prefix, the rest → street). Tuples whose street carries
|
|
16
|
-
* no recognized FR type word are skipped — the failing class is precisely the prefix-led street.
|
|
17
|
-
*
|
|
18
|
-
* ⚠ Convention loss-mask: this recipe TEACHES FR `street_prefix`. The conventions loss-mask forbids it
|
|
19
|
-
* for FR and will `-inf` these gold labels (the v1.6.0 ~7M-loss blow-up). Disable that mask for any
|
|
20
|
-
* run including this shard.
|
|
21
|
-
*/
|
|
22
|
-
import { type ShardRecipe } from "./scaffold.ts";
|
|
23
|
-
/**
|
|
24
|
-
* Shard recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
|
|
25
|
-
* and `description` below for the surface form it generates.
|
|
26
|
-
*/
|
|
27
|
-
export declare const frBareStreetRecipe: ShardRecipe;
|
|
28
|
-
//# sourceMappingURL=fr-bare-street.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"fr-bare-street.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/fr-bare-street.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;GAoBG;AAKH,OAAO,EAA6C,KAAK,WAAW,EAAiB,MAAM,eAAe,CAAA;AAgB1G;;;GAGG;AACH,eAAO,MAAM,kBAAkB,EAAE,WA4HhC,CAAA"}
|
|
@@ -1,143 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* `fr-bare-street` shard recipe (#251) — the postcode-anchoring-imbalance lever. BAN (and every
|
|
7
|
-
* other comprehensive FR source) is postcode-COMPLETE, so the model learned the French
|
|
8
|
-
* street→locality boundary as "the token after the 5-digit postcode," never as "comma + city." Strip
|
|
9
|
-
* the postcode and it leaks the street's proper-noun tokens into the following locality ("Rue René
|
|
10
|
-
* Cassin, Paris" → street="Rue Ren", locality="Cassin"). This recipe mints the MISSING distribution:
|
|
11
|
-
* the BARE comma form, NO postcode, real `(street, number, city)` tuples from BAN (Licence Ouverte —
|
|
12
|
-
* permissive; the model stays clean of ODbL, unlike the opt-in OSM rooftop shards).
|
|
13
|
-
*
|
|
14
|
-
* Each tuple → `<n> <Rue/Avenue/…> <proper-noun name>, <City>` with the FR prefix split
|
|
15
|
-
* ({@link decomposeFrStreet}: "Rue" → street_prefix, the rest → street). Tuples whose street carries
|
|
16
|
-
* no recognized FR type word are skipped — the failing class is precisely the prefix-led street.
|
|
17
|
-
*
|
|
18
|
-
* ⚠ Convention loss-mask: this recipe TEACHES FR `street_prefix`. The conventions loss-mask forbids it
|
|
19
|
-
* for FR and will `-inf` these gold labels (the v1.6.0 ~7M-loss blow-up). Disable that mask for any
|
|
20
|
-
* run including this shard.
|
|
21
|
-
*/
|
|
22
|
-
import { FR_VOIE_TYPES } from "@mailwoman/codex/fr";
|
|
23
|
-
import { decomposeFrStreet } from "../adapters/ban/street-decompose.js";
|
|
24
|
-
import { alignAndWrite, makeMulberry32, readTuples, shardSourceID } from "./scaffold.js";
|
|
25
|
-
/**
|
|
26
|
-
* Canonical voie type (lowercase, accent-kept) → its most common written abbreviation, from the codex table's first
|
|
27
|
-
* entry. Types with no attested abbreviation stay canonical in the abbreviated form.
|
|
28
|
-
*/
|
|
29
|
-
const FR_VOIE_ABBREV = Object.fromEntries(Object.entries(FR_VOIE_TYPES).flatMap(([canonical, abbrevs]) => (abbrevs[0] ? [[canonical, abbrevs[0]]] : [])));
|
|
30
|
-
/**
|
|
31
|
-
* The order-cycle slot for the BARE-STREET-ONLY form (`«voie» «name»`, no number, no locality) — the absence
|
|
32
|
-
* counterweight to the locality-terminated comma-free forms (see the cycle comment).
|
|
33
|
-
*/
|
|
34
|
-
const BARE_STREET_ONLY_FORM = 3;
|
|
35
|
-
/**
|
|
36
|
-
* Shard recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
|
|
37
|
-
* and `description` below for the surface form it generates.
|
|
38
|
-
*/
|
|
39
|
-
export const frBareStreetRecipe = {
|
|
40
|
-
name: "fr-bare-street",
|
|
41
|
-
description: "FR bare street+city, NO postcode (#251): comma / comma-free / abbreviated-voie surfaces over one label set",
|
|
42
|
-
mode: "tuples",
|
|
43
|
-
async run(opts, write) {
|
|
44
|
-
// Seeded for parity with the other recipes; unused beyond reproducibility (the tuples drive the content).
|
|
45
|
-
makeMulberry32(opts.seed);
|
|
46
|
-
let read = 0;
|
|
47
|
-
let emitted = 0;
|
|
48
|
-
let skipped = 0;
|
|
49
|
-
for await (const t of readTuples(opts.input)) {
|
|
50
|
-
read++;
|
|
51
|
-
const fullStreet = String(t.street ?? "").trim();
|
|
52
|
-
const number = String(t.number ?? "").trim();
|
|
53
|
-
const locality = String(t.locality ?? "").trim();
|
|
54
|
-
if (!fullStreet || !number || !locality) {
|
|
55
|
-
skipped++;
|
|
56
|
-
continue;
|
|
57
|
-
}
|
|
58
|
-
const { prefix, street } = decomposeFrStreet(fullStreet);
|
|
59
|
-
// A no-prefix nom_voie ("La Ville Mois") is not the prefix-led class the numbered forms
|
|
60
|
-
// exercise — but as a BARE surface it is exactly the non-voie-led counterweight the v4.5.1
|
|
61
|
-
// probe showed missing ('Savile Row'-shaped spans still fell to the trailing-locality
|
|
62
|
-
// prior; the voie-led bare form protected only voie-led spans). Alternate rows emit the
|
|
63
|
-
// whole span as a bare street; the rest skip as before.
|
|
64
|
-
if (!prefix || !street) {
|
|
65
|
-
if (read % 2 === 0 && fullStreet.split(" ").length >= 2) {
|
|
66
|
-
const bare = {
|
|
67
|
-
raw: fullStreet,
|
|
68
|
-
components: { street: fullStreet },
|
|
69
|
-
country: "FR",
|
|
70
|
-
locale: "fr-FR",
|
|
71
|
-
source: "synth-fr-bare-street",
|
|
72
|
-
source_id: shardSourceID("synth-fr-bare-street", {
|
|
73
|
-
street: fullStreet,
|
|
74
|
-
f: "bare-nonvoie",
|
|
75
|
-
v: String(read),
|
|
76
|
-
}),
|
|
77
|
-
corpus_version: "0.9.4",
|
|
78
|
-
license: "Synthetic — fr-bare-street; (street, number, city) from BAN (Base Adresse Nationale, Licence Ouverte)",
|
|
79
|
-
};
|
|
80
|
-
if (alignAndWrite(write, bare, "fr-bare-street")) {
|
|
81
|
-
emitted++;
|
|
82
|
-
}
|
|
83
|
-
else {
|
|
84
|
-
skipped++;
|
|
85
|
-
}
|
|
86
|
-
continue;
|
|
87
|
-
}
|
|
88
|
-
skipped++;
|
|
89
|
-
continue;
|
|
90
|
-
}
|
|
91
|
-
// Four surfaces over the same tuple, cycled deterministically. The comma form was the
|
|
92
|
-
// original lever; the COMMA-FREE form is the colloquial register users actually type
|
|
93
|
-
// ('12 rue de Rome Paris' — the street↔locality boundary with NO delimiter, the fr-fr
|
|
94
|
-
// panel's named loss); the ABBREVIATED form is the typeahead register the geocoder-tester
|
|
95
|
-
// FR slice attests at scale; and the BARE-STREET-ONLY form is the absence counterweight —
|
|
96
|
-
// without it, every delimiter-free surface in the mix ENDS in a locality, the model learns
|
|
97
|
-
// "trailing span = locality" as categorical, and bare street names across locales flip to
|
|
98
|
-
// locality wholesale (the v4.5.0 no-promote's measured erosion: 'Calle de Alcalá',
|
|
99
|
-
// 'Madison Square West', and COMER's fork all fell to that prior). Tags are identical
|
|
100
|
-
// where present; each component VALUE is the span as written (BIO alignment binds value
|
|
101
|
-
// to surface); the bare form carries NO number and NO locality because the surface has
|
|
102
|
-
// neither.
|
|
103
|
-
const form = read % 4;
|
|
104
|
-
const prefixSurface = form >= 2 ? (FR_VOIE_ABBREV[prefix.toLowerCase()] ?? prefix) : prefix;
|
|
105
|
-
const components = form === BARE_STREET_ONLY_FORM
|
|
106
|
-
? { street_prefix: prefix, street }
|
|
107
|
-
: { house_number: number, street_prefix: prefixSurface, street, locality };
|
|
108
|
-
// When the tuple carries a WOF-attested neighbourhood, the comma slot renders the
|
|
109
|
-
// THREE-slot middle surface — the dependent-locality counterweight (the v4.5.1 erosion's
|
|
110
|
-
// untouched half: the two-slot comma-free endings squeezed the middle tag out).
|
|
111
|
-
const hood = String(t.neighbourhood ?? "").trim();
|
|
112
|
-
if (form === 0 && hood) {
|
|
113
|
-
components.dependent_locality = hood;
|
|
114
|
-
}
|
|
115
|
-
const raw = form === 0
|
|
116
|
-
? hood
|
|
117
|
-
? `${number} ${prefix} ${street}, ${hood}, ${locality}`
|
|
118
|
-
: `${number} ${prefix} ${street}, ${locality}`
|
|
119
|
-
: form === BARE_STREET_ONLY_FORM
|
|
120
|
-
? `${prefix} ${street}`
|
|
121
|
-
: `${number} ${prefixSurface} ${street} ${locality}`;
|
|
122
|
-
const source_id = shardSourceID("synth-fr-bare-street", { ...components, f: String(form), v: String(read) });
|
|
123
|
-
const canonical = {
|
|
124
|
-
raw,
|
|
125
|
-
components,
|
|
126
|
-
country: "FR",
|
|
127
|
-
locale: "fr-FR",
|
|
128
|
-
source: "synth-fr-bare-street",
|
|
129
|
-
source_id,
|
|
130
|
-
corpus_version: "0.9.4",
|
|
131
|
-
license: "Synthetic — fr-bare-street; (street, number, city) from BAN (Base Adresse Nationale, Licence Ouverte)",
|
|
132
|
-
};
|
|
133
|
-
if (alignAndWrite(write, canonical, "fr-bare-street")) {
|
|
134
|
-
emitted++;
|
|
135
|
-
}
|
|
136
|
-
else {
|
|
137
|
-
skipped++;
|
|
138
|
-
}
|
|
139
|
-
}
|
|
140
|
-
return { read, emitted, skipped };
|
|
141
|
-
},
|
|
142
|
-
};
|
|
143
|
-
//# sourceMappingURL=fr-bare-street.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"fr-bare-street.js","sourceRoot":"","sources":["../../../src/shard-recipes/fr-bare-street.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;GAoBG;AAEH,OAAO,EAAE,aAAa,EAAE,MAAM,qBAAqB,CAAA;AAEnD,OAAO,EAAE,iBAAiB,EAAE,MAAM,qCAAqC,CAAA;AACvE,OAAO,EAAE,aAAa,EAAE,cAAc,EAAE,UAAU,EAAoB,aAAa,EAAE,MAAM,eAAe,CAAA;AAE1G;;;GAGG;AACH,MAAM,cAAc,GAA2B,MAAM,CAAC,WAAW,CAChE,MAAM,CAAC,OAAO,CAAC,aAAa,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,SAAS,EAAE,OAAO,CAAC,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,EAAE,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAC9G,CAAA;AAED;;;GAGG;AACH,MAAM,qBAAqB,GAAG,CAAC,CAAA;AAE/B;;;GAGG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAgB;IAC9C,IAAI,EAAE,gBAAgB;IACtB,WAAW,EACV,4GAA4G;IAC7G,IAAI,EAAE,QAAQ;IACd,KAAK,CAAC,GAAG,CAAC,IAAI,EAAE,KAAK;QACpB,0GAA0G;QAC1G,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACzB,IAAI,IAAI,GAAG,CAAC,CAAA;QACZ,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,OAAO,GAAG,CAAC,CAAA;QAEf,IAAI,KAAK,EAAE,MAAM,CAAC,IAAI,UAAU,CAAC,IAAI,CAAC,KAAM,CAAC,EAAE,CAAC;YAC/C,IAAI,EAAE,CAAA;YACN,MAAM,UAAU,GAAG,MAAM,CAAC,CAAC,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAChD,MAAM,MAAM,GAAG,MAAM,CAAC,CAAC,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAC5C,MAAM,QAAQ,GAAG,MAAM,CAAC,CAAC,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAEhD,IAAI,CAAC,UAAU,IAAI,CAAC,MAAM,IAAI,CAAC,QAAQ,EAAE,CAAC;gBACzC,OAAO,EAAE,CAAA;gBAET,SAAQ;YACT,CAAC;YAED,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,iBAAiB,CAAC,UAAU,CAAC,CAAA;YAExD,wFAAwF;YACxF,2FAA2F;YAC3F,sFAAsF;YACtF,wFAAwF;YACxF,wDAAwD;YACxD,IAAI,CAAC,MAAM,IAAI,CAAC,MAAM,EAAE,CAAC;gBACxB,IAAI,IAAI,GAAG,CAAC,KAAK,CAAC,IAAI,UAAU,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,MAAM,IAAI,CAAC,EAAE,CAAC;oBACzD,MAAM,IAAI,GAAG;wBACZ,GAAG,EAAE,UAAU;wBACf,UAAU,EAAE,EAAE,MAAM,EAAE,UAAU,EAA4B;wBAC5D,OAAO,EAAE,IAAI;wBACb,MAAM,EAAE,OAAO;wBACf,MAAM,EAAE,sBAAsB;wBAC9B,SAAS,EAAE,aAAa,CAAC,sBAAsB,EAAE;4BAChD,MAAM,EAAE,UAAU;4BAClB,CAAC,EAAE,cAAc;4BACjB,CAAC,EAAE,MAAM,CAAC,IAAI,CAAC;yBACf,CAAC;wBACF,cAAc,EAAE,OAAO;wBACvB,OAAO,EACN,uGAAuG;qBACxG,CAAA;oBAED,IAAI,aAAa,CAAC,KAAK,EAAE,IAAI,EAAE,gBAAgB,CAAC,EAAE,CAAC;wBAClD,OAAO,EAAE,CAAA;oBACV,CAAC;yBAAM,CAAC;wBACP,OAAO,EAAE,CAAA;oBACV,CAAC;oBAED,SAAQ;gBACT,CAAC;gBAED,OAAO,EAAE,CAAA;gBAET,SAAQ;YACT,CAAC;YAED,sFAAsF;YACtF,qFAAqF;YACrF,sFAAsF;YACtF,0FAA0F;YAC1F,0FAA0F;YAC1F,2FAA2F;YAC3F,0FAA0F;YAC1F,mFAAmF;YACnF,sFAAsF;YACtF,wFAAwF;YACxF,uFAAuF;YACvF,WAAW;YACX,MAAM,IAAI,GAAG,IAAI,GAAG,CAAC,CAAA;YACrB,MAAM,aAAa,GAAG,IAAI,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,cAAc,CAAC,MAAM,CAAC,WAAW,EAAE,CAAC,IAAI,MAAM,CAAC,CAAC,CAAC,CAAC,MAAM,CAAA;YAE3F,MAAM,UAAU,GACf,IAAI,KAAK,qBAAqB;gBAC7B,CAAC,CAAC,EAAE,aAAa,EAAE,MAAM,EAAE,MAAM,EAAE;gBACnC,CAAC,CAAC,EAAE,YAAY,EAAE,MAAM,EAAE,aAAa,EAAE,aAAa,EAAE,MAAM,EAAE,QAAQ,EAAE,CAAA;YAE5E,kFAAkF;YAClF,yFAAyF;YACzF,gFAAgF;YAChF,MAAM,IAAI,GAAG,MAAM,CAAC,CAAC,CAAC,aAAa,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAEjD,IAAI,IAAI,KAAK,CAAC,IAAI,IAAI,EAAE,CAAC;gBACxB,UAAU,CAAC,kBAAkB,GAAG,IAAI,CAAA;YACrC,CAAC;YAED,MAAM,GAAG,GACR,IAAI,KAAK,CAAC;gBACT,CAAC,CAAC,IAAI;oBACL,CAAC,CAAC,GAAG,MAAM,IAAI,MAAM,IAAI,MAAM,KAAK,IAAI,KAAK,QAAQ,EAAE;oBACvD,CAAC,CAAC,GAAG,MAAM,IAAI,MAAM,IAAI,MAAM,KAAK,QAAQ,EAAE;gBAC/C,CAAC,CAAC,IAAI,KAAK,qBAAqB;oBAC/B,CAAC,CAAC,GAAG,MAAM,IAAI,MAAM,EAAE;oBACvB,CAAC,CAAC,GAAG,MAAM,IAAI,aAAa,IAAI,MAAM,IAAI,QAAQ,EAAE,CAAA;YAEvD,MAAM,SAAS,GAAG,aAAa,CAAC,sBAAsB,EAAE,EAAE,GAAG,UAAU,EAAE,CAAC,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;YAE5G,MAAM,SAAS,GAAG;gBACjB,GAAG;gBACH,UAAU;gBACV,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,OAAO;gBACf,MAAM,EAAE,sBAAsB;gBAC9B,SAAS;gBACT,cAAc,EAAE,OAAO;gBACvB,OAAO,EACN,uGAAuG;aACxG,CAAA;YAED,IAAI,aAAa,CAAC,KAAK,EAAE,SAAS,EAAE,gBAAgB,CAAC,EAAE,CAAC;gBACvD,OAAO,EAAE,CAAA;YACV,CAAC;iBAAM,CAAC;gBACP,OAAO,EAAE,CAAA;YACV,CAAC;QACF,CAAC;QAED,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,CAAA;IAClC,CAAC;CACD,CAAA"}
|
|
@@ -1,77 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* `fr-fragment` shard recipe (#727 T2) — the HOUSE-NUMBER-LICENCE lever.
|
|
7
|
-
*
|
|
8
|
-
* The measured problem (T1c, `2026-07-16-t1c-fragment-board-verdict.md`): the shipped model scores
|
|
9
|
-
* **0.925** on `<n> Rue X` and **0.215** on `Rue X`. Same streets, same model; the only difference
|
|
10
|
-
* is a leading number. It has learned that a house number LICENSES a street reading at all — strip
|
|
11
|
-
* it and a designator-led phrase parses as a locality, designator included:
|
|
12
|
-
*
|
|
13
|
-
* ```
|
|
14
|
-
* "Rue Montmartre" -> locality="Rue Montmartre"
|
|
15
|
-
* "Allee Poque" -> locality="Allee Poque"
|
|
16
|
-
* ```
|
|
17
|
-
*
|
|
18
|
-
* `Rue` can only mean street in French. The model is not mislabelling an ambiguous toponym; it is
|
|
19
|
-
* mislabelling `Rue`.
|
|
20
|
-
*
|
|
21
|
-
* WHY THE EXISTING RECIPE DOESN'T COVER THIS. {@link frBareStreetRecipe} (#251) mints
|
|
22
|
-
* `<n> Rue <name>, <City>` — the bare COMMA form, no postcode. It targets postcode-anchoring
|
|
23
|
-
* imbalance, and every row it emits still carries a house number AND a locality. It cannot teach
|
|
24
|
-
* the class above, because it never shows the model a street standing alone.
|
|
25
|
-
*
|
|
26
|
-
* WHAT THIS MINTS. Five street forms — the first three carry NO house number and NO locality, which
|
|
27
|
-
* is the whole point — plus the counter-distribution:
|
|
28
|
-
*
|
|
29
|
-
* 1. `bare-street` "Rue Montmartre"
|
|
30
|
-
* 2. `street-particle` "Rue de la Paix"
|
|
31
|
-
* 3. `date-name` "Allee du 11 Novembre 1918"
|
|
32
|
-
* 4. `street-housenumber` "12 Rue Montmartre" — the anchor, so the licence isn't UNLEARNED
|
|
33
|
-
* 5. `alnum-housenumber` "12 bis Rue Montmartre"
|
|
34
|
-
* 6. `bare-locality` "Mery-sur-Oise" — NEGATIVE: a bare toponym that IS a locality
|
|
35
|
-
*
|
|
36
|
-
* The **admin/street homonym** class (`Rue de Rome`) is deliberately NOT a separate form. Homonym
|
|
37
|
-
* streets are already in the register and fall into 1–3 naturally; there is nothing different to
|
|
38
|
-
* teach about them, because the lesson is identical — the designator makes it a street. The
|
|
39
|
-
* fragment board scores them separately (they need measuring, not minting), and the recipe would
|
|
40
|
-
* need a second pass over the commune set to label them, buying a `synth_method` string and no
|
|
41
|
-
* training signal.
|
|
42
|
-
*
|
|
43
|
-
* FORM 7 IS NOT OPTIONAL. T1c's standing prediction: the board's `bare-locality` cell reads 0.980
|
|
44
|
-
* for the WRONG REASON — the model calls everything without a house number a locality, and on bare
|
|
45
|
-
* localities that is accidentally right. Teach bare streets alone and the model has every incentive
|
|
46
|
-
* to flip that default rather than learn the distinction, trading a 0.215 for a 0.980. The shard
|
|
47
|
-
* must show BOTH bare forms so the discriminating evidence is the designator, which is the only
|
|
48
|
-
* thing that actually distinguishes them. This is the same counter-distribution principle
|
|
49
|
-
* {@link noStreetRecipe} established after synth-street pushed the model into "decompose mode".
|
|
50
|
-
*
|
|
51
|
-
* SPLIT. `--exclude-surfaces` takes the fragment board's reserved surface list
|
|
52
|
-
* (`mailwoman/eval-harness/fixtures/ban-fragments-fr.surfaces.txt`). Every listed surface is
|
|
53
|
-
* skipped — source-disjoint by normalized street SURFACE, never by record row. Row-disjoint leaks
|
|
54
|
-
* the surface across the boundary and measures memorization of `Rue de Rivoli` while claiming
|
|
55
|
-
* generalization to unseen streets. The recipe REFUSES to run without the list rather than
|
|
56
|
-
* silently minting a contaminated shard.
|
|
57
|
-
*
|
|
58
|
-
* MIX. The shard is ~145K rows off a 120K-tuple draw; the intended corpus mix is **5–10%**, set at
|
|
59
|
-
* assembly time by shard weight rather than by row count. Keep the cap: a shard that fixes fragments
|
|
60
|
-
* by degrading full addresses has moved the failure, not fixed it — which is what
|
|
61
|
-
* `street-housenumber` / `alnum-housenumber` on the fragment board and the global parity floor are
|
|
62
|
-
* there to catch. `date-name` is ~0.5% of the shard because BAN only holds ~1,418 date-name streets
|
|
63
|
-
* after filtering and the tuple extractor already takes every one; if that class needs more, the
|
|
64
|
-
* lever is shard weight, not invented data.
|
|
65
|
-
*
|
|
66
|
-
* ⚠ Convention loss-mask: like {@link frBareStreetRecipe}, this recipe TEACHES FR `street_prefix`.
|
|
67
|
-
* The conventions loss-mask forbids it for FR and will `-inf` these gold labels (the v1.6.0 ~7M-loss
|
|
68
|
-
* blow-up). Disable that mask for any run including this shard.
|
|
69
|
-
*/
|
|
70
|
-
import { type ShardRecipe } from "./scaffold.ts";
|
|
71
|
-
export declare function frTitleCase(value: string): string;
|
|
72
|
-
/**
|
|
73
|
-
* Shard recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
|
|
74
|
-
* and `description` below for the surface form it generates.
|
|
75
|
-
*/
|
|
76
|
-
export declare const frFragmentRecipe: ShardRecipe;
|
|
77
|
-
//# sourceMappingURL=fr-fragment.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"fr-fragment.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/fr-fragment.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAoEG;AAKH,OAAO,EAA6C,KAAK,WAAW,EAAiB,MAAM,eAAe,CAAA;AAuD1G,wBAAgB,WAAW,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAajD;AAaD;;;GAGG;AACH,eAAO,MAAM,gBAAgB,EAAE,WA6K9B,CAAA"}
|