@mailwoman/corpus 4.12.0 → 4.14.0

package/out/src/adapters/gnaf/adapter.d.ts

@@ -0,0 +1,40 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   G-NAF (Australia) corpus adapter — the parser-teaching half of #208.
+ *
+ *   The model mis-parses Australian addresses in their native postcode-first / house-number-last
+ *   order: it tags a leading 4-digit postcode as a house number (its US/EU prior) and swaps street
+ *   ↔ locality with it. `scripts/eval/au-order-probe.ts` proved this is a word-ORDER coverage gap,
+ *   not capability — the same addresses parse perfectly in canonical order (65% → 87% @25km if the
+ *   parse were order-robust). EU survives the same eval because its postcodes are
+ *   format-distinctive (a hyphenated `26-300` reads as a postcode anywhere); a bare AU `3053` only
+ *   disambiguates by position.
+ *
+ *   So this adapter renders each assembled G-NAF tuple (from {@link ./assemble}) in one of three real
+ *   AU layouts — real-AU canonical (number-first, postcode-trailing), postcode-first,
+ *   locality-first — ROTATED by row index (`i % 3`), so the locality + postcode each land in every
+ *   position across the shard. This is the exact mechanism that fixed #148's v1.9.0 order-overfit
+ *   for the 16 EU locales (`scripts/rerender-overture-multiorder.mjs`, v1.9.1 → shipped v4.13.0);
+ *   AU was simply never in that train (`country_weights` had no AU, and `data_loader.py` excludes
+ *   unlisted countries). Rotating one order per row (rather than emitting all three) keeps this a
+ *   clean single-variable extension of the proven recipe + matches its source-mass structure. The
+ *   corpus aligner BIO-labels each (every component surface form occurs verbatim in `raw`, so
+ *   alignment lands).
+ *
+ *   Input: the assembled component JSONL (one `{house_number,street,locality,region,postcode}` per
+ *   line). Open G-NAF licence — attribute "Geoscape Australia".
+ */
+import type { CorpusAdapter } from "../../types.js";
+export declare const GNAF_ADAPTER_ID = "gnaf";
+/** Open G-NAF is freely redistributable with attribution to Geoscape Australia (CC-BY-style). */
+export declare const GNAF_DEFAULT_LICENSE = "CC-BY-4.0";
+/**
+ * Build the G-NAF adapter. `inputPath` is the assembled component JSONL (see {@link ./assemble}); it
+ * is country-pinned to AU regardless of `opts.country` (G-NAF is Australia-only).
+ */
+export declare function createGnafAdapter(): CorpusAdapter;
+export declare const gnafAdapter: CorpusAdapter;
+//# sourceMappingURL=adapter.d.ts.map

package/out/src/adapters/gnaf/adapter.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../src/adapters/gnaf/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAMH,OAAO,KAAK,EAAgC,aAAa,EAAE,MAAM,gBAAgB,CAAA;AAEjF,eAAO,MAAM,eAAe,SAAS,CAAA;AACrC,iGAAiG;AACjG,eAAO,MAAM,oBAAoB,cAAc,CAAA;AA2B/C;;;GAGG;AACH,wBAAgB,iBAAiB,IAAI,aAAa,CA4DjD;AAED,eAAO,MAAM,WAAW,eAAsB,CAAA"}

package/out/src/adapters/gnaf/adapter.js

@@ -0,0 +1,122 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   G-NAF (Australia) corpus adapter — the parser-teaching half of #208.
+ *
+ *   The model mis-parses Australian addresses in their native postcode-first / house-number-last
+ *   order: it tags a leading 4-digit postcode as a house number (its US/EU prior) and swaps street
+ *   ↔ locality with it. `scripts/eval/au-order-probe.ts` proved this is a word-ORDER coverage gap,
+ *   not capability — the same addresses parse perfectly in canonical order (65% → 87% @25km if the
+ *   parse were order-robust). EU survives the same eval because its postcodes are
+ *   format-distinctive (a hyphenated `26-300` reads as a postcode anywhere); a bare AU `3053` only
+ *   disambiguates by position.
+ *
+ *   So this adapter renders each assembled G-NAF tuple (from {@link ./assemble}) in one of three real
+ *   AU layouts — real-AU canonical (number-first, postcode-trailing), postcode-first,
+ *   locality-first — ROTATED by row index (`i % 3`), so the locality + postcode each land in every
+ *   position across the shard. This is the exact mechanism that fixed #148's v1.9.0 order-overfit
+ *   for the 16 EU locales (`scripts/rerender-overture-multiorder.mjs`, v1.9.1 → shipped v4.13.0);
+ *   AU was simply never in that train (`country_weights` had no AU, and `data_loader.py` excludes
+ *   unlisted countries). Rotating one order per row (rather than emitting all three) keeps this a
+ *   clean single-variable extension of the proven recipe + matches its source-mass structure. The
+ *   corpus aligner BIO-labels each (every component surface form occurs verbatim in `raw`, so
+ *   alignment lands).
+ *
+ *   Input: the assembled component JSONL (one `{house_number,street,locality,region,postcode}` per
+ *   line). Open G-NAF licence — attribute "Geoscape Australia".
+ */
+import { createReadStream } from "node:fs";
+import { createInterface } from "node:readline";
+import { stableSourceId } from "../../adapter.js";
+import { reconcileComponents } from "../../format.js";
+export const GNAF_ADAPTER_ID = "gnaf";
+/** Open G-NAF is freely redistributable with attribution to Geoscape Australia (CC-BY-style). */
+export const GNAF_DEFAULT_LICENSE = "CC-BY-4.0";
+/**
+ * The address layouts an AU address actually arrives in. The model already handles
+ * postcode-TRAILING (canonical); the two postcode-LEADING forms are the ones it fails, so they
+ * carry the lever. We keep the canonical form too so the retrain doesn't forget it.
+ */
+function renderOrders(c) {
+    const region = c.region ? ` ${c.region}` : "";
+    return [
+        // real-AU canonical: number-first, street, suburb [state] postcode — "50 Barry Street, Carlton NSW 2000"
+        `${c.house_number} ${c.street}, ${c.locality}${region} ${c.postcode}`,
+        // postcode-first (the dominant failure): "2000 Carlton, Barry Street 50"
+        `${c.postcode} ${c.locality}, ${c.street} ${c.house_number}`,
+        // locality-first: "Carlton, 2000, Barry Street 50"
+        `${c.locality}, ${c.postcode}, ${c.street} ${c.house_number}`,
+    ];
+}
+/**
+ * Build the G-NAF adapter. `inputPath` is the assembled component JSONL (see {@link ./assemble}); it
+ * is country-pinned to AU regardless of `opts.country` (G-NAF is Australia-only).
+ */
+export function createGnafAdapter() {
+    return {
+        id: GNAF_ADAPTER_ID,
+        defaultLicense: GNAF_DEFAULT_LICENSE,
+        description: "G-NAF (Australia): assembled address tuples rendered in multiple word orders (canonical / postcode-first / locality-first) — teaches the model AU's postcode-first layout.",
+        async *rows(opts) {
+            const stream = createReadStream(opts.inputPath, { encoding: "utf8" });
+            const lines = createInterface({ input: stream, crlfDelay: Infinity });
+            let emitted = 0;
+            let idx = 0; // rotates the render order (i % 3), matching v1.9.1's rerender
+            try {
+                for await (const line of lines) {
+                    if (opts.signal?.aborted)
+                        break;
+                    if (opts.limit !== undefined && emitted >= opts.limit)
+                        break;
+                    if (!line.trim())
+                        continue;
+                    let t;
+                    try {
+                        t = JSON.parse(line);
+                    }
+                    catch {
+                        continue;
+                    }
+                    if (!t.house_number || !t.street || !t.locality || !t.postcode)
+                        continue;
+                    const orders = renderOrders(t);
+                    const order = idx % orders.length;
+                    idx++;
+                    const raw = orders[order];
+                    const components = {
+                        house_number: t.house_number,
+                        street: t.street,
+                        locality: t.locality,
+                        postcode: t.postcode,
+                    };
+                    // region (state) rides only the canonical render (order 0); the postcode-leading layouts
+                    // omit it (matching the eval's serialization) so it never breaks verbatim alignment.
+                    if (order === 0 && t.region)
+                        components.region = t.region;
+                    const aligned = reconcileComponents(components, raw);
+                    if (Object.keys(aligned).length === 0)
+                        continue;
+                    yield {
+                        raw,
+                        components: aligned,
+                        country: "AU",
+                        locale: "en-AU",
+                        source: GNAF_ADAPTER_ID,
+                        source_id: `${stableSourceId(GNAF_ADAPTER_ID, aligned)}-o${order}`,
+                        corpus_version: "",
+                        license: GNAF_DEFAULT_LICENSE,
+                    };
+                    emitted++;
+                }
+            }
+            finally {
+                lines.close();
+                stream.destroy();
+            }
+        },
+    };
+}
+export const gnafAdapter = createGnafAdapter();
+//# sourceMappingURL=adapter.js.map

package/out/src/adapters/gnaf/adapter.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../src/adapters/gnaf/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,SAAS,CAAA;AAC1C,OAAO,EAAE,eAAe,EAAE,MAAM,eAAe,CAAA;AAC/C,OAAO,EAAE,cAAc,EAAE,MAAM,kBAAkB,CAAA;AACjD,OAAO,EAAE,mBAAmB,EAAE,MAAM,iBAAiB,CAAA;AAGrD,MAAM,CAAC,MAAM,eAAe,GAAG,MAAM,CAAA;AACrC,iGAAiG;AACjG,MAAM,CAAC,MAAM,oBAAoB,GAAG,WAAW,CAAA;AAU/C;;;;GAIG;AACH,SAAS,YAAY,CAAC,CAAY;IACjC,MAAM,MAAM,GAAG,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAA;IAC7C,OAAO;QACN,yGAAyG;QACzG,GAAG,CAAC,CAAC,YAAY,IAAI,CAAC,CAAC,MAAM,KAAK,CAAC,CAAC,QAAQ,GAAG,MAAM,IAAI,CAAC,CAAC,QAAQ,EAAE;QACrE,yEAAyE;QACzE,GAAG,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,YAAY,EAAE;QAC5D,mDAAmD;QACnD,GAAG,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,YAAY,EAAE;KAC7D,CAAA;AACF,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,iBAAiB;IAChC,OAAO;QACN,EAAE,EAAE,eAAe;QACnB,cAAc,EAAE,oBAAoB;QACpC,WAAW,EACV,4KAA4K;QAE7K,KAAK,CAAC,CAAC,IAAI,CAAC,IAAoB;YAC/B,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;YACrE,MAAM,KAAK,GAAG,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAA;YACrE,IAAI,OAAO,GAAG,CAAC,CAAA;YACf,IAAI,GAAG,GAAG,CAAC,CAAA,CAAC,+DAA+D;YAC3E,IAAI,CAAC;gBACJ,IAAI,KAAK,EAAE,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;oBAChC,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;wBAAE,MAAK;oBAC/B,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;wBAAE,MAAK;oBAC5D,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;wBAAE,SAAQ;oBAE1B,IAAI,CAAY,CAAA;oBAChB,IAAI,CAAC;wBACJ,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAc,CAAA;oBAClC,CAAC;oBAAC,MAAM,CAAC;wBACR,SAAQ;oBACT,CAAC;oBACD,IAAI,CAAC,CAAC,CAAC,YAAY,IAAI,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,CAAC,QAAQ;wBAAE,SAAQ;oBAExE,MAAM,MAAM,GAAG,YAAY,CAAC,CAAC,CAAC,CAAA;oBAC9B,MAAM,KAAK,GAAG,GAAG,GAAG,MAAM,CAAC,MAAM,CAAA;oBACjC,GAAG,EAAE,CAAA;oBACL,MAAM,GAAG,GAAG,MAAM,CAAC,KAAK,CAAE,CAAA;oBAC1B,MAAM,UAAU,GAA+B;wBAC9C,YAAY,EAAE,CAAC,CAAC,YAAY;wBAC5B,MAAM,EAAE,CAAC,CAAC,MAAM;wBAChB,QAAQ,EAAE,CAAC,CAAC,QAAQ;wBACpB,QAAQ,EAAE,CAAC,CAAC,QAAQ;qBACpB,CAAA;oBACD,yFAAyF;oBACzF,qFAAqF;oBACrF,IAAI,KAAK,KAAK,CAAC,IAAI,CAAC,CAAC,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,CAAC,CAAC,MAAM,CAAA;oBAEzD,MAAM,OAAO,GAAG,mBAAmB,CAAC,UAAU,EAAE,GAAG,CAAC,CAAA;oBACpD,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAQ;oBAC/C,MAAM;wBACL,GAAG;wBACH,UAAU,EAAE,OAAO;wBACnB,OAAO,EAAE,IAAI;wBACb,MAAM,EAAE,OAAO;wBACf,MAAM,EAAE,eAAe;wBACvB,SAAS,EAAE,GAAG,cAAc,CAAC,eAAe,EAAE,OAAO,CAAC,KAAK,KAAK,EAAE;wBAClE,cAAc,EAAE,EAAE;wBAClB,OAAO,EAAE,oBAAoB;qBAC7B,CAAA;oBACD,OAAO,EAAE,CAAA;gBACV,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,KAAK,CAAC,KAAK,EAAE,CAAA;gBACb,MAAM,CAAC,OAAO,EAAE,CAAA;YACjB,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED,MAAM,CAAC,MAAM,WAAW,GAAG,iBAAiB,EAAE,CAAA"}

package/out/src/adapters/gnaf/assemble.d.ts

@@ -0,0 +1,52 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Assemble a sampled, component-labeled Australian address set from G-NAF (the Geocoded National
+ *   Address File — Geoscape Australia, Open G-NAF licence; any derived artifact must attribute
+ *   "Geoscape Australia"). G-NAF is a relational PSV distribution (~16.9M addresses);
+ *   reconstructing a street address joins three tables — ADDRESS_DETAIL (number, postcode, the
+ *   PIDs) → STREET_LOCALITY (street name + type) → LOCALITY (suburb). State is the per-file prefix
+ *   (ACT/NSW/…).
+ *
+ *   Streaming + in-memory join via the house {@link PSVSpliterator} (pipe-separated; `mode: "object"`
+ *   keys each row by its header) — NOT raw `read_csv` SQL, which a flat-file join doesn't need and
+ *   which the #183–190 cleanup retired. The two lookup tables (STREET_LOCALITY ~765k rows, LOCALITY
+ *   ~16k) fit as Maps; ADDRESS_DETAIL is streamed once and reservoir-sampled, so memory stays
+ *   bounded (the OOM lesson from the Overture ingest).
+ *
+ *   No coordinates: the output feeds the PARSER ({@link ../gnaf/adapter}, #208) — teaching the model
+ *   AU's postcode-first / house-number-last word order, the gap `scripts/eval/au-order-probe.ts`
+ *   pinned (65%→87% if the parse were order-robust). The parser needs the address string +
+ *   component labels, not lat/lon.
+ *
+ *   Output: component tuples as JSONL, consumed by the `gnaf` corpus adapter (which renders them in
+ *   multiple orders + the corpus aligner BIO-labels them). An optional held-out eval set is
+ *   excluded by (street, locality, postcode) so the training shard never overlaps the benchmark.
+ */
+export interface GnafAssembleOptions {
+    /** G-NAF `Standard` directory (holds the per-state `*_psv.psv` tables). */
+    standardDir: string;
+    /** Target sample size (uniform reservoir → population-proportional across states). */
+    sampleSize: number;
+    /** Output JSONL path. */
+    out: string;
+    /** Optional held-out eval JSONL (rows with a `components` field) — its (street,locality,postcode)
+are excluded. */
+    holdoutPath?: string;
+    /** Progress sink (the CLI passes a setter). */
+    onProgress?: (message: string) => void;
+}
+export interface GnafAssembleResult {
+    written: number;
+    seen: number;
+    heldOut: number;
+    byState: Record<string, number>;
+}
+/** UPPERCASE → Title Case, preserving intra-word apostrophes/hyphens (O'Brien, Coff's Harbour). */
+export declare function titlecase(s: string): string;
+/** Holdout/dedup key: a street within a locality+postcode (house-number-agnostic — conservative). */
+export declare function gnafHoldoutKey(street: string, locality: string, postcode: string): string;
+export declare function assembleGnaf(opts: GnafAssembleOptions): Promise<GnafAssembleResult>;
+//# sourceMappingURL=assemble.d.ts.map

package/out/src/adapters/gnaf/assemble.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"assemble.d.ts","sourceRoot":"","sources":["../../../../src/adapters/gnaf/assemble.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AAOH,MAAM,WAAW,mBAAmB;IACnC,2EAA2E;IAC3E,WAAW,EAAE,MAAM,CAAA;IACnB,sFAAsF;IACtF,UAAU,EAAE,MAAM,CAAA;IAClB,yBAAyB;IACzB,GAAG,EAAE,MAAM,CAAA;IACX;gBACe;IACf,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB,+CAA+C;IAC/C,UAAU,CAAC,EAAE,CAAC,OAAO,EAAE,MAAM,KAAK,IAAI,CAAA;CACtC;AAED,MAAM,WAAW,kBAAkB;IAClC,OAAO,EAAE,MAAM,CAAA;IACf,IAAI,EAAE,MAAM,CAAA;IACZ,OAAO,EAAE,MAAM,CAAA;IACf,OAAO,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;CAC/B;AAED,mGAAmG;AACnG,wBAAgB,SAAS,CAAC,CAAC,EAAE,MAAM,GAAG,MAAM,CAK3C;AAED,qGAAqG;AACrG,wBAAgB,cAAc,CAAC,MAAM,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,MAAM,CAEzF;AAmCD,wBAAsB,YAAY,CAAC,IAAI,EAAE,mBAAmB,GAAG,OAAO,CAAC,kBAAkB,CAAC,CAiEzF"}

package/out/src/adapters/gnaf/assemble.js

@@ -0,0 +1,142 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Assemble a sampled, component-labeled Australian address set from G-NAF (the Geocoded National
+ *   Address File — Geoscape Australia, Open G-NAF licence; any derived artifact must attribute
+ *   "Geoscape Australia"). G-NAF is a relational PSV distribution (~16.9M addresses);
+ *   reconstructing a street address joins three tables — ADDRESS_DETAIL (number, postcode, the
+ *   PIDs) → STREET_LOCALITY (street name + type) → LOCALITY (suburb). State is the per-file prefix
+ *   (ACT/NSW/…).
+ *
+ *   Streaming + in-memory join via the house {@link PSVSpliterator} (pipe-separated; `mode: "object"`
+ *   keys each row by its header) — NOT raw `read_csv` SQL, which a flat-file join doesn't need and
+ *   which the #183–190 cleanup retired. The two lookup tables (STREET_LOCALITY ~765k rows, LOCALITY
+ *   ~16k) fit as Maps; ADDRESS_DETAIL is streamed once and reservoir-sampled, so memory stays
+ *   bounded (the OOM lesson from the Overture ingest).
+ *
+ *   No coordinates: the output feeds the PARSER ({@link ../gnaf/adapter}, #208) — teaching the model
+ *   AU's postcode-first / house-number-last word order, the gap `scripts/eval/au-order-probe.ts`
+ *   pinned (65%→87% if the parse were order-robust). The parser needs the address string +
+ *   component labels, not lat/lon.
+ *
+ *   Output: component tuples as JSONL, consumed by the `gnaf` corpus adapter (which renders them in
+ *   multiple orders + the corpus aligner BIO-labels them). An optional held-out eval set is
+ *   excluded by (street, locality, postcode) so the training shard never overlaps the benchmark.
+ */
+import { createWriteStream } from "node:fs";
+import { readdir, readFile } from "node:fs/promises";
+import { join } from "node:path";
+import { PSVSpliterator } from "spliterator";
+/** UPPERCASE → Title Case, preserving intra-word apostrophes/hyphens (O'Brien, Coff's Harbour). */
+export function titlecase(s) {
+    return s
+        .toLowerCase()
+        .replace(/(^|[\s'\-/])([a-z])/g, (_m, p, c) => p + c.toUpperCase())
+        .trim();
+}
+/** Holdout/dedup key: a street within a locality+postcode (house-number-agnostic — conservative). */
+export function gnafHoldoutKey(street, locality, postcode) {
+    return `${street}|${locality}|${postcode}`.toLowerCase();
+}
+async function* psvObjects(path) {
+    yield* PSVSpliterator.fromAsync(path, { mode: "object", header: true });
+}
+/** Load a small lookup table fully into a Map keyed by `keyCol`. */
+async function loadMap(paths, keyCol, pick) {
+    const m = new Map();
+    for (const p of paths) {
+        for await (const r of psvObjects(p)) {
+            const k = r[keyCol];
+            if (k != null && k !== "")
+                m.set(String(k), pick(r));
+        }
+    }
+    return m;
+}
+/** Build the held-out key set from an eval JSONL whose rows carry a `components` object. */
+async function loadHoldout(path) {
+    const keys = new Set();
+    const text = await readFile(path, "utf8");
+    for (const line of text.split("\n")) {
+        if (!line.trim())
+            continue;
+        try {
+            const c = JSON.parse(line).components;
+            if (c?.street && c?.locality && c?.postcode)
+                keys.add(gnafHoldoutKey(c.street, c.locality, c.postcode));
+        }
+        catch {
+            /* skip malformed */
+        }
+    }
+    return keys;
+}
+export async function assembleGnaf(opts) {
+    const progress = opts.onProgress ?? (() => { });
+    const files = await readdir(opts.standardDir);
+    const pick = (re, exclude) => files.filter((f) => re.test(f) && !(exclude && exclude.test(f))).map((f) => join(opts.standardDir, f));
+    // `*_LOCALITY_psv.psv` also globs `*_STREET_LOCALITY_psv.psv` — exclude the latter explicitly.
+    const streetPaths = pick(/_STREET_LOCALITY_psv\.psv$/);
+    const localityPaths = pick(/_LOCALITY_psv\.psv$/, /_STREET_LOCALITY_psv\.psv$/);
+    const addressPaths = pick(/_ADDRESS_DETAIL_psv\.psv$/);
+    const holdout = opts.holdoutPath ? await loadHoldout(opts.holdoutPath) : new Set();
+    if (opts.holdoutPath)
+        progress(`held-out eval keys: ${holdout.size}`);
+    progress(`loading STREET_LOCALITY (${streetPaths.length} files) + LOCALITY (${localityPaths.length})…`);
+    const streetMap = await loadMap(streetPaths, "STREET_LOCALITY_PID", (r) => ({
+        name: String(r.STREET_NAME ?? ""),
+        type: String(r.STREET_TYPE_CODE ?? ""),
+        suffix: String(r.STREET_SUFFIX_CODE ?? ""),
+    }));
+    const localityMap = await loadMap(localityPaths, "LOCALITY_PID", (r) => String(r.LOCALITY_NAME ?? ""));
+    progress(`streets=${streetMap.size.toLocaleString()} localities=${localityMap.size.toLocaleString()}`);
+    const reservoir = [];
+    let seen = 0;
+    let heldOut = 0;
+    for (const p of addressPaths) {
+        const state = (p.match(/\/([A-Z]+)_ADDRESS_DETAIL/) ?? [])[1] ?? "";
+        for await (const r of psvObjects(p)) {
+            const numberFirst = String(r.NUMBER_FIRST ?? "");
+            if (!numberFirst || r.DATE_RETIRED || !r.POSTCODE)
+                continue;
+            const st = streetMap.get(String(r.STREET_LOCALITY_PID ?? ""));
+            const suburbRaw = localityMap.get(String(r.LOCALITY_PID ?? ""));
+            if (!st?.name || !suburbRaw)
+                continue;
+            const street = `${titlecase(st.name)} ${titlecase(st.type)}${st.suffix ? " " + titlecase(st.suffix) : ""}`.trim();
+            const locality = titlecase(suburbRaw);
+            const postcode = String(r.POSTCODE);
+            if (holdout.has(gnafHoldoutKey(street, locality, postcode))) {
+                heldOut++;
+                continue;
+            }
+            let house = numberFirst + (r.NUMBER_FIRST_SUFFIX ? String(r.NUMBER_FIRST_SUFFIX) : "");
+            if (r.NUMBER_LAST)
+                house = `${house}-${String(r.NUMBER_LAST)}`;
+            if (r.FLAT_NUMBER)
+                house = `${String(r.FLAT_NUMBER)}/${house}`;
+            const tuple = { house_number: house, street, locality, region: state, postcode };
+            seen++;
+            if (reservoir.length < opts.sampleSize)
+                reservoir.push(tuple);
+            else {
+                const j = Math.floor(Math.random() * seen);
+                if (j < opts.sampleSize)
+                    reservoir[j] = tuple;
+            }
+        }
+        progress(`${state}: ${seen.toLocaleString()} valid joinable seen`);
+    }
+    const out = createWriteStream(opts.out);
+    const byState = {};
+    for (const t of reservoir) {
+        out.write(JSON.stringify(t) + "\n");
+        byState[t.region] = (byState[t.region] ?? 0) + 1;
+    }
+    await new Promise((res) => out.end(res));
+    progress(`wrote ${reservoir.length.toLocaleString()} tuples → ${opts.out}`);
+    return { written: reservoir.length, seen, heldOut, byState };
+}
+//# sourceMappingURL=assemble.js.map

package/out/src/adapters/gnaf/assemble.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"assemble.js","sourceRoot":"","sources":["../../../../src/adapters/gnaf/assemble.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,SAAS,CAAA;AAC3C,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAA;AACpD,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAA;AAChC,OAAO,EAAE,cAAc,EAAE,MAAM,aAAa,CAAA;AAuB5C,mGAAmG;AACnG,MAAM,UAAU,SAAS,CAAC,CAAS;IAClC,OAAO,CAAC;SACN,WAAW,EAAE;SACb,OAAO,CAAC,sBAAsB,EAAE,CAAC,EAAE,EAAE,CAAS,EAAE,CAAS,EAAE,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC,WAAW,EAAE,CAAC;SAClF,IAAI,EAAE,CAAA;AACT,CAAC;AAED,qGAAqG;AACrG,MAAM,UAAU,cAAc,CAAC,MAAc,EAAE,QAAgB,EAAE,QAAgB;IAChF,OAAO,GAAG,MAAM,IAAI,QAAQ,IAAI,QAAQ,EAAE,CAAC,WAAW,EAAE,CAAA;AACzD,CAAC;AAGD,KAAK,SAAS,CAAC,CAAC,UAAU,CAAC,IAAY;IACtC,KAAK,CAAC,CAAC,cAAc,CAAC,SAAS,CAAC,IAAI,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,MAAM,EAAE,IAAI,EAAE,CAAuB,CAAA;AAC9F,CAAC;AAED,oEAAoE;AACpE,KAAK,UAAU,OAAO,CAAI,KAAe,EAAE,MAAc,EAAE,IAAmB;IAC7E,MAAM,CAAC,GAAG,IAAI,GAAG,EAAa,CAAA;IAC9B,KAAK,MAAM,CAAC,IAAI,KAAK,EAAE,CAAC;QACvB,IAAI,KAAK,EAAE,MAAM,CAAC,IAAI,UAAU,CAAC,CAAC,CAAC,EAAE,CAAC;YACrC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;YACnB,IAAI,CAAC,IAAI,IAAI,IAAI,CAAC,KAAK,EAAE;gBAAE,CAAC,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC,CAAC,CAAC,CAAA;QACrD,CAAC;IACF,CAAC;IACD,OAAO,CAAC,CAAA;AACT,CAAC;AAED,4FAA4F;AAC5F,KAAK,UAAU,WAAW,CAAC,IAAY;IACtC,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAA;IAC9B,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IACzC,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC;QACrC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;YAAE,SAAQ;QAC1B,IAAI,CAAC;YACJ,MAAM,CAAC,GAAI,IAAI,CAAC,KAAK,CAAC,IAAI,CAA6C,CAAC,UAAU,CAAA;YAClF,IAAI,CAAC,EAAE,MAAM,IAAI,CAAC,EAAE,QAAQ,IAAI,CAAC,EAAE,QAAQ;gBAAE,IAAI,CAAC,GAAG,CAAC,cAAc,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAA;QACxG,CAAC;QAAC,MAAM,CAAC;YACR,oBAAoB;QACrB,CAAC;IACF,CAAC;IACD,OAAO,IAAI,CAAA;AACZ,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,YAAY,CAAC,IAAyB;IAC3D,MAAM,QAAQ,GAAG,IAAI,CAAC,UAAU,IAAI,CAAC,GAAG,EAAE,GAAE,CAAC,CAAC,CAAA;IAC9C,MAAM,KAAK,GAAG,MAAM,OAAO,CAAC,IAAI,CAAC,WAAW,CAAC,CAAA;IAC7C,MAAM,IAAI,GAAG,CAAC,EAAU,EAAE,OAAgB,EAAE,EAAE,CAC7C,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,EAAE,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,OAAO,IAAI,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,CAAA;IAEvG,+FAA+F;IAC/F,MAAM,WAAW,GAAG,IAAI,CAAC,4BAA4B,CAAC,CAAA;IACtD,MAAM,aAAa,GAAG,IAAI,CAAC,qBAAqB,EAAE,4BAA4B,CAAC,CAAA;IAC/E,MAAM,YAAY,GAAG,IAAI,CAAC,2BAA2B,CAAC,CAAA;IAEtD,MAAM,OAAO,GAAG,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,MAAM,WAAW,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC,IAAI,GAAG,EAAU,CAAA;IAC1F,IAAI,IAAI,CAAC,WAAW;QAAE,QAAQ,CAAC,uBAAuB,OAAO,CAAC,IAAI,EAAE,CAAC,CAAA;IAErE,QAAQ,CAAC,4BAA4B,WAAW,CAAC,MAAM,uBAAuB,aAAa,CAAC,MAAM,IAAI,CAAC,CAAA;IACvG,MAAM,SAAS,GAAG,MAAM,OAAO,CAAC,WAAW,EAAE,qBAAqB,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;QAC3E,IAAI,EAAE,MAAM,CAAC,CAAC,CAAC,WAAW,IAAI,EAAE,CAAC;QACjC,IAAI,EAAE,MAAM,CAAC,CAAC,CAAC,gBAAgB,IAAI,EAAE,CAAC;QACtC,MAAM,EAAE,MAAM,CAAC,CAAC,CAAC,kBAAkB,IAAI,EAAE,CAAC;KAC1C,CAAC,CAAC,CAAA;IACH,MAAM,WAAW,GAAG,MAAM,OAAO,CAAC,aAAa,EAAE,cAAc,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,aAAa,IAAI,EAAE,CAAC,CAAC,CAAA;IACtG,QAAQ,CAAC,WAAW,SAAS,CAAC,IAAI,CAAC,cAAc,EAAE,eAAe,WAAW,CAAC,IAAI,CAAC,cAAc,EAAE,EAAE,CAAC,CAAA;IAEtG,MAAM,SAAS,GACd,EAAE,CAAA;IACH,IAAI,IAAI,GAAG,CAAC,CAAA;IACZ,IAAI,OAAO,GAAG,CAAC,CAAA;IACf,KAAK,MAAM,CAAC,IAAI,YAAY,EAAE,CAAC;QAC9B,MAAM,KAAK,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,2BAA2B,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;QACnE,IAAI,KAAK,EAAE,MAAM,CAAC,IAAI,UAAU,CAAC,CAAC,CAAC,EAAE,CAAC;YACrC,MAAM,WAAW,GAAG,MAAM,CAAC,CAAC,CAAC,YAAY,IAAI,EAAE,CAAC,CAAA;YAChD,IAAI,CAAC,WAAW,IAAI,CAAC,CAAC,YAAY,IAAI,CAAC,CAAC,CAAC,QAAQ;gBAAE,SAAQ;YAC3D,MAAM,EAAE,GAAG,SAAS,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,mBAAmB,IAAI,EAAE,CAAC,CAAC,CAAA;YAC7D,MAAM,SAAS,GAAG,WAAW,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,YAAY,IAAI,EAAE,CAAC,CAAC,CAAA;YAC/D,IAAI,CAAC,EAAE,EAAE,IAAI,IAAI,CAAC,SAAS;gBAAE,SAAQ;YACrC,MAAM,MAAM,GAAG,GAAG,SAAS,CAAC,EAAE,CAAC,IAAI,CAAC,IAAI,SAAS,CAAC,EAAE,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,GAAG,GAAG,SAAS,CAAC,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,IAAI,EAAE,CAAA;YACjH,MAAM,QAAQ,GAAG,SAAS,CAAC,SAAS,CAAC,CAAA;YACrC,MAAM,QAAQ,GAAG,MAAM,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAA;YACnC,IAAI,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC,MAAM,EAAE,QAAQ,EAAE,QAAQ,CAAC,CAAC,EAAE,CAAC;gBAC7D,OAAO,EAAE,CAAA;gBACT,SAAQ;YACT,CAAC;YACD,IAAI,KAAK,GAAG,WAAW,GAAG,CAAC,CAAC,CAAC,mBAAmB,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,mBAAmB,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAA;YACtF,IAAI,CAAC,CAAC,WAAW;gBAAE,KAAK,GAAG,GAAG,KAAK,IAAI,MAAM,CAAC,CAAC,CAAC,WAAW,CAAC,EAAE,CAAA;YAC9D,IAAI,CAAC,CAAC,WAAW;gBAAE,KAAK,GAAG,GAAG,MAAM,CAAC,CAAC,CAAC,WAAW,CAAC,IAAI,KAAK,EAAE,CAAA;YAC9D,MAAM,KAAK,GAAG,EAAE,YAAY,EAAE,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,EAAE,KAAK,EAAE,QAAQ,EAAE,CAAA;YAChF,IAAI,EAAE,CAAA;YACN,IAAI,SAAS,CAAC,MAAM,GAAG,IAAI,CAAC,UAAU;gBAAE,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;iBACxD,CAAC;gBACL,MAAM,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,CAAA;gBAC1C,IAAI,CAAC,GAAG,IAAI,CAAC,UAAU;oBAAE,SAAS,CAAC,CAAC,CAAC,GAAG,KAAK,CAAA;YAC9C,CAAC;QACF,CAAC;QACD,QAAQ,CAAC,GAAG,KAAK,KAAK,IAAI,CAAC,cAAc,EAAE,sBAAsB,CAAC,CAAA;IACnE,CAAC;IAED,MAAM,GAAG,GAAG,iBAAiB,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;IACvC,MAAM,OAAO,GAA2B,EAAE,CAAA;IAC1C,KAAK,MAAM,CAAC,IAAI,SAAS,EAAE,CAAC;QAC3B,GAAG,CAAC,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,CAAA;QACnC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAA;IACjD,CAAC;IACD,MAAM,IAAI,OAAO,CAAO,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,CAAA;IAC9C,QAAQ,CAAC,SAAS,SAAS,CAAC,MAAM,CAAC,cAAc,EAAE,aAAa,IAAI,CAAC,GAAG,EAAE,CAAC,CAAA;IAC3E,OAAO,EAAE,OAAO,EAAE,SAAS,CAAC,MAAM,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,CAAA;AAC7D,CAAC"}

package/out/src/adapters/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/adapters/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAGH,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,aAAa,CAAA;AAoBhD;;;;;;;;;;GAUG;AACH,eAAO,MAAM,gBAAgB,EAAE,SAAS,aAAa,EAmBpD,CAAA;AAQD,OAAO,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7D,OAAO,EAAE,kBAAkB,EAAE,uBAAuB,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACjG,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,qBAAqB,GACrB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EAAE,mBAAmB,EAAE,wBAAwB,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACtG,OAAO,EACN,wBAAwB,EACxB,6BAA6B,EAC7B,oBAAoB,GACpB,MAAM,4BAA4B,CAAA;AACnC,OAAO,EACN,2BAA2B,EAC3B,gCAAgC,EAChC,qBAAqB,GACrB,MAAM,+BAA+B,CAAA;AACtC,OAAO,EACN,+BAA+B,EAC/B,oCAAoC,EACpC,yBAAyB,GACzB,MAAM,mCAAmC,CAAA;AAC1C,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAC1F,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,oBAAoB,GACpB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EACN,yBAAyB,EACzB,8BAA8B,EAC9B,mBAAmB,GACnB,MAAM,6BAA6B,CAAA;AACpC,OAAO,EAAE,wBAAwB,EAAE,6BAA6B,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AACxH,OAAO,EAAE,oBAAoB,EAAE,yBAAyB,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACzG,OAAO,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AACjH,OAAO,EACN,uBAAuB,EACvB,4BAA4B,EAC5B,kCAAkC,GAClC,MAAM,6CAA6C,CAAA;AACpD,OAAO,EAAE,oBAAoB,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AACnF,OAAO,EAAE,yBAAyB,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/adapters/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAGH,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,aAAa,CAAA;AAqBhD;;;;;;;;;;GAUG;AACH,eAAO,MAAM,gBAAgB,EAAE,SAAS,aAAa,EAoBpD,CAAA;AAQD,OAAO,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7D,OAAO,EAAE,kBAAkB,EAAE,uBAAuB,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACjG,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,qBAAqB,GACrB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EAAE,mBAAmB,EAAE,wBAAwB,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACtG,OAAO,EACN,wBAAwB,EACxB,6BAA6B,EAC7B,oBAAoB,GACpB,MAAM,4BAA4B,CAAA;AACnC,OAAO,EACN,2BAA2B,EAC3B,gCAAgC,EAChC,qBAAqB,GACrB,MAAM,+BAA+B,CAAA;AACtC,OAAO,EACN,+BAA+B,EAC/B,oCAAoC,EACpC,yBAAyB,GACzB,MAAM,mCAAmC,CAAA;AAC1C,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAC1F,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,oBAAoB,GACpB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EACN,yBAAyB,EACzB,8BAA8B,EAC9B,mBAAmB,GACnB,MAAM,6BAA6B,CAAA;AACpC,OAAO,EAAE,wBAAwB,EAAE,6BAA6B,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AACxH,OAAO,EAAE,oBAAoB,EAAE,yBAAyB,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACzG,OAAO,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AACjH,OAAO,EACN,uBAAuB,EACvB,4BAA4B,EAC5B,kCAAkC,GAClC,MAAM,6CAA6C,CAAA;AACpD,OAAO,EAAE,oBAAoB,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AACnF,OAAO,EAAE,yBAAyB,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA"}

package/out/src/adapters/index.js

@@ -23,6 +23,7 @@ import { banAdapter } from "./ban/adapter.js";
 import { fccBdcAdapter } from "./fcc-bdc/adapter.js";
 import { geonamesPostalAdapter } from "./geonames-postal/adapter.js";
 import { geonamesAdapter } from "./geonames/adapter.js";
+import { gnafAdapter } from "./gnaf/adapter.js";
 import { openaddressesAdapter } from "./openaddresses/adapter.js";
 import { overtureAdapter } from "./overture/adapter.js";
 import { stateHiSchoolsAdapter } from "./state-hi-schools/adapter.js";
@@ -57,6 +58,7 @@ export const BUILTIN_ADAPTERS = [
     tigerAdapter,
     openaddressesAdapter,
     overtureAdapter,
+    gnafAdapter,
     fccBdcAdapter,
     usgovHrsaFqhcAdapter,
     usgovNppesAdapter,

package/out/src/adapters/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/adapters/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAEH,OAAO,EAAE,sBAAsB,EAAE,MAAM,eAAe,CAAA;AAEtD,OAAO,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7C,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACpD,OAAO,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAA;AACpE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,oBAAoB,EAAE,MAAM,4BAA4B,CAAA;AACjE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,qBAAqB,EAAE,MAAM,+BAA+B,CAAA;AACrE,OAAO,EAAE,yBAAyB,EAAE,MAAM,mCAAmC,CAAA;AAC7E,OAAO,EAAE,sBAAsB,EAAE,MAAM,gCAAgC,CAAA;AACvE,OAAO,EAAE,sBAAsB,EAAE,MAAM,gCAAgC,CAAA;AACvE,OAAO,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AACjD,OAAO,EAAE,oBAAoB,EAAE,MAAM,8BAA8B,CAAA;AACnE,OAAO,EAAE,mBAAmB,EAAE,MAAM,6BAA6B,CAAA;AACjE,OAAO,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AAC/D,OAAO,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACxD,OAAO,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AAC5D,OAAO,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AAC7D,OAAO,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA;AAEvE;;;;;;;;;;GAUG;AACH,MAAM,CAAC,MAAM,gBAAgB,GAA6B;IACzD,eAAe;IACf,oBAAoB;IACpB,eAAe;IACf,qBAAqB;IACrB,UAAU;IACV,YAAY;IACZ,oBAAoB;IACpB,eAAe;IACf,aAAa;IACb,oBAAoB;IACpB,iBAAiB;IACjB,eAAe;IACf,mBAAmB;IACnB,kBAAkB;IAClB,yBAAyB;IACzB,sBAAsB;IACtB,sBAAsB;IACtB,qBAAqB;CACrB,CAAA;AAED,KAAK,MAAM,OAAO,IAAI,gBAAgB,EAAE,CAAC;IACxC,IAAI,CAAC,sBAAsB,CAAC,GAAG,CAAC,OAAO,CAAC,EAAE,CAAC,EAAE,CAAC;QAC7C,sBAAsB,CAAC,QAAQ,CAAC,OAAO,CAAC,CAAA;IACzC,CAAC;AACF,CAAC;AAED,OAAO,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7D,OAAO,EAAE,kBAAkB,EAAE,uBAAuB,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACjG,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,qBAAqB,GACrB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EAAE,mBAAmB,EAAE,wBAAwB,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACtG,OAAO,EACN,wBAAwB,EACxB,6BAA6B,EAC7B,oBAAoB,GACpB,MAAM,4BAA4B,CAAA;AACnC,OAAO,EACN,2BAA2B,EAC3B,gCAAgC,EAChC,qBAAqB,GACrB,MAAM,+BAA+B,CAAA;AACtC,OAAO,EACN,+BAA+B,EAC/B,oCAAoC,EACpC,yBAAyB,GACzB,MAAM,mCAAmC,CAAA;AAC1C,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAC1F,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,oBAAoB,GACpB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EACN,yBAAyB,EACzB,8BAA8B,EAC9B,mBAAmB,GACnB,MAAM,6BAA6B,CAAA;AACpC,OAAO,EAAE,wBAAwB,EAAE,6BAA6B,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AACxH,OAAO,EAAE,oBAAoB,EAAE,yBAAyB,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACzG,OAAO,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AACjH,OAAO,EACN,uBAAuB,EACvB,4BAA4B,EAC5B,kCAAkC,GAClC,MAAM,6CAA6C,CAAA;AACpD,OAAO,EAAE,oBAAoB,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AACnF,OAAO,EAAE,yBAAyB,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/adapters/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AAEH,OAAO,EAAE,sBAAsB,EAAE,MAAM,eAAe,CAAA;AAEtD,OAAO,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7C,OAAO,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACpD,OAAO,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAA;AACpE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,WAAW,EAAE,MAAM,mBAAmB,CAAA;AAC/C,OAAO,EAAE,oBAAoB,EAAE,MAAM,4BAA4B,CAAA;AACjE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,qBAAqB,EAAE,MAAM,+BAA+B,CAAA;AACrE,OAAO,EAAE,yBAAyB,EAAE,MAAM,mCAAmC,CAAA;AAC7E,OAAO,EAAE,sBAAsB,EAAE,MAAM,gCAAgC,CAAA;AACvE,OAAO,EAAE,sBAAsB,EAAE,MAAM,gCAAgC,CAAA;AACvE,OAAO,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AACjD,OAAO,EAAE,oBAAoB,EAAE,MAAM,8BAA8B,CAAA;AACnE,OAAO,EAAE,mBAAmB,EAAE,MAAM,6BAA6B,CAAA;AACjE,OAAO,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AAC/D,OAAO,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACxD,OAAO,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AAC5D,OAAO,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AAC7D,OAAO,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA;AAEvE;;;;;;;;;;GAUG;AACH,MAAM,CAAC,MAAM,gBAAgB,GAA6B;IACzD,eAAe;IACf,oBAAoB;IACpB,eAAe;IACf,qBAAqB;IACrB,UAAU;IACV,YAAY;IACZ,oBAAoB;IACpB,eAAe;IACf,WAAW;IACX,aAAa;IACb,oBAAoB;IACpB,iBAAiB;IACjB,eAAe;IACf,mBAAmB;IACnB,kBAAkB;IAClB,yBAAyB;IACzB,sBAAsB;IACtB,sBAAsB;IACtB,qBAAqB;CACrB,CAAA;AAED,KAAK,MAAM,OAAO,IAAI,gBAAgB,EAAE,CAAC;IACxC,IAAI,CAAC,sBAAsB,CAAC,GAAG,CAAC,OAAO,CAAC,EAAE,CAAC,EAAE,CAAC;QAC7C,sBAAsB,CAAC,QAAQ,CAAC,OAAO,CAAC,CAAA;IACzC,CAAC;AACF,CAAC;AAED,OAAO,EAAE,cAAc,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAA;AAC7D,OAAO,EAAE,kBAAkB,EAAE,uBAAuB,EAAE,aAAa,EAAE,MAAM,sBAAsB,CAAA;AACjG,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,qBAAqB,GACrB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EAAE,mBAAmB,EAAE,wBAAwB,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACtG,OAAO,EACN,wBAAwB,EACxB,6BAA6B,EAC7B,oBAAoB,GACpB,MAAM,4BAA4B,CAAA;AACnC,OAAO,EACN,2BAA2B,EAC3B,gCAAgC,EAChC,qBAAqB,GACrB,MAAM,+BAA+B,CAAA;AACtC,OAAO,EACN,+BAA+B,EAC/B,oCAAoC,EACpC,yBAAyB,GACzB,MAAM,mCAAmC,CAAA;AAC1C,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EACN,4BAA4B,EAC5B,iCAAiC,EACjC,sBAAsB,GACtB,MAAM,gCAAgC,CAAA;AACvC,OAAO,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAC1F,OAAO,EACN,0BAA0B,EAC1B,+BAA+B,EAC/B,oBAAoB,GACpB,MAAM,8BAA8B,CAAA;AACrC,OAAO,EACN,yBAAyB,EACzB,8BAA8B,EAC9B,mBAAmB,GACnB,MAAM,6BAA6B,CAAA;AACpC,OAAO,EAAE,wBAAwB,EAAE,6BAA6B,EAAE,kBAAkB,EAAE,MAAM,4BAA4B,CAAA;AACxH,OAAO,EAAE,oBAAoB,EAAE,yBAAyB,EAAE,eAAe,EAAE,MAAM,wBAAwB,CAAA;AACzG,OAAO,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,iBAAiB,EAAE,MAAM,0BAA0B,CAAA;AACjH,OAAO,EACN,uBAAuB,EACvB,4BAA4B,EAC5B,kCAAkC,GAClC,MAAM,6CAA6C,CAAA;AACpD,OAAO,EAAE,oBAAoB,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAA;AACnF,OAAO,EAAE,yBAAyB,EAAE,oBAAoB,EAAE,MAAM,kCAAkC,CAAA"}

package/out/src/index.d.ts

@@ -4,6 +4,7 @@
  * @author Teffen Ellis, et al.
  */
 export * from "./adapter.js";
+export * from "./adapters/gnaf/assemble.js";
 export * from "./adapters/index.js";
 export * from "./align.js";
 export * from "./build.js";

package/out/src/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH,cAAc,cAAc,CAAA;AAC5B,cAAc,qBAAqB,CAAA;AACnC,cAAc,YAAY,CAAA;AAC1B,cAAc,YAAY,CAAA;AAC1B,cAAc,aAAa,CAAA;AAC3B,cAAc,aAAa,CAAA;AAC3B,cAAc,cAAc,CAAA;AAC5B,cAAc,aAAa,CAAA;AAC3B,cAAc,YAAY,CAAA;AAC1B,cAAc,wBAAwB,CAAA;AACtC,cAAc,6BAA6B,CAAA;AAC3C,cAAc,8BAA8B,CAAA;AAC5C,cAAc,2BAA2B,CAAA;AACzC,cAAc,wBAAwB,CAAA;AACtC,cAAc,wBAAwB,CAAA;AACtC,cAAc,iBAAiB,CAAA;AAC/B,cAAc,eAAe,CAAA;AAC7B,cAAc,YAAY,CAAA"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH,cAAc,cAAc,CAAA;AAC5B,cAAc,6BAA6B,CAAA;AAC3C,cAAc,qBAAqB,CAAA;AACnC,cAAc,YAAY,CAAA;AAC1B,cAAc,YAAY,CAAA;AAC1B,cAAc,aAAa,CAAA;AAC3B,cAAc,aAAa,CAAA;AAC3B,cAAc,cAAc,CAAA;AAC5B,cAAc,aAAa,CAAA;AAC3B,cAAc,YAAY,CAAA;AAC1B,cAAc,wBAAwB,CAAA;AACtC,cAAc,6BAA6B,CAAA;AAC3C,cAAc,8BAA8B,CAAA;AAC5C,cAAc,2BAA2B,CAAA;AACzC,cAAc,wBAAwB,CAAA;AACtC,cAAc,wBAAwB,CAAA;AACtC,cAAc,iBAAiB,CAAA;AAC/B,cAAc,eAAe,CAAA;AAC7B,cAAc,YAAY,CAAA"}

package/out/src/index.js

@@ -4,6 +4,7 @@
  * @author Teffen Ellis, et al.
  */
 export * from "./adapter.js";
+export * from "./adapters/gnaf/assemble.js";
 export * from "./adapters/index.js";
 export * from "./align.js";
 export * from "./build.js";

package/out/src/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH,cAAc,cAAc,CAAA;AAC5B,cAAc,qBAAqB,CAAA;AACnC,cAAc,YAAY,CAAA;AAC1B,cAAc,YAAY,CAAA;AAC1B,cAAc,aAAa,CAAA;AAC3B,cAAc,aAAa,CAAA;AAC3B,cAAc,cAAc,CAAA;AAC5B,cAAc,aAAa,CAAA;AAC3B,cAAc,YAAY,CAAA;AAC1B,cAAc,wBAAwB,CAAA;AACtC,cAAc,6BAA6B,CAAA;AAC3C,cAAc,8BAA8B,CAAA;AAC5C,cAAc,2BAA2B,CAAA;AACzC,cAAc,wBAAwB,CAAA;AACtC,cAAc,wBAAwB,CAAA;AACtC,cAAc,iBAAiB,CAAA;AAC/B,cAAc,eAAe,CAAA;AAC7B,cAAc,YAAY,CAAA"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH,cAAc,cAAc,CAAA;AAC5B,cAAc,6BAA6B,CAAA;AAC3C,cAAc,qBAAqB,CAAA;AACnC,cAAc,YAAY,CAAA;AAC1B,cAAc,YAAY,CAAA;AAC1B,cAAc,aAAa,CAAA;AAC3B,cAAc,aAAa,CAAA;AAC3B,cAAc,cAAc,CAAA;AAC5B,cAAc,aAAa,CAAA;AAC3B,cAAc,YAAY,CAAA;AAC1B,cAAc,wBAAwB,CAAA;AACtC,cAAc,6BAA6B,CAAA;AAC3C,cAAc,8BAA8B,CAAA;AAC5C,cAAc,2BAA2B,CAAA;AACzC,cAAc,wBAAwB,CAAA;AACtC,cAAc,wBAAwB,CAAA;AACtC,cAAc,iBAAiB,CAAA;AAC/B,cAAc,eAAe,CAAA;AAC7B,cAAc,YAAY,CAAA"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mailwoman/corpus",
-  "version": "4.12.0",
+  "version": "4.14.0",
   "description": "Mailwoman corpus pipeline: BIO-labeled dataset builder for the neural classifier.",
   "license": "AGPL-3.0-only",
   "repository": {
@@ -17,14 +17,14 @@
   },
   "dependencies": {
     "@dsnp/parquetjs": "1.8.7",
-    "@mailwoman/codex": "4.12.0",
-    "@mailwoman/core": "4.12.0",
-    "@mailwoman/formatter": "4.12.0",
+    "@mailwoman/codex": "4.14.0",
+    "@mailwoman/core": "4.14.0",
+    "@mailwoman/formatter": "4.14.0",
     "csv-parse": "^5.6.0",
     "fast-glob": "^3.3.3",
     "fastest-levenshtein": "^1.0.16",
     "lru-cache": "^10.4.3",
-    "spliterator": "^2.0.0"
+    "spliterator": "^2.2.1"
   },
   "devDependencies": {
     "@types/node": "^25.9.2"