@mailwoman/corpus 9.4.0 → 10.1.0

package/out/us/adapters/state/ny-notaries/adapter.d.ts

@@ -0,0 +1,33 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `state-ny-notaries`: New York Commissioned Notaries CSV consumer.
+ *
+ *   The New York Department of State publishes a registry of commissioned notaries public. Each row
+ *   optionally includes a business name and business address (~1-5% fill rate).
+ *
+ *   The adapter consumes the CSV the operator pre-downloads via `fetch-state-sources.ts`. Column
+ *   names match the data.ny.gov export header (note: some columns have leading spaces).
+ *
+ *   License: stamped `"Public Domain"` per New York state government open-data terms.
+ */
+import { type CorpusAdapter } from "#types";
+/**
+ * Registry id for this adapter.
+ *
+ * Stamped into every row it emits, so a corpus record can be traced back to the dataset it came from.
+ */
+export declare const STATE_NY_NOTARIES_ADAPTER_ID = "state-ny-notaries";
+/**
+ * License assigned by this source (Public Domain), attached to each row so downstream
+ * consumers inherit the terms rather than having to look them up.
+ */
+export declare const STATE_NY_NOTARIES_DEFAULT_LICENSE = "Public Domain";
+export declare function createStateNyNotariesAdapter(): CorpusAdapter;
+/**
+ * The configured adapter instance registered with the corpus builder.
+ */
+export declare const stateNyNotariesAdapter: CorpusAdapter;
+//# sourceMappingURL=adapter.d.ts.map

package/out/us/adapters/state/ny-notaries/adapter.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../../lib/us/adapters/state/ny-notaries/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AASH,OAAO,EAAuD,KAAK,aAAa,EAAiB,MAAM,QAAQ,CAAA;AAG/G;;;;GAIG;AACH,eAAO,MAAM,4BAA4B,sBAAsB,CAAA;AAC/D;;;GAGG;AACH,eAAO,MAAM,iCAAiC,kBAAkB,CAAA;AAEhE,wBAAgB,4BAA4B,IAAI,aAAa,CAwF5D;AAED;;GAEG;AACH,eAAO,MAAM,sBAAsB,eAAiC,CAAA"}

package/out/us/adapters/state/ny-notaries/adapter.js

@@ -0,0 +1,114 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `state-ny-notaries`: New York Commissioned Notaries CSV consumer.
+ *
+ *   The New York Department of State publishes a registry of commissioned notaries public. Each row
+ *   optionally includes a business name and business address (~1-5% fill rate).
+ *
+ *   The adapter consumes the CSV the operator pre-downloads via `fetch-state-sources.ts`. Column
+ *   names match the data.ny.gov export header (note: some columns have leading spaces).
+ *
+ *   License: stamped `"Public Domain"` per New York state government open-data terms.
+ */
+import { formatAddressRow } from "@mailwoman/codex/address/format";
+import { isPresent } from "@mailwoman/core/objects";
+import { CSVSpliterator } from "spliterator";
+import { stableSourceID } from "#adapters/source-id";
+import { splitStreetLine } from "#adapters/street-line";
+import { SourceRegister } from "#registers";
+import { AddressRole, SurfaceOrigin } from "#types";
+import { lookupStateAbbreviation } from "#us/fips-state";
+/**
+ * Registry id for this adapter.
+ *
+ * Stamped into every row it emits, so a corpus record can be traced back to the dataset it came from.
+ */
+export const STATE_NY_NOTARIES_ADAPTER_ID = "state-ny-notaries";
+/**
+ * License assigned by this source (Public Domain), attached to each row so downstream
+ * consumers inherit the terms rather than having to look them up.
+ */
+export const STATE_NY_NOTARIES_DEFAULT_LICENSE = "Public Domain";
+export function createStateNyNotariesAdapter() {
+    return {
+        id: STATE_NY_NOTARIES_ADAPTER_ID,
+        defaultLicense: STATE_NY_NOTARIES_DEFAULT_LICENSE,
+        addressRole: AddressRole.Practice,
+        register: SourceRegister.NewYorkNotaries,
+        surface: SurfaceOrigin.Rendered,
+        description: "New York Commissioned Notaries — name + optional business address (public-domain).",
+        async *rows(opts) {
+            if (opts.country && opts.country !== "US") {
+                throw new Error(`state-ny-notaries adapter: only US supported, got country=${opts.country}`);
+            }
+            const rows = CSVSpliterator.fromAsync(opts.inputPath, {
+                normalizeKeys: false,
+            });
+            let emitted = 0;
+            for await (const record of rows) {
+                if (opts.signal?.aborted)
+                    break;
+                if (opts.limit !== undefined && emitted >= opts.limit)
+                    break;
+                const holderName = record["Commission Holder Name"] ?? "";
+                const businessName = record["Business Name (if available)"] ?? "";
+                const address1 = record["Business Address 1 (if available)"] ?? "";
+                const address2 = record["Business Address 2 (if available)"] ?? "";
+                const city = record["Business City (if available)"] ?? "";
+                const stateAbbr = record["Business State (if available)"] ?? "";
+                const zip = record["Business Zip (if available)"] ?? "";
+                const county = record["Commissioned County"] ?? "";
+                if (!city || !stateAbbr || !zip)
+                    continue;
+                if (!address1 && !address2)
+                    continue;
+                const state = lookupStateAbbreviation(stateAbbr);
+                if (!state)
+                    continue;
+                const fullAddress = [address1, address2].filter(isPresent).join(" ");
+                const split = splitStreetLine(fullAddress);
+                if (!split)
+                    continue;
+                const venue = businessName || holderName || undefined;
+                const components = {
+                    ...(venue ? { venue } : {}),
+                    ...(split.house_number ? { house_number: split.house_number } : {}),
+                    street: split.street,
+                    locality: city,
+                    region: state.abbreviation,
+                    postcode: zip,
+                    ...(county ? { subregion: county } : {}),
+                };
+                const rendered = formatAddressRow(components, "US", { singleLine: true });
+                if (!rendered)
+                    continue;
+                const { raw, components: aligned } = rendered;
+                if (Object.keys(aligned).length <= 2)
+                    continue;
+                const commNum = record["Commission Number (UID)"] ?? "";
+                const sourceID = commNum
+                    ? `${STATE_NY_NOTARIES_ADAPTER_ID}-${commNum}`
+                    : stableSourceID(STATE_NY_NOTARIES_ADAPTER_ID, aligned);
+                yield {
+                    raw,
+                    components: aligned,
+                    country: "US",
+                    locale: "en-US",
+                    source: STATE_NY_NOTARIES_ADAPTER_ID,
+                    source_id: sourceID,
+                    corpus_version: "",
+                    license: STATE_NY_NOTARIES_DEFAULT_LICENSE,
+                };
+                emitted++;
+            }
+        },
+    };
+}
+/**
+ * The configured adapter instance registered with the corpus builder.
+ */
+export const stateNyNotariesAdapter = createStateNyNotariesAdapter();
+//# sourceMappingURL=adapter.js.map

package/out/us/adapters/state/ny-notaries/adapter.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../../lib/us/adapters/state/ny-notaries/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,iCAAiC,CAAA;AAClE,OAAO,EAAE,SAAS,EAAE,MAAM,yBAAyB,CAAA;AACnD,OAAO,EAAE,cAAc,EAAE,MAAM,aAAa,CAAA;AAE5C,OAAO,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAA;AACpD,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,cAAc,EAAE,MAAM,YAAY,CAAA;AAC3C,OAAO,EAAE,WAAW,EAA8D,aAAa,EAAE,MAAM,QAAQ,CAAA;AAC/G,OAAO,EAAE,uBAAuB,EAAE,MAAM,gBAAgB,CAAA;AAExD;;;;GAIG;AACH,MAAM,CAAC,MAAM,4BAA4B,GAAG,mBAAmB,CAAA;AAC/D;;;GAGG;AACH,MAAM,CAAC,MAAM,iCAAiC,GAAG,eAAe,CAAA;AAEhE,MAAM,UAAU,4BAA4B;IAC3C,OAAO;QACN,EAAE,EAAE,4BAA4B;QAChC,cAAc,EAAE,iCAAiC;QACjD,WAAW,EAAE,WAAW,CAAC,QAAQ;QACjC,QAAQ,EAAE,cAAc,CAAC,eAAe;QACxC,OAAO,EAAE,aAAa,CAAC,QAAQ;QAC/B,WAAW,EAAE,oFAAoF;QAEjG,KAAK,CAAC,CAAC,IAAI,CAAC,IAAoB;YAC/B,IAAI,IAAI,CAAC,OAAO,IAAI,IAAI,CAAC,OAAO,KAAK,IAAI,EAAE,CAAC;gBAC3C,MAAM,IAAI,KAAK,CAAC,6DAA6D,IAAI,CAAC,OAAO,EAAE,CAAC,CAAA;YAC7F,CAAC;YAED,MAAM,IAAI,GAAG,cAAc,CAAC,SAAS,CAAC,IAAI,CAAC,SAAS,EAAE;gBACrD,aAAa,EAAE,KAAK;aACpB,CAAC,CAAA;YAEF,IAAI,OAAO,GAAG,CAAC,CAAA;YAEf,IAAI,KAAK,EAAE,MAAM,MAAM,IAAI,IAA6C,EAAE,CAAC;gBAC1E,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;oBAAE,MAAK;gBAE/B,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;oBAAE,MAAK;gBAE5D,MAAM,UAAU,GAAG,MAAM,CAAC,wBAAwB,CAAC,IAAI,EAAE,CAAA;gBACzD,MAAM,YAAY,GAAG,MAAM,CAAC,8BAA8B,CAAC,IAAI,EAAE,CAAA;gBACjE,MAAM,QAAQ,GAAG,MAAM,CAAC,mCAAmC,CAAC,IAAI,EAAE,CAAA;gBAClE,MAAM,QAAQ,GAAG,MAAM,CAAC,mCAAmC,CAAC,IAAI,EAAE,CAAA;gBAClE,MAAM,IAAI,GAAG,MAAM,CAAC,8BAA8B,CAAC,IAAI,EAAE,CAAA;gBACzD,MAAM,SAAS,GAAG,MAAM,CAAC,+BAA+B,CAAC,IAAI,EAAE,CAAA;gBAC/D,MAAM,GAAG,GAAG,MAAM,CAAC,6BAA6B,CAAC,IAAI,EAAE,CAAA;gBACvD,MAAM,MAAM,GAAG,MAAM,CAAC,qBAAqB,CAAC,IAAI,EAAE,CAAA;gBAElD,IAAI,CAAC,IAAI,IAAI,CAAC,SAAS,IAAI,CAAC,GAAG;oBAAE,SAAQ;gBAEzC,IAAI,CAAC,QAAQ,IAAI,CAAC,QAAQ;oBAAE,SAAQ;gBAEpC,MAAM,KAAK,GAAG,uBAAuB,CAAC,SAAS,CAAC,CAAA;gBAEhD,IAAI,CAAC,KAAK;oBAAE,SAAQ;gBAEpB,MAAM,WAAW,GAAG,CAAC,QAAQ,EAAE,QAAQ,CAAC,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;gBACpE,MAAM,KAAK,GAAG,eAAe,CAAC,WAAW,CAAC,CAAA;gBAE1C,IAAI,CAAC,KAAK;oBAAE,SAAQ;gBAEpB,MAAM,KAAK,GAAG,YAAY,IAAI,UAAU,IAAI,SAAS,CAAA;gBAErD,MAAM,UAAU,GAA+B;oBAC9C,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;oBAC3B,GAAG,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,KAAK,CAAC,YAAY,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;oBACnE,MAAM,EAAE,KAAK,CAAC,MAAM;oBACpB,QAAQ,EAAE,IAAI;oBACd,MAAM,EAAE,KAAK,CAAC,YAAY;oBAC1B,QAAQ,EAAE,GAAG;oBACb,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,SAAS,EAAE,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;iBACxC,CAAA;gBAED,MAAM,QAAQ,GAAG,gBAAgB,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;gBAEzE,IAAI,CAAC,QAAQ;oBAAE,SAAQ;gBAEvB,MAAM,EAAE,GAAG,EAAE,UAAU,EAAE,OAAO,EAAE,GAAG,QAAQ,CAAA;gBAE7C,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,IAAI,CAAC;oBAAE,SAAQ;gBAE9C,MAAM,OAAO,GAAG,MAAM,CAAC,yBAAyB,CAAC,IAAI,EAAE,CAAA;gBAEvD,MAAM,QAAQ,GAAG,OAAO;oBACvB,CAAC,CAAC,GAAG,4BAA4B,IAAI,OAAO,EAAE;oBAC9C,CAAC,CAAC,cAAc,CAAC,4BAA4B,EAAE,OAAO,CAAC,CAAA;gBAExD,MAAM;oBACL,GAAG;oBACH,UAAU,EAAE,OAAO;oBACnB,OAAO,EAAE,IAAI;oBACb,MAAM,EAAE,OAAO;oBACf,MAAM,EAAE,4BAA4B;oBACpC,SAAS,EAAE,QAAQ;oBACnB,cAAc,EAAE,EAAE;oBAClB,OAAO,EAAE,iCAAiC;iBAC1C,CAAA;gBAED,OAAO,EAAE,CAAA;YACV,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,MAAM,sBAAsB,GAAG,4BAA4B,EAAE,CAAA"}

package/out/us/adapters/state/tx-notaries/adapter.d.ts

@@ -0,0 +1,34 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `state-tx-notaries`: Texas Notary Public Commissions CSV consumer.
+ *
+ *   The Texas Secretary of State publishes a registry of commissioned notaries public. Each row
+ *   optionally includes a mailing address in free-form text (often multi-line with embedded
+ *   city/state/zip). Address fill rate is ~5-10%.
+ *
+ *   The adapter parses the embedded `Address` field for city/state/zip using a trailing `"city, ST
+ *   ZIP"` pattern.
+ *
+ *   License: stamped `"Public Domain"` per Texas state government open-data terms.
+ */
+import { type CorpusAdapter } from "#types";
+/**
+ * Registry id for this adapter.
+ *
+ * Stamped into every row it emits, so a corpus record can be traced back to the dataset it came from.
+ */
+export declare const STATE_TX_NOTARIES_ADAPTER_ID = "state-tx-notaries";
+/**
+ * License assigned by this source (Public Domain), attached to each row so downstream
+ * consumers inherit the terms rather than having to look them up.
+ */
+export declare const STATE_TX_NOTARIES_DEFAULT_LICENSE = "Public Domain";
+export declare function createStateTxNotariesAdapter(): CorpusAdapter;
+/**
+ * The configured adapter instance registered with the corpus builder.
+ */
+export declare const stateTxNotariesAdapter: CorpusAdapter;
+//# sourceMappingURL=adapter.d.ts.map

package/out/us/adapters/state/tx-notaries/adapter.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../../lib/us/adapters/state/tx-notaries/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AASH,OAAO,EAAuD,KAAK,aAAa,EAAiB,MAAM,QAAQ,CAAA;AAG/G;;;;GAIG;AACH,eAAO,MAAM,4BAA4B,sBAAsB,CAAA;AAC/D;;;GAGG;AACH,eAAO,MAAM,iCAAiC,kBAAkB,CAAA;AAchE,wBAAgB,4BAA4B,IAAI,aAAa,CAkG5D;AAED;;GAEG;AACH,eAAO,MAAM,sBAAsB,eAAiC,CAAA"}

package/out/us/adapters/state/tx-notaries/adapter.js

@@ -0,0 +1,125 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `state-tx-notaries`: Texas Notary Public Commissions CSV consumer.
+ *
+ *   The Texas Secretary of State publishes a registry of commissioned notaries public. Each row
+ *   optionally includes a mailing address in free-form text (often multi-line with embedded
+ *   city/state/zip). Address fill rate is ~5-10%.
+ *
+ *   The adapter parses the embedded `Address` field for city/state/zip using a trailing `"city, ST
+ *   ZIP"` pattern.
+ *
+ *   License: stamped `"Public Domain"` per Texas state government open-data terms.
+ */
+import { formatAddressRow } from "@mailwoman/codex/address/format";
+import { formatPersonName } from "@mailwoman/record/name";
+import { CSVSpliterator } from "spliterator";
+import { stableSourceID } from "#adapters/source-id";
+import { splitStreetLine } from "#adapters/street-line";
+import { SourceRegister } from "#registers";
+import { AddressRole, SurfaceOrigin } from "#types";
+import { lookupStateAbbreviation } from "#us/fips-state";
+/**
+ * Registry id for this adapter.
+ *
+ * Stamped into every row it emits, so a corpus record can be traced back to the dataset it came from.
+ */
+export const STATE_TX_NOTARIES_ADAPTER_ID = "state-tx-notaries";
+/**
+ * License assigned by this source (Public Domain), attached to each row so downstream
+ * consumers inherit the terms rather than having to look them up.
+ */
+export const STATE_TX_NOTARIES_DEFAULT_LICENSE = "Public Domain";
+/**
+ * Match trailing "city, ST ZIP" or "city, ST" at the end of an address line.
+ */
+const CITY_STATE_ZIP_SUFFIX = /[,]?\s*([^,]+),\s*([A-Z]{2})\s*(\d{5}(?:-\d{4})?)?\s*$/i;
+export function createStateTxNotariesAdapter() {
+    return {
+        id: STATE_TX_NOTARIES_ADAPTER_ID,
+        defaultLicense: STATE_TX_NOTARIES_DEFAULT_LICENSE,
+        addressRole: AddressRole.Mailing,
+        register: SourceRegister.TexasNotaries,
+        surface: SurfaceOrigin.Rendered,
+        description: "Texas Notary Public Commissions — name + mailing address with embedded city/state/zip (public-domain).",
+        async *rows(opts) {
+            if (opts.country && opts.country !== "US") {
+                throw new Error(`state-tx-notaries adapter: only US supported, got country=${opts.country}`);
+            }
+            const rows = CSVSpliterator.fromAsync(opts.inputPath, {
+                normalizeKeys: false,
+            });
+            let emitted = 0;
+            for await (const record of rows) {
+                if (opts.signal?.aborted)
+                    break;
+                if (opts.limit !== undefined && emitted >= opts.limit)
+                    break;
+                const rawAddress = record.Address ?? "";
+                if (!rawAddress)
+                    continue;
+                const firstName = record["First Name"] ?? "";
+                const lastName = record["Last Name"] ?? "";
+                const notaryID = record["Notary ID"] ?? "";
+                // Parse embedded city/state/zip from the trailing portion of the address.
+                // Addresses look like: "1215 mcmillan DR\nCEDAR hill, TX 75104"
+                const addrSingleLine = rawAddress.replaceAll("\n", ", ");
+                const cszMatch = CITY_STATE_ZIP_SUFFIX.exec(addrSingleLine);
+                if (!cszMatch)
+                    continue;
+                const city = (cszMatch[1] ?? "").trim();
+                const stateAbbr = (cszMatch[2] ?? "").trim();
+                const zip = (cszMatch[3] ?? "").trim();
+                if (!city || !stateAbbr)
+                    continue;
+                const state = lookupStateAbbreviation(stateAbbr);
+                if (!state)
+                    continue;
+                // Extract the street portion (everything before the city/state/zip)
+                const streetPortion = addrSingleLine.slice(0, cszMatch.index).replace(/,\s*$/, "").trim();
+                if (!streetPortion)
+                    continue;
+                const split = splitStreetLine(streetPortion);
+                if (!split)
+                    continue;
+                const venue = formatPersonName({ given: firstName, family: lastName }, "short") || undefined;
+                const components = {
+                    ...(venue ? { venue } : {}),
+                    ...(split.house_number ? { house_number: split.house_number } : {}),
+                    street: split.street,
+                    locality: city,
+                    region: state.abbreviation,
+                    ...(zip ? { postcode: zip } : {}),
+                };
+                const rendered = formatAddressRow(components, "US", { singleLine: true });
+                if (!rendered)
+                    continue;
+                const { raw, components: aligned } = rendered;
+                if (Object.keys(aligned).length <= 2)
+                    continue;
+                const sourceID = notaryID
+                    ? `${STATE_TX_NOTARIES_ADAPTER_ID}-${notaryID}`
+                    : stableSourceID(STATE_TX_NOTARIES_ADAPTER_ID, aligned);
+                yield {
+                    raw,
+                    components: aligned,
+                    country: "US",
+                    locale: "en-US",
+                    source: STATE_TX_NOTARIES_ADAPTER_ID,
+                    source_id: sourceID,
+                    corpus_version: "",
+                    license: STATE_TX_NOTARIES_DEFAULT_LICENSE,
+                };
+                emitted++;
+            }
+        },
+    };
+}
+/**
+ * The configured adapter instance registered with the corpus builder.
+ */
+export const stateTxNotariesAdapter = createStateTxNotariesAdapter();
+//# sourceMappingURL=adapter.js.map

package/out/us/adapters/state/tx-notaries/adapter.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../../lib/us/adapters/state/tx-notaries/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;GAeG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,iCAAiC,CAAA;AAClE,OAAO,EAAE,gBAAgB,EAAE,MAAM,wBAAwB,CAAA;AACzD,OAAO,EAAE,cAAc,EAAE,MAAM,aAAa,CAAA;AAE5C,OAAO,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAA;AACpD,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAA;AACvD,OAAO,EAAE,cAAc,EAAE,MAAM,YAAY,CAAA;AAC3C,OAAO,EAAE,WAAW,EAA8D,aAAa,EAAE,MAAM,QAAQ,CAAA;AAC/G,OAAO,EAAE,uBAAuB,EAAE,MAAM,gBAAgB,CAAA;AAExD;;;;GAIG;AACH,MAAM,CAAC,MAAM,4BAA4B,GAAG,mBAAmB,CAAA;AAC/D;;;GAGG;AACH,MAAM,CAAC,MAAM,iCAAiC,GAAG,eAAe,CAAA;AAEhE;;GAEG;AACH,MAAM,qBAAqB,GAAG,yDAAyD,CAAA;AASvF,MAAM,UAAU,4BAA4B;IAC3C,OAAO;QACN,EAAE,EAAE,4BAA4B;QAChC,cAAc,EAAE,iCAAiC;QACjD,WAAW,EAAE,WAAW,CAAC,OAAO;QAChC,QAAQ,EAAE,cAAc,CAAC,aAAa;QACtC,OAAO,EAAE,aAAa,CAAC,QAAQ;QAC/B,WAAW,EACV,wGAAwG;QAEzG,KAAK,CAAC,CAAC,IAAI,CAAC,IAAoB;YAC/B,IAAI,IAAI,CAAC,OAAO,IAAI,IAAI,CAAC,OAAO,KAAK,IAAI,EAAE,CAAC;gBAC3C,MAAM,IAAI,KAAK,CAAC,6DAA6D,IAAI,CAAC,OAAO,EAAE,CAAC,CAAA;YAC7F,CAAC;YAED,MAAM,IAAI,GAAG,cAAc,CAAC,SAAS,CAAC,IAAI,CAAC,SAAS,EAAE;gBACrD,aAAa,EAAE,KAAK;aACpB,CAAC,CAAA;YAEF,IAAI,OAAO,GAAG,CAAC,CAAA;YAEf,IAAI,KAAK,EAAE,MAAM,MAAM,IAAI,IAAkC,EAAE,CAAC;gBAC/D,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;oBAAE,MAAK;gBAE/B,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;oBAAE,MAAK;gBAE5D,MAAM,UAAU,GAAG,MAAM,CAAC,OAAO,IAAI,EAAE,CAAA;gBAEvC,IAAI,CAAC,UAAU;oBAAE,SAAQ;gBAEzB,MAAM,SAAS,GAAG,MAAM,CAAC,YAAY,CAAC,IAAI,EAAE,CAAA;gBAC5C,MAAM,QAAQ,GAAG,MAAM,CAAC,WAAW,CAAC,IAAI,EAAE,CAAA;gBAC1C,MAAM,QAAQ,GAAG,MAAM,CAAC,WAAW,CAAC,IAAI,EAAE,CAAA;gBAE1C,0EAA0E;gBAC1E,gEAAgE;gBAChE,MAAM,cAAc,GAAG,UAAU,CAAC,UAAU,CAAC,IAAI,EAAE,IAAI,CAAC,CAAA;gBACxD,MAAM,QAAQ,GAAG,qBAAqB,CAAC,IAAI,CAAC,cAAc,CAAC,CAAA;gBAE3D,IAAI,CAAC,QAAQ;oBAAE,SAAQ;gBAEvB,MAAM,IAAI,GAAG,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;gBACvC,MAAM,SAAS,GAAG,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;gBAC5C,MAAM,GAAG,GAAG,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;gBAEtC,IAAI,CAAC,IAAI,IAAI,CAAC,SAAS;oBAAE,SAAQ;gBAEjC,MAAM,KAAK,GAAG,uBAAuB,CAAC,SAAS,CAAC,CAAA;gBAEhD,IAAI,CAAC,KAAK;oBAAE,SAAQ;gBAEpB,oEAAoE;gBACpE,MAAM,aAAa,GAAG,cAAc,CAAC,KAAK,CAAC,CAAC,EAAE,QAAQ,CAAC,KAAK,CAAC,CAAC,OAAO,CAAC,OAAO,EAAE,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;gBAEzF,IAAI,CAAC,aAAa;oBAAE,SAAQ;gBAE5B,MAAM,KAAK,GAAG,eAAe,CAAC,aAAa,CAAC,CAAA;gBAE5C,IAAI,CAAC,KAAK;oBAAE,SAAQ;gBAEpB,MAAM,KAAK,GAAG,gBAAgB,CAAC,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,EAAE,QAAQ,EAAE,EAAE,OAAO,CAAC,IAAI,SAAS,CAAA;gBAE5F,MAAM,UAAU,GAA+B;oBAC9C,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;oBAC3B,GAAG,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,KAAK,CAAC,YAAY,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;oBACnE,MAAM,EAAE,KAAK,CAAC,MAAM;oBACpB,QAAQ,EAAE,IAAI;oBACd,MAAM,EAAE,KAAK,CAAC,YAAY;oBAC1B,GAAG,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;iBACjC,CAAA;gBAED,MAAM,QAAQ,GAAG,gBAAgB,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;gBAEzE,IAAI,CAAC,QAAQ;oBAAE,SAAQ;gBAEvB,MAAM,EAAE,GAAG,EAAE,UAAU,EAAE,OAAO,EAAE,GAAG,QAAQ,CAAA;gBAE7C,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,IAAI,CAAC;oBAAE,SAAQ;gBAE9C,MAAM,QAAQ,GAAG,QAAQ;oBACxB,CAAC,CAAC,GAAG,4BAA4B,IAAI,QAAQ,EAAE;oBAC/C,CAAC,CAAC,cAAc,CAAC,4BAA4B,EAAE,OAAO,CAAC,CAAA;gBAExD,MAAM;oBACL,GAAG;oBACH,UAAU,EAAE,OAAO;oBACnB,OAAO,EAAE,IAAI;oBACb,MAAM,EAAE,OAAO;oBACf,MAAM,EAAE,4BAA4B;oBACpC,SAAS,EAAE,QAAQ;oBACnB,cAAc,EAAE,EAAE;oBAClB,OAAO,EAAE,iCAAiC;iBAC1C,CAAA;gBAED,OAAO,EAAE,CAAA;YACV,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,MAAM,sBAAsB,GAAG,4BAA4B,EAAE,CAAA"}

package/out/us/adapters/tiger/adapter.d.ts

@@ -0,0 +1,31 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ * `tiger`: Adapter for US Census TIGER/Line data.
+ *
+ * This reads a prebuilt SQLite database (from TIGER shapefiles) and emits:
+ * - Street rows from `tiger_streets`
+ * - Locality rows from `tiger_places`
+ *
+ * TIGER is public domain, so every emitted row is stamped `"Public Domain"`.
+ */
+import { type CorpusAdapter } from "#types";
+/**
+ * Registry ID for this adapter.
+ */
+export declare const TIGER_ADAPTER_ID = "tiger";
+/**
+ * Default license for rows from this source.
+ */
+export declare const TIGER_DEFAULT_LICENSE = "Public Domain";
+/**
+ * Build a TIGER adapter.
+ */
+export declare function createTigerAdapter(): CorpusAdapter;
+/**
+ * Configured adapter instance.
+ */
+export declare const tigerAdapter: CorpusAdapter;
+//# sourceMappingURL=adapter.d.ts.map

package/out/us/adapters/tiger/adapter.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../lib/us/adapters/tiger/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AAOH,OAAO,EAAuD,KAAK,aAAa,EAAiB,MAAM,QAAQ,CAAA;AAI/G;;GAEG;AACH,eAAO,MAAM,gBAAgB,UAAU,CAAA;AACvC;;GAEG;AACH,eAAO,MAAM,qBAAqB,kBAAkB,CAAA;AA4GpD;;GAEG;AACH,wBAAgB,kBAAkB,IAAI,aAAa,CA8ElD;AAED;;GAEG;AACH,eAAO,MAAM,YAAY,eAAuB,CAAA"}

package/out/us/adapters/tiger/adapter.js

@@ -0,0 +1,173 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ * `tiger`: Adapter for US Census TIGER/Line data.
+ *
+ * This reads a prebuilt SQLite database (from TIGER shapefiles) and emits:
+ * - Street rows from `tiger_streets`
+ * - Locality rows from `tiger_places`
+ *
+ * TIGER is public domain, so every emitted row is stamped `"Public Domain"`.
+ */
+import { formatAddressRow } from "@mailwoman/codex/address/format";
+import { DatabaseClient } from "@mailwoman/sqlite/client";
+import { SourceRegister } from "#registers";
+import { AddressRole, SurfaceOrigin } from "#types";
+import { decomposeStreet } from "#us/adapters/tiger/street-decompose";
+import { lookupFipsState } from "#us/fips-state";
+/**
+ * Registry ID for this adapter.
+ */
+export const TIGER_ADAPTER_ID = "tiger";
+/**
+ * Default license for rows from this source.
+ */
+export const TIGER_DEFAULT_LICENSE = "Public Domain";
+/**
+ * Country display name used by `formatAddress`.
+ */
+const US_COUNTRY_DISPLAY = "United States of America";
+/**
+ * Build one or more street variants from a TIGER street segment.
+ */
+function* streetVariants(row) {
+    const fullname = row.fullname.trim();
+    if (!fullname)
+        return;
+    const state = lookupFipsState(row.statefp);
+    if (!state)
+        return;
+    const zipl = row.zipl?.trim() ?? "";
+    const zipr = row.zipr?.trim() ?? "";
+    const decomposed = decomposeStreet(fullname);
+    const baseComponents = {
+        region: state.abbreviation,
+        street: decomposed.street,
+    };
+    if (decomposed.prefix) {
+        baseComponents.street_prefix = decomposed.prefix;
+    }
+    if (decomposed.suffix) {
+        baseComponents.street_suffix = decomposed.suffix;
+    }
+    if (!zipl && !zipr) {
+        yield { components: baseComponents, variantKey: "no-zip" };
+        return;
+    }
+    if (zipl && zipr && zipl === zipr) {
+        yield {
+            components: { ...baseComponents, postcode: zipl },
+            variantKey: `zip-${zipl}`,
+        };
+        return;
+    }
+    if (zipl) {
+        yield { components: { ...baseComponents, postcode: zipl }, variantKey: `zipl-${zipl}` };
+    }
+    if (zipr && zipr !== zipl) {
+        yield { components: { ...baseComponents, postcode: zipr }, variantKey: `zipr-${zipr}` };
+    }
+}
+/**
+ * Build locality variants for a TIGER place.
+ */
+function* placeVariants(row) {
+    const name = row.name.trim();
+    if (!name)
+        return;
+    const state = lookupFipsState(row.statefp);
+    if (!state)
+        return;
+    yield {
+        components: { locality: name },
+        variantKey: "locality-only",
+    };
+    yield {
+        components: { locality: name, region: state.abbreviation },
+        variantKey: "with-region",
+    };
+    yield {
+        components: { locality: name, region: state.abbreviation, country: US_COUNTRY_DISPLAY },
+        variantKey: "with-region-country",
+    };
+}
+/**
+ * Build a TIGER adapter.
+ */
+export function createTigerAdapter() {
+    return {
+        id: TIGER_ADAPTER_ID,
+        defaultLicense: TIGER_DEFAULT_LICENSE,
+        addressRole: AddressRole.Premise,
+        register: SourceRegister.CensusTIGER,
+        surface: SurfaceOrigin.Rendered,
+        description: "US Census TIGER/Line streets + places consumer (public-domain); SQLite DB built via `mailwoman tiger fetch`.",
+        async *rows(opts) {
+            if (opts.country && opts.country !== "US") {
+                throw new Error(`tiger adapter: only US supported, got country=${opts.country}`);
+            }
+            const db = new DatabaseClient(opts.inputPath, { readOnly: true });
+            let emitted = 0;
+            try {
+                const streetStmt = db.prepare(`SELECT linearid, fullname, zipl, zipr, statefp FROM tiger_streets`);
+                const placeStmt = db.prepare(`SELECT geoid, name, statefp, lsad FROM tiger_places`);
+                for (const row of streetStmt.iterate()) {
+                    if (opts.signal?.aborted)
+                        return;
+                    for (const variant of streetVariants(row)) {
+                        if (opts.limit !== undefined && emitted >= opts.limit)
+                            return;
+                        const rendered = formatAddressRow(variant.components, "US", { singleLine: true });
+                        if (!rendered)
+                            continue;
+                        const { raw, components: aligned } = rendered;
+                        yield {
+                            raw,
+                            components: aligned,
+                            country: "US",
+                            locale: "en-US",
+                            source: TIGER_ADAPTER_ID,
+                            source_id: `${TIGER_ADAPTER_ID}-st-${row.linearid}-${variant.variantKey}`,
+                            corpus_version: "",
+                            license: TIGER_DEFAULT_LICENSE,
+                        };
+                        emitted++;
+                    }
+                }
+                for (const row of placeStmt.iterate()) {
+                    if (opts.signal?.aborted)
+                        return;
+                    for (const variant of placeVariants(row)) {
+                        if (opts.limit !== undefined && emitted >= opts.limit)
+                            return;
+                        const rendered = formatAddressRow(variant.components, "US", { singleLine: true });
+                        if (!rendered)
+                            continue;
+                        const { raw, components: aligned } = rendered;
+                        yield {
+                            raw,
+                            components: aligned,
+                            country: "US",
+                            locale: "en-US",
+                            source: TIGER_ADAPTER_ID,
+                            source_id: `${TIGER_ADAPTER_ID}-pl-${row.geoid}-${variant.variantKey}`,
+                            corpus_version: "",
+                            license: TIGER_DEFAULT_LICENSE,
+                        };
+                        emitted++;
+                    }
+                }
+            }
+            finally {
+                await db.destroy();
+            }
+        },
+    };
+}
+/**
+ * Configured adapter instance.
+ */
+export const tigerAdapter = createTigerAdapter();
+//# sourceMappingURL=adapter.js.map

package/out/us/adapters/tiger/adapter.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../lib/us/adapters/tiger/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,iCAAiC,CAAA;AAClE,OAAO,EAAE,cAAc,EAAE,MAAM,0BAA0B,CAAA;AAGzD,OAAO,EAAE,cAAc,EAAE,MAAM,YAAY,CAAA;AAC3C,OAAO,EAAE,WAAW,EAA8D,aAAa,EAAE,MAAM,QAAQ,CAAA;AAC/G,OAAO,EAAE,eAAe,EAAE,MAAM,qCAAqC,CAAA;AACrE,OAAO,EAAE,eAAe,EAAE,MAAM,gBAAgB,CAAA;AAEhD;;GAEG;AACH,MAAM,CAAC,MAAM,gBAAgB,GAAG,OAAO,CAAA;AACvC;;GAEG;AACH,MAAM,CAAC,MAAM,qBAAqB,GAAG,eAAe,CAAA;AAEpD;;GAEG;AACH,MAAM,kBAAkB,GAAG,0BAA0B,CAAA;AAiBrD;;GAEG;AACH,QAAQ,CAAC,CAAC,cAAc,CAAC,GAAmB;IAI3C,MAAM,QAAQ,GAAG,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAA;IAEpC,IAAI,CAAC,QAAQ;QAAE,OAAM;IACrB,MAAM,KAAK,GAAG,eAAe,CAAC,GAAG,CAAC,OAAO,CAAC,CAAA;IAE1C,IAAI,CAAC,KAAK;QAAE,OAAM;IAElB,MAAM,IAAI,GAAG,GAAG,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;IACnC,MAAM,IAAI,GAAG,GAAG,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;IAEnC,MAAM,UAAU,GAAG,eAAe,CAAC,QAAQ,CAAC,CAAA;IAE5C,MAAM,cAAc,GAA+B;QAClD,MAAM,EAAE,KAAK,CAAC,YAAY;QAC1B,MAAM,EAAE,UAAU,CAAC,MAAM;KACzB,CAAA;IAED,IAAI,UAAU,CAAC,MAAM,EAAE,CAAC;QACvB,cAAc,CAAC,aAAa,GAAG,UAAU,CAAC,MAAM,CAAA;IACjD,CAAC;IAED,IAAI,UAAU,CAAC,MAAM,EAAE,CAAC;QACvB,cAAc,CAAC,aAAa,GAAG,UAAU,CAAC,MAAM,CAAA;IACjD,CAAC;IAED,IAAI,CAAC,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC;QACpB,MAAM,EAAE,UAAU,EAAE,cAAc,EAAE,UAAU,EAAE,QAAQ,EAAE,CAAA;QAE1D,OAAM;IACP,CAAC;IAED,IAAI,IAAI,IAAI,IAAI,IAAI,IAAI,KAAK,IAAI,EAAE,CAAC;QACnC,MAAM;YACL,UAAU,EAAE,EAAE,GAAG,cAAc,EAAE,QAAQ,EAAE,IAAI,EAAE;YACjD,UAAU,EAAE,OAAO,IAAI,EAAE;SACzB,CAAA;QAED,OAAM;IACP,CAAC;IAED,IAAI,IAAI,EAAE,CAAC;QACV,MAAM,EAAE,UAAU,EAAE,EAAE,GAAG,cAAc,EAAE,QAAQ,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,QAAQ,IAAI,EAAE,EAAE,CAAA;IACxF,CAAC;IAED,IAAI,IAAI,IAAI,IAAI,KAAK,IAAI,EAAE,CAAC;QAC3B,MAAM,EAAE,UAAU,EAAE,EAAE,GAAG,cAAc,EAAE,QAAQ,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,QAAQ,IAAI,EAAE,EAAE,CAAA;IACxF,CAAC;AACF,CAAC;AAED;;GAEG;AACH,QAAQ,CAAC,CAAC,aAAa,CAAC,GAAkB;IAIzC,MAAM,IAAI,GAAG,GAAG,CAAC,IAAI,CAAC,IAAI,EAAE,CAAA;IAE5B,IAAI,CAAC,IAAI;QAAE,OAAM;IACjB,MAAM,KAAK,GAAG,eAAe,CAAC,GAAG,CAAC,OAAO,CAAC,CAAA;IAE1C,IAAI,CAAC,KAAK;QAAE,OAAM;IAElB,MAAM;QACL,UAAU,EAAE,EAAE,QAAQ,EAAE,IAAI,EAAE;QAC9B,UAAU,EAAE,eAAe;KAC3B,CAAA;IAED,MAAM;QACL,UAAU,EAAE,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,CAAC,YAAY,EAAE;QAC1D,UAAU,EAAE,aAAa;KACzB,CAAA;IAED,MAAM;QACL,UAAU,EAAE,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,CAAC,YAAY,EAAE,OAAO,EAAE,kBAAkB,EAAE;QACvF,UAAU,EAAE,qBAAqB;KACjC,CAAA;AACF,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,kBAAkB;IACjC,OAAO;QACN,EAAE,EAAE,gBAAgB;QACpB,cAAc,EAAE,qBAAqB;QACrC,WAAW,EAAE,WAAW,CAAC,OAAO;QAChC,QAAQ,EAAE,cAAc,CAAC,WAAW;QACpC,OAAO,EAAE,aAAa,CAAC,QAAQ;QAC/B,WAAW,EACV,8GAA8G;QAE/G,KAAK,CAAC,CAAC,IAAI,CAAC,IAAoB;YAC/B,IAAI,IAAI,CAAC,OAAO,IAAI,IAAI,CAAC,OAAO,KAAK,IAAI,EAAE,CAAC;gBAC3C,MAAM,IAAI,KAAK,CAAC,iDAAiD,IAAI,CAAC,OAAO,EAAE,CAAC,CAAA;YACjF,CAAC;YAED,MAAM,EAAE,GAAG,IAAI,cAAc,CAAgB,IAAI,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAA;YAChF,IAAI,OAAO,GAAG,CAAC,CAAA;YAEf,IAAI,CAAC;gBACJ,MAAM,UAAU,GAAG,EAAE,CAAC,OAAO,CAAC,mEAAmE,CAAC,CAAA;gBAClG,MAAM,SAAS,GAAG,EAAE,CAAC,OAAO,CAAC,qDAAqD,CAAC,CAAA;gBAEnF,KAAK,MAAM,GAAG,IAAI,UAAU,CAAC,OAAO,EAAsC,EAAE,CAAC;oBAC5E,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;wBAAE,OAAM;oBAEhC,KAAK,MAAM,OAAO,IAAI,cAAc,CAAC,GAAG,CAAC,EAAE,CAAC;wBAC3C,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;4BAAE,OAAM;wBAC7D,MAAM,QAAQ,GAAG,gBAAgB,CAAC,OAAO,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;wBAEjF,IAAI,CAAC,QAAQ;4BAAE,SAAQ;wBAEvB,MAAM,EAAE,GAAG,EAAE,UAAU,EAAE,OAAO,EAAE,GAAG,QAAQ,CAAA;wBAE7C,MAAM;4BACL,GAAG;4BACH,UAAU,EAAE,OAAO;4BACnB,OAAO,EAAE,IAAI;4BACb,MAAM,EAAE,OAAO;4BACf,MAAM,EAAE,gBAAgB;4BACxB,SAAS,EAAE,GAAG,gBAAgB,OAAO,GAAG,CAAC,QAAQ,IAAI,OAAO,CAAC,UAAU,EAAE;4BACzE,cAAc,EAAE,EAAE;4BAClB,OAAO,EAAE,qBAAqB;yBAC9B,CAAA;wBAED,OAAO,EAAE,CAAA;oBACV,CAAC;gBACF,CAAC;gBAED,KAAK,MAAM,GAAG,IAAI,SAAS,CAAC,OAAO,EAAqC,EAAE,CAAC;oBAC1E,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;wBAAE,OAAM;oBAEhC,KAAK,MAAM,OAAO,IAAI,aAAa,CAAC,GAAG,CAAC,EAAE,CAAC;wBAC1C,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;4BAAE,OAAM;wBAC7D,MAAM,QAAQ,GAAG,gBAAgB,CAAC,OAAO,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,UAAU,EAAE,IAAI,EAAE,CAAC,CAAA;wBAEjF,IAAI,CAAC,QAAQ;4BAAE,SAAQ;wBAEvB,MAAM,EAAE,GAAG,EAAE,UAAU,EAAE,OAAO,EAAE,GAAG,QAAQ,CAAA;wBAE7C,MAAM;4BACL,GAAG;4BACH,UAAU,EAAE,OAAO;4BACnB,OAAO,EAAE,IAAI;4BACb,MAAM,EAAE,OAAO;4BACf,MAAM,EAAE,gBAAgB;4BACxB,SAAS,EAAE,GAAG,gBAAgB,OAAO,GAAG,CAAC,KAAK,IAAI,OAAO,CAAC,UAAU,EAAE;4BACtE,cAAc,EAAE,EAAE;4BAClB,OAAO,EAAE,qBAAqB;yBAC9B,CAAA;wBAED,OAAO,EAAE,CAAA;oBACV,CAAC;gBACF,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,MAAM,EAAE,CAAC,OAAO,EAAE,CAAA;YACnB,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED;;GAEG;AACH,MAAM,CAAC,MAAM,YAAY,GAAG,kBAAkB,EAAE,CAAA"}

package/out/us/adapters/tiger/street-decompose.d.ts

@@ -0,0 +1,23 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ * Decompose a US street name into Stage 3 components (street_prefix, street, street_suffix) from the
+ * curated libpostal/en directionals and street_types dictionaries — the same ones the runtime
+ * StreetPrefixClassifier and StreetSuffixClassifier use, so corpus labels and runtime classifications
+ * agree on the vocabulary.
+ */
+export interface DecomposedStreet {
+    prefix: string | null;
+    street: string;
+    suffix: string | null;
+}
+/**
+ * Decompose a US street name into prefix/name/suffix components.
+ *
+ * Conservative, emitting a prefix or suffix only for a clear directional or street-type keyword.
+ * It returns the original as `street` when no pattern matches.
+ */
+export declare function decomposeStreet(fullname: string): DecomposedStreet;
+//# sourceMappingURL=street-decompose.d.ts.map

package/out/us/adapters/tiger/street-decompose.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"street-decompose.d.ts","sourceRoot":"","sources":["../../../../lib/us/adapters/tiger/street-decompose.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AASH,MAAM,WAAW,gBAAgB;IAChC,MAAM,EAAE,MAAM,GAAG,IAAI,CAAA;IACrB,MAAM,EAAE,MAAM,CAAA;IACd,MAAM,EAAE,MAAM,GAAG,IAAI,CAAA;CACrB;AAED;;;;;GAKG;AACH,wBAAgB,eAAe,CAAC,QAAQ,EAAE,MAAM,GAAG,gBAAgB,CA0ClE"}

package/out/us/adapters/tiger/street-decompose.js

@@ -0,0 +1,58 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ * Decompose a US street name into Stage 3 components (street_prefix, street, street_suffix) from the
+ * curated libpostal/en directionals and street_types dictionaries — the same ones the runtime
+ * StreetPrefixClassifier and StreetSuffixClassifier use, so corpus labels and runtime classifications
+ * agree on the vocabulary.
+ */
+import { loadLibpostalDictionary } from "#adapters/libpostal-dictionary";
+const [DIRECTIONALS, STREET_TYPES] = await Promise.all([
+    loadLibpostalDictionary("en", "directionals.txt"),
+    loadLibpostalDictionary("en", "street_types.txt"),
+]);
+/**
+ * Decompose a US street name into prefix/name/suffix components.
+ *
+ * Conservative, emitting a prefix or suffix only for a clear directional or street-type keyword.
+ * It returns the original as `street` when no pattern matches.
+ */
+export function decomposeStreet(fullname) {
+    const trimmed = fullname.trim();
+    if (!trimmed)
+        return { prefix: null, street: "", suffix: null };
+    const tokens = trimmed.split(/\s+/);
+    if (tokens.length === 1)
+        return { prefix: null, street: trimmed, suffix: null };
+    const norm = (s) => s.toLowerCase().replace(/\.$/, "");
+    let prefix = null;
+    let suffix = null;
+    let startIdx = 0;
+    let endIdx = tokens.length;
+    if (DIRECTIONALS.has(norm(tokens[0])) && tokens.length >= 2) {
+        prefix = tokens[0];
+        startIdx = 1;
+    }
+    const last = norm(tokens[endIdx - 1]);
+    const secondLast = endIdx >= 2 ? norm(tokens[endIdx - 2]) : "";
+    if (DIRECTIONALS.has(last) && STREET_TYPES.has(secondLast)) {
+        suffix = tokens.slice(endIdx - 2, endIdx).join(" ");
+        endIdx -= 2;
+    }
+    else if (STREET_TYPES.has(last) && endIdx - startIdx >= 2) {
+        suffix = tokens[endIdx - 1];
+        endIdx -= 1;
+    }
+    else if (DIRECTIONALS.has(last) && endIdx - startIdx >= 2) {
+        suffix = tokens[endIdx - 1];
+        endIdx -= 1;
+    }
+    const street = tokens.slice(startIdx, endIdx).join(" ").trim();
+    if (!street) {
+        return { prefix: null, street: trimmed, suffix: null };
+    }
+    return { prefix, street, suffix };
+}
+//# sourceMappingURL=street-decompose.js.map

package/out/us/adapters/tiger/street-decompose.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"street-decompose.js","sourceRoot":"","sources":["../../../../lib/us/adapters/tiger/street-decompose.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,EAAE,uBAAuB,EAAE,MAAM,gCAAgC,CAAA;AAExE,MAAM,CAAC,YAAY,EAAE,YAAY,CAAC,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC;IACtD,uBAAuB,CAAC,IAAI,EAAE,kBAAkB,CAAC;IACjD,uBAAuB,CAAC,IAAI,EAAE,kBAAkB,CAAC;CACjD,CAAC,CAAA;AAQF;;;;;GAKG;AACH,MAAM,UAAU,eAAe,CAAC,QAAgB;IAC/C,MAAM,OAAO,GAAG,QAAQ,CAAC,IAAI,EAAE,CAAA;IAE/B,IAAI,CAAC,OAAO;QAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAA;IAE/D,MAAM,MAAM,GAAG,OAAO,CAAC,KAAK,CAAC,KAAK,CAAC,CAAA;IAEnC,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,CAAA;IAE/E,MAAM,IAAI,GAAG,CAAC,CAAS,EAAE,EAAE,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,OAAO,CAAC,KAAK,EAAE,EAAE,CAAC,CAAA;IAE9D,IAAI,MAAM,GAAkB,IAAI,CAAA;IAChC,IAAI,MAAM,GAAkB,IAAI,CAAA;IAChC,IAAI,QAAQ,GAAG,CAAC,CAAA;IAChB,IAAI,MAAM,GAAG,MAAM,CAAC,MAAM,CAAA;IAE1B,IAAI,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,CAAC,IAAI,MAAM,CAAC,MAAM,IAAI,CAAC,EAAE,CAAC;QAC9D,MAAM,GAAG,MAAM,CAAC,CAAC,CAAE,CAAA;QACnB,QAAQ,GAAG,CAAC,CAAA;IACb,CAAC;IAED,MAAM,IAAI,GAAG,IAAI,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAE,CAAC,CAAA;IACtC,MAAM,UAAU,GAAG,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAE,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;IAE/D,IAAI,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,YAAY,CAAC,GAAG,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5D,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,MAAM,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QACnD,MAAM,IAAI,CAAC,CAAA;IACZ,CAAC;SAAM,IAAI,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,MAAM,GAAG,QAAQ,IAAI,CAAC,EAAE,CAAC;QAC7D,MAAM,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,CAAE,CAAA;QAC5B,MAAM,IAAI,CAAC,CAAA;IACZ,CAAC;SAAM,IAAI,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,MAAM,GAAG,QAAQ,IAAI,CAAC,EAAE,CAAC;QAC7D,MAAM,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,CAAE,CAAA;QAC5B,MAAM,IAAI,CAAC,CAAA;IACZ,CAAC;IAED,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IAE9D,IAAI,CAAC,MAAM,EAAE,CAAC;QACb,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,CAAA;IACvD,CAAC;IAED,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,CAAA;AAClC,CAAC"}