@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
|
@@ -1,148 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* German address synthesizer — multi-locale coverage (night-shift 2026-06-02, DE-1).
|
|
7
|
-
*
|
|
8
|
-
* The neural model is out-of-distribution on German: it truncates `Straußstraße`→`Strau` (exits at
|
|
9
|
-
* the ß-piece boundary), absorbs the house number into the street (`Hauptstraße 5` → one span),
|
|
10
|
-
* and mis-tags the native-order house number as a postcode (`Prenzlauer Allee 36, 10405 Berlin` →
|
|
11
|
-
* postcode `36`). The cause is ORDER: the model was trained US+FR (house-number-FIRST,
|
|
12
|
-
* postcode-AFTER-city), and never saw the German convention (house-number-AFTER-street,
|
|
13
|
-
* postcode-BEFORE-city). DE-0 confirmed the tokenizer round-trips German orthography cleanly, so
|
|
14
|
-
* this is a coverage gap, not a tokenizer ceiling.
|
|
15
|
-
*
|
|
16
|
-
* This generator produces the missing signal as a small targeted supplement shard
|
|
17
|
-
* (synthesis-as-supplement discipline: weight < 0.25, one-and-done). It does NOT synthesize
|
|
18
|
-
* German street names (German morphology is hard to fake) — it takes REAL German component tuples
|
|
19
|
-
* (from OpenAddresses Berlin/Saxony) and renders them in idiomatic German order via the OpenCage
|
|
20
|
-
* `DE` template (`formatAddress(..., "DE")` → `"Straußstraße 27, 12623 Berlin"`). The corpus
|
|
21
|
-
* aligner turns the row into BIO labels; every emitted component surface form occurs verbatim in
|
|
22
|
-
* `raw` so alignment lands.
|
|
23
|
-
*/
|
|
24
|
-
import { formatAddress } from "@mailwoman/formatter";
|
|
25
|
-
/**
|
|
26
|
-
* ISO-3166 alpha-2 → BCP-47 tag for the emitted rows (primary language per country).
|
|
27
|
-
*/
|
|
28
|
-
const LOCALE_TAG = {
|
|
29
|
-
DE: "de-DE",
|
|
30
|
-
ES: "es-ES",
|
|
31
|
-
IT: "it-IT",
|
|
32
|
-
NL: "nl-NL",
|
|
33
|
-
GB: "en-GB",
|
|
34
|
-
FR: "fr-FR",
|
|
35
|
-
US: "en-US",
|
|
36
|
-
NZ: "en-NZ",
|
|
37
|
-
};
|
|
38
|
-
/**
|
|
39
|
-
* Canonicalize a postcode to the form the country's template renders, so the stored component aligns verbatim against
|
|
40
|
-
* `raw`. NL is the case that needs it: OA stores `1011AB` but the OpenCage NL template emits the conventional spaced
|
|
41
|
-
* `1011 AB` (4 digits + space + 2 letters), which otherwise fails verbatim alignment and drops the row. Other countries
|
|
42
|
-
* pass through unchanged.
|
|
43
|
-
*/
|
|
44
|
-
function normalizePostcode(postcode, country) {
|
|
45
|
-
if (country === "NL") {
|
|
46
|
-
const m = /^(\d{4})\s*([A-Za-z]{2})$/.exec(postcode);
|
|
47
|
-
if (m)
|
|
48
|
-
return `${m[1]} ${m[2].toUpperCase()}`;
|
|
49
|
-
}
|
|
50
|
-
return postcode;
|
|
51
|
-
}
|
|
52
|
-
/**
|
|
53
|
-
* True when `value` appears verbatim AND as a standalone token (so BIO alignment lands cleanly).
|
|
54
|
-
*/
|
|
55
|
-
function tokenPresent(raw, value) {
|
|
56
|
-
if (!raw.includes(value))
|
|
57
|
-
return false;
|
|
58
|
-
// Reject substring-of-a-larger-number collisions (e.g. house "2" inside postcode "12623").
|
|
59
|
-
const i = raw.indexOf(value);
|
|
60
|
-
const before = raw[i - 1];
|
|
61
|
-
const after = raw[i + value.length];
|
|
62
|
-
const isDigit = (c) => c !== undefined && c >= "0" && c <= "9";
|
|
63
|
-
if (/^\d+$/.test(value) && (isDigit(before) || isDigit(after)))
|
|
64
|
-
return false;
|
|
65
|
-
return true;
|
|
66
|
-
}
|
|
67
|
-
/**
|
|
68
|
-
* Render one real tuple into an idiomatic, locale-ordered `{raw, components}` row via the OpenCage `country` template
|
|
69
|
-
* (DE → house-after-street + postcode-before-city; ES/IT the same; GB house-first; NL carries the `1012 LM` postcode),
|
|
70
|
-
* with light variation (drop house number / postcode some of the time). Returns `null` when the tuple is too thin or a
|
|
71
|
-
* component wouldn't align cleanly.
|
|
72
|
-
*
|
|
73
|
-
* Region handling is order-dependent: NATIVE order omits it (the native template absorbs the admin region into the
|
|
74
|
-
* postcode/city line, so it rarely renders verbatim and would break BIO alignment), while INTERNATIONAL order includes
|
|
75
|
-
* it in the tail ("City, Region Postcode" — the US/feed layout the eval uses; v0.9.3 / #327).
|
|
76
|
-
*
|
|
77
|
-
* Pass `opts.order: "international"` to render the same components house-first / postcode-after-city instead (see
|
|
78
|
-
* {@link LocaleSynthesisOpts.order}) — the layout international feeds impose on foreign addresses, and the one a
|
|
79
|
-
* native-order-trained model treats as a "collapse."
|
|
80
|
-
*/
|
|
81
|
-
export function synthesizeLocaleRow(base, country, opts = {}) {
|
|
82
|
-
const random = opts.random ?? Math.random;
|
|
83
|
-
const order = opts.order ?? "native";
|
|
84
|
-
if (!base.street || !base.locality)
|
|
85
|
-
return null;
|
|
86
|
-
const components = { street: base.street, locality: base.locality };
|
|
87
|
-
// Sub-locality (suburb / district) sits between street and locality and renders in BOTH orders — it's part
|
|
88
|
-
// of the address body, not the admin-region tail that native order drops. NZ needs it (suburb + city both on
|
|
89
|
-
// the envelope). The tokenPresent gate below drops the row if the template didn't surface it verbatim.
|
|
90
|
-
if (base.dependent_locality) {
|
|
91
|
-
components.dependent_locality = base.dependent_locality;
|
|
92
|
-
}
|
|
93
|
-
// ~80% keep the house number (the rest are street-only forms, also idiomatic).
|
|
94
|
-
if (base.house_number && random() < 0.8) {
|
|
95
|
-
components.house_number = base.house_number;
|
|
96
|
-
}
|
|
97
|
-
// ~85% keep the postcode (canonicalized to the country's rendered form — NL spaces it). The
|
|
98
|
-
// `postcodeShape: "as-source"` rewrite happens AFTER the render: the OpenCage NL template
|
|
99
|
-
// normalizes the postcode itself, so a glued input can't survive rendering directly.
|
|
100
|
-
if (base.postcode && random() < 0.85) {
|
|
101
|
-
components.postcode = normalizePostcode(base.postcode, country);
|
|
102
|
-
}
|
|
103
|
-
// International order carries the REGION in the tail ("City, Region Postcode") — the layout real
|
|
104
|
-
// US/feed renderings (and our OA eval) use. v0.9.2 rendered international order WITHOUT the region,
|
|
105
|
-
// so the model never learned to segment the tail and mangled it at eval (region absorbed into the
|
|
106
|
-
// locality / locality dropped); v0.9.3 closes that gap (#327). Native order still drops the region
|
|
107
|
-
// (the native template absorbs it into the city line, which would break verbatim alignment).
|
|
108
|
-
if (order === "international" && base.region) {
|
|
109
|
-
components.region = base.region;
|
|
110
|
-
}
|
|
111
|
-
// Native order uses the address's own country template; international order uses the US template —
|
|
112
|
-
// house-first, postcode-after-city, with a region slot for the tail. Neither branch consumes a
|
|
113
|
-
// `random()` draw for the template, so the RNG sequence existing callers/tests depend on is stable.
|
|
114
|
-
const renderCountry = order === "international" ? "US" : country;
|
|
115
|
-
let raw = formatAddress(components, renderCountry, { separator: ", " });
|
|
116
|
-
if (!raw)
|
|
117
|
-
return null;
|
|
118
|
-
// Native-order space-join (see {@link LocaleSynthesisOpts.nativeHouseJoin}): collapse the template's
|
|
119
|
-
// `<street>, <hn>` comma to a space — the OA/feed layout. A no-op for templates that already
|
|
120
|
-
// space-join (the substring isn't present), and skipped when the house number was dropped above.
|
|
121
|
-
if (order === "native" && opts.nativeHouseJoin === "space" && components.house_number) {
|
|
122
|
-
raw = raw.replace(`${components.street}, ${components.house_number}`, `${components.street} ${components.house_number}`);
|
|
123
|
-
}
|
|
124
|
-
// `postcodeShape: "as-source"` (see {@link LocaleSynthesisOpts.postcodeShape}): rewrite BOTH raw and
|
|
125
|
-
// the component back to the source's own surface (NL glued `1011AB`). Post-render because the
|
|
126
|
-
// OpenCage NL template normalizes the postcode to the spaced form no matter what it's given.
|
|
127
|
-
if (opts.postcodeShape === "as-source" && components.postcode && base.postcode) {
|
|
128
|
-
const sourceForm = base.postcode.trim();
|
|
129
|
-
if (sourceForm && sourceForm !== components.postcode && raw.includes(components.postcode)) {
|
|
130
|
-
raw = raw.replace(components.postcode, sourceForm);
|
|
131
|
-
components.postcode = sourceForm;
|
|
132
|
-
}
|
|
133
|
-
}
|
|
134
|
-
// Every component must align — drop the row if the template didn't surface one verbatim, or a
|
|
135
|
-
// numeric component collides with a neighbouring digit run.
|
|
136
|
-
for (const value of Object.values(components)) {
|
|
137
|
-
if (!value || !tokenPresent(raw, value))
|
|
138
|
-
return null;
|
|
139
|
-
}
|
|
140
|
-
return { raw, components, locale: LOCALE_TAG[country] ?? country.toLowerCase() };
|
|
141
|
-
}
|
|
142
|
-
/**
|
|
143
|
-
* German wrapper over {@link synthesizeLocaleRow}. Kept for the build-german-shard caller + tests.
|
|
144
|
-
*/
|
|
145
|
-
export function synthesizeGermanRow(base, opts = {}) {
|
|
146
|
-
return synthesizeLocaleRow(base, "DE", opts);
|
|
147
|
-
}
|
|
148
|
-
//# sourceMappingURL=german.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"german.js","sourceRoot":"","sources":["../../../src/synthesizers/german.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;GAsBG;AAEH,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AA+EpD;;GAEG;AACH,MAAM,UAAU,GAA2B;IAC1C,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;IACX,EAAE,EAAE,OAAO;CACX,CAAA;AAED;;;;;GAKG;AACH,SAAS,iBAAiB,CAAC,QAAgB,EAAE,OAAe;IAC3D,IAAI,OAAO,KAAK,IAAI,EAAE,CAAC;QACtB,MAAM,CAAC,GAAG,2BAA2B,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;QAEpD,IAAI,CAAC;YAAE,OAAO,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAE,CAAC,WAAW,EAAE,EAAE,CAAA;IAC/C,CAAC;IAED,OAAO,QAAQ,CAAA;AAChB,CAAC;AAED;;GAEG;AACH,SAAS,YAAY,CAAC,GAAW,EAAE,KAAa;IAC/C,IAAI,CAAC,GAAG,CAAC,QAAQ,CAAC,KAAK,CAAC;QAAE,OAAO,KAAK,CAAA;IACtC,2FAA2F;IAC3F,MAAM,CAAC,GAAG,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,CAAA;IAC5B,MAAM,MAAM,GAAG,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;IACzB,MAAM,KAAK,GAAG,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC,MAAM,CAAC,CAAA;IACnC,MAAM,OAAO,GAAG,CAAC,CAAqB,EAAE,EAAE,CAAC,CAAC,KAAK,SAAS,IAAI,CAAC,IAAI,GAAG,IAAI,CAAC,IAAI,GAAG,CAAA;IAElF,IAAI,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,IAAI,OAAO,CAAC,KAAK,CAAC,CAAC;QAAE,OAAO,KAAK,CAAA;IAE5E,OAAO,IAAI,CAAA;AACZ,CAAC;AAED;;;;;;;;;;;;;GAaG;AACH,MAAM,UAAU,mBAAmB,CAClC,IAAqB,EACrB,OAAe,EACf,OAA4B,EAAE;IAE9B,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,IAAI,QAAQ,CAAA;IAEpC,IAAI,CAAC,IAAI,CAAC,MAAM,IAAI,CAAC,IAAI,CAAC,QAAQ;QAAE,OAAO,IAAI,CAAA;IAE/C,MAAM,UAAU,GAA+B,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,QAAQ,EAAE,IAAI,CAAC,QAAQ,EAAE,CAAA;IAE/F,2GAA2G;IAC3G,6GAA6G;IAC7G,uGAAuG;IACvG,IAAI,IAAI,CAAC,kBAAkB,EAAE,CAAC;QAC7B,UAAU,CAAC,kBAAkB,GAAG,IAAI,CAAC,kBAAkB,CAAA;IACxD,CAAC;IAED,+EAA+E;IAC/E,IAAI,IAAI,CAAC,YAAY,IAAI,MAAM,EAAE,GAAG,GAAG,EAAE,CAAC;QACzC,UAAU,CAAC,YAAY,GAAG,IAAI,CAAC,YAAY,CAAA;IAC5C,CAAC;IAED,4FAA4F;IAC5F,0FAA0F;IAC1F,qFAAqF;IACrF,IAAI,IAAI,CAAC,QAAQ,IAAI,MAAM,EAAE,GAAG,IAAI,EAAE,CAAC;QACtC,UAAU,CAAC,QAAQ,GAAG,iBAAiB,CAAC,IAAI,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAA;IAChE,CAAC;IAED,iGAAiG;IACjG,oGAAoG;IACpG,kGAAkG;IAClG,mGAAmG;IACnG,6FAA6F;IAC7F,IAAI,KAAK,KAAK,eAAe,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC;QAC9C,UAAU,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAA;IAChC,CAAC;IAED,mGAAmG;IACnG,+FAA+F;IAC/F,oGAAoG;IACpG,MAAM,aAAa,GAAG,KAAK,KAAK,eAAe,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,OAAO,CAAA;IAChE,IAAI,GAAG,GAAG,aAAa,CAAC,UAAU,EAAE,aAAa,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAEvE,IAAI,CAAC,GAAG;QAAE,OAAO,IAAI,CAAA;IAErB,qGAAqG;IACrG,6FAA6F;IAC7F,iGAAiG;IACjG,IAAI,KAAK,KAAK,QAAQ,IAAI,IAAI,CAAC,eAAe,KAAK,OAAO,IAAI,UAAU,CAAC,YAAY,EAAE,CAAC;QACvF,GAAG,GAAG,GAAG,CAAC,OAAO,CAChB,GAAG,UAAU,CAAC,MAAM,KAAK,UAAU,CAAC,YAAY,EAAE,EAClD,GAAG,UAAU,CAAC,MAAM,IAAI,UAAU,CAAC,YAAY,EAAE,CACjD,CAAA;IACF,CAAC;IAED,qGAAqG;IACrG,8FAA8F;IAC9F,6FAA6F;IAC7F,IAAI,IAAI,CAAC,aAAa,KAAK,WAAW,IAAI,UAAU,CAAC,QAAQ,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAC;QAChF,MAAM,UAAU,GAAG,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAA;QAEvC,IAAI,UAAU,IAAI,UAAU,KAAK,UAAU,CAAC,QAAQ,IAAI,GAAG,CAAC,QAAQ,CAAC,UAAU,CAAC,QAAQ,CAAC,EAAE,CAAC;YAC3F,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,UAAU,CAAC,QAAQ,EAAE,UAAU,CAAC,CAAA;YAClD,UAAU,CAAC,QAAQ,GAAG,UAAU,CAAA;QACjC,CAAC;IACF,CAAC;IAED,8FAA8F;IAC9F,4DAA4D;IAC5D,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,MAAM,CAAC,UAAU,CAAC,EAAE,CAAC;QAC/C,IAAI,CAAC,KAAK,IAAI,CAAC,YAAY,CAAC,GAAG,EAAE,KAAK,CAAC;YAAE,OAAO,IAAI,CAAA;IACrD,CAAC;IAED,OAAO,EAAE,GAAG,EAAE,UAAU,EAAE,MAAM,EAAE,UAAU,CAAC,OAAO,CAAC,IAAI,OAAO,CAAC,WAAW,EAAE,EAAE,CAAA;AACjF,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,mBAAmB,CAClC,IAAqB,EACrB,OAA4B,EAAE;IAE9B,OAAO,mBAAmB,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,CAAC,CAAA;AAC7C,CAAC"}
|
|
@@ -1,56 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* House-number + venue + street co-occurrence synthesizer. The v0.6.3 corrective shard.
|
|
7
|
-
*
|
|
8
|
-
* The v0.6.2 step-20K diagnostic showed that adding synth-no-street counter-distribution regressed
|
|
9
|
-
* house_number recall by ~4-5pp. DeepSeek's turn-8 root-cause:
|
|
10
|
-
*
|
|
11
|
-
* 1. Direct: `5th Avenue Theatre`-style adversarial venues teach the model that tokens like "5th"
|
|
12
|
-
* belong to venues, not house_numbers. (Fixed in `no-street.ts` by removing
|
|
13
|
-
* digit+ordinal venue patterns.)
|
|
14
|
-
* 2. Distributional dilution: synth-no-street adds 122K rows where house_number is absent. The model's
|
|
15
|
-
* training distribution shifts toward "house_number is rare," and it under-emits the tag at
|
|
16
|
-
* inference.
|
|
17
|
-
*
|
|
18
|
-
* This synthesizer fixes #2 directly. Each emitted row has ALL of: house_number, street, venue,
|
|
19
|
-
* locality, region, postcode — a counter-example to "house_number is rare." Used as a companion
|
|
20
|
-
* shard to synth-no-street; the v0.6.3 config weights synth-no-street at 0.5 and
|
|
21
|
-
* synth-house-venue at 1.0 to recover the lost house_number signal.
|
|
22
|
-
*
|
|
23
|
-
* Real-world shape: business cards, mailing labels, store directories — `"123 Main St, Sunrise
|
|
24
|
-
* Bakery, Springfield, IL 62701"` is a perfectly ordinary address form.
|
|
25
|
-
*
|
|
26
|
-
* Venue pool: PLAIN_VENUES from `no-street.ts` (re-exported here). Adversarial venues
|
|
27
|
-
* are deliberately NOT used here — the point is to teach co-occurrence, not to re-introduce
|
|
28
|
-
* decompose-mode pressure.
|
|
29
|
-
*/
|
|
30
|
-
import type { CanonicalRow } from "#types";
|
|
31
|
-
export interface HouseVenueBaseTuple {
|
|
32
|
-
locality: string;
|
|
33
|
-
region: string;
|
|
34
|
-
postcode: string;
|
|
35
|
-
country: string;
|
|
36
|
-
street?: string;
|
|
37
|
-
houseNumber?: string;
|
|
38
|
-
}
|
|
39
|
-
export type HouseVenueTemplate = "venue-after-street" | "venue-before-street";
|
|
40
|
-
export interface HouseVenueSynthesisOpts {
|
|
41
|
-
random?: () => number;
|
|
42
|
-
forceTemplate?: HouseVenueTemplate;
|
|
43
|
-
}
|
|
44
|
-
export interface SynthesizedHouseVenueRow {
|
|
45
|
-
raw: string;
|
|
46
|
-
components: CanonicalRow["components"];
|
|
47
|
-
locale: string;
|
|
48
|
-
template: HouseVenueTemplate;
|
|
49
|
-
}
|
|
50
|
-
export declare function synthesizeHouseVenueRow(base: HouseVenueBaseTuple, opts?: HouseVenueSynthesisOpts): SynthesizedHouseVenueRow | null;
|
|
51
|
-
/**
|
|
52
|
-
* Contract: every synthesized row carries BOTH house_number AND venue (the co-occurrence signal that synth-no-street's
|
|
53
|
-
* distributional shift cost the model). Used by tests + downstream consumers.
|
|
54
|
-
*/
|
|
55
|
-
export declare function hasHouseNumberAndVenue(components: CanonicalRow["components"]): boolean;
|
|
56
|
-
//# sourceMappingURL=house-venue.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"house-venue.d.ts","sourceRoot":"","sources":["../../../src/synthesizers/house-venue.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAEH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,QAAQ,CAAA;AAE1C,MAAM,WAAW,mBAAmB;IACnC,QAAQ,EAAE,MAAM,CAAA;IAChB,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,OAAO,EAAE,MAAM,CAAA;IACf,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,WAAW,CAAC,EAAE,MAAM,CAAA;CACpB;AAED,MAAM,MAAM,kBAAkB,GAC3B,oBAAoB,GACpB,qBAAqB,CAAA;AAIxB,MAAM,WAAW,uBAAuB;IACvC,MAAM,CAAC,EAAE,MAAM,MAAM,CAAA;IACrB,aAAa,CAAC,EAAE,kBAAkB,CAAA;CAClC;AAED,MAAM,WAAW,wBAAwB;IACxC,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,CAAA;IACtC,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,kBAAkB,CAAA;CAC5B;AAkLD,wBAAgB,uBAAuB,CACtC,IAAI,EAAE,mBAAmB,EACzB,IAAI,GAAE,uBAA4B,GAChC,wBAAwB,GAAG,IAAI,CAiFjC;AAED;;;GAGG;AACH,wBAAgB,sBAAsB,CAAC,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,GAAG,OAAO,CAEtF"}
|
|
@@ -1,267 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* House-number + venue + street co-occurrence synthesizer. The v0.6.3 corrective shard.
|
|
7
|
-
*
|
|
8
|
-
* The v0.6.2 step-20K diagnostic showed that adding synth-no-street counter-distribution regressed
|
|
9
|
-
* house_number recall by ~4-5pp. DeepSeek's turn-8 root-cause:
|
|
10
|
-
*
|
|
11
|
-
* 1. Direct: `5th Avenue Theatre`-style adversarial venues teach the model that tokens like "5th"
|
|
12
|
-
* belong to venues, not house_numbers. (Fixed in `no-street.ts` by removing
|
|
13
|
-
* digit+ordinal venue patterns.)
|
|
14
|
-
* 2. Distributional dilution: synth-no-street adds 122K rows where house_number is absent. The model's
|
|
15
|
-
* training distribution shifts toward "house_number is rare," and it under-emits the tag at
|
|
16
|
-
* inference.
|
|
17
|
-
*
|
|
18
|
-
* This synthesizer fixes #2 directly. Each emitted row has ALL of: house_number, street, venue,
|
|
19
|
-
* locality, region, postcode — a counter-example to "house_number is rare." Used as a companion
|
|
20
|
-
* shard to synth-no-street; the v0.6.3 config weights synth-no-street at 0.5 and
|
|
21
|
-
* synth-house-venue at 1.0 to recover the lost house_number signal.
|
|
22
|
-
*
|
|
23
|
-
* Real-world shape: business cards, mailing labels, store directories — `"123 Main St, Sunrise
|
|
24
|
-
* Bakery, Springfield, IL 62701"` is a perfectly ordinary address form.
|
|
25
|
-
*
|
|
26
|
-
* Venue pool: PLAIN_VENUES from `no-street.ts` (re-exported here). Adversarial venues
|
|
27
|
-
* are deliberately NOT used here — the point is to teach co-occurrence, not to re-introduce
|
|
28
|
-
* decompose-mode pressure.
|
|
29
|
-
*/
|
|
30
|
-
//#region Venue pool
|
|
31
|
-
/**
|
|
32
|
-
* PLAIN venue names, carrying no street-typing tokens. This shard teaches house_number + venue coexistence, NOT
|
|
33
|
-
* decompose-mode pressure — adversarial venue names live in `no-street.ts`.
|
|
34
|
-
*/
|
|
35
|
-
const PLAIN_VENUES = [
|
|
36
|
-
"Bob's Pizza",
|
|
37
|
-
"Acme Corporation",
|
|
38
|
-
"Joe's Diner",
|
|
39
|
-
"Sunrise Bakery",
|
|
40
|
-
"Maple Leaf Cafe",
|
|
41
|
-
"Riverside Garden Center",
|
|
42
|
-
"Tech Solutions Inc",
|
|
43
|
-
// FR-flavored venue names (the run-2 contingency): the failing gauntlet fixtures carry
|
|
44
|
-
// international/English names at FR addresses, but native forms must appear too — the register
|
|
45
|
-
// mixes both in real Paris data.
|
|
46
|
-
"Café de la Poste",
|
|
47
|
-
"Boulangerie Saint-Michel",
|
|
48
|
-
"Le Petit Bistrot",
|
|
49
|
-
"Brasserie du Marché",
|
|
50
|
-
"Chez Marcel",
|
|
51
|
-
"La Belle Époque",
|
|
52
|
-
"Restaurant du Port",
|
|
53
|
-
"Pharmacie Centrale",
|
|
54
|
-
"Pacific Industries",
|
|
55
|
-
"Atlantic Holdings",
|
|
56
|
-
"Stellar Consulting",
|
|
57
|
-
"Greenfield Partners",
|
|
58
|
-
"Mountain View Studio",
|
|
59
|
-
"The Daily Grind",
|
|
60
|
-
"Sunset Bistro",
|
|
61
|
-
"Harvest Moon Florist",
|
|
62
|
-
"Iron Forge Brewing",
|
|
63
|
-
"Crescent City Bookstore",
|
|
64
|
-
"Lighthouse Insurance Group",
|
|
65
|
-
"Pinecrest Veterinary",
|
|
66
|
-
"Westwood Realty",
|
|
67
|
-
];
|
|
68
|
-
/**
|
|
69
|
-
* GB-flavored venue names (#1366): institutional forms (Club/Centre/House/Arms/Station), the "Ye" archaic register, and
|
|
70
|
-
* brand–dash–place compounds — INCLUDING directional-led names, because the target class is venues that open with
|
|
71
|
-
* compass words ("New North Health Centre", "Southfields Station") and the base model reads those as locality/street
|
|
72
|
-
* evidence. The six #1366 gauntlet fixtures' own venue names are deliberately ABSENT — the fixtures stay held-out.
|
|
73
|
-
*/
|
|
74
|
-
const GB_VENUES = [
|
|
75
|
-
"Ye Olde Cheshire Cheese",
|
|
76
|
-
"Ye Old Mitre",
|
|
77
|
-
"The Red Lion",
|
|
78
|
-
"The Crown & Anchor",
|
|
79
|
-
"The King's Arms",
|
|
80
|
-
"The White Hart",
|
|
81
|
-
"The Royal Oak",
|
|
82
|
-
"North End Road Market",
|
|
83
|
-
"Northfields Community Centre",
|
|
84
|
-
"South Bank Tavern",
|
|
85
|
-
"Southgate Dental Practice",
|
|
86
|
-
"West End Barbers",
|
|
87
|
-
"East Street Pharmacy",
|
|
88
|
-
"New Cross Learning Centre",
|
|
89
|
-
"Old Street Works",
|
|
90
|
-
"Upper Crust - Waterloo",
|
|
91
|
-
"Pret a Manger - Leadenhall",
|
|
92
|
-
"Greggs - Camden High Street",
|
|
93
|
-
"The Ivy - Chelsea Garden",
|
|
94
|
-
"Marks & Spencer Simply Food",
|
|
95
|
-
"Chapel Market Fishmongers",
|
|
96
|
-
"Victoria Coach Station",
|
|
97
|
-
"Highbury Fields Tennis Club",
|
|
98
|
-
"The Carpenters Arms",
|
|
99
|
-
"St Bride's Institute",
|
|
100
|
-
"Albion House",
|
|
101
|
-
"Imperial Dry Cleaners",
|
|
102
|
-
"Golden Dragon 金龍饭店",
|
|
103
|
-
];
|
|
104
|
-
//#endregion
|
|
105
|
-
//#region Fallback street pool
|
|
106
|
-
/**
|
|
107
|
-
* Stand-in streets for tuples that carried no `street` field. Plain names, no typing-token ambiguity.
|
|
108
|
-
*/
|
|
109
|
-
const FALLBACK_STREETS = [
|
|
110
|
-
"Main St",
|
|
111
|
-
"Oak Ave",
|
|
112
|
-
"Pine Rd",
|
|
113
|
-
"Elm Dr",
|
|
114
|
-
"Cedar Ln",
|
|
115
|
-
"Maple Blvd",
|
|
116
|
-
"Birch Ct",
|
|
117
|
-
"Walnut Pl",
|
|
118
|
-
"Cherry Way",
|
|
119
|
-
"Spruce St",
|
|
120
|
-
"Park Ave",
|
|
121
|
-
"Lake Dr",
|
|
122
|
-
"Hill Rd",
|
|
123
|
-
"River Ln",
|
|
124
|
-
"Forest Blvd",
|
|
125
|
-
];
|
|
126
|
-
//#endregion
|
|
127
|
-
//#region House-number generator
|
|
128
|
-
function randomHouseNumber(random) {
|
|
129
|
-
// Generate a plain numeric house number 1-9999. No fractions/ranges — those land in
|
|
130
|
-
// `data/eval/falsehoods/numbers.jsonl` as known edge cases, not training material.
|
|
131
|
-
const digits = Math.floor(random() * 4) + 1;
|
|
132
|
-
const max = Math.pow(10, digits);
|
|
133
|
-
const n = Math.floor(random() * max) + 1;
|
|
134
|
-
return String(n);
|
|
135
|
-
}
|
|
136
|
-
function pick(arr, random) {
|
|
137
|
-
return arr[Math.floor(random() * arr.length)];
|
|
138
|
-
}
|
|
139
|
-
function countryToLocale(country) {
|
|
140
|
-
switch (country) {
|
|
141
|
-
case "US":
|
|
142
|
-
return "en-US";
|
|
143
|
-
case "CA":
|
|
144
|
-
return "en-CA";
|
|
145
|
-
case "GB":
|
|
146
|
-
return "en-GB";
|
|
147
|
-
case "AU":
|
|
148
|
-
return "en-AU";
|
|
149
|
-
case "FR":
|
|
150
|
-
return "fr-FR";
|
|
151
|
-
case "DE":
|
|
152
|
-
return "de-DE";
|
|
153
|
-
default:
|
|
154
|
-
return "en-US";
|
|
155
|
-
}
|
|
156
|
-
}
|
|
157
|
-
//#endregion
|
|
158
|
-
//#region Synthesis
|
|
159
|
-
/**
|
|
160
|
-
* Fraction of GB rows drawing from {@link GB_VENUES} instead of the shared pool. 0.7 mirrors the register mix in real
|
|
161
|
-
* GB listings data (institutional names dominate, international/generic names still appear) — pre-registered in the
|
|
162
|
-
* #1366 memo.
|
|
163
|
-
*/
|
|
164
|
-
const GB_VENUE_POOL_RATE = 0.7;
|
|
165
|
-
/**
|
|
166
|
-
* Fraction of GB rows whose house number widens into a range ("287-293"). Real GB venue addresses frequently span
|
|
167
|
-
* buildings; 0.15 keeps ranges a minority register — pre-registered in the #1366 memo.
|
|
168
|
-
*/
|
|
169
|
-
const GB_RANGE_NUMBER_RATE = 0.15;
|
|
170
|
-
/**
|
|
171
|
-
* Fraction of rows (EVERY template order) rendered with a trailing country surface, tagged `country`. The 2026-08-01
|
|
172
|
-
* operator probe set proved the mechanism: the FR control row ("…, 75004 Paris, France") fails on a model trained only
|
|
173
|
-
* on country-less venue rows while its country-less twin passes — a trailing country makes the whole template OOD
|
|
174
|
-
* (Addendum 3 of the #1366 pre-registration). 0.3 keeps the country-less register dominant.
|
|
175
|
-
*/
|
|
176
|
-
const COUNTRY_APPEND_RATE = 0.3;
|
|
177
|
-
/**
|
|
178
|
-
* Trailing country surfaces by tuple country — the register mixes formal and short forms where both are common.
|
|
179
|
-
*/
|
|
180
|
-
const COUNTRY_SURFACES = {
|
|
181
|
-
US: ["United States", "USA"],
|
|
182
|
-
GB: ["United Kingdom", "United Kingdom", "United Kingdom", "UK"],
|
|
183
|
-
FR: ["France"],
|
|
184
|
-
CA: ["Canada"],
|
|
185
|
-
AU: ["Australia"],
|
|
186
|
-
DE: ["Germany", "Deutschland"],
|
|
187
|
-
};
|
|
188
|
-
export function synthesizeHouseVenueRow(base, opts = {}) {
|
|
189
|
-
const random = opts.random ?? Math.random;
|
|
190
|
-
const locale = countryToLocale(base.country);
|
|
191
|
-
const template = opts.forceTemplate ?? (random() < 0.5 ? "venue-after-street" : "venue-before-street");
|
|
192
|
-
// FR renders postcode-before-locality with NO region ("MR & MRS CRAB, 20 Rue de la Huchette,
|
|
193
|
-
// 75005 Paris" — the v4.0.0 gauntlet's venue-led failure family, the run-2 contingency's exact
|
|
194
|
-
// target shape). GB (#1366) renders locality-then-postcode with NO region and NO comma between
|
|
195
|
-
// them ("Ye Three Lords, 27 Minories, London EC3N 1DE" — the third tail the shard must teach).
|
|
196
|
-
// VE (#1821) renders locality-then-postcode and KEEPS the region after it. Every other country
|
|
197
|
-
// keeps the original US-order tail, which is itself the `after_region` surface — `Springfield, IL
|
|
198
|
-
// 02101` and `Bengaluru, Karnataka 560038` are the same shape.
|
|
199
|
-
const frOrder = base.country === "FR";
|
|
200
|
-
const gbOrder = base.country === "GB";
|
|
201
|
-
// VE writes the code on the LOCALITY segment with the region still after it — `…, Barcelona 6001, Anzoátegui,
|
|
202
|
-
// Venezuela`, the four `ve_city_postcode_trailing_state` board rows. That is neither GB's form (which drops the
|
|
203
|
-
// region) nor the default (which puts the region before the code), so it needs its own branch.
|
|
204
|
-
//
|
|
205
|
-
// It belongs HERE rather than in a standalone admin shard, and that is measured: three trailing-region shards
|
|
206
|
-
// carrying only admin segments all graded DO-NOT-SHIP, and the way they failed was by damaging the classes they did
|
|
207
|
-
// not contain — v4.8.0 turned `Ye Three Lords, 27 Minories, London EC3N 1DE` into `locality: "Ye Three Lords"`,
|
|
208
|
-
// losing the venue and the street. Every row this synthesizer emits carries a venue, a street AND a house number,
|
|
209
|
-
// so the surface is taught with the alternatives present rather than against them.
|
|
210
|
-
const veOrder = base.country === "VE";
|
|
211
|
-
// GB rows draw from the GB pool 70% of the time (institutional/archaic/brand-dash-place forms,
|
|
212
|
-
// incl. directional-led names — the #1366 target class) and the shared pool otherwise; real GB
|
|
213
|
-
// registers mix both. Other locales keep the shared pool (which already carries the FR flavor).
|
|
214
|
-
const venue = gbOrder && random() < GB_VENUE_POOL_RATE ? pick(GB_VENUES, random) : pick(PLAIN_VENUES, random);
|
|
215
|
-
const street = base.street ?? pick(FALLBACK_STREETS, random);
|
|
216
|
-
let houseNumber = base.houseNumber ?? randomHouseNumber(random);
|
|
217
|
-
// GB range numbers ("287-293 New N Rd"): real GB venue addresses frequently span buildings.
|
|
218
|
-
// 15% of GB rows widen the number into a range (same parity, small span — the register's real
|
|
219
|
-
// shape). Pre-registered in the #1366 memo; the base pool's no-ranges stance stays for other
|
|
220
|
-
// locales.
|
|
221
|
-
if (gbOrder && random() < GB_RANGE_NUMBER_RATE && /^\d+$/.test(houseNumber)) {
|
|
222
|
-
const start = Number.parseInt(houseNumber, 10);
|
|
223
|
-
const span = (1 + Math.floor(random() * 4)) * 2;
|
|
224
|
-
houseNumber = `${start}-${start + span}`;
|
|
225
|
-
}
|
|
226
|
-
const components = {
|
|
227
|
-
house_number: houseNumber,
|
|
228
|
-
street,
|
|
229
|
-
venue,
|
|
230
|
-
locality: base.locality,
|
|
231
|
-
...(frOrder || gbOrder ? {} : { region: base.region }),
|
|
232
|
-
postcode: base.postcode,
|
|
233
|
-
};
|
|
234
|
-
let tail = frOrder
|
|
235
|
-
? `${base.postcode} ${base.locality}`
|
|
236
|
-
: gbOrder
|
|
237
|
-
? `${base.locality} ${base.postcode}`
|
|
238
|
-
: veOrder
|
|
239
|
-
? `${base.locality} ${base.postcode}, ${base.region}`
|
|
240
|
-
: `${base.locality}, ${base.region} ${base.postcode}`;
|
|
241
|
-
// Trailing country surface (Addendum 3): appended AFTER the tail in every order, tagged.
|
|
242
|
-
const countrySurfaces = COUNTRY_SURFACES[base.country];
|
|
243
|
-
if (countrySurfaces && random() < COUNTRY_APPEND_RATE) {
|
|
244
|
-
const countrySurface = pick(countrySurfaces, random);
|
|
245
|
-
components.country = countrySurface;
|
|
246
|
-
tail = `${tail}, ${countrySurface}`;
|
|
247
|
-
}
|
|
248
|
-
let raw;
|
|
249
|
-
switch (template) {
|
|
250
|
-
case "venue-after-street":
|
|
251
|
-
raw = `${houseNumber} ${street}, ${venue}, ${tail}`;
|
|
252
|
-
break;
|
|
253
|
-
case "venue-before-street":
|
|
254
|
-
raw = `${venue}, ${houseNumber} ${street}, ${tail}`;
|
|
255
|
-
break;
|
|
256
|
-
}
|
|
257
|
-
return { raw, components, locale, template };
|
|
258
|
-
}
|
|
259
|
-
/**
|
|
260
|
-
* Contract: every synthesized row carries BOTH house_number AND venue (the co-occurrence signal that synth-no-street's
|
|
261
|
-
* distributional shift cost the model). Used by tests + downstream consumers.
|
|
262
|
-
*/
|
|
263
|
-
export function hasHouseNumberAndVenue(components) {
|
|
264
|
-
return components.house_number !== undefined && components.venue !== undefined;
|
|
265
|
-
}
|
|
266
|
-
//#endregion
|
|
267
|
-
//# sourceMappingURL=house-venue.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"house-venue.js","sourceRoot":"","sources":["../../../src/synthesizers/house-venue.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AA+BH,oBAAoB;AAEpB;;;GAGG;AACH,MAAM,YAAY,GAA0B;IAC3C,aAAa;IACb,kBAAkB;IAClB,aAAa;IACb,gBAAgB;IAChB,iBAAiB;IACjB,yBAAyB;IACzB,oBAAoB;IACpB,uFAAuF;IACvF,+FAA+F;IAC/F,iCAAiC;IACjC,kBAAkB;IAClB,0BAA0B;IAC1B,kBAAkB;IAClB,qBAAqB;IACrB,aAAa;IACb,iBAAiB;IACjB,oBAAoB;IACpB,oBAAoB;IACpB,oBAAoB;IACpB,mBAAmB;IACnB,oBAAoB;IACpB,qBAAqB;IACrB,sBAAsB;IACtB,iBAAiB;IACjB,eAAe;IACf,sBAAsB;IACtB,oBAAoB;IACpB,yBAAyB;IACzB,4BAA4B;IAC5B,sBAAsB;IACtB,iBAAiB;CACjB,CAAA;AAED;;;;;GAKG;AACH,MAAM,SAAS,GAA0B;IACxC,yBAAyB;IACzB,cAAc;IACd,cAAc;IACd,oBAAoB;IACpB,iBAAiB;IACjB,gBAAgB;IAChB,eAAe;IACf,uBAAuB;IACvB,8BAA8B;IAC9B,mBAAmB;IACnB,2BAA2B;IAC3B,kBAAkB;IAClB,sBAAsB;IACtB,2BAA2B;IAC3B,kBAAkB;IAClB,wBAAwB;IACxB,4BAA4B;IAC5B,6BAA6B;IAC7B,0BAA0B;IAC1B,6BAA6B;IAC7B,2BAA2B;IAC3B,wBAAwB;IACxB,6BAA6B;IAC7B,qBAAqB;IACrB,sBAAsB;IACtB,cAAc;IACd,uBAAuB;IACvB,oBAAoB;CACpB,CAAA;AAED,YAAY;AAEZ,8BAA8B;AAE9B;;GAEG;AACH,MAAM,gBAAgB,GAA0B;IAC/C,SAAS;IACT,SAAS;IACT,SAAS;IACT,QAAQ;IACR,UAAU;IACV,YAAY;IACZ,UAAU;IACV,WAAW;IACX,YAAY;IACZ,WAAW;IACX,UAAU;IACV,SAAS;IACT,SAAS;IACT,UAAU;IACV,aAAa;CACb,CAAA;AAED,YAAY;AAEZ,gCAAgC;AAEhC,SAAS,iBAAiB,CAAC,MAAoB;IAC9C,oFAAoF;IACpF,mFAAmF;IACnF,MAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,GAAG,CAAC,CAAA;IAC3C,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,EAAE,EAAE,MAAM,CAAC,CAAA;IAChC,MAAM,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,GAAG,CAAC,CAAA;IAExC,OAAO,MAAM,CAAC,CAAC,CAAC,CAAA;AACjB,CAAC;AAED,SAAS,IAAI,CAAI,GAAqB,EAAE,MAAoB;IAC3D,OAAO,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,MAAM,CAAC,CAAE,CAAA;AAC/C,CAAC;AAED,SAAS,eAAe,CAAC,OAAe;IACvC,QAAQ,OAAO,EAAE,CAAC;QACjB,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf,KAAK,IAAI;YACR,OAAO,OAAO,CAAA;QACf;YACC,OAAO,OAAO,CAAA;IAChB,CAAC;AACF,CAAC;AAED,YAAY;AAEZ,mBAAmB;AAEnB;;;;GAIG;AACH,MAAM,kBAAkB,GAAG,GAAG,CAAA;AAE9B;;;GAGG;AACH,MAAM,oBAAoB,GAAG,IAAI,CAAA;AAEjC;;;;;GAKG;AACH,MAAM,mBAAmB,GAAG,GAAG,CAAA;AAE/B;;GAEG;AACH,MAAM,gBAAgB,GAAoD;IACzE,EAAE,EAAE,CAAC,eAAe,EAAE,KAAK,CAAC;IAC5B,EAAE,EAAE,CAAC,gBAAgB,EAAE,gBAAgB,EAAE,gBAAgB,EAAE,IAAI,CAAC;IAChE,EAAE,EAAE,CAAC,QAAQ,CAAC;IACd,EAAE,EAAE,CAAC,QAAQ,CAAC;IACd,EAAE,EAAE,CAAC,WAAW,CAAC;IACjB,EAAE,EAAE,CAAC,SAAS,EAAE,aAAa,CAAC;CAC9B,CAAA;AAED,MAAM,UAAU,uBAAuB,CACtC,IAAyB,EACzB,OAAgC,EAAE;IAElC,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,MAAM,GAAG,eAAe,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;IAC5C,MAAM,QAAQ,GAAG,IAAI,CAAC,aAAa,IAAI,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,oBAAoB,CAAC,CAAC,CAAC,qBAAqB,CAAC,CAAA;IAEtG,6FAA6F;IAC7F,+FAA+F;IAC/F,+FAA+F;IAC/F,+FAA+F;IAC/F,+FAA+F;IAC/F,kGAAkG;IAClG,+DAA+D;IAC/D,MAAM,OAAO,GAAG,IAAI,CAAC,OAAO,KAAK,IAAI,CAAA;IACrC,MAAM,OAAO,GAAG,IAAI,CAAC,OAAO,KAAK,IAAI,CAAA;IACrC,8GAA8G;IAC9G,gHAAgH;IAChH,+FAA+F;IAC/F,EAAE;IACF,8GAA8G;IAC9G,oHAAoH;IACpH,gHAAgH;IAChH,kHAAkH;IAClH,mFAAmF;IACnF,MAAM,OAAO,GAAG,IAAI,CAAC,OAAO,KAAK,IAAI,CAAA;IAErC,+FAA+F;IAC/F,+FAA+F;IAC/F,gGAAgG;IAChG,MAAM,KAAK,GAAG,OAAO,IAAI,MAAM,EAAE,GAAG,kBAAkB,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAA;IAC7G,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,gBAAgB,EAAE,MAAM,CAAC,CAAA;IAC5D,IAAI,WAAW,GAAG,IAAI,CAAC,WAAW,IAAI,iBAAiB,CAAC,MAAM,CAAC,CAAA;IAE/D,4FAA4F;IAC5F,8FAA8F;IAC9F,6FAA6F;IAC7F,WAAW;IACX,IAAI,OAAO,IAAI,MAAM,EAAE,GAAG,oBAAoB,IAAI,OAAO,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,CAAC;QAC7E,MAAM,KAAK,GAAG,MAAM,CAAC,QAAQ,CAAC,WAAW,EAAE,EAAE,CAAC,CAAA;QAC9C,MAAM,IAAI,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAA;QAE/C,WAAW,GAAG,GAAG,KAAK,IAAI,KAAK,GAAG,IAAI,EAAE,CAAA;IACzC,CAAC;IAED,MAAM,UAAU,GAA+B;QAC9C,YAAY,EAAE,WAAW;QACzB,MAAM;QACN,KAAK;QACL,QAAQ,EAAE,IAAI,CAAC,QAAQ;QACvB,GAAG,CAAC,OAAO,IAAI,OAAO,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,CAAC;QACtD,QAAQ,EAAE,IAAI,CAAC,QAAQ;KACvB,CAAA;IAED,IAAI,IAAI,GAAG,OAAO;QACjB,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,IAAI,IAAI,CAAC,QAAQ,EAAE;QACrC,CAAC,CAAC,OAAO;YACR,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,IAAI,IAAI,CAAC,QAAQ,EAAE;YACrC,CAAC,CAAC,OAAO;gBACR,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,IAAI,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,EAAE;gBACrD,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAA;IAExD,yFAAyF;IACzF,MAAM,eAAe,GAAG,gBAAgB,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;IAEtD,IAAI,eAAe,IAAI,MAAM,EAAE,GAAG,mBAAmB,EAAE,CAAC;QACvD,MAAM,cAAc,GAAG,IAAI,CAAC,eAAe,EAAE,MAAM,CAAC,CAAA;QACpD,UAAU,CAAC,OAAO,GAAG,cAAc,CAAA;QACnC,IAAI,GAAG,GAAG,IAAI,KAAK,cAAc,EAAE,CAAA;IACpC,CAAC;IAED,IAAI,GAAW,CAAA;IAEf,QAAQ,QAAQ,EAAE,CAAC;QAClB,KAAK,oBAAoB;YACxB,GAAG,GAAG,GAAG,WAAW,IAAI,MAAM,KAAK,KAAK,KAAK,IAAI,EAAE,CAAA;YACnD,MAAK;QACN,KAAK,qBAAqB;YACzB,GAAG,GAAG,GAAG,KAAK,KAAK,WAAW,IAAI,MAAM,KAAK,IAAI,EAAE,CAAA;YACnD,MAAK;IACP,CAAC;IAED,OAAO,EAAE,GAAG,EAAE,UAAU,EAAE,MAAM,EAAE,QAAQ,EAAE,CAAA;AAC7C,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,sBAAsB,CAAC,UAAsC;IAC5E,OAAO,UAAU,CAAC,YAAY,KAAK,SAAS,IAAI,UAAU,CAAC,KAAK,KAAK,SAAS,CAAA;AAC/E,CAAC;AAED,YAAY"}
|
|
@@ -1,52 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* @copyright Sister Software
|
|
3
|
-
* @license AGPL-3.0
|
|
4
|
-
* @author Teffen Ellis, et al.
|
|
5
|
-
*
|
|
6
|
-
* Intersection synthesizer — v0.7 coverage fix (night-3, DeepSeek-decided).
|
|
7
|
-
*
|
|
8
|
-
* The 2026-05-29 harness diagnostic found the neural model emits `intersection_a`/`intersection_b`
|
|
9
|
-
* with ~0.0001 probability on canonical intersections ("Broadway & W 42nd St") — it never learned
|
|
10
|
-
* the tags, because the corpus has NO intersection training signal (no generator, and real-data
|
|
11
|
-
* adapters don't emit intersection-formatted rows). Intersections are 65 of the 376 harness
|
|
12
|
-
* assertions (17%), all 0% neural. This generator produces the missing signal as a small targeted
|
|
13
|
-
* supplement shard (synthesis-as-supplement discipline: weight < 0.25, one-and-done).
|
|
14
|
-
*
|
|
15
|
-
* Output is a `CanonicalRow` ({raw, components}); the corpus aligner turns it into BIO labels
|
|
16
|
-
* (B-/I-intersection_a, O on the connector, B-/I-intersection_b). Surface forms of both streets
|
|
17
|
-
* MUST occur verbatim in `raw` so alignment lands.
|
|
18
|
-
*
|
|
19
|
-
* US-idiomatic only (the harness intersection cases are US: "X & Y, City, ST ZIP").
|
|
20
|
-
*/
|
|
21
|
-
import type { CanonicalRow } from "#types";
|
|
22
|
-
export interface IntersectionBaseTuple {
|
|
23
|
-
locality: string;
|
|
24
|
-
region: string;
|
|
25
|
-
/**
|
|
26
|
-
* ZIP — optional; ~30% of synthetic intersections omit it (idiomatic).
|
|
27
|
-
*/
|
|
28
|
-
postcode?: string;
|
|
29
|
-
country: string;
|
|
30
|
-
}
|
|
31
|
-
export interface SynthesizedIntersectionRow {
|
|
32
|
-
raw: string;
|
|
33
|
-
components: CanonicalRow["components"];
|
|
34
|
-
locale: string;
|
|
35
|
-
}
|
|
36
|
-
export interface IntersectionSynthesisOpts {
|
|
37
|
-
random?: () => number;
|
|
38
|
-
}
|
|
39
|
-
/**
|
|
40
|
-
* Synthesize one US intersection row. Returns null on the rare degenerate case where the two streets collide (so
|
|
41
|
-
* alignment never has two identical surface forms to disambiguate).
|
|
42
|
-
*/
|
|
43
|
-
export declare function synthesizeIntersectionRow(base: IntersectionBaseTuple, opts?: IntersectionSynthesisOpts): SynthesizedIntersectionRow | null;
|
|
44
|
-
/**
|
|
45
|
-
* A small built-in US city/region/zip pool for standalone shard generation + tests.
|
|
46
|
-
*/
|
|
47
|
-
export declare const DEFAULT_US_BASES: ReadonlyArray<IntersectionBaseTuple>;
|
|
48
|
-
/**
|
|
49
|
-
* Generate `count` intersection rows over the provided bases (round-robin).
|
|
50
|
-
*/
|
|
51
|
-
export declare function generateIntersectionRows(count: number, bases?: ReadonlyArray<IntersectionBaseTuple>, opts?: IntersectionSynthesisOpts): SynthesizedIntersectionRow[];
|
|
52
|
-
//# sourceMappingURL=intersection.d.ts.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"intersection.d.ts","sourceRoot":"","sources":["../../../src/synthesizers/intersection.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAEH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,QAAQ,CAAA;AAyE1C,MAAM,WAAW,qBAAqB;IACrC,QAAQ,EAAE,MAAM,CAAA;IAChB,MAAM,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,EAAE,MAAM,CAAA;CACf;AAED,MAAM,WAAW,0BAA0B;IAC1C,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,CAAA;IACtC,MAAM,EAAE,MAAM,CAAA;CACd;AAED,MAAM,WAAW,yBAAyB;IACzC,MAAM,CAAC,EAAE,MAAM,MAAM,CAAA;CACrB;AAyBD;;;GAGG;AACH,wBAAgB,yBAAyB,CACxC,IAAI,EAAE,qBAAqB,EAC3B,IAAI,GAAE,yBAA8B,GAClC,0BAA0B,GAAG,IAAI,CA+CnC;AAED;;GAEG;AACH,eAAO,MAAM,gBAAgB,EAAE,aAAa,CAAC,qBAAqB,CAWjE,CAAA;AAED;;GAEG;AACH,wBAAgB,wBAAwB,CACvC,KAAK,EAAE,MAAM,EACb,KAAK,GAAE,aAAa,CAAC,qBAAqB,CAAoB,EAC9D,IAAI,GAAE,yBAA8B,GAClC,0BAA0B,EAAE,CAe9B"}
|