@mailwoman/corpus 4.10.0 → 4.11.0

package/README.md

@@ -0,0 +1,79 @@
+# @mailwoman/corpus
+
+**BIO-labeled training-corpus pipeline** for the Mailwoman address parser.
+
+Generates sequence-labeling training data from reference sources
+(OpenAddresses, libpostal dictionaries, synthetic shards) and assembles them into
+the TSV format consumed by the Modal training pipeline. This package produces
+the data that trains `@mailwoman/neural-weights-*`.
+
+```ts
+// The corpus pipeline is primarily build-time CLI tooling.
+// Key entry points:
+import { expandGolden } from "@mailwoman/corpus" // Expand reference addresses
+import { synthesizeShard } from "@mailwoman/corpus" // Generate synthetic training shards
+import { alignRow } from "@mailwoman/corpus" // Align raw address → BIO tokens
+import { validateCorpus } from "@mailwoman/corpus" // Validate corpus integrity
+```
+
+## What it produces
+
+The corpus pipeline assembles training data from multiple sources:
+
+| Source               | Description                                                              |
+| -------------------- | ------------------------------------------------------------------------ |
+| **OpenAddresses**    | Real government address point data (US, FR, DE, …)                       |
+| **NAD**              | National Address Database (US-specific)                                  |
+| **libpostal**        | Multilingual street/place name dictionaries                              |
+| **Synthetic shards** | Generated address variations (boundary stress, order variants, all-caps) |
+| **Overture Maps**    | Address theme ingestion (alpha)                                          |
+
+Output format: TSV rows with `raw<TAB>BIO_labels` consumed by the Python
+training pipeline (`corpus-python/`).
+
+## Key modules
+
+| Module                  | Purpose                                                            |
+| ----------------------- | ------------------------------------------------------------------ |
+| **`expand-golden.ts`**  | Expand reference addresses into training rows with alignment       |
+| **`align.ts`**          | Tokenize raw address → BIO label sequence                          |
+| **`validate.ts`**       | Validate corpus integrity, label coverage, shard balance           |
+| **`synthesize-*.ts`**   | Synthetic shard generators (boundary stress, order variants, etc.) |
+| **`ingest/`**           | Overture Maps + NAD ingestion                                      |
+| **`shard-registry.ts`** | Shard metadata and composition                                     |
+| **`stats.ts`**          | Per-shard and per-tag statistics                                   |
+
+## Build-time tooling
+
+The corpus is assembled via scripts in `scripts/`:
+
+```bash
+# Validate the corpus
+node scripts/validate-corpus.mjs
+
+# Rebuild shards
+node scripts/build-boundary-stress-shard.mjs
+
+# Corpus statistics
+node scripts/corpus-stats.mjs
+```
+
+## Design
+
+- **BIO (Begin/Inside/Outside) labeling** over SentencePiece tokens.
+- **Character-offset aligned** — labels track the raw string, not the
+  normalized form, so the model learns real input distributions.
+- **Source-homogeneous shards** — each shard comes from one source, ordered by
+  type, so eval splits are honest (no bleed between train and held-out).
+
+## Related
+
+- [`@mailwoman/neural`](../neural) — the runtime that loads and runs the trained model
+- [`@mailwoman/neural-weights-en-us`](../neural-weights-en-us) — the trained model itself
+- [Corpus Construction concepts](https://mailwoman.sister.software/articles/concepts/corpus-construction/)
+- [Training Pipeline concepts](https://mailwoman.sister.software/articles/concepts/training-pipeline/)
+- [CONTRIBUTING_MODEL_WORK](https://mailwoman.sister.software/articles/plan/CONTRIBUTING_MODEL_WORK/)
+
+## License
+
+[AGPL-3.0-only](https://www.gnu.org/licenses/agpl-3.0.html)

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

@@ -18,8 +18,12 @@
  *   left for the wof-postalcode + wof-admin cross-reference at corpus build time (a future pass;
  *   for Phase 1 the row's region is simply absent).
  *
- *   License: ODbL / Licence Ouverte. Stamped onto every row as `ODbL-1.0` (the OpenStreetMap-
- *   compatible label most consumers expect).
+ *   License: the official BAN (adresse.data.gouv.fr) is DUAL-licensed — Licence Ouverte 2.0 (Etalab,
+ *   attribution-only) OR ODbL (share-alike). We ELECT Licence Ouverte 2.0 (issue #26 Tier B:
+ *   allowed for training with attribution; the ODbL option's share-alike obligation would defeat
+ *   the proprietary-weights goal). Stamped onto every row as `Licence Ouverte 2.0` — NOT the older
+ *   conservative `ODbL-1.0` label, which wrongly read as Tier-C-denied in the corpus license audit.
+ *   The model card MUST carry the BAN attribution (Tier B obligation).
  *
  *   The adapter is streaming-aware: it uses `csv-parse` in streaming mode so a 25M-row dump never
  *   sits in memory. Honors `opts.limit` for fixture / smoke runs, `opts.signal` for cancellation,

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

@@ -1 +1 @@
-{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../src/adapters/ban/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AAMH,OAAO,KAAK,EAAgC,aAAa,EAAE,MAAM,gBAAgB,CAAA;AAGjF,eAAO,MAAM,cAAc,QAAQ,CAAA;AA4CnC,wBAAgB,gBAAgB,IAAI,aAAa,CAwEhD;AAED,eAAO,MAAM,UAAU,eAAqB,CAAA"}
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../src/adapters/ban/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AAMH,OAAO,KAAK,EAAgC,aAAa,EAAE,MAAM,gBAAgB,CAAA;AAGjF,eAAO,MAAM,cAAc,QAAQ,CAAA;AA4CnC,wBAAgB,gBAAgB,IAAI,aAAa,CAwEhD;AAED,eAAO,MAAM,UAAU,eAAqB,CAAA"}

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

@@ -18,8 +18,12 @@
  *   left for the wof-postalcode + wof-admin cross-reference at corpus build time (a future pass;
  *   for Phase 1 the row's region is simply absent).
  *
- *   License: ODbL / Licence Ouverte. Stamped onto every row as `ODbL-1.0` (the OpenStreetMap-
- *   compatible label most consumers expect).
+ *   License: the official BAN (adresse.data.gouv.fr) is DUAL-licensed — Licence Ouverte 2.0 (Etalab,
+ *   attribution-only) OR ODbL (share-alike). We ELECT Licence Ouverte 2.0 (issue #26 Tier B:
+ *   allowed for training with attribution; the ODbL option's share-alike obligation would defeat
+ *   the proprietary-weights goal). Stamped onto every row as `Licence Ouverte 2.0` — NOT the older
+ *   conservative `ODbL-1.0` label, which wrongly read as Tier-C-denied in the corpus license audit.
+ *   The model card MUST carry the BAN attribution (Tier B obligation).
  *
  *   The adapter is streaming-aware: it uses `csv-parse` in streaming mode so a 25M-row dump never
  *   sits in memory. Honors `opts.limit` for fixture / smoke runs, `opts.signal` for cancellation,
@@ -64,7 +68,7 @@ function composeRaw(house, street, postcode, locality) {
 export function createBanAdapter() {
     return {
         id: BAN_ADAPTER_ID,
-        defaultLicense: "ODbL-1.0",
+        defaultLicense: "Licence Ouverte 2.0",
         description: "Base Adresse Nationale (FR): house-number-level street addresses (~25M rows).",
         async *rows(opts) {
             if (opts.country && opts.country !== "FR") {
@@ -122,7 +126,7 @@ export function createBanAdapter() {
                         source: BAN_ADAPTER_ID,
                         source_id: sourceId,
                         corpus_version: "",
-                        license: "ODbL-1.0",
+                        license: "Licence Ouverte 2.0",
                     };
                     emitted++;
                 }

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

@@ -1 +1 @@
-{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../src/adapters/ban/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AAEH,OAAO,EAAE,KAAK,IAAI,QAAQ,EAAE,MAAM,WAAW,CAAA;AAC7C,OAAO,EAAE,gBAAgB,EAAE,MAAM,SAAS,CAAA;AAC1C,OAAO,EAAE,cAAc,EAAE,MAAM,kBAAkB,CAAA;AACjD,OAAO,EAAE,mBAAmB,EAAE,MAAM,iBAAiB,CAAA;AAErD,OAAO,EAAE,iBAAiB,EAAE,MAAM,uBAAuB,CAAA;AAEzD,MAAM,CAAC,MAAM,cAAc,GAAG,KAAK,CAAA;AAenC;;;GAGG;AACH,SAAS,kBAAkB,CAAC,MAAc,EAAE,GAAW;IACtD,MAAM,CAAC,GAAG,MAAM,CAAC,IAAI,EAAE,CAAA;IACvB,MAAM,CAAC,GAAG,GAAG,CAAC,IAAI,EAAE,CAAA;IACpB,IAAI,CAAC,CAAC;QAAE,OAAO,EAAE,CAAA;IACjB,OAAO,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAA;AAC3B,CAAC;AAED;;;;;;;;GAQG;AACH,SAAS,UAAU,CAAC,KAAa,EAAE,MAAc,EAAE,QAAgB,EAAE,QAAgB;IACpF,MAAM,KAAK,GAAa,EAAE,CAAA;IAC1B,MAAM,UAAU,GAAG,CAAC,KAAK,EAAE,MAAM,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IACnE,IAAI,UAAU;QAAE,KAAK,CAAC,IAAI,CAAC,UAAU,CAAC,CAAA;IACtC,MAAM,QAAQ,GAAG,CAAC,QAAQ,EAAE,QAAQ,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IACtE,IAAI,QAAQ;QAAE,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;IAClC,OAAO,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AACpD,CAAC;AAED,MAAM,UAAU,gBAAgB;IAC/B,OAAO;QACN,EAAE,EAAE,cAAc;QAClB,cAAc,EAAE,UAAU;QAC1B,WAAW,EAAE,+EAA+E;QAE5F,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,+CAA+C,IAAI,CAAC,OAAO,EAAE,CAAC,CAAA;YAC/E,CAAC;YAED,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;YACrE,MAAM,MAAM,GAAG,MAAM,CAAC,IAAI,CACzB,QAAQ,CAAC;gBACR,SAAS,EAAE,GAAG;gBACd,OAAO,EAAE,IAAI;gBACb,gBAAgB,EAAE,IAAI;gBACtB,YAAY,EAAE,IAAI;gBAClB,kBAAkB,EAAE,IAAI;aACxB,CAAC,CACF,CAAA;YAED,IAAI,OAAO,GAAG,CAAC,CAAA;YACf,IAAI,CAAC;gBACJ,IAAI,KAAK,EAAE,MAAM,MAAM,IAAI,MAA+B,EAAE,CAAC;oBAC5D,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;wBAAE,MAAK;oBAC/B,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;wBAAE,MAAK;oBAE5D,MAAM,KAAK,GAAG,kBAAkB,CAAC,MAAM,CAAC,MAAM,IAAI,EAAE,EAAE,MAAM,CAAC,GAAG,IAAI,EAAE,CAAC,CAAA;oBACvE,MAAM,MAAM,GAAG,CAAC,MAAM,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAC7C,MAAM,QAAQ,GAAG,CAAC,MAAM,CAAC,WAAW,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAClD,MAAM,QAAQ,GAAG,CAAC,MAAM,CAAC,WAAW,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAElD,IAAI,CAAC,MAAM,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAClC,IAAI,CAAC,KAAK,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAEjC,MAAM,UAAU,GAAG,iBAAiB,CAAC,MAAM,CAAC,CAAA;oBAE5C,MAAM,UAAU,GAA+B,EAAE,CAAA;oBACjD,IAAI,KAAK;wBAAE,UAAU,CAAC,YAAY,GAAG,KAAK,CAAA;oBAC1C,IAAI,UAAU,CAAC,MAAM;wBAAE,UAAU,CAAC,aAAa,GAAG,UAAU,CAAC,MAAM,CAAA;oBACnE,IAAI,UAAU,CAAC,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,UAAU,CAAC,MAAM,CAAA;oBAC5D,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAC5C,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAE5C,MAAM,GAAG,GAAG,UAAU,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,QAAQ,CAAC,CAAA;oBACzD,IAAI,CAAC,GAAG;wBAAE,SAAQ;oBAElB,MAAM,OAAO,GAAG,mBAAmB,CAAC,UAAU,EAAE,GAAG,CAAC,CAAA;oBACpD,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAQ;oBAE/C,MAAM,QAAQ,GAAG,MAAM,CAAC,EAAE,EAAE,IAAI,EAAE;wBACjC,CAAC,CAAC,GAAG,cAAc,IAAI,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;wBACzC,CAAC,CAAC,cAAc,CAAC,cAAc,EAAE,OAAO,CAAC,CAAA;oBAE1C,MAAM;wBACL,GAAG;wBACH,UAAU,EAAE,OAAO;wBACnB,OAAO,EAAE,IAAI;wBACb,MAAM,EAAE,OAAO;wBACf,MAAM,EAAE,cAAc;wBACtB,SAAS,EAAE,QAAQ;wBACnB,cAAc,EAAE,EAAE;wBAClB,OAAO,EAAE,UAAU;qBACnB,CAAA;oBACD,OAAO,EAAE,CAAA;gBACV,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,MAAM,CAAC,OAAO,EAAE,CAAA;YACjB,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED,MAAM,CAAC,MAAM,UAAU,GAAG,gBAAgB,EAAE,CAAA"}
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../src/adapters/ban/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AAEH,OAAO,EAAE,KAAK,IAAI,QAAQ,EAAE,MAAM,WAAW,CAAA;AAC7C,OAAO,EAAE,gBAAgB,EAAE,MAAM,SAAS,CAAA;AAC1C,OAAO,EAAE,cAAc,EAAE,MAAM,kBAAkB,CAAA;AACjD,OAAO,EAAE,mBAAmB,EAAE,MAAM,iBAAiB,CAAA;AAErD,OAAO,EAAE,iBAAiB,EAAE,MAAM,uBAAuB,CAAA;AAEzD,MAAM,CAAC,MAAM,cAAc,GAAG,KAAK,CAAA;AAenC;;;GAGG;AACH,SAAS,kBAAkB,CAAC,MAAc,EAAE,GAAW;IACtD,MAAM,CAAC,GAAG,MAAM,CAAC,IAAI,EAAE,CAAA;IACvB,MAAM,CAAC,GAAG,GAAG,CAAC,IAAI,EAAE,CAAA;IACpB,IAAI,CAAC,CAAC;QAAE,OAAO,EAAE,CAAA;IACjB,OAAO,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAA;AAC3B,CAAC;AAED;;;;;;;;GAQG;AACH,SAAS,UAAU,CAAC,KAAa,EAAE,MAAc,EAAE,QAAgB,EAAE,QAAgB;IACpF,MAAM,KAAK,GAAa,EAAE,CAAA;IAC1B,MAAM,UAAU,GAAG,CAAC,KAAK,EAAE,MAAM,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IACnE,IAAI,UAAU;QAAE,KAAK,CAAC,IAAI,CAAC,UAAU,CAAC,CAAA;IACtC,MAAM,QAAQ,GAAG,CAAC,QAAQ,EAAE,QAAQ,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IACtE,IAAI,QAAQ;QAAE,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;IAClC,OAAO,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AACpD,CAAC;AAED,MAAM,UAAU,gBAAgB;IAC/B,OAAO;QACN,EAAE,EAAE,cAAc;QAClB,cAAc,EAAE,qBAAqB;QACrC,WAAW,EAAE,+EAA+E;QAE5F,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,+CAA+C,IAAI,CAAC,OAAO,EAAE,CAAC,CAAA;YAC/E,CAAC;YAED,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;YACrE,MAAM,MAAM,GAAG,MAAM,CAAC,IAAI,CACzB,QAAQ,CAAC;gBACR,SAAS,EAAE,GAAG;gBACd,OAAO,EAAE,IAAI;gBACb,gBAAgB,EAAE,IAAI;gBACtB,YAAY,EAAE,IAAI;gBAClB,kBAAkB,EAAE,IAAI;aACxB,CAAC,CACF,CAAA;YAED,IAAI,OAAO,GAAG,CAAC,CAAA;YACf,IAAI,CAAC;gBACJ,IAAI,KAAK,EAAE,MAAM,MAAM,IAAI,MAA+B,EAAE,CAAC;oBAC5D,IAAI,IAAI,CAAC,MAAM,EAAE,OAAO;wBAAE,MAAK;oBAC/B,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,IAAI,CAAC,KAAK;wBAAE,MAAK;oBAE5D,MAAM,KAAK,GAAG,kBAAkB,CAAC,MAAM,CAAC,MAAM,IAAI,EAAE,EAAE,MAAM,CAAC,GAAG,IAAI,EAAE,CAAC,CAAA;oBACvE,MAAM,MAAM,GAAG,CAAC,MAAM,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAC7C,MAAM,QAAQ,GAAG,CAAC,MAAM,CAAC,WAAW,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAClD,MAAM,QAAQ,GAAG,CAAC,MAAM,CAAC,WAAW,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;oBAElD,IAAI,CAAC,MAAM,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAClC,IAAI,CAAC,KAAK,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAEjC,MAAM,UAAU,GAAG,iBAAiB,CAAC,MAAM,CAAC,CAAA;oBAE5C,MAAM,UAAU,GAA+B,EAAE,CAAA;oBACjD,IAAI,KAAK;wBAAE,UAAU,CAAC,YAAY,GAAG,KAAK,CAAA;oBAC1C,IAAI,UAAU,CAAC,MAAM;wBAAE,UAAU,CAAC,aAAa,GAAG,UAAU,CAAC,MAAM,CAAA;oBACnE,IAAI,UAAU,CAAC,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,UAAU,CAAC,MAAM,CAAA;oBAC5D,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAC5C,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAE5C,MAAM,GAAG,GAAG,UAAU,CAAC,KAAK,EAAE,MAAM,EAAE,QAAQ,EAAE,QAAQ,CAAC,CAAA;oBACzD,IAAI,CAAC,GAAG;wBAAE,SAAQ;oBAElB,MAAM,OAAO,GAAG,mBAAmB,CAAC,UAAU,EAAE,GAAG,CAAC,CAAA;oBACpD,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAQ;oBAE/C,MAAM,QAAQ,GAAG,MAAM,CAAC,EAAE,EAAE,IAAI,EAAE;wBACjC,CAAC,CAAC,GAAG,cAAc,IAAI,MAAM,CAAC,EAAE,CAAC,IAAI,EAAE,EAAE;wBACzC,CAAC,CAAC,cAAc,CAAC,cAAc,EAAE,OAAO,CAAC,CAAA;oBAE1C,MAAM;wBACL,GAAG;wBACH,UAAU,EAAE,OAAO;wBACnB,OAAO,EAAE,IAAI;wBACb,MAAM,EAAE,OAAO;wBACf,MAAM,EAAE,cAAc;wBACtB,SAAS,EAAE,QAAQ;wBACnB,cAAc,EAAE,EAAE;wBAClB,OAAO,EAAE,qBAAqB;qBAC9B,CAAA;oBACD,OAAO,EAAE,CAAA;gBACV,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,MAAM,CAAC,OAAO,EAAE,CAAA;YACjB,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED,MAAM,CAAC,MAAM,UAAU,GAAG,gBAAgB,EAAE,CAAA"}

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

@@ -46,8 +46,11 @@ export interface OpenaddressesAdapterOptions {
      */
     defaultLicense?: string;
     /**
-     * When true, rows with share-alike licenses (ODbL, CC-BY-SA, CC-SA) are NOT filtered. Default
-     * false — share-alike rows are dropped per the licensing strategy (#26).
+     * Per-adapter share-alike drop. Default **true** (include) as of 2026-06-19: exclusion is a
+     * deliberate BUILD-level act (`buildCorpus({ excludeLicenses })` / `--exclude-share-alike`), NOT
+     * a silent adapter default (#26 — "purposely exclude, don't opt in to include"). Set false only
+     * for an explicit adapter-scoped drop; the build-level `--exclude-share-alike` is the normal
+     * path.
      */
     allowShareAlike?: boolean;
 }

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

@@ -1 +1 @@
-{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../src/adapters/openaddresses/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAoCG;AAMH,OAAO,KAAK,EAAgC,aAAa,EAAE,MAAM,gBAAgB,CAAA;AAEjF,eAAO,MAAM,wBAAwB,kBAAkB,CAAA;AACvD,eAAO,MAAM,6BAA6B,cAAc,CAAA;AAqDxD,MAAM,WAAW,2BAA2B;IAC3C;;;;OAIG;IACH,cAAc,CAAC,EAAE,MAAM,CAAA;IAEvB;;;OAGG;IACH,eAAe,CAAC,EAAE,OAAO,CAAA;CACzB;AAED;;;GAGG;AACH,wBAAgB,0BAA0B,CAAC,IAAI,GAAE,2BAAgC,GAAG,aAAa,CAwFhG;AAED,eAAO,MAAM,oBAAoB,eAA+B,CAAA"}
+{"version":3,"file":"adapter.d.ts","sourceRoot":"","sources":["../../../../src/adapters/openaddresses/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAoCG;AAOH,OAAO,KAAK,EAAgC,aAAa,EAAE,MAAM,gBAAgB,CAAA;AAEjF,eAAO,MAAM,wBAAwB,kBAAkB,CAAA;AACvD,eAAO,MAAM,6BAA6B,cAAc,CAAA;AA8CxD,MAAM,WAAW,2BAA2B;IAC3C;;;;OAIG;IACH,cAAc,CAAC,EAAE,MAAM,CAAA;IAEvB;;;;;;OAMG;IACH,eAAe,CAAC,EAAE,OAAO,CAAA;CACzB;AAED;;;GAGG;AACH,wBAAgB,0BAA0B,CAAC,IAAI,GAAE,2BAAgC,GAAG,aAAa,CAwFhG;AAED,eAAO,MAAM,oBAAoB,eAA+B,CAAA"}

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

@@ -39,6 +39,7 @@ import { createReadStream } from "node:fs";
 import { createInterface } from "node:readline";
 import { stableSourceId } from "../../adapter.js";
 import { formatAddress, reconcileComponents } from "../../format.js";
+import { SHARE_ALIKE_PATTERN } from "../../license.js";
 export const OPENADDRESSES_ADAPTER_ID = "openaddresses";
 export const OPENADDRESSES_DEFAULT_LICENSE = "CC-BY-4.0";
 /** Return a lowercase-keyed view of a Feature's properties so case variants both work. */
@@ -73,19 +74,13 @@ function parseFeatureLine(line) {
         return null;
     return normalizeProperties(obj.properties);
 }
-/**
- * Licenses that require share-alike or create a copyleft obligation on derived works. Rows carrying
- * any of these licenses are dropped by default because they would contaminate proprietary-weights
- * training. Per the licensing strategy (#26).
- */
-const SHARE_ALIKE_PATTERN = /^ODbL|^Open Database License|^CC-BY-SA|^CC-SA/i;
 /**
  * Build an OpenAddresses adapter. The optional `defaultLicense` lets callers stamp a non-default
  * fallback for dumps known to carry a single license throughout (e.g. a PDDL-only state slice).
  */
 export function createOpenaddressesAdapter(opts = {}) {
     const defaultLicense = opts.defaultLicense ?? OPENADDRESSES_DEFAULT_LICENSE;
-    const allowShareAlike = opts.allowShareAlike ?? false;
+    const allowShareAlike = opts.allowShareAlike ?? true;
     return {
         id: OPENADDRESSES_ADAPTER_ID,
         defaultLicense,

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

@@ -1 +1 @@
-{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../src/adapters/openaddresses/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAoCG;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,aAAa,EAAE,mBAAmB,EAAE,MAAM,iBAAiB,CAAA;AAGpE,MAAM,CAAC,MAAM,wBAAwB,GAAG,eAAe,CAAA;AACvD,MAAM,CAAC,MAAM,6BAA6B,GAAG,WAAW,CAAA;AAmBxD,0FAA0F;AAC1F,SAAS,mBAAmB,CAAC,GAAY;IACxC,IAAI,CAAC,GAAG,IAAI,OAAO,GAAG,KAAK,QAAQ;QAAE,OAAO,EAAE,CAAA;IAC9C,MAAM,GAAG,GAA2B,EAAE,CAAA;IACtC,KAAK,MAAM,CAAC,CAAC,EAAE,CAAC,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,GAA8B,CAAC,EAAE,CAAC;QACrE,IAAI,OAAO,CAAC,KAAK,QAAQ;YAAE,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,GAAG,CAAC,CAAA;aAC9C,IAAI,OAAO,CAAC,KAAK,QAAQ;YAAE,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;IACjE,CAAC;IACD,OAAO,GAAmB,CAAA;AAC3B,CAAC;AAED,+FAA+F;AAC/F,SAAS,gBAAgB,CAAC,IAAY;IACrC,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;IAC3B,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC;QAAE,OAAO,IAAI,CAAA;IACpD,IAAI,MAAe,CAAA;IACnB,IAAI,CAAC;QACJ,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,CAAA;IAC7B,CAAC;IAAC,MAAM,CAAC;QACR,OAAO,IAAI,CAAA;IACZ,CAAC;IACD,IAAI,CAAC,MAAM,IAAI,OAAO,MAAM,KAAK,QAAQ;QAAE,OAAO,IAAI,CAAA;IACtD,MAAM,GAAG,GAAG,MAAiD,CAAA;IAC7D,IAAI,GAAG,CAAC,IAAI,KAAK,SAAS;QAAE,OAAO,IAAI,CAAA;IACvC,OAAO,mBAAmB,CAAC,GAAG,CAAC,UAAU,CAAC,CAAA;AAC3C,CAAC;AAED;;;;GAIG;AACH,MAAM,mBAAmB,GAAG,gDAAgD,CAAA;AAiB5E;;;GAGG;AACH,MAAM,UAAU,0BAA0B,CAAC,OAAoC,EAAE;IAChF,MAAM,cAAc,GAAG,IAAI,CAAC,cAAc,IAAI,6BAA6B,CAAA;IAC3E,MAAM,eAAe,GAAG,IAAI,CAAC,eAAe,IAAI,KAAK,CAAA;IAErD,OAAO;QACN,EAAE,EAAE,wBAAwB;QAC5B,cAAc;QACd,WAAW,EAAE,6EAA6E;QAE1F,KAAK,CAAC,CAAC,IAAI,CAAC,WAA2B;YACtC,IAAI,CAAC,WAAW,CAAC,OAAO,EAAE,CAAC;gBAC1B,MAAM,IAAI,KAAK,CACd,0HAA0H,CAC1H,CAAA;YACF,CAAC;YACD,MAAM,OAAO,GAAG,WAAW,CAAC,OAAO,CAAA;YAEnC,MAAM,MAAM,GAAG,gBAAgB,CAAC,WAAW,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;YAC5E,MAAM,KAAK,GAAG,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAA;YAErE,IAAI,OAAO,GAAG,CAAC,CAAA;YACf,IAAI,iBAAiB,GAAG,CAAC,CAAA;YACzB,IAAI,CAAC;gBACJ,IAAI,KAAK,EAAE,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;oBAChC,IAAI,WAAW,CAAC,MAAM,EAAE,OAAO;wBAAE,MAAK;oBACtC,IAAI,WAAW,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,WAAW,CAAC,KAAK;wBAAE,MAAK;oBAE1E,MAAM,KAAK,GAAG,gBAAgB,CAAC,IAAI,CAAC,CAAA;oBACpC,IAAI,CAAC,KAAK;wBAAE,SAAQ;oBAEpB,MAAM,WAAW,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBAC9C,MAAM,MAAM,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACzC,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACrC,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACrC,MAAM,MAAM,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACzC,MAAM,QAAQ,GAAG,KAAK,CAAC,QAAQ,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBAE7C,iFAAiF;oBACjF,wDAAwD;oBACxD,IAAI,CAAC,MAAM;wBAAE,SAAQ;oBACrB,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAEhC,MAAM,OAAO,GAAG,CAAC,KAAK,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,cAAc,CAAC,CAAC,IAAI,EAAE,CAAA;oBAEhE,IAAI,CAAC,eAAe,IAAI,mBAAmB,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;wBAC3D,iBAAiB,EAAE,CAAA;wBACnB,SAAQ;oBACT,CAAC;oBAED,MAAM,UAAU,GAA+B,EAAE,CAAA;oBACjD,IAAI,WAAW;wBAAE,UAAU,CAAC,YAAY,GAAG,WAAW,CAAA;oBACtD,IAAI,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,MAAM,CAAA;oBACtC,IAAI,IAAI;wBAAE,UAAU,CAAC,IAAI,GAAG,IAAI,CAAA;oBAChC,IAAI,IAAI;wBAAE,UAAU,CAAC,QAAQ,GAAG,IAAI,CAAA;oBACpC,IAAI,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,MAAM,CAAA;oBACtC,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAE5C,MAAM,GAAG,GAAG,aAAa,CAAC,UAAU,EAAE,OAAO,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;oBACnE,IAAI,CAAC,GAAG;wBAAE,SAAQ;oBAElB,MAAM,OAAO,GAAG,mBAAmB,CAAC,UAAU,EAAE,GAAG,CAAC,CAAA;oBACpD,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAQ;oBAE/C,MAAM,YAAY,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,KAAK,CAAC,EAAE,EAAE,IAAI,EAAE,CAAA;oBAC3D,MAAM,QAAQ,GAAG,YAAY;wBAC5B,CAAC,CAAC,GAAG,wBAAwB,IAAI,YAAY,EAAE;wBAC/C,CAAC,CAAC,cAAc,CAAC,wBAAwB,EAAE,OAAO,CAAC,CAAA;oBAEpD,MAAM;wBACL,GAAG;wBACH,UAAU,EAAE,OAAO;wBACnB,OAAO;wBACP,MAAM,EAAE,wBAAwB;wBAChC,SAAS,EAAE,QAAQ;wBACnB,cAAc,EAAE,EAAE;wBAClB,OAAO;qBACP,CAAA;oBACD,OAAO,EAAE,CAAA;gBACV,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,KAAK,CAAC,KAAK,EAAE,CAAA;gBACb,MAAM,CAAC,OAAO,EAAE,CAAA;gBAChB,IAAI,iBAAiB,GAAG,CAAC,EAAE,CAAC;oBAC3B,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,oBAAoB,iBAAiB,8BAA8B,OAAO,SAAS,CAAC,CAAA;gBAC1G,CAAC;YACF,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED,MAAM,CAAC,MAAM,oBAAoB,GAAG,0BAA0B,EAAE,CAAA"}
+{"version":3,"file":"adapter.js","sourceRoot":"","sources":["../../../../src/adapters/openaddresses/adapter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAoCG;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,aAAa,EAAE,mBAAmB,EAAE,MAAM,iBAAiB,CAAA;AACpE,OAAO,EAAE,mBAAmB,EAAE,MAAM,kBAAkB,CAAA;AAGtD,MAAM,CAAC,MAAM,wBAAwB,GAAG,eAAe,CAAA;AACvD,MAAM,CAAC,MAAM,6BAA6B,GAAG,WAAW,CAAA;AAmBxD,0FAA0F;AAC1F,SAAS,mBAAmB,CAAC,GAAY;IACxC,IAAI,CAAC,GAAG,IAAI,OAAO,GAAG,KAAK,QAAQ;QAAE,OAAO,EAAE,CAAA;IAC9C,MAAM,GAAG,GAA2B,EAAE,CAAA;IACtC,KAAK,MAAM,CAAC,CAAC,EAAE,CAAC,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,GAA8B,CAAC,EAAE,CAAC;QACrE,IAAI,OAAO,CAAC,KAAK,QAAQ;YAAE,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,GAAG,CAAC,CAAA;aAC9C,IAAI,OAAO,CAAC,KAAK,QAAQ;YAAE,GAAG,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;IACjE,CAAC;IACD,OAAO,GAAmB,CAAA;AAC3B,CAAC;AAED,+FAA+F;AAC/F,SAAS,gBAAgB,CAAC,IAAY;IACrC,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;IAC3B,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC;QAAE,OAAO,IAAI,CAAA;IACpD,IAAI,MAAe,CAAA;IACnB,IAAI,CAAC;QACJ,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,CAAA;IAC7B,CAAC;IAAC,MAAM,CAAC;QACR,OAAO,IAAI,CAAA;IACZ,CAAC;IACD,IAAI,CAAC,MAAM,IAAI,OAAO,MAAM,KAAK,QAAQ;QAAE,OAAO,IAAI,CAAA;IACtD,MAAM,GAAG,GAAG,MAAiD,CAAA;IAC7D,IAAI,GAAG,CAAC,IAAI,KAAK,SAAS;QAAE,OAAO,IAAI,CAAA;IACvC,OAAO,mBAAmB,CAAC,GAAG,CAAC,UAAU,CAAC,CAAA;AAC3C,CAAC;AAoBD;;;GAGG;AACH,MAAM,UAAU,0BAA0B,CAAC,OAAoC,EAAE;IAChF,MAAM,cAAc,GAAG,IAAI,CAAC,cAAc,IAAI,6BAA6B,CAAA;IAC3E,MAAM,eAAe,GAAG,IAAI,CAAC,eAAe,IAAI,IAAI,CAAA;IAEpD,OAAO;QACN,EAAE,EAAE,wBAAwB;QAC5B,cAAc;QACd,WAAW,EAAE,6EAA6E;QAE1F,KAAK,CAAC,CAAC,IAAI,CAAC,WAA2B;YACtC,IAAI,CAAC,WAAW,CAAC,OAAO,EAAE,CAAC;gBAC1B,MAAM,IAAI,KAAK,CACd,0HAA0H,CAC1H,CAAA;YACF,CAAC;YACD,MAAM,OAAO,GAAG,WAAW,CAAC,OAAO,CAAA;YAEnC,MAAM,MAAM,GAAG,gBAAgB,CAAC,WAAW,CAAC,SAAS,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;YAC5E,MAAM,KAAK,GAAG,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAA;YAErE,IAAI,OAAO,GAAG,CAAC,CAAA;YACf,IAAI,iBAAiB,GAAG,CAAC,CAAA;YACzB,IAAI,CAAC;gBACJ,IAAI,KAAK,EAAE,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;oBAChC,IAAI,WAAW,CAAC,MAAM,EAAE,OAAO;wBAAE,MAAK;oBACtC,IAAI,WAAW,CAAC,KAAK,KAAK,SAAS,IAAI,OAAO,IAAI,WAAW,CAAC,KAAK;wBAAE,MAAK;oBAE1E,MAAM,KAAK,GAAG,gBAAgB,CAAC,IAAI,CAAC,CAAA;oBACpC,IAAI,CAAC,KAAK;wBAAE,SAAQ;oBAEpB,MAAM,WAAW,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBAC9C,MAAM,MAAM,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACzC,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACrC,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACrC,MAAM,MAAM,GAAG,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBACzC,MAAM,QAAQ,GAAG,KAAK,CAAC,QAAQ,EAAE,IAAI,EAAE,IAAI,EAAE,CAAA;oBAE7C,iFAAiF;oBACjF,wDAAwD;oBACxD,IAAI,CAAC,MAAM;wBAAE,SAAQ;oBACrB,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ;wBAAE,SAAQ;oBAEhC,MAAM,OAAO,GAAG,CAAC,KAAK,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,cAAc,CAAC,CAAC,IAAI,EAAE,CAAA;oBAEhE,IAAI,CAAC,eAAe,IAAI,mBAAmB,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;wBAC3D,iBAAiB,EAAE,CAAA;wBACnB,SAAQ;oBACT,CAAC;oBAED,MAAM,UAAU,GAA+B,EAAE,CAAA;oBACjD,IAAI,WAAW;wBAAE,UAAU,CAAC,YAAY,GAAG,WAAW,CAAA;oBACtD,IAAI,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,MAAM,CAAA;oBACtC,IAAI,IAAI;wBAAE,UAAU,CAAC,IAAI,GAAG,IAAI,CAAA;oBAChC,IAAI,IAAI;wBAAE,UAAU,CAAC,QAAQ,GAAG,IAAI,CAAA;oBACpC,IAAI,MAAM;wBAAE,UAAU,CAAC,MAAM,GAAG,MAAM,CAAA;oBACtC,IAAI,QAAQ;wBAAE,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;oBAE5C,MAAM,GAAG,GAAG,aAAa,CAAC,UAAU,EAAE,OAAO,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;oBACnE,IAAI,CAAC,GAAG;wBAAE,SAAQ;oBAElB,MAAM,OAAO,GAAG,mBAAmB,CAAC,UAAU,EAAE,GAAG,CAAC,CAAA;oBACpD,IAAI,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAQ;oBAE/C,MAAM,YAAY,GAAG,KAAK,CAAC,IAAI,EAAE,IAAI,EAAE,IAAI,KAAK,CAAC,EAAE,EAAE,IAAI,EAAE,CAAA;oBAC3D,MAAM,QAAQ,GAAG,YAAY;wBAC5B,CAAC,CAAC,GAAG,wBAAwB,IAAI,YAAY,EAAE;wBAC/C,CAAC,CAAC,cAAc,CAAC,wBAAwB,EAAE,OAAO,CAAC,CAAA;oBAEpD,MAAM;wBACL,GAAG;wBACH,UAAU,EAAE,OAAO;wBACnB,OAAO;wBACP,MAAM,EAAE,wBAAwB;wBAChC,SAAS,EAAE,QAAQ;wBACnB,cAAc,EAAE,EAAE;wBAClB,OAAO;qBACP,CAAA;oBACD,OAAO,EAAE,CAAA;gBACV,CAAC;YACF,CAAC;oBAAS,CAAC;gBACV,KAAK,CAAC,KAAK,EAAE,CAAA;gBACb,MAAM,CAAC,OAAO,EAAE,CAAA;gBAChB,IAAI,iBAAiB,GAAG,CAAC,EAAE,CAAC;oBAC3B,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,oBAAoB,iBAAiB,8BAA8B,OAAO,SAAS,CAAC,CAAA;gBAC1G,CAAC;YACF,CAAC;QACF,CAAC;KACD,CAAA;AACF,CAAC;AAED,MAAM,CAAC,MAAM,oBAAoB,GAAG,0BAA0B,EAAE,CAAA"}

package/out/src/build.d.ts

@@ -74,6 +74,14 @@ export interface BuildCorpusOptions {
     rowsPerShard?: number;
     /** Progress hook. Errors thrown abort the build. */
     onProgress?: (stage: BuildStage, message: string) => void;
+    /**
+     * License kinds to PURPOSELY exclude from this build (#26). Compiled patterns (see
+     * `compileLicenseExcludes` / `SHARE_ALIKE_PATTERN` in `license.ts`); a row whose `license`
+     * matches any is dropped at ingest. Default (omitted) includes EVERYTHING — exclusion is a
+     * deliberate act, not a silent default. A proprietary-weights build passes the share-alike set
+     * (`--exclude-share-alike`).
+     */
+    excludeLicenses?: readonly RegExp[];
 }
 /** Top-level manifest tying every stage together. */
 export interface BuildCorpusManifest {
@@ -91,6 +99,12 @@ export interface BuildCorpusManifest {
     };
     quarantine_count: number;
     total_aligned_rows: number;
+    /**
+     * Resolved license set across all INCLUDED rows (license string → row count), + the count dropped
+     * by `excludeLicenses` (#26). The model card derives its data-attribution table from `licenses`.
+     */
+    licenses: Record<string, number>;
+    excluded_by_license: number;
 }
 /**
  * Drive the full corpus build to completion.

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

@@ -1 +1 @@
-{"version":3,"file":"build.d.ts","sourceRoot":"","sources":["../../src/build.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA+CG;AAQH,OAAO,EAAe,KAAK,aAAa,EAAE,MAAM,cAAc,CAAA;AAC9D,OAAO,EAAc,KAAK,kBAAkB,EAAE,MAAM,aAAa,CAAA;AACjE,OAAO,EAIN,KAAK,aAAa,EAElB,MAAM,YAAY,CAAA;AAEnB,OAAO,KAAK,EAAE,cAAc,EAAgB,aAAa,EAAc,MAAM,YAAY,CAAA;AAEzF,2CAA2C;AAC3C,MAAM,MAAM,UAAU,GAAG,aAAa,GAAG,OAAO,GAAG,OAAO,GAAG,OAAO,GAAG,UAAU,CAAA;AAEjF,gDAAgD;AAChD,MAAM,WAAW,kBAAkB;IAClC,kEAAkE;IAClE,SAAS,EAAE,MAAM,CAAA;IAEjB,0FAA0F;IAC1F,aAAa,EAAE,MAAM,CAAA;IAErB;;;OAGG;IACH,QAAQ,CAAC,EAAE,SAAS,aAAa,EAAE,CAAA;IAEnC;;;OAGG;IACH,aAAa,EAAE,MAAM,CAAC,MAAM,EAAE,cAAc,CAAC,CAAA;IAE7C,yFAAyF;IACzF,UAAU,CAAC,EAAE,OAAO,CAAA;IAEpB,qDAAqD;IACrD,YAAY,CAAC,EAAE,MAAM,CAAA;IAErB,oDAAoD;IACpD,UAAU,CAAC,EAAE,CAAC,KAAK,EAAE,UAAU,EAAE,OAAO,EAAE,MAAM,KAAK,IAAI,CAAA;CACzD;AAED,qDAAqD;AACrD,MAAM,WAAW,mBAAmB;IACnC,cAAc,EAAE,MAAM,CAAA;IACtB,QAAQ,EAAE,MAAM,CAAA;IAChB,QAAQ,EAAE,kBAAkB,EAAE,CAAA;IAC9B,gBAAgB,EAAE,MAAM,EAAE,CAAA;IAC1B,MAAM,EAAE;QAAE,MAAM,EAAE,aAAa,CAAC,QAAQ,CAAC,CAAC;QAAC,QAAQ,EAAE,aAAa,CAAC,UAAU,CAAC,CAAA;KAAE,CAAA;IAChF,MAAM,EAAE;QAAE,MAAM,EAAE,aAAa,CAAC,QAAQ,CAAC,CAAC;QAAC,UAAU,EAAE,MAAM,CAAA;KAAE,CAAA;IAC/D,gBAAgB,EAAE,MAAM,CAAA;IACxB,kBAAkB,EAAE,MAAM,CAAA;CAC1B;AAED;;;;;;;GAOG;AACH,wBAAsB,WAAW,CAAC,IAAI,EAAE,kBAAkB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CA4JxF"}
+{"version":3,"file":"build.d.ts","sourceRoot":"","sources":["../../src/build.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA+CG;AASH,OAAO,EAAe,KAAK,aAAa,EAAE,MAAM,cAAc,CAAA;AAC9D,OAAO,EAAc,KAAK,kBAAkB,EAAE,MAAM,aAAa,CAAA;AACjE,OAAO,EAIN,KAAK,aAAa,EAElB,MAAM,YAAY,CAAA;AAEnB,OAAO,KAAK,EAAE,cAAc,EAAgB,aAAa,EAAc,MAAM,YAAY,CAAA;AAEzF,2CAA2C;AAC3C,MAAM,MAAM,UAAU,GAAG,aAAa,GAAG,OAAO,GAAG,OAAO,GAAG,OAAO,GAAG,UAAU,CAAA;AAEjF,gDAAgD;AAChD,MAAM,WAAW,kBAAkB;IAClC,kEAAkE;IAClE,SAAS,EAAE,MAAM,CAAA;IAEjB,0FAA0F;IAC1F,aAAa,EAAE,MAAM,CAAA;IAErB;;;OAGG;IACH,QAAQ,CAAC,EAAE,SAAS,aAAa,EAAE,CAAA;IAEnC;;;OAGG;IACH,aAAa,EAAE,MAAM,CAAC,MAAM,EAAE,cAAc,CAAC,CAAA;IAE7C,yFAAyF;IACzF,UAAU,CAAC,EAAE,OAAO,CAAA;IAEpB,qDAAqD;IACrD,YAAY,CAAC,EAAE,MAAM,CAAA;IAErB,oDAAoD;IACpD,UAAU,CAAC,EAAE,CAAC,KAAK,EAAE,UAAU,EAAE,OAAO,EAAE,MAAM,KAAK,IAAI,CAAA;IAEzD;;;;;;OAMG;IACH,eAAe,CAAC,EAAE,SAAS,MAAM,EAAE,CAAA;CACnC;AAED,qDAAqD;AACrD,MAAM,WAAW,mBAAmB;IACnC,cAAc,EAAE,MAAM,CAAA;IACtB,QAAQ,EAAE,MAAM,CAAA;IAChB,QAAQ,EAAE,kBAAkB,EAAE,CAAA;IAC9B,gBAAgB,EAAE,MAAM,EAAE,CAAA;IAC1B,MAAM,EAAE;QAAE,MAAM,EAAE,aAAa,CAAC,QAAQ,CAAC,CAAC;QAAC,QAAQ,EAAE,aAAa,CAAC,UAAU,CAAC,CAAA;KAAE,CAAA;IAChF,MAAM,EAAE;QAAE,MAAM,EAAE,aAAa,CAAC,QAAQ,CAAC,CAAC;QAAC,UAAU,EAAE,MAAM,CAAA;KAAE,CAAA;IAC/D,gBAAgB,EAAE,MAAM,CAAA;IACxB,kBAAkB,EAAE,MAAM,CAAA;IAC1B;;;OAGG;IACH,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IAChC,mBAAmB,EAAE,MAAM,CAAA;CAC3B;AAED;;;;;;;GAOG;AACH,wBAAsB,WAAW,CAAC,IAAI,EAAE,kBAAkB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAwLxF"}

package/out/src/build.js

@@ -52,6 +52,7 @@ import { join } from "node:path";
 import { createInterface } from "node:readline";
 import { defaultAdapterRegistry } from "./adapter.js";
 import { alignRow } from "./align.js";
+import { licenseExcluded } from "./license.js";
 import { writeShards } from "./parquet.js";
 import { runAdapter } from "./runner.js";
 import { defaultHoldouts, splitForRow, writeSplitManifestsFromLabeledFiles, } from "./split.js";
@@ -127,12 +128,27 @@ export async function buildCorpus(opts) {
     let quarantined = 0;
     const counts = { train: 0, val: 0, test: 0 };
     const holdouts = defaultHoldouts();
+    // License accounting + the deliberate exclusion filter (#26). `licenseCounts` is the resolved
+    // license set (→ manifest + model-card attribution); `excludeLicenses` (empty by default → include
+    // everything) is the operator's PURPOSEFUL exclusion, never a silent drop.
+    const excludeLicenses = opts.excludeLicenses ?? [];
+    const licenseCounts = new Map();
+    let excludedByLicense = 0;
     const writeQuarantine = (row, reason) => {
         quarantineStream.write(`${JSON.stringify({ row, reason })}\n`);
     };
     for (const adapterRun of adapterRuns) {
         opts.onProgress?.("align", `aligning ${adapterRun.adapter_id}`);
         for await (const row of streamJsonl(adapterRun.jsonl_path)) {
+            licenseCounts.set(row.license, (licenseCounts.get(row.license) ?? 0) + 1);
+            // Deliberate license exclusion (#26): drop a row ONLY when the operator named its license
+            // kind via `excludeLicenses`. Default (no patterns) keeps everything — exclusion is a
+            // purposeful act, not a silent default. Counted BEFORE the drop so the manifest's license
+            // set reflects what the corpus actually CONTAINED, and `excluded_by_license` what was cut.
+            if (licenseExcluded(row.license, excludeLicenses)) {
+                excludedByLicense++;
+                continue;
+            }
             const fanned = [row];
             if (synthesize) {
                 for (const aug of synthesizeRow(row, defaultAugmentationsForCountry(row.country))) {
@@ -194,6 +210,13 @@ export async function buildCorpus(opts) {
         corpusVersion: opts.corpusVersion,
         rowsPerShard,
     });
+    // License-set visibility (#26): loudly report the resolved license set so a build is an obvious
+    // deliberate act — especially a proprietary-weights build (did you pass --exclude-share-alike?).
+    const licenseSummary = [...licenseCounts.entries()].sort((a, b) => b[1] - a[1]);
+    opts.onProgress?.("manifest", `license set: ${licenseSummary.map(([l, c]) => `${l}=${c}`).join(", ")}` +
+        (excludedByLicense > 0
+            ? ` | EXCLUDED ${excludedByLicense} rows by --exclude-licenses`
+            : " | NO license exclusion applied (all rows kept)"));
     // 6. Top-level manifest.
     opts.onProgress?.("manifest", "writing top-level MANIFEST.json");
     const manifest = {
@@ -205,6 +228,8 @@ export async function buildCorpus(opts) {
         shards: { counts: shardManifest.counts, total_rows: shardManifest.total_rows },
         quarantine_count: quarantined,
         total_aligned_rows: aligned,
+        licenses: Object.fromEntries(licenseSummary),
+        excluded_by_license: excludedByLicense,
     };
     await writeFile(join(opts.outputDir, "MANIFEST.json"), `${JSON.stringify(manifest, null, 2)}\n`, "utf8");
     return manifest;

package/out/src/build.js.map

@@ -1 +1 @@
-{"version":3,"file":"build.js","sourceRoot":"","sources":["../../src/build.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA+CG;AAEH,OAAO,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,UAAU,EAAE,YAAY,EAAoB,MAAM,SAAS,CAAA;AACzG,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACnD,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAA;AAChC,OAAO,EAAE,eAAe,EAAE,MAAM,eAAe,CAAA;AAC/C,OAAO,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAA;AACrD,OAAO,EAAE,QAAQ,EAAE,MAAM,YAAY,CAAA;AACrC,OAAO,EAAE,WAAW,EAAsB,MAAM,cAAc,CAAA;AAC9D,OAAO,EAAE,UAAU,EAA2B,MAAM,aAAa,CAAA;AACjE,OAAO,EACN,eAAe,EACf,WAAW,EACX,mCAAmC,GAGnC,MAAM,YAAY,CAAA;AACnB,OAAO,EAAE,8BAA8B,EAAE,aAAa,EAAE,MAAM,iBAAiB,CAAA;AAgD/E;;;;;;;GAOG;AACH,MAAM,CAAC,KAAK,UAAU,WAAW,CAAC,IAAwB;IACzD,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,IAAI,sBAAsB,CAAC,IAAI,EAAE,CAAA;IAC/D,MAAM,UAAU,GAAG,IAAI,CAAC,UAAU,IAAI,IAAI,CAAA;IAC1C,MAAM,YAAY,GAAG,IAAI,CAAC,YAAY,IAAI,SAAS,CAAA;IACnD,MAAM,QAAQ,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAA;IAEzC,MAAM,KAAK,CAAC,IAAI,CAAC,SAAS,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAChD,MAAM,eAAe,GAAG,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,cAAc,CAAC,CAAA;IAC5D,MAAM,KAAK,CAAC,eAAe,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAEjD,mBAAmB;IACnB,MAAM,WAAW,GAAyB,EAAE,CAAA;IAC5C,MAAM,OAAO,GAAa,EAAE,CAAA;IAC5B,KAAK,MAAM,OAAO,IAAI,QAAQ,EAAE,CAAC;QAChC,MAAM,cAAc,GAAG,IAAI,CAAC,aAAa,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;QACrD,IAAI,CAAC,cAAc,EAAE,CAAC;YACrB,OAAO,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;YACxB,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,wBAAwB,CAAC,CAAA;YAC/E,SAAQ;QACT,CAAC;QACD,gGAAgG;QAChG,2FAA2F;QAC3F,+FAA+F;QAC/F,8FAA8F;QAC9F,sGAAsG;QACtG,MAAM,UAAU,GAAG,IAAI,CAAC,eAAe,EAAE,OAAO,CAAC,EAAE,CAAC,CAAA;QACpD,MAAM,cAAc,GAAG,IAAI,CAAC,UAAU,EAAE,eAAe,CAAC,CAAA;QACxD,IACC,OAAO,CAAC,GAAG,CAAC,gBAAgB,KAAK,GAAG;YACpC,UAAU,CAAC,cAAc,CAAC;YAC1B,UAAU,CAAC,IAAI,CAAC,UAAU,EAAE,iBAAiB,CAAC,CAAC,EAC9C,CAAC;YACF,MAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,YAAY,CAAC,cAAc,EAAE,MAAM,CAAC,CAAuB,CAAA;YACrF,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,YAAY,MAAM,CAAC,OAAO,kBAAkB,CAAC,CAAA;YACnG,WAAW,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;YACxB,SAAQ;QACT,CAAC;QACD,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,EAAE,CAAC,CAAA;QACzD,MAAM,CAAC,GAAG,MAAM,UAAU,CAAC;YAC1B,OAAO;YACP,cAAc;YACd,SAAS,EAAE,eAAe;YAC1B,aAAa,EAAE,IAAI,CAAC,aAAa;SACjC,CAAC,CAAA;QACF,WAAW,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;IACpB,CAAC;IAED,yFAAyF;IACzF,8FAA8F;IAC9F,0FAA0F;IAC1F,0FAA0F;IAC1F,0EAA0E;IAC1E,MAAM,YAAY,GAA8B;QAC/C,KAAK,EAAE,IAAI,CAAC,eAAe,EAAE,qBAAqB,CAAC;QACnD,GAAG,EAAE,IAAI,CAAC,eAAe,EAAE,mBAAmB,CAAC;QAC/C,IAAI,EAAE,IAAI,CAAC,eAAe,EAAE,oBAAoB,CAAC;KACjD,CAAA;IACD,MAAM,cAAc,GAAmC;QACtD,KAAK,EAAE,iBAAiB,CAAC,YAAY,CAAC,KAAK,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;QAClE,GAAG,EAAE,iBAAiB,CAAC,YAAY,CAAC,GAAG,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;QAC9D,IAAI,EAAE,iBAAiB,CAAC,YAAY,CAAC,IAAI,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;KAChE,CAAA;IACD,MAAM,cAAc,GAAG,IAAI,CAAC,eAAe,EAAE,kBAAkB,CAAC,CAAA;IAChE,MAAM,gBAAgB,GAAG,iBAAiB,CAAC,cAAc,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;IAEhF,IAAI,OAAO,GAAG,CAAC,CAAA;IACf,IAAI,WAAW,GAAG,CAAC,CAAA;IACnB,MAAM,MAAM,GAA8B,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,IAAI,EAAE,CAAC,EAAE,CAAA;IACvE,MAAM,QAAQ,GAAG,eAAe,EAAE,CAAA;IAElC,MAAM,eAAe,GAAG,CAAC,GAAiB,EAAE,MAAc,EAAQ,EAAE;QACnE,gBAAgB,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC,SAAS,CAAC,EAAE,GAAG,EAAE,MAAM,EAAE,CAAC,IAAI,CAAC,CAAA;IAC/D,CAAC,CAAA;IAED,KAAK,MAAM,UAAU,IAAI,WAAW,EAAE,CAAC;QACtC,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,YAAY,UAAU,CAAC,UAAU,EAAE,CAAC,CAAA;QAC/D,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,WAAW,CAAe,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;YAC1E,MAAM,MAAM,GAAmB,CAAC,GAAG,CAAC,CAAA;YACpC,IAAI,UAAU,EAAE,CAAC;gBAChB,KAAK,MAAM,GAAG,IAAI,aAAa,CAAC,GAAG,EAAE,8BAA8B,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC;oBACnF,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;gBACjB,CAAC;YACF,CAAC;YACD,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;gBACxB,IAAI,MAAmC,CAAA;gBACvC,IAAI,CAAC;oBACJ,MAAM,GAAG,QAAQ,CAAC,CAAC,CAAC,CAAA;gBACrB,CAAC;gBAAC,OAAO,GAAG,EAAE,CAAC;oBACd,mFAAmF;oBACnF,qFAAqF;oBACrF,kFAAkF;oBAClF,iFAAiF;oBACjF,eAAe,CAAC,CAAC,EAAE,eAAgB,GAAa,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,CAAA;oBACzE,WAAW,EAAE,CAAA;oBACb,SAAQ;gBACT,CAAC;gBACD,IAAI,MAAM,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;oBAC/B,MAAM,KAAK,GAAG,WAAW,CAAC,MAAM,CAAC,GAAG,EAAE,QAAQ,CAAC,CAAA;oBAC/C,cAAc,CAAC,KAAK,CAAC,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC,SAAS,CAAC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,CAAA;oBAC9D,MAAM,CAAC,KAAK,CAAC,EAAE,CAAA;oBACf,OAAO,EAAE,CAAA;gBACV,CAAC;qBAAM,CAAC;oBACP,eAAe,CAAC,CAAC,EAAE,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;oBACrC,WAAW,EAAE,CAAA;gBACd,CAAC;YACF,CAAC;QACF,CAAC;IACF,CAAC;IAED,KAAK,MAAM,CAAC,IAAI,MAAM,CAAC,MAAM,CAAC,cAAc,CAAC;QAAE,CAAC,CAAC,GAAG,EAAE,CAAA;IACtD,gBAAgB,CAAC,GAAG,EAAE,CAAA;IACtB,MAAM,OAAO,CAAC,GAAG,CAAC,CAAC,GAAG,MAAM,CAAC,MAAM,CAAC,cAAc,CAAC,CAAC,GAAG,CAAC,SAAS,CAAC,EAAE,SAAS,CAAC,gBAAgB,CAAC,CAAC,CAAC,CAAA;IAEjG,sFAAsF;IACtF,uFAAuF;IACvF,yEAAyE;IACzE,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,aAAa,OAAO,eAAe,CAAC,CAAA;IAC/D,MAAM,SAAS,GAAG,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,QAAQ,CAAC,CAAA;IAChD,MAAM,WAAW,GAAG,MAAM,mCAAmC,CAAC;QAC7D,YAAY;QACZ,SAAS,EAAE,SAAS;QACpB,aAAa,EAAE,IAAI,CAAC,aAAa;QACjC,MAAM;QACN,QAAQ;KACR,CAAC,CAAA;IAEF,4FAA4F;IAC5F,0FAA0F;IAC1F,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,wBAAwB,CAAC,CAAA;IACpD,MAAM,aAAa,GAAG,MAAM,WAAW,CACtC;QACC,KAAK,EAAE,WAAW,CAAa,YAAY,CAAC,KAAK,CAAC;QAClD,GAAG,EAAE,WAAW,CAAa,YAAY,CAAC,GAAG,CAAC;QAC9C,IAAI,EAAE,WAAW,CAAa,YAAY,CAAC,IAAI,CAAC;KAChD,EACD;QACC,SAAS,EAAE,IAAI,CAAC,SAAS;QACzB,aAAa,EAAE,IAAI,CAAC,aAAa;QACjC,YAAY;KACZ,CACD,CAAA;IAED,yBAAyB;IACzB,IAAI,CAAC,UAAU,EAAE,CAAC,UAAU,EAAE,iCAAiC,CAAC,CAAA;IAChE,MAAM,QAAQ,GAAwB;QACrC,cAAc,EAAE,IAAI,CAAC,aAAa;QAClC,QAAQ;QACR,QAAQ,EAAE,WAAW;QACrB,gBAAgB,EAAE,OAAO;QACzB,MAAM,EAAE,EAAE,MAAM,EAAE,WAAW,EAAE,QAAQ,EAAE;QACzC,MAAM,EAAE,EAAE,MAAM,EAAE,aAAa,CAAC,MAAM,EAAE,UAAU,EAAE,aAAa,CAAC,UAAU,EAAE;QAC9E,gBAAgB,EAAE,WAAW;QAC7B,kBAAkB,EAAE,OAAO;KAC3B,CAAA;IACD,MAAM,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,eAAe,CAAC,EAAE,GAAG,IAAI,CAAC,SAAS,CAAC,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IACxG,OAAO,QAAQ,CAAA;AAChB,CAAC;AAED,KAAK,SAAS,CAAC,CAAC,WAAW,CAAI,IAAY;IAC1C,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;IAC3D,MAAM,EAAE,GAAG,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAA;IAClE,IAAI,KAAK,EAAE,MAAM,IAAI,IAAI,EAAE,EAAE,CAAC;QAC7B,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;QAC3B,IAAI,CAAC,OAAO;YAAE,SAAQ;QACtB,MAAM,IAAI,CAAC,KAAK,CAAC,OAAO,CAAM,CAAA;IAC/B,CAAC;AACF,CAAC;AAED,SAAS,SAAS,CAAC,CAAc;IAChC,OAAO,IAAI,OAAO,CAAC,CAAC,OAAO,EAAE,MAAM,EAAE,EAAE;QACtC,CAAC,CAAC,IAAI,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;QACxB,CAAC,CAAC,IAAI,CAAC,OAAO,EAAE,MAAM,CAAC,CAAA;IACxB,CAAC,CAAC,CAAA;AACH,CAAC"}
+{"version":3,"file":"build.js","sourceRoot":"","sources":["../../src/build.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA+CG;AAEH,OAAO,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,UAAU,EAAE,YAAY,EAAoB,MAAM,SAAS,CAAA;AACzG,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAA;AACnD,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAA;AAChC,OAAO,EAAE,eAAe,EAAE,MAAM,eAAe,CAAA;AAC/C,OAAO,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAA;AACrD,OAAO,EAAE,QAAQ,EAAE,MAAM,YAAY,CAAA;AACrC,OAAO,EAAE,eAAe,EAAE,MAAM,cAAc,CAAA;AAC9C,OAAO,EAAE,WAAW,EAAsB,MAAM,cAAc,CAAA;AAC9D,OAAO,EAAE,UAAU,EAA2B,MAAM,aAAa,CAAA;AACjE,OAAO,EACN,eAAe,EACf,WAAW,EACX,mCAAmC,GAGnC,MAAM,YAAY,CAAA;AACnB,OAAO,EAAE,8BAA8B,EAAE,aAAa,EAAE,MAAM,iBAAiB,CAAA;AA+D/E;;;;;;;GAOG;AACH,MAAM,CAAC,KAAK,UAAU,WAAW,CAAC,IAAwB;IACzD,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,IAAI,sBAAsB,CAAC,IAAI,EAAE,CAAA;IAC/D,MAAM,UAAU,GAAG,IAAI,CAAC,UAAU,IAAI,IAAI,CAAA;IAC1C,MAAM,YAAY,GAAG,IAAI,CAAC,YAAY,IAAI,SAAS,CAAA;IACnD,MAAM,QAAQ,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAA;IAEzC,MAAM,KAAK,CAAC,IAAI,CAAC,SAAS,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAChD,MAAM,eAAe,GAAG,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,cAAc,CAAC,CAAA;IAC5D,MAAM,KAAK,CAAC,eAAe,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAA;IAEjD,mBAAmB;IACnB,MAAM,WAAW,GAAyB,EAAE,CAAA;IAC5C,MAAM,OAAO,GAAa,EAAE,CAAA;IAC5B,KAAK,MAAM,OAAO,IAAI,QAAQ,EAAE,CAAC;QAChC,MAAM,cAAc,GAAG,IAAI,CAAC,aAAa,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;QACrD,IAAI,CAAC,cAAc,EAAE,CAAC;YACrB,OAAO,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;YACxB,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,wBAAwB,CAAC,CAAA;YAC/E,SAAQ;QACT,CAAC;QACD,gGAAgG;QAChG,2FAA2F;QAC3F,+FAA+F;QAC/F,8FAA8F;QAC9F,sGAAsG;QACtG,MAAM,UAAU,GAAG,IAAI,CAAC,eAAe,EAAE,OAAO,CAAC,EAAE,CAAC,CAAA;QACpD,MAAM,cAAc,GAAG,IAAI,CAAC,UAAU,EAAE,eAAe,CAAC,CAAA;QACxD,IACC,OAAO,CAAC,GAAG,CAAC,gBAAgB,KAAK,GAAG;YACpC,UAAU,CAAC,cAAc,CAAC;YAC1B,UAAU,CAAC,IAAI,CAAC,UAAU,EAAE,iBAAiB,CAAC,CAAC,EAC9C,CAAC;YACF,MAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,YAAY,CAAC,cAAc,EAAE,MAAM,CAAC,CAAuB,CAAA;YACrF,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,YAAY,MAAM,CAAC,OAAO,kBAAkB,CAAC,CAAA;YACnG,WAAW,CAAC,IAAI,CAAC,MAAM,CAAC,CAAA;YACxB,SAAQ;QACT,CAAC;QACD,IAAI,CAAC,UAAU,EAAE,CAAC,aAAa,EAAE,WAAW,OAAO,CAAC,EAAE,EAAE,CAAC,CAAA;QACzD,MAAM,CAAC,GAAG,MAAM,UAAU,CAAC;YAC1B,OAAO;YACP,cAAc;YACd,SAAS,EAAE,eAAe;YAC1B,aAAa,EAAE,IAAI,CAAC,aAAa;SACjC,CAAC,CAAA;QACF,WAAW,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;IACpB,CAAC;IAED,yFAAyF;IACzF,8FAA8F;IAC9F,0FAA0F;IAC1F,0FAA0F;IAC1F,0EAA0E;IAC1E,MAAM,YAAY,GAA8B;QAC/C,KAAK,EAAE,IAAI,CAAC,eAAe,EAAE,qBAAqB,CAAC;QACnD,GAAG,EAAE,IAAI,CAAC,eAAe,EAAE,mBAAmB,CAAC;QAC/C,IAAI,EAAE,IAAI,CAAC,eAAe,EAAE,oBAAoB,CAAC;KACjD,CAAA;IACD,MAAM,cAAc,GAAmC;QACtD,KAAK,EAAE,iBAAiB,CAAC,YAAY,CAAC,KAAK,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;QAClE,GAAG,EAAE,iBAAiB,CAAC,YAAY,CAAC,GAAG,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;QAC9D,IAAI,EAAE,iBAAiB,CAAC,YAAY,CAAC,IAAI,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;KAChE,CAAA;IACD,MAAM,cAAc,GAAG,IAAI,CAAC,eAAe,EAAE,kBAAkB,CAAC,CAAA;IAChE,MAAM,gBAAgB,GAAG,iBAAiB,CAAC,cAAc,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;IAEhF,IAAI,OAAO,GAAG,CAAC,CAAA;IACf,IAAI,WAAW,GAAG,CAAC,CAAA;IACnB,MAAM,MAAM,GAA8B,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,IAAI,EAAE,CAAC,EAAE,CAAA;IACvE,MAAM,QAAQ,GAAG,eAAe,EAAE,CAAA;IAClC,8FAA8F;IAC9F,mGAAmG;IACnG,2EAA2E;IAC3E,MAAM,eAAe,GAAG,IAAI,CAAC,eAAe,IAAI,EAAE,CAAA;IAClD,MAAM,aAAa,GAAG,IAAI,GAAG,EAAkB,CAAA;IAC/C,IAAI,iBAAiB,GAAG,CAAC,CAAA;IAEzB,MAAM,eAAe,GAAG,CAAC,GAAiB,EAAE,MAAc,EAAQ,EAAE;QACnE,gBAAgB,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC,SAAS,CAAC,EAAE,GAAG,EAAE,MAAM,EAAE,CAAC,IAAI,CAAC,CAAA;IAC/D,CAAC,CAAA;IAED,KAAK,MAAM,UAAU,IAAI,WAAW,EAAE,CAAC;QACtC,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,YAAY,UAAU,CAAC,UAAU,EAAE,CAAC,CAAA;QAC/D,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,WAAW,CAAe,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;YAC1E,aAAa,CAAC,GAAG,CAAC,GAAG,CAAC,OAAO,EAAE,CAAC,aAAa,CAAC,GAAG,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;YACzE,0FAA0F;YAC1F,sFAAsF;YACtF,0FAA0F;YAC1F,2FAA2F;YAC3F,IAAI,eAAe,CAAC,GAAG,CAAC,OAAO,EAAE,eAAe,CAAC,EAAE,CAAC;gBACnD,iBAAiB,EAAE,CAAA;gBACnB,SAAQ;YACT,CAAC;YACD,MAAM,MAAM,GAAmB,CAAC,GAAG,CAAC,CAAA;YACpC,IAAI,UAAU,EAAE,CAAC;gBAChB,KAAK,MAAM,GAAG,IAAI,aAAa,CAAC,GAAG,EAAE,8BAA8B,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC;oBACnF,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;gBACjB,CAAC;YACF,CAAC;YACD,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;gBACxB,IAAI,MAAmC,CAAA;gBACvC,IAAI,CAAC;oBACJ,MAAM,GAAG,QAAQ,CAAC,CAAC,CAAC,CAAA;gBACrB,CAAC;gBAAC,OAAO,GAAG,EAAE,CAAC;oBACd,mFAAmF;oBACnF,qFAAqF;oBACrF,kFAAkF;oBAClF,iFAAiF;oBACjF,eAAe,CAAC,CAAC,EAAE,eAAgB,GAAa,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,CAAC,CAAA;oBACzE,WAAW,EAAE,CAAA;oBACb,SAAQ;gBACT,CAAC;gBACD,IAAI,MAAM,CAAC,IAAI,KAAK,SAAS,EAAE,CAAC;oBAC/B,MAAM,KAAK,GAAG,WAAW,CAAC,MAAM,CAAC,GAAG,EAAE,QAAQ,CAAC,CAAA;oBAC/C,cAAc,CAAC,KAAK,CAAC,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC,SAAS,CAAC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,CAAA;oBAC9D,MAAM,CAAC,KAAK,CAAC,EAAE,CAAA;oBACf,OAAO,EAAE,CAAA;gBACV,CAAC;qBAAM,CAAC;oBACP,eAAe,CAAC,CAAC,EAAE,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;oBACrC,WAAW,EAAE,CAAA;gBACd,CAAC;YACF,CAAC;QACF,CAAC;IACF,CAAC;IAED,KAAK,MAAM,CAAC,IAAI,MAAM,CAAC,MAAM,CAAC,cAAc,CAAC;QAAE,CAAC,CAAC,GAAG,EAAE,CAAA;IACtD,gBAAgB,CAAC,GAAG,EAAE,CAAA;IACtB,MAAM,OAAO,CAAC,GAAG,CAAC,CAAC,GAAG,MAAM,CAAC,MAAM,CAAC,cAAc,CAAC,CAAC,GAAG,CAAC,SAAS,CAAC,EAAE,SAAS,CAAC,gBAAgB,CAAC,CAAC,CAAC,CAAA;IAEjG,sFAAsF;IACtF,uFAAuF;IACvF,yEAAyE;IACzE,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,aAAa,OAAO,eAAe,CAAC,CAAA;IAC/D,MAAM,SAAS,GAAG,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,QAAQ,CAAC,CAAA;IAChD,MAAM,WAAW,GAAG,MAAM,mCAAmC,CAAC;QAC7D,YAAY;QACZ,SAAS,EAAE,SAAS;QACpB,aAAa,EAAE,IAAI,CAAC,aAAa;QACjC,MAAM;QACN,QAAQ;KACR,CAAC,CAAA;IAEF,4FAA4F;IAC5F,0FAA0F;IAC1F,IAAI,CAAC,UAAU,EAAE,CAAC,OAAO,EAAE,wBAAwB,CAAC,CAAA;IACpD,MAAM,aAAa,GAAG,MAAM,WAAW,CACtC;QACC,KAAK,EAAE,WAAW,CAAa,YAAY,CAAC,KAAK,CAAC;QAClD,GAAG,EAAE,WAAW,CAAa,YAAY,CAAC,GAAG,CAAC;QAC9C,IAAI,EAAE,WAAW,CAAa,YAAY,CAAC,IAAI,CAAC;KAChD,EACD;QACC,SAAS,EAAE,IAAI,CAAC,SAAS;QACzB,aAAa,EAAE,IAAI,CAAC,aAAa;QACjC,YAAY;KACZ,CACD,CAAA;IAED,gGAAgG;IAChG,iGAAiG;IACjG,MAAM,cAAc,GAAG,CAAC,GAAG,aAAa,CAAC,OAAO,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAA;IAC/E,IAAI,CAAC,UAAU,EAAE,CAChB,UAAU,EACV,gBAAgB,cAAc,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE;QACvE,CAAC,iBAAiB,GAAG,CAAC;YACrB,CAAC,CAAC,eAAe,iBAAiB,6BAA6B;YAC/D,CAAC,CAAC,iDAAiD,CAAC,CACtD,CAAA;IAED,yBAAyB;IACzB,IAAI,CAAC,UAAU,EAAE,CAAC,UAAU,EAAE,iCAAiC,CAAC,CAAA;IAChE,MAAM,QAAQ,GAAwB;QACrC,cAAc,EAAE,IAAI,CAAC,aAAa;QAClC,QAAQ;QACR,QAAQ,EAAE,WAAW;QACrB,gBAAgB,EAAE,OAAO;QACzB,MAAM,EAAE,EAAE,MAAM,EAAE,WAAW,EAAE,QAAQ,EAAE;QACzC,MAAM,EAAE,EAAE,MAAM,EAAE,aAAa,CAAC,MAAM,EAAE,UAAU,EAAE,aAAa,CAAC,UAAU,EAAE;QAC9E,gBAAgB,EAAE,WAAW;QAC7B,kBAAkB,EAAE,OAAO;QAC3B,QAAQ,EAAE,MAAM,CAAC,WAAW,CAAC,cAAc,CAAC;QAC5C,mBAAmB,EAAE,iBAAiB;KACtC,CAAA;IACD,MAAM,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,eAAe,CAAC,EAAE,GAAG,IAAI,CAAC,SAAS,CAAC,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,MAAM,CAAC,CAAA;IACxG,OAAO,QAAQ,CAAA;AAChB,CAAC;AAED,KAAK,SAAS,CAAC,CAAC,WAAW,CAAI,IAAY;IAC1C,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,EAAE,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC,CAAA;IAC3D,MAAM,EAAE,GAAG,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAA;IAClE,IAAI,KAAK,EAAE,MAAM,IAAI,IAAI,EAAE,EAAE,CAAC;QAC7B,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAA;QAC3B,IAAI,CAAC,OAAO;YAAE,SAAQ;QACtB,MAAM,IAAI,CAAC,KAAK,CAAC,OAAO,CAAM,CAAA;IAC/B,CAAC;AACF,CAAC;AAED,SAAS,SAAS,CAAC,CAAc;IAChC,OAAO,IAAI,OAAO,CAAC,CAAC,OAAO,EAAE,MAAM,EAAE,EAAE;QACtC,CAAC,CAAC,IAAI,CAAC,OAAO,EAAE,OAAO,CAAC,CAAA;QACxB,CAAC,CAAC,IAAI,CAAC,OAAO,EAAE,MAAM,CAAC,CAAA;IACxB,CAAC,CAAC,CAAA;AACH,CAAC"}

package/out/src/license.d.ts

@@ -0,0 +1,37 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Corpus licensing — the single source of truth for the training-data license policy (#26).
+ *
+ *   Posture (operator, 2026-06-19): **exclusion is a deliberate act, not a silent default.** The
+ *   build INCLUDES every row an adapter yields (stamping its `license`); a build that needs a clean
+ *   license set — e.g. the proprietary `@mailwoman/neural-weights-*` weights, which must not
+ *   inherit a share-alike obligation — PURPOSELY excludes kinds via `buildCorpus({ excludeLicenses
+ *   })` (CLI `--exclude-licenses` / `--exclude-share-alike`). Nothing is dropped on a license
+ *   string unless the operator named it. This avoids the trap of silently dropping allowed data
+ *   mis-stamped with a conservative license (e.g. BAN, which is dual-licensed Licence Ouverte OR
+ *   ODbL — we elect Licence Ouverte; a default-deny on the old `ODbL` stamp would have wrongly
+ *   dropped 48M allowed rows).
+ *
+ *   Tier reference (#26): A = PD/CC0 (allowed); B = CC-BY / Licence Ouverte (allowed WITH attribution
+ *   — the model card must carry it); C = share-alike (ODbL, CC-BY-SA, CC-SA) — exclude for a
+ *   proprietary-weights build via `--exclude-share-alike`.
+ */
+/**
+ * Licenses that require share-alike / create a copyleft obligation on derived works (Tier C). The
+ * `--exclude-share-alike` convenience expands to this; `allowShareAlike: false` adapters also use
+ * it.
+ */
+export declare const SHARE_ALIKE_PATTERN: RegExp;
+/**
+ * Compile a `--exclude-licenses` spec (comma-separated, e.g. `"ODbL,CC-BY-SA"`) into anchored,
+ * case-insensitive prefix patterns. Each entry matches a license string that STARTS with it, so
+ * `CC-BY-SA` catches `CC-BY-SA-3.0`, `CC-BY-SA-4.0`, etc. Regex metacharacters are escaped — the
+ * spec is a literal license prefix, not a user-supplied regex.
+ */
+export declare function compileLicenseExcludes(spec: string): RegExp[];
+/** True iff `license` matches any of the exclude `patterns` (empty patterns → never excluded). */
+export declare function licenseExcluded(license: string | undefined, patterns: readonly RegExp[]): boolean;
+//# sourceMappingURL=license.d.ts.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"license.d.ts","sourceRoot":"","sources":["../../src/license.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;GAoBG;AAEH;;;;GAIG;AACH,eAAO,MAAM,mBAAmB,QAAmD,CAAA;AAEnF;;;;;GAKG;AACH,wBAAgB,sBAAsB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,EAAE,CAM7D;AAED,kGAAkG;AAClG,wBAAgB,eAAe,CAAC,OAAO,EAAE,MAAM,GAAG,SAAS,EAAE,QAAQ,EAAE,SAAS,MAAM,EAAE,GAAG,OAAO,CAGjG"}

package/out/src/license.js

@@ -0,0 +1,46 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Corpus licensing — the single source of truth for the training-data license policy (#26).
+ *
+ *   Posture (operator, 2026-06-19): **exclusion is a deliberate act, not a silent default.** The
+ *   build INCLUDES every row an adapter yields (stamping its `license`); a build that needs a clean
+ *   license set — e.g. the proprietary `@mailwoman/neural-weights-*` weights, which must not
+ *   inherit a share-alike obligation — PURPOSELY excludes kinds via `buildCorpus({ excludeLicenses
+ *   })` (CLI `--exclude-licenses` / `--exclude-share-alike`). Nothing is dropped on a license
+ *   string unless the operator named it. This avoids the trap of silently dropping allowed data
+ *   mis-stamped with a conservative license (e.g. BAN, which is dual-licensed Licence Ouverte OR
+ *   ODbL — we elect Licence Ouverte; a default-deny on the old `ODbL` stamp would have wrongly
+ *   dropped 48M allowed rows).
+ *
+ *   Tier reference (#26): A = PD/CC0 (allowed); B = CC-BY / Licence Ouverte (allowed WITH attribution
+ *   — the model card must carry it); C = share-alike (ODbL, CC-BY-SA, CC-SA) — exclude for a
+ *   proprietary-weights build via `--exclude-share-alike`.
+ */
+/**
+ * Licenses that require share-alike / create a copyleft obligation on derived works (Tier C). The
+ * `--exclude-share-alike` convenience expands to this; `allowShareAlike: false` adapters also use
+ * it.
+ */
+export const SHARE_ALIKE_PATTERN = /^ODbL|^Open Database License|^CC-BY-SA|^CC-SA/i;
+/**
+ * Compile a `--exclude-licenses` spec (comma-separated, e.g. `"ODbL,CC-BY-SA"`) into anchored,
+ * case-insensitive prefix patterns. Each entry matches a license string that STARTS with it, so
+ * `CC-BY-SA` catches `CC-BY-SA-3.0`, `CC-BY-SA-4.0`, etc. Regex metacharacters are escaped — the
+ * spec is a literal license prefix, not a user-supplied regex.
+ */
+export function compileLicenseExcludes(spec) {
+    return spec
+        .split(",")
+        .map((s) => s.trim())
+        .filter(Boolean)
+        .map((s) => new RegExp("^" + s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&"), "i"));
+}
+/** True iff `license` matches any of the exclude `patterns` (empty patterns → never excluded). */
+export function licenseExcluded(license, patterns) {
+    const l = license ?? "";
+    return patterns.some((p) => p.test(l));
+}
+//# sourceMappingURL=license.js.map

package/out/src/license.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"license.js","sourceRoot":"","sources":["../../src/license.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;GAoBG;AAEH;;;;GAIG;AACH,MAAM,CAAC,MAAM,mBAAmB,GAAG,gDAAgD,CAAA;AAEnF;;;;;GAKG;AACH,MAAM,UAAU,sBAAsB,CAAC,IAAY;IAClD,OAAO,IAAI;SACT,KAAK,CAAC,GAAG,CAAC;SACV,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;SACpB,MAAM,CAAC,OAAO,CAAC;SACf,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,IAAI,MAAM,CAAC,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC,qBAAqB,EAAE,MAAM,CAAC,EAAE,GAAG,CAAC,CAAC,CAAA;AAC9E,CAAC;AAED,kGAAkG;AAClG,MAAM,UAAU,eAAe,CAAC,OAA2B,EAAE,QAA2B;IACvF,MAAM,CAAC,GAAG,OAAO,IAAI,EAAE,CAAA;IACvB,OAAO,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAA;AACvC,CAAC"}

package/out/src/synthesize-boundary-stress.d.ts

@@ -0,0 +1,77 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Boundary-instability synthesizer (#375 — the highest-leverage parser lever). The failure taxonomy
+ *
+ *   The within-token-punctuation decomposition (#702) found one failure FAMILY surfacing under many
+ *   names: the model mis-places token boundaries between adjacent components when the boundary is
+ *   ambiguous or unmarked. This generator emits diverse BIO-labeled rows that put the gold boundary
+ *   exactly where the model wobbles, so a retrain learns the boundary from context, not the
+ *   lexeme.
+ *
+ *   The four token-aligned stress shapes, all in BASE LOCALES (US/FR/DE) so the shard never
+ *   introduces tokens the base corpus lacks (the #511 base-consistency lint flagged an earlier
+ *   AU-bearing draft: AU 4-digit postcodes collide with US house numbers, and AU localities are
+ *   absent from the US/FR/DE base — a real contradiction). Each component is a whitespace-separated
+ *   token run, so `alignRow` labels it cleanly:
+ *
+ *   1. `street-eats-affix` — multi-word street + suffix (`Country Club Rd` → street + street_suffix),
+ *        the #1 wobble: the model keeps the suffix in the street.
+ *   2. `comma-less-city-state` — no comma between street / locality / region (`100 Main St Springfield
+ *        IL 62701`), the #694 family: concatenated input loses the segmentation cue. US-only (US
+ *        zips are base-consistent; the boundary is locale-agnostic).
+ *   3. `fr-prefix` — FR street-type prefix split from the name (`Rue Jean-Baptiste Lebas` →
+ *        street_prefix
+ *
+ *        - Street), postcode-first order.
+ *   4. `house-number-after-street` — FR/DE number-follows-street (`Neuve-des-Capucines 5` → street +
+ *        house_number), the model absorbs the number into the street.
+ *
+ *   Two BALANCING shapes (added 2026-06-18 after the v1.6.0 probes). The first pass at weight 1.0
+ *   lifted the boundaries but over-fit a NARROW distribution — every row was a clean, full,
+ *   structured address — so the model regressed on out-of-distribution real rows (held-out US
+ *   locality 66.3→58.2%). These two widen the distribution the shard teaches over, per the
+ *   diagnosis (scripts/eval/locality-regression-probe): 5. `bare-locality` — locality with NO
+ *   street (`Public Library, Lisbon ND`, `75003 Paris`), the ship-blocker: 84% of the v1.6.0
+ *   locality regression was DROPPED locality on bare "City, STATE" rows, because every other shape
+ *   placed a street before the city. Bare / comma-less / postcode'd / venue-prefixed forms, US +
+ *   FR. 6. `house-number-before-street` — the confounding mirror of #4 (same FR vocab, number
+ *   BEFORE the street). A balanced before:after mix breaks the positional shortcut behind the #4
+ *   order-bias.
+ *
+ *   EXCLUDED: the region+postcode glue (`NY14201` — sub-token, no punctuation to split) and the
+ *   AU/NZ/UK slash unit-convention (`4/2A` → unit+house_number). The slash labels cleanly (the
+ *   tokenizer splits `/`) and is the worst within-token class — but it inherently requires non-base
+ *   AU/NZ/UK locales, which contradict the US/FR/DE base (the lint catch). It belongs in a
+ *   separately-scoped AU/NZ/UK boundary-coverage shard that ALSO adds AU base coverage, not in this
+ *   base-locale shard. `synthesize-boundary-stress.test.ts` proves the alignments.
+ */
+import type { CanonicalRow } from "./types.js";
+export type BoundaryStressTemplate = "street-eats-affix" | "comma-less-city-state" | "fr-prefix" | "house-number-after-street" | "bare-locality" | "house-number-before-street";
+export interface BoundaryStressBaseTuple {
+    locality: string;
+    region: string;
+    postcode: string;
+    country: string;
+}
+export interface BoundaryStressSynthesisOpts {
+    random?: () => number;
+    /** Force a specific shape (tests + balanced shard composition). */
+    forceTemplate?: BoundaryStressTemplate;
+}
+export interface SynthesizedBoundaryStressRow {
+    raw: string;
+    components: CanonicalRow["components"];
+    locale: string;
+    template: BoundaryStressTemplate;
+}
+/**
+ * Synthesize one boundary-stress row. `base` is optional — when omitted, a locale-appropriate tuple
+ * is drawn from the internal pools (so the generator is self-contained; a build script can pass
+ * real tuples for scale + diversity). Every component value is a verbatim substring of `raw`, so
+ * `alignRow` locates + BIO-labels it.
+ */
+export declare function synthesizeBoundaryStressRow(base: BoundaryStressBaseTuple | undefined, opts?: BoundaryStressSynthesisOpts): SynthesizedBoundaryStressRow;
+//# sourceMappingURL=synthesize-boundary-stress.d.ts.map

package/out/src/synthesize-boundary-stress.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"synthesize-boundary-stress.d.ts","sourceRoot":"","sources":["../../src/synthesize-boundary-stress.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAiDG;AAEH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,YAAY,CAAA;AAE9C,MAAM,MAAM,sBAAsB,GAC/B,mBAAmB,GACnB,uBAAuB,GACvB,WAAW,GACX,2BAA2B,GAG3B,eAAe,GACf,4BAA4B,CAAA;AAE/B,MAAM,WAAW,uBAAuB;IACvC,QAAQ,EAAE,MAAM,CAAA;IAChB,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,OAAO,EAAE,MAAM,CAAA;CACf;AAED,MAAM,WAAW,2BAA2B;IAC3C,MAAM,CAAC,EAAE,MAAM,MAAM,CAAA;IACrB,mEAAmE;IACnE,aAAa,CAAC,EAAE,sBAAsB,CAAA;CACtC;AAED,MAAM,WAAW,4BAA4B;IAC5C,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,CAAA;IACtC,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,sBAAsB,CAAA;CAChC;AAyRD;;;;;GAKG;AACH,wBAAgB,2BAA2B,CAC1C,IAAI,EAAE,uBAAuB,GAAG,SAAS,EACzC,IAAI,GAAE,2BAAgC,GACpC,4BAA4B,CA4H9B"}

package/out/src/synthesize-boundary-stress.js

@@ -0,0 +1,449 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Boundary-instability synthesizer (#375 — the highest-leverage parser lever). The failure taxonomy
+ *
+ *   The within-token-punctuation decomposition (#702) found one failure FAMILY surfacing under many
+ *   names: the model mis-places token boundaries between adjacent components when the boundary is
+ *   ambiguous or unmarked. This generator emits diverse BIO-labeled rows that put the gold boundary
+ *   exactly where the model wobbles, so a retrain learns the boundary from context, not the
+ *   lexeme.
+ *
+ *   The four token-aligned stress shapes, all in BASE LOCALES (US/FR/DE) so the shard never
+ *   introduces tokens the base corpus lacks (the #511 base-consistency lint flagged an earlier
+ *   AU-bearing draft: AU 4-digit postcodes collide with US house numbers, and AU localities are
+ *   absent from the US/FR/DE base — a real contradiction). Each component is a whitespace-separated
+ *   token run, so `alignRow` labels it cleanly:
+ *
+ *   1. `street-eats-affix` — multi-word street + suffix (`Country Club Rd` → street + street_suffix),
+ *        the #1 wobble: the model keeps the suffix in the street.
+ *   2. `comma-less-city-state` — no comma between street / locality / region (`100 Main St Springfield
+ *        IL 62701`), the #694 family: concatenated input loses the segmentation cue. US-only (US
+ *        zips are base-consistent; the boundary is locale-agnostic).
+ *   3. `fr-prefix` — FR street-type prefix split from the name (`Rue Jean-Baptiste Lebas` →
+ *        street_prefix
+ *
+ *        - Street), postcode-first order.
+ *   4. `house-number-after-street` — FR/DE number-follows-street (`Neuve-des-Capucines 5` → street +
+ *        house_number), the model absorbs the number into the street.
+ *
+ *   Two BALANCING shapes (added 2026-06-18 after the v1.6.0 probes). The first pass at weight 1.0
+ *   lifted the boundaries but over-fit a NARROW distribution — every row was a clean, full,
+ *   structured address — so the model regressed on out-of-distribution real rows (held-out US
+ *   locality 66.3→58.2%). These two widen the distribution the shard teaches over, per the
+ *   diagnosis (scripts/eval/locality-regression-probe): 5. `bare-locality` — locality with NO
+ *   street (`Public Library, Lisbon ND`, `75003 Paris`), the ship-blocker: 84% of the v1.6.0
+ *   locality regression was DROPPED locality on bare "City, STATE" rows, because every other shape
+ *   placed a street before the city. Bare / comma-less / postcode'd / venue-prefixed forms, US +
+ *   FR. 6. `house-number-before-street` — the confounding mirror of #4 (same FR vocab, number
+ *   BEFORE the street). A balanced before:after mix breaks the positional shortcut behind the #4
+ *   order-bias.
+ *
+ *   EXCLUDED: the region+postcode glue (`NY14201` — sub-token, no punctuation to split) and the
+ *   AU/NZ/UK slash unit-convention (`4/2A` → unit+house_number). The slash labels cleanly (the
+ *   tokenizer splits `/`) and is the worst within-token class — but it inherently requires non-base
+ *   AU/NZ/UK locales, which contradict the US/FR/DE base (the lint catch). It belongs in a
+ *   separately-scoped AU/NZ/UK boundary-coverage shard that ALSO adds AU base coverage, not in this
+ *   base-locale shard. `synthesize-boundary-stress.test.ts` proves the alignments.
+ */
+function pick(arr, random) {
+    return arr[Math.floor(random() * arr.length)];
+}
+// Multi-word street names — the suffix boundary only bites when "Club" could be read as part of the
+// name. Single-word names alone teach nothing about the suffix edge.
+// Multi-word names are what make the suffix boundary BITE (the model must not read the trailing
+// suffix word as part of the name). Kept diverse so the shard teaches the boundary, not the lexeme.
+const MULTIWORD_STREETS = [
+    "Country Club",
+    "Martin Luther King",
+    "Forest Hill",
+    "Lake View",
+    "Spring Valley",
+    "Cedar Ridge",
+    "Old Mill",
+    "Sunset Park",
+    "Maple Grove",
+    "Stone Creek",
+    "Glen Cove",
+    "Pine Bluff",
+    "Fox Hollow",
+    "Briar Patch",
+    "West End",
+    "College Station",
+    "Quail Hollow",
+    "Eagle Ridge",
+    "Deer Run",
+    "Bear Creek",
+    "Willow Bend",
+    "Cypress Point",
+    "Laurel Oak",
+    "Magnolia Park",
+    "Cherry Hill",
+    "Walnut Grove",
+    "Birch Hollow",
+    "Aspen Grove",
+    "Juniper Ridge",
+    "Hidden Valley",
+    "Rolling Hills",
+    "Tanglewood",
+    "Meadow Brook",
+    "Clover Field",
+    "Sunrise Point",
+    "Harbor View",
+    "Bay Shore",
+    "Ocean Breeze",
+    "Mountain View",
+    "Valley Forge",
+    "Liberty Square",
+    "Washington Crossing",
+    "Kings Highway",
+    "Queens Gate",
+    "Princeton Junction",
+];
+const SINGLE_STREETS = [
+    "Main",
+    "Oak",
+    "Maple",
+    "Park",
+    "Washington",
+    "Lincoln",
+    "Church",
+    "River",
+    "Pine",
+    "Cedar",
+    "Elm",
+    "Jefferson",
+    "Madison",
+    "Adams",
+    "Jackson",
+    "Franklin",
+    "Highland",
+    "Sunset",
+    "Lakeview",
+    "Hillcrest",
+    "Cambridge",
+    "Devonshire",
+    "Sherwood",
+    "Kingston",
+    "Berkshire",
+    "Aberdeen",
+    "Belmont",
+    "Carlisle",
+    "Dover",
+    "Easton",
+    "Fairfax",
+    "Greenwood",
+    "1st",
+    "2nd",
+    "3rd",
+    "4th",
+    "5th",
+    "12th",
+    "42nd",
+];
+const SUFFIXES = [
+    "St",
+    "Street",
+    "Ave",
+    "Avenue",
+    "Rd",
+    "Road",
+    "Blvd",
+    "Boulevard",
+    "Ln",
+    "Lane",
+    "Dr",
+    "Drive",
+    "Pkwy",
+    "Parkway",
+    "Way",
+    "Ct",
+    "Court",
+    "Pl",
+    "Place",
+    "Cir",
+    "Circle",
+    "Ter",
+    "Terrace",
+    "Hwy",
+    "Trail",
+    "Loop",
+    "Cres",
+    "Crescent",
+    "Row",
+    "Walk",
+];
+const DIRECTIONALS = ["N", "S", "E", "W", "NE", "NW", "SE", "SW"];
+// FR street-type prefixes + hyphenated honorific street names (the hyphen is incidental; the boundary
+// stress is the prefix↔name split + the number-after-street order).
+const FR_PREFIXES = [
+    "Rue",
+    "Avenue",
+    "Boulevard",
+    "Place",
+    "Impasse",
+    "Chemin",
+    "Quai",
+    "Cours",
+    "Allée",
+    "Passage",
+    "Square",
+    "Villa",
+    "Sentier",
+    "Promenade",
+];
+const FR_NAMES = [
+    "Jean-Baptiste Lebas",
+    "Neuve-des-Capucines",
+    "Charles-de-Gaulle",
+    "du Général-Leclerc",
+    "de la République",
+    "des Trois-Frères",
+    "Victor-Hugo",
+    "Jean-Jaurès",
+    "de l'Abreuvoir",
+    "Émile-Zola",
+    "Gambetta",
+    "Jean-Moulin",
+    "des Martyrs-de-la-Résistance",
+    "du Maréchal-Foch",
+    "Pierre-et-Marie-Curie",
+    "Antoine-de-Saint-Exupéry",
+    "de la Liberté",
+    "des Quatre-Vents",
+    "du Faubourg-Saint-Antoine",
+    "Saint-Honoré",
+    "de la Pompe",
+    "des Petits-Champs",
+    "Léon-Blum",
+    "Aristide-Briand",
+];
+// Org/venue prefixes for the bare-locality shape — the v1.6.0 locality drop hit org-PREFIXED real rows
+// hardest ("LISBON PUBLIC LIBRARY, …, Lisbon ND"; "Alburg Health Center"). Teaching the locality WITH a
+// leading venue keeps the model emitting it on facility-style addresses (NPPES/HRSA shapes). `venue` is a
+// base ComponentTag.
+//
+// #511-LINTED 2026-06-18 (scripts/lint-venue-vocab — scan of nppes/hrsa/tiger/nad/wof-admin): every token
+// here is venue-DOMINANT in the base, so the shard agrees with it. The first draft was naive — 9 terms
+// were dropped because their tokens are dominantly street/locality and would CONTRADICT the base the way
+// Madison-as-street did (#511): "Fire" 93% street, "Veterans" 94% street, "City" 68% locality, "Hall" 63%
+// street, "Memorial"/"Hospital" 62-63% street, "Recreation" 79% street, "Town" 48% locality, "Library" 51%
+// street, "County" 68% street, "Arts"/"Courthouse"/"Municipal" dependent_locality. Kept tokens: Clinic 98%,
+// Practice 98%, Dental 100%, Health 99%, Medical 88%, Community 92%, Department 90%, Group 87%, Center 65%,
+// School 70%, Public 89%, Elementary/Family 97% (all venue).
+const VENUES = [
+    "Community Center",
+    "Health Center",
+    "Medical Center",
+    "Medical Clinic",
+    "Family Clinic",
+    "Community Clinic",
+    "Dental Clinic",
+    "Family Practice",
+    "Medical Practice",
+    "Dental Group",
+    "Medical Group",
+    "Health Department",
+    "Elementary School",
+    "Public School",
+];
+// Localities DERIVED from the base corpus (#511): every name here is verified locality-DOMINANT in the
+// training data (B-locality ≫ I-street), so the shard agrees with the base instead of fighting it. The
+// night's targeted scan caught the prior vocab (Madison, Portland, Springfield IL…) at 92–100% STREET
+// in the base ("Madison Ave"), the "5th Avenue Theatre" #511 trap. See 2026-06-17-locality-vocab-fix.
+const US_TUPLES = [
+    { locality: "Albuquerque", region: "NM", postcode: "87102", country: "US" },
+    { locality: "Indianapolis", region: "IN", postcode: "46203", country: "US" },
+    { locality: "Sacramento", region: "CA", postcode: "95823", country: "US" },
+    { locality: "Rochester", region: "NY", postcode: "14606", country: "US" },
+    { locality: "Jacksonville", region: "FL", postcode: "32209", country: "US" },
+    { locality: "Portsmouth", region: "VA", postcode: "23704", country: "US" },
+    { locality: "Merced", region: "CA", postcode: "95340", country: "US" },
+    { locality: "Miami", region: "FL", postcode: "33125", country: "US" },
+    { locality: "Tampa", region: "FL", postcode: "33624", country: "US" },
+    { locality: "Orlando", region: "FL", postcode: "32827", country: "US" },
+    { locality: "Tulsa", region: "OK", postcode: "74133", country: "US" },
+    { locality: "Louisville", region: "KY", postcode: "40203", country: "US" },
+    { locality: "Nashville", region: "TN", postcode: "37207", country: "US" },
+    { locality: "Spokane", region: "WA", postcode: "99202", country: "US" },
+    { locality: "Akron", region: "OH", postcode: "44313", country: "US" },
+    { locality: "Fairbanks", region: "AK", postcode: "99701", country: "US" },
+    { locality: "Plano", region: "TX", postcode: "75024", country: "US" },
+    { locality: "Shreveport", region: "LA", postcode: "71103", country: "US" },
+    { locality: "Southfield", region: "MI", postcode: "48034", country: "US" },
+    { locality: "Glendale", region: "CA", postcode: "91203", country: "US" },
+    { locality: "Philadelphia", region: "PA", postcode: "19104", country: "US" },
+    { locality: "Brooklyn", region: "NY", postcode: "11230", country: "US" },
+    { locality: "Bronx", region: "NY", postcode: "10461", country: "US" },
+    { locality: "Fairport", region: "NY", postcode: "14450", country: "US" },
+    { locality: "Syracuse", region: "NE", postcode: "68446", country: "US" },
+    { locality: "Marion", region: "AR", postcode: "72364", country: "US" },
+    { locality: "Chicago", region: "IL", postcode: "60625", country: "US" },
+    { locality: "Springfield", region: "MA", postcode: "01108", country: "US" },
+];
+// FR localities DERIVED from the FR (ban) shards specifically — where these famous cities are 95–99%
+// locality-DOMINANT (Paris 515605/24789, Marseille 247014/1752, Lyon 106239/3114). NB: the all-shard
+// scan falsely flagged them street-dominant by undersampling the FR block (parts 180–209) and mixing in
+// US street-contexts; the FR-block scan is the honest distribution. Dept-diverse (28 depts), region
+// empty (French addresses carry no region token; the generator's region-optional path handles it).
+const FR_TUPLES = [
+    { locality: "Paris", region: "", postcode: "75003", country: "FR" },
+    { locality: "Marseille", region: "", postcode: "13016", country: "FR" },
+    { locality: "Lyon", region: "", postcode: "69009", country: "FR" },
+    { locality: "Perpignan", region: "", postcode: "66000", country: "FR" },
+    { locality: "Toulon", region: "", postcode: "83100", country: "FR" },
+    { locality: "Avignon", region: "", postcode: "84140", country: "FR" },
+    { locality: "Poitiers", region: "", postcode: "86000", country: "FR" },
+    { locality: "Arles", region: "", postcode: "13280", country: "FR" },
+    { locality: "Annecy", region: "", postcode: "74940", country: "FR" },
+    { locality: "Mulhouse", region: "", postcode: "68200", country: "FR" },
+    { locality: "Carpentras", region: "", postcode: "84200", country: "FR" },
+    { locality: "Antony", region: "", postcode: "92160", country: "FR" },
+    { locality: "Sartrouville", region: "", postcode: "78500", country: "FR" },
+    { locality: "Épinal", region: "", postcode: "88000", country: "FR" },
+    { locality: "Meyzieu", region: "", postcode: "69330", country: "FR" },
+    { locality: "Sens", region: "", postcode: "89100", country: "FR" },
+    { locality: "Brunoy", region: "", postcode: "91800", country: "FR" },
+    { locality: "Rambouillet", region: "", postcode: "78120", country: "FR" },
+];
+// NB: no DE_TUPLES — German cities are street-dominated too ("Berliner Straße"), and the base yielded
+// zero locality-dominant DE towns in the scan, so house-number-after-street is FR-only here. DE's
+// native-order number-after-street is covered by the dedicated synth-german shard.
+const houseNumber = (random) => String(1 + Math.floor(random() * 4999));
+const localeFor = { US: "en-US", FR: "fr-FR", DE: "de-DE" };
+const ALL_TEMPLATES = [
+    "street-eats-affix",
+    "comma-less-city-state",
+    "fr-prefix",
+    "house-number-after-street",
+    "bare-locality",
+    "house-number-before-street",
+];
+/**
+ * Synthesize one boundary-stress row. `base` is optional — when omitted, a locale-appropriate tuple
+ * is drawn from the internal pools (so the generator is self-contained; a build script can pass
+ * real tuples for scale + diversity). Every component value is a verbatim substring of `raw`, so
+ * `alignRow` locates + BIO-labels it.
+ */
+export function synthesizeBoundaryStressRow(base, opts = {}) {
+    const random = opts.random ?? Math.random;
+    const template = opts.forceTemplate ?? pick(ALL_TEMPLATES, random);
+    if (template === "bare-locality") {
+        // The v1.6.0 ship-blocker fix: locality was DROPPED on bare/short "City, STATE" rows (84% of the
+        // regression) because every prior shape placed a street before the city, so the model learned "the
+        // city follows a street" and stopped emitting locality without one. Teach the locality with NO street,
+        // across the forms real data carries it — bare, comma-LESS, postcode'd, and venue/org-prefixed.
+        const b = base ?? (random() < 0.3 ? pick(FR_TUPLES, random) : pick(US_TUPLES, random));
+        const venue = random() < 0.45 ? pick(VENUES, random) : "";
+        // ~12% carry a trailing country token — the v1.7.1 country patch (DeepSeek 2026-06-18). The pure
+        // "City, STATE" bare rows carry NO country token, which cost ~4pp on us.country_homograph in v1.7.0;
+        // teaching "…, USA"/"…, France" recovers it as a single-variable additive without diluting locality.
+        const withCountry = random() < 0.12;
+        if (b.country === "FR") {
+            // FR carries no region token; "{postcode} {locality}" is the bare FR form.
+            const core = `${b.postcode} ${b.locality}${withCountry ? ", France" : ""}`;
+            return {
+                raw: venue ? `${venue}, ${core}` : core,
+                components: {
+                    ...(venue ? { venue } : {}),
+                    postcode: b.postcode,
+                    locality: b.locality,
+                    ...(withCountry ? { country: "France" } : {}),
+                },
+                locale: "fr-FR",
+                template,
+            };
+        }
+        const withZip = random() < 0.5;
+        const comma = random() < 0.6 ? "," : ""; // include the comma-LESS "City STATE" form too
+        // "United States" (United 98% / States 98% country in the base), NOT "USA" — the #511 lint found
+        // "USA" is locality-DOMINANT (75%, only 6% country) in the base; labeling it country would contradict.
+        const countryName = "United States";
+        const core = `${b.locality}${comma} ${b.region}${withZip ? ` ${b.postcode}` : ""}${withCountry ? `, ${countryName}` : ""}`;
+        return {
+            raw: venue ? `${venue}, ${core}` : core,
+            components: {
+                ...(venue ? { venue } : {}),
+                locality: b.locality,
+                region: b.region,
+                ...(withZip ? { postcode: b.postcode } : {}),
+                ...(withCountry ? { country: countryName } : {}),
+            },
+            locale: "en-US",
+            template,
+        };
+    }
+    if (template === "fr-prefix" ||
+        template === "house-number-after-street" ||
+        template === "house-number-before-street") {
+        // FR-only (no base-consistent DE locality vocab; see the DE_TUPLES note above).
+        const b = base ?? pick(FR_TUPLES, random);
+        const name = pick(FR_NAMES, random);
+        const hn = houseNumber(random);
+        if (template === "house-number-before-street") {
+            // The confounding MIRROR of house-number-after-street: the SAME FR street vocab with the number
+            // BEFORE the name. A balanced before:after mix (the build/recipe sets the ratio, ~7:3 to keep US
+            // house_number 99.8% safe) teaches the model a street-adjacent number is a house_number by FORM,
+            // not position — the probe found v1.6.0 confidently absorbs the TRAILING number into street (I-street
+            // P=0.96), the order-bias.
+            const raw = `${hn} ${name}, ${b.postcode} ${b.locality}`;
+            return {
+                raw,
+                components: { house_number: hn, street: name, postcode: b.postcode, locality: b.locality },
+                locale: localeFor[b.country] ?? "fr-FR",
+                template,
+            };
+        }
+        if (template === "fr-prefix") {
+            const prefix = pick(FR_PREFIXES, random);
+            // "{hn} {prefix} {name}, {postcode} {locality}" — postcode-first, prefix split from the name.
+            const raw = `${hn} ${prefix} ${name}, ${b.postcode} ${b.locality}`;
+            return {
+                raw,
+                components: {
+                    house_number: hn,
+                    street_prefix: prefix,
+                    street: name,
+                    postcode: b.postcode,
+                    locality: b.locality,
+                },
+                locale: localeFor[b.country] ?? "fr-FR",
+                template,
+            };
+        }
+        // house-number-after-street: "{name} {hn}, {postcode} {locality}" — number FOLLOWS the street.
+        const raw = `${name} ${hn}, ${b.postcode} ${b.locality}`;
+        return {
+            raw,
+            components: { street: name, house_number: hn, postcode: b.postcode, locality: b.locality },
+            locale: localeFor[b.country] ?? "fr-FR",
+            template,
+        };
+    }
+    // en-US street shapes (street-eats-affix + comma-less). US-only — US zips are base-consistent and
+    // the boundary these teach is locale-agnostic; no need to introduce a non-base locale.
+    const b = base ?? pick(US_TUPLES, random);
+    const hn = houseNumber(random);
+    const dir = random() < 0.4 ? pick(DIRECTIONALS, random) : "";
+    const name = random() < 0.7 ? pick(MULTIWORD_STREETS, random) : pick(SINGLE_STREETS, random);
+    const suffix = pick(SUFFIXES, random);
+    const streetCore = `${dir ? `${dir} ` : ""}${name} ${suffix}`;
+    const components = {
+        house_number: hn,
+        ...(dir ? { street_prefix: dir } : {}),
+        street: name,
+        street_suffix: suffix,
+        locality: b.locality,
+        region: b.region,
+        postcode: b.postcode,
+    };
+    const raw = template === "comma-less-city-state"
+        ? // no commas anywhere — the segmentation cue is gone
+            `${hn} ${streetCore} ${b.locality} ${b.region} ${b.postcode}`
+        : // standard delimited, multi-word street stresses the suffix boundary
+            `${hn} ${streetCore}, ${b.locality}, ${b.region} ${b.postcode}`;
+    return { raw, components, locale: localeFor[b.country] ?? "en-US", template };
+}
+//# sourceMappingURL=synthesize-boundary-stress.js.map

package/out/src/synthesize-boundary-stress.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"synthesize-boundary-stress.js","sourceRoot":"","sources":["../../src/synthesize-boundary-stress.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAiDG;AAkCH,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,oGAAoG;AACpG,qEAAqE;AACrE,gGAAgG;AAChG,oGAAoG;AACpG,MAAM,iBAAiB,GAAG;IACzB,cAAc;IACd,oBAAoB;IACpB,aAAa;IACb,WAAW;IACX,eAAe;IACf,aAAa;IACb,UAAU;IACV,aAAa;IACb,aAAa;IACb,aAAa;IACb,WAAW;IACX,YAAY;IACZ,YAAY;IACZ,aAAa;IACb,UAAU;IACV,iBAAiB;IACjB,cAAc;IACd,aAAa;IACb,UAAU;IACV,YAAY;IACZ,aAAa;IACb,eAAe;IACf,YAAY;IACZ,eAAe;IACf,aAAa;IACb,cAAc;IACd,cAAc;IACd,aAAa;IACb,eAAe;IACf,eAAe;IACf,eAAe;IACf,YAAY;IACZ,cAAc;IACd,cAAc;IACd,eAAe;IACf,aAAa;IACb,WAAW;IACX,cAAc;IACd,eAAe;IACf,cAAc;IACd,gBAAgB;IAChB,qBAAqB;IACrB,eAAe;IACf,aAAa;IACb,oBAAoB;CACX,CAAA;AACV,MAAM,cAAc,GAAG;IACtB,MAAM;IACN,KAAK;IACL,OAAO;IACP,MAAM;IACN,YAAY;IACZ,SAAS;IACT,QAAQ;IACR,OAAO;IACP,MAAM;IACN,OAAO;IACP,KAAK;IACL,WAAW;IACX,SAAS;IACT,OAAO;IACP,SAAS;IACT,UAAU;IACV,UAAU;IACV,QAAQ;IACR,UAAU;IACV,WAAW;IACX,WAAW;IACX,YAAY;IACZ,UAAU;IACV,UAAU;IACV,WAAW;IACX,UAAU;IACV,SAAS;IACT,UAAU;IACV,OAAO;IACP,QAAQ;IACR,SAAS;IACT,WAAW;IACX,KAAK;IACL,KAAK;IACL,KAAK;IACL,KAAK;IACL,KAAK;IACL,MAAM;IACN,MAAM;CACG,CAAA;AACV,MAAM,QAAQ,GAAG;IAChB,IAAI;IACJ,QAAQ;IACR,KAAK;IACL,QAAQ;IACR,IAAI;IACJ,MAAM;IACN,MAAM;IACN,WAAW;IACX,IAAI;IACJ,MAAM;IACN,IAAI;IACJ,OAAO;IACP,MAAM;IACN,SAAS;IACT,KAAK;IACL,IAAI;IACJ,OAAO;IACP,IAAI;IACJ,OAAO;IACP,KAAK;IACL,QAAQ;IACR,KAAK;IACL,SAAS;IACT,KAAK;IACL,OAAO;IACP,MAAM;IACN,MAAM;IACN,UAAU;IACV,KAAK;IACL,MAAM;CACG,CAAA;AACV,MAAM,YAAY,GAAG,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,CAAU,CAAA;AAE1E,sGAAsG;AACtG,oEAAoE;AACpE,MAAM,WAAW,GAAG;IACnB,KAAK;IACL,QAAQ;IACR,WAAW;IACX,OAAO;IACP,SAAS;IACT,QAAQ;IACR,MAAM;IACN,OAAO;IACP,OAAO;IACP,SAAS;IACT,QAAQ;IACR,OAAO;IACP,SAAS;IACT,WAAW;CACF,CAAA;AACV,MAAM,QAAQ,GAAG;IAChB,qBAAqB;IACrB,qBAAqB;IACrB,mBAAmB;IACnB,oBAAoB;IACpB,kBAAkB;IAClB,kBAAkB;IAClB,aAAa;IACb,aAAa;IACb,gBAAgB;IAChB,YAAY;IACZ,UAAU;IACV,aAAa;IACb,8BAA8B;IAC9B,kBAAkB;IAClB,uBAAuB;IACvB,0BAA0B;IAC1B,eAAe;IACf,kBAAkB;IAClB,2BAA2B;IAC3B,cAAc;IACd,aAAa;IACb,mBAAmB;IACnB,WAAW;IACX,iBAAiB;CACR,CAAA;AAEV,uGAAuG;AACvG,wGAAwG;AACxG,0GAA0G;AAC1G,qBAAqB;AACrB,EAAE;AACF,0GAA0G;AAC1G,uGAAuG;AACvG,yGAAyG;AACzG,0GAA0G;AAC1G,2GAA2G;AAC3G,4GAA4G;AAC5G,4GAA4G;AAC5G,6DAA6D;AAC7D,MAAM,MAAM,GAAG;IACd,kBAAkB;IAClB,eAAe;IACf,gBAAgB;IAChB,gBAAgB;IAChB,eAAe;IACf,kBAAkB;IAClB,eAAe;IACf,iBAAiB;IACjB,kBAAkB;IAClB,cAAc;IACd,eAAe;IACf,mBAAmB;IACnB,mBAAmB;IACnB,eAAe;CACN,CAAA;AAEV,uGAAuG;AACvG,uGAAuG;AACvG,sGAAsG;AACtG,sGAAsG;AACtG,MAAM,SAAS,GAA2C;IACzD,EAAE,QAAQ,EAAE,aAAa,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC3E,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC5E,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACzE,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC5E,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACtE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACvE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACzE,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACvE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACzE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACxE,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC5E,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACxE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACxE,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACxE,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACtE,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACvE,EAAE,QAAQ,EAAE,aAAa,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;CAC3E,CAAA;AACD,qGAAqG;AACrG,qGAAqG;AACrG,wGAAwG;AACxG,oGAAoG;AACpG,mGAAmG;AACnG,MAAM,SAAS,GAA2C;IACzD,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACnE,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACvE,EAAE,QAAQ,EAAE,MAAM,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAClE,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACvE,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACpE,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACtE,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACnE,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACpE,EAAE,QAAQ,EAAE,UAAU,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACtE,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACxE,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACpE,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAC1E,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACpE,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACrE,EAAE,QAAQ,EAAE,MAAM,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IAClE,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;IACpE,EAAE,QAAQ,EAAE,aAAa,EAAE,MAAM,EAAE,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE;CACzE,CAAA;AACD,sGAAsG;AACtG,kGAAkG;AAClG,mFAAmF;AACnF,MAAM,WAAW,GAAG,CAAC,MAAoB,EAAU,EAAE,CAAC,MAAM,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,CAAC,CAAA;AAC7F,MAAM,SAAS,GAA2B,EAAE,EAAE,EAAE,OAAO,EAAE,EAAE,EAAE,OAAO,EAAE,EAAE,EAAE,OAAO,EAAE,CAAA;AAEnF,MAAM,aAAa,GAAsC;IACxD,mBAAmB;IACnB,uBAAuB;IACvB,WAAW;IACX,2BAA2B;IAC3B,eAAe;IACf,4BAA4B;CAC5B,CAAA;AAED;;;;;GAKG;AACH,MAAM,UAAU,2BAA2B,CAC1C,IAAyC,EACzC,OAAoC,EAAE;IAEtC,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,QAAQ,GAAG,IAAI,CAAC,aAAa,IAAI,IAAI,CAAC,aAAa,EAAE,MAAM,CAAC,CAAA;IAElE,IAAI,QAAQ,KAAK,eAAe,EAAE,CAAC;QAClC,iGAAiG;QACjG,mGAAmG;QACnG,uGAAuG;QACvG,gGAAgG;QAChG,MAAM,CAAC,GAAG,IAAI,IAAI,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC,CAAA;QACtF,MAAM,KAAK,GAAG,MAAM,EAAE,GAAG,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;QACzD,iGAAiG;QACjG,qGAAqG;QACrG,qGAAqG;QACrG,MAAM,WAAW,GAAG,MAAM,EAAE,GAAG,IAAI,CAAA;QACnC,IAAI,CAAC,CAAC,OAAO,KAAK,IAAI,EAAE,CAAC;YACxB,2EAA2E;YAC3E,MAAM,IAAI,GAAG,GAAG,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,QAAQ,GAAG,WAAW,CAAC,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,EAAE,EAAE,CAAA;YAC1E,OAAO;gBACN,GAAG,EAAE,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,KAAK,IAAI,EAAE,CAAC,CAAC,CAAC,IAAI;gBACvC,UAAU,EAAE;oBACX,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;oBAC3B,QAAQ,EAAE,CAAC,CAAC,QAAQ;oBACpB,QAAQ,EAAE,CAAC,CAAC,QAAQ;oBACpB,GAAG,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,OAAO,EAAE,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;iBAC7C;gBACD,MAAM,EAAE,OAAO;gBACf,QAAQ;aACR,CAAA;QACF,CAAC;QACD,MAAM,OAAO,GAAG,MAAM,EAAE,GAAG,GAAG,CAAA;QAC9B,MAAM,KAAK,GAAG,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAA,CAAC,+CAA+C;QACvF,iGAAiG;QACjG,uGAAuG;QACvG,MAAM,WAAW,GAAG,eAAe,CAAA;QACnC,MAAM,IAAI,GAAG,GAAG,CAAC,CAAC,QAAQ,GAAG,KAAK,IAAI,CAAC,CAAC,MAAM,GAAG,OAAO,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,GAAG,WAAW,CAAC,CAAC,CAAC,KAAK,WAAW,EAAE,CAAC,CAAC,CAAC,EAAE,EAAE,CAAA;QAC1H,OAAO;YACN,GAAG,EAAE,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,KAAK,IAAI,EAAE,CAAC,CAAC,CAAC,IAAI;YACvC,UAAU,EAAE;gBACX,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;gBAC3B,QAAQ,EAAE,CAAC,CAAC,QAAQ;gBACpB,MAAM,EAAE,CAAC,CAAC,MAAM;gBAChB,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;gBAC5C,GAAG,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,OAAO,EAAE,WAAW,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;aAChD;YACD,MAAM,EAAE,OAAO;YACf,QAAQ;SACR,CAAA;IACF,CAAC;IAED,IACC,QAAQ,KAAK,WAAW;QACxB,QAAQ,KAAK,2BAA2B;QACxC,QAAQ,KAAK,4BAA4B,EACxC,CAAC;QACF,gFAAgF;QAChF,MAAM,CAAC,GAAG,IAAI,IAAI,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;QACzC,MAAM,IAAI,GAAG,IAAI,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAA;QACnC,MAAM,EAAE,GAAG,WAAW,CAAC,MAAM,CAAC,CAAA;QAC9B,IAAI,QAAQ,KAAK,4BAA4B,EAAE,CAAC;YAC/C,gGAAgG;YAChG,iGAAiG;YACjG,iGAAiG;YACjG,sGAAsG;YACtG,2BAA2B;YAC3B,MAAM,GAAG,GAAG,GAAG,EAAE,IAAI,IAAI,KAAK,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,QAAQ,EAAE,CAAA;YACxD,OAAO;gBACN,GAAG;gBACH,UAAU,EAAE,EAAE,YAAY,EAAE,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC,CAAC,QAAQ,EAAE,QAAQ,EAAE,CAAC,CAAC,QAAQ,EAAE;gBAC1F,MAAM,EAAE,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,IAAI,OAAO;gBACvC,QAAQ;aACR,CAAA;QACF,CAAC;QACD,IAAI,QAAQ,KAAK,WAAW,EAAE,CAAC;YAC9B,MAAM,MAAM,GAAG,IAAI,CAAC,WAAW,EAAE,MAAM,CAAC,CAAA;YACxC,8FAA8F;YAC9F,MAAM,GAAG,GAAG,GAAG,EAAE,IAAI,MAAM,IAAI,IAAI,KAAK,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,QAAQ,EAAE,CAAA;YAClE,OAAO;gBACN,GAAG;gBACH,UAAU,EAAE;oBACX,YAAY,EAAE,EAAE;oBAChB,aAAa,EAAE,MAAM;oBACrB,MAAM,EAAE,IAAI;oBACZ,QAAQ,EAAE,CAAC,CAAC,QAAQ;oBACpB,QAAQ,EAAE,CAAC,CAAC,QAAQ;iBACpB;gBACD,MAAM,EAAE,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,IAAI,OAAO;gBACvC,QAAQ;aACR,CAAA;QACF,CAAC;QACD,+FAA+F;QAC/F,MAAM,GAAG,GAAG,GAAG,IAAI,IAAI,EAAE,KAAK,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,QAAQ,EAAE,CAAA;QACxD,OAAO;YACN,GAAG;YACH,UAAU,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,YAAY,EAAE,EAAE,EAAE,QAAQ,EAAE,CAAC,CAAC,QAAQ,EAAE,QAAQ,EAAE,CAAC,CAAC,QAAQ,EAAE;YAC1F,MAAM,EAAE,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,IAAI,OAAO;YACvC,QAAQ;SACR,CAAA;IACF,CAAC;IAED,kGAAkG;IAClG,uFAAuF;IACvF,MAAM,CAAC,GAAG,IAAI,IAAI,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;IACzC,MAAM,EAAE,GAAG,WAAW,CAAC,MAAM,CAAC,CAAA;IAC9B,MAAM,GAAG,GAAG,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,YAAY,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;IAC5D,MAAM,IAAI,GAAG,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,iBAAiB,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,cAAc,EAAE,MAAM,CAAC,CAAA;IAC5F,MAAM,MAAM,GAAG,IAAI,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAA;IACrC,MAAM,UAAU,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,EAAE,GAAG,IAAI,IAAI,MAAM,EAAE,CAAA;IAC7D,MAAM,UAAU,GAA+B;QAC9C,YAAY,EAAE,EAAE;QAChB,GAAG,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,aAAa,EAAE,GAAG,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;QACtC,MAAM,EAAE,IAAI;QACZ,aAAa,EAAE,MAAM;QACrB,QAAQ,EAAE,CAAC,CAAC,QAAQ;QACpB,MAAM,EAAE,CAAC,CAAC,MAAM;QAChB,QAAQ,EAAE,CAAC,CAAC,QAAQ;KACpB,CAAA;IACD,MAAM,GAAG,GACR,QAAQ,KAAK,uBAAuB;QACnC,CAAC,CAAC,oDAAoD;YACrD,GAAG,EAAE,IAAI,UAAU,IAAI,CAAC,CAAC,QAAQ,IAAI,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,QAAQ,EAAE;QAC9D,CAAC,CAAC,qEAAqE;YACtE,GAAG,EAAE,IAAI,UAAU,KAAK,CAAC,CAAC,QAAQ,KAAK,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,QAAQ,EAAE,CAAA;IAClE,OAAO,EAAE,GAAG,EAAE,UAAU,EAAE,MAAM,EAAE,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,IAAI,OAAO,EAAE,QAAQ,EAAE,CAAA;AAC9E,CAAC"}

package/out/src/synthesize-po-box.d.ts

@@ -58,7 +58,7 @@ export interface SynthesizedPoBoxRow {
     raw: string;
     components: CanonicalRow["components"];
     locale: string;
-    template: "po-box" | "pmb-with-street";
+    template: "po-box" | "pmb-with-street" | "military-po-box";
 }
 export interface PoBoxSynthesisOpts {
     /** Random function — pass deterministic seed for tests. Default Math.random. */
@@ -77,6 +77,8 @@ export declare function synthesizePoBoxRow(base: PoBoxBaseTuple & {
     street?: string;
     houseNumber?: string;
 }, opts?: PoBoxSynthesisOpts): SynthesizedPoBoxRow | null;
+/** Generate one US military/diplomatic PO-box row (#517). Self-contained — draws no base tuple. */
+export declare function synthesizeMilitaryPoBoxRow(opts?: PoBoxSynthesisOpts): SynthesizedPoBoxRow;
 /**
  * Map a country code (ISO-3166-1 alpha-2 or alpha-3, or country display name) to the locale code we
  * have a PO box template for.

package/out/src/synthesize-po-box.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"synthesize-po-box.d.ts","sourceRoot":"","sources":["../../src/synthesize-po-box.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAEH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,YAAY,CAAA;AAE9C,MAAM,WAAW,cAAc;IAC9B,QAAQ,EAAE,MAAM,CAAA;IAChB,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,OAAO,EAAE,MAAM,CAAA;CACf;AAED,MAAM,WAAW,cAAc;IAC9B,MAAM,EAAE,MAAM,CAAA;IACd,OAAO,EAAE,aAAa,CAAC,MAAM,CAAC,CAAA;IAE9B,GAAG,CAAC,EAAE,aAAa,CAAC,MAAM,CAAC,CAAA;CAC3B;AAED;;;;;GAKG;AACH,eAAO,MAAM,uBAAuB,EAAE,aAAa,CAAC,cAAc,CAuCjE,CAAA;AAID;;;GAGG;AACH,wBAAgB,qBAAqB,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,MAAM,GAAG,MAAM,CAY/E;AAED;;;;;GAKG;AACH,wBAAgB,kBAAkB,CAAC,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,MAAM,CAEzE;AAED,MAAM,WAAW,mBAAmB;IACnC,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,CAAA;IACtC,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,QAAQ,GAAG,iBAAiB,CAAA;CACtC;AAED,MAAM,WAAW,kBAAkB;IAClC,gFAAgF;IAChF,MAAM,CAAC,EAAE,MAAM,MAAM,CAAA;IACrB,uDAAuD;IACvD,UAAU,CAAC,EAAE,CAAC,MAAM,EAAE,MAAM,MAAM,KAAK,MAAM,CAAA;IAC7C,4FAA4F;IAC5F,QAAQ,CAAC,EAAE,MAAM,CAAA;CACjB;AAWD;;;;GAIG;AACH,wBAAgB,kBAAkB,CACjC,IAAI,EAAE,cAAc,GAAG;IAAE,MAAM,CAAC,EAAE,MAAM,CAAC;IAAC,WAAW,CAAC,EAAE,MAAM,CAAA;CAAE,EAChE,IAAI,GAAE,kBAAuB,GAC3B,mBAAmB,GAAG,IAAI,CAkD5B;AAED;;;GAGG;AACH,wBAAgB,eAAe,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,CAWvD;AAED,qFAAqF;AACrF,wBAAgB,gBAAgB,IAAI,aAAa,CAAC,MAAM,CAAC,CAExD"}
+{"version":3,"file":"synthesize-po-box.d.ts","sourceRoot":"","sources":["../../src/synthesize-po-box.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAEH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,YAAY,CAAA;AAE9C,MAAM,WAAW,cAAc;IAC9B,QAAQ,EAAE,MAAM,CAAA;IAChB,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,OAAO,EAAE,MAAM,CAAA;CACf;AAED,MAAM,WAAW,cAAc;IAC9B,MAAM,EAAE,MAAM,CAAA;IACd,OAAO,EAAE,aAAa,CAAC,MAAM,CAAC,CAAA;IAE9B,GAAG,CAAC,EAAE,aAAa,CAAC,MAAM,CAAC,CAAA;CAC3B;AAED;;;;;GAKG;AACH,eAAO,MAAM,uBAAuB,EAAE,aAAa,CAAC,cAAc,CA2CjE,CAAA;AAID;;;GAGG;AACH,wBAAgB,qBAAqB,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,MAAM,GAAG,MAAM,CAY/E;AAED;;;;;GAKG;AACH,wBAAgB,kBAAkB,CAAC,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,MAAM,CAEzE;AAED,MAAM,WAAW,mBAAmB;IACnC,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,YAAY,CAAC,YAAY,CAAC,CAAA;IACtC,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,QAAQ,GAAG,iBAAiB,GAAG,iBAAiB,CAAA;CAC1D;AAED,MAAM,WAAW,kBAAkB;IAClC,gFAAgF;IAChF,MAAM,CAAC,EAAE,MAAM,MAAM,CAAA;IACrB,uDAAuD;IACvD,UAAU,CAAC,EAAE,CAAC,MAAM,EAAE,MAAM,MAAM,KAAK,MAAM,CAAA;IAC7C,4FAA4F;IAC5F,QAAQ,CAAC,EAAE,MAAM,CAAA;CACjB;AAWD;;;;GAIG;AACH,wBAAgB,kBAAkB,CACjC,IAAI,EAAE,cAAc,GAAG;IAAE,MAAM,CAAC,EAAE,MAAM,CAAC;IAAC,WAAW,CAAC,EAAE,MAAM,CAAA;CAAE,EAChE,IAAI,GAAE,kBAAuB,GAC3B,mBAAmB,GAAG,IAAI,CAqD5B;AAuBD,mGAAmG;AACnG,wBAAgB,0BAA0B,CAAC,IAAI,GAAE,kBAAuB,GAAG,mBAAmB,CAsB7F;AAED;;;GAGG;AACH,wBAAgB,eAAe,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,CAYvD;AAED,qFAAqF;AACrF,wBAAgB,gBAAgB,IAAI,aAAa,CAAC,MAAM,CAAC,CAExD"}

package/out/src/synthesize-po-box.js

@@ -48,6 +48,10 @@ export const PO_BOX_LOCALE_TEMPLATES = [
         locale: "en-AU",
         leaders: ["PO Box", "P.O. Box", "Post Office Box", "GPO Box", "Locked Bag"],
     },
+    {
+        locale: "en-NZ",
+        leaders: ["PO Box", "P.O. Box", "Post Office Box", "Private Bag", "Private Box"],
+    },
     {
         locale: "fr-FR",
         leaders: ["BP", "B.P.", "Boîte Postale", "BP."],
@@ -146,14 +150,17 @@ export function synthesizePoBoxRow(base, opts = {}) {
             template: "pmb-with-street",
         };
     }
-    // Standard PO box: replaces the street line entirely.
-    const raw = `${poBoxPhrase}, ${base.locality}, ${base.region} ${base.postcode}`;
+    // Standard PO box: replaces the street line entirely. Region-optional — NZ (and other region-less
+    // locales) read "Private Bag 12, Auckland 1010" with no region token between locality and postcode.
+    const hasRegion = Boolean(base.region && base.region.trim());
+    const tail = hasRegion ? `${base.locality}, ${base.region} ${base.postcode}` : `${base.locality} ${base.postcode}`;
+    const raw = `${poBoxPhrase}, ${tail}`;
     return {
         raw,
         components: {
             po_box: poBoxPhrase,
             locality: base.locality,
-            region: base.region,
+            ...(hasRegion ? { region: base.region } : {}),
             postcode: base.postcode,
             country: base.country,
         },
@@ -161,6 +168,50 @@ export function synthesizePoBoxRow(base, opts = {}) {
         template: "po-box",
     };
 }
+/**
+ * The US military/diplomatic PO-box class (#517). A distinct shape the leader-based locale
+ * templates can't express: a unit line (`PSC <id> Box <box>`, `CMR <id> Box <box>`, `Unit <id> [Box
+ * <box>]`) tagged `po_box`, then the post-office code (APO/FPO/DPO) as the locality and the
+ * armed-forces region (AA/AE/AP) as the region, with a theatre-specific ZIP. Authoritative
+ * reference + citations: `@mailwoman/codex` `codex/us/military-address.ts`; the small constants are
+ * inlined here so the generator is self-contained.
+ */
+const MIL_UNITS = [
+    { code: "PSC", boxRequired: true },
+    { code: "CMR", boxRequired: true },
+    { code: "Unit", boxRequired: false },
+];
+const MIL_PO_CODES = ["APO", "FPO", "DPO"];
+// region → plausible ZIP prefix (AE Europe 09xxx, AP Pacific 962-966xx, AA Americas 340xx).
+const MIL_REGION_ZIP = [
+    { region: "AE", zip: (r) => `09${String(Math.floor(r() * 1000)).padStart(3, "0")}` },
+    { region: "AP", zip: (r) => `96${String(200 + Math.floor(r() * 100)).padStart(3, "0")}` },
+    { region: "AA", zip: (r) => `340${String(Math.floor(r() * 100)).padStart(2, "0")}` },
+];
+/** Generate one US military/diplomatic PO-box row (#517). Self-contained — draws no base tuple. */
+export function synthesizeMilitaryPoBoxRow(opts = {}) {
+    const random = opts.random ?? Math.random;
+    const unit = MIL_UNITS[Math.floor(random() * MIL_UNITS.length)];
+    const unitId = String(1 + Math.floor(random() * 9999));
+    const { region, zip } = MIL_REGION_ZIP[Math.floor(random() * MIL_REGION_ZIP.length)];
+    const zipStr = zip(random);
+    const po = MIL_PO_CODES[Math.floor(random() * MIL_PO_CODES.length)];
+    const hasBox = unit.boxRequired || random() < 0.5;
+    const unitLine = hasBox ? `${unit.code} ${unitId} Box ${1 + Math.floor(random() * 9999)}` : `${unit.code} ${unitId}`;
+    const raw = `${unitLine}, ${po} ${region} ${zipStr}`;
+    return {
+        raw,
+        components: {
+            po_box: unitLine,
+            locality: po,
+            region,
+            postcode: zipStr,
+            country: "US",
+        },
+        locale: "en-US",
+        template: "military-po-box",
+    };
+}
 /**
  * Map a country code (ISO-3166-1 alpha-2 or alpha-3, or country display name) to the locale code we
  * have a PO box template for.
@@ -175,6 +226,8 @@ export function countryToLocale(country) {
         return "en-GB";
     if (c === "AU" || c === "AUS" || c === "AUSTRALIA")
         return "en-AU";
+    if (c === "NZ" || c === "NZL" || c === "NEW ZEALAND")
+        return "en-NZ";
     if (c === "FR" || c === "FRA" || c === "FRANCE")
         return "fr-FR";
     if (c === "ES" || c === "ESP" || c === "SPAIN")

package/out/src/synthesize-po-box.js.map

@@ -1 +1 @@
-{"version":3,"file":"synthesize-po-box.js","sourceRoot":"","sources":["../../src/synthesize-po-box.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAkBH;;;;;GAKG;AACH,MAAM,CAAC,MAAM,uBAAuB,GAAkC;IACrE;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,SAAS,EAAE,QAAQ,EAAE,KAAK,EAAE,iBAAiB,EAAE,KAAK,CAAC;QACrF,GAAG,EAAE,CAAC,KAAK,EAAE,GAAG,CAAC;KACjB;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,KAAK,EAAE,iBAAiB,CAAC;QACzD,GAAG,EAAE,CAAC,KAAK,EAAE,GAAG,CAAC;KACjB;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,iBAAiB,CAAC;KAClD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,iBAAiB,EAAE,SAAS,EAAE,YAAY,CAAC;KAC3E;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,IAAI,EAAE,MAAM,EAAE,eAAe,EAAE,KAAK,CAAC;KAC/C;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,IAAI,EAAE,MAAM,EAAE,cAAc,EAAE,IAAI,EAAE,MAAM,CAAC;KACrD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,UAAU,EAAE,qBAAqB,CAAC;KAC7D;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,IAAI,CAAC;KACvD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,SAAS,EAAE,mBAAmB,EAAE,IAAI,CAAC;KAC/C;CACD,CAAA;AAED,MAAM,iBAAiB,GAAG,IAAI,GAAG,CAAyB,uBAAuB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,CAAC,CAAA;AAE5G;;;GAGG;AACH,MAAM,UAAU,qBAAqB,CAAC,GAAW,EAAE,MAAoB;IACtE,IAAI,MAAM,EAAE,GAAG,GAAG;QAAE,OAAO,GAAG,CAAA;IAC9B,MAAM,QAAQ,GAAiC;QAC9C,qDAAqD;QACrD,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;QAC/D,wCAAwC;QACxC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;QAC/D,mDAAmD;QACnD,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,CAAC,EAAE,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC;KAC5B,CAAA;IACD,MAAM,CAAC,GAAG,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,QAAQ,CAAC,MAAM,CAAC,CAAE,CAAA;IAC3D,OAAO,CAAC,CAAC,GAAG,CAAC,CAAA;AACd,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,kBAAkB,CAAC,MAAc,EAAE,MAAc;IAChE,OAAO,GAAG,MAAM,IAAI,MAAM,EAAE,CAAA;AAC7B,CAAC;AAkBD,SAAS,iBAAiB,CAAC,MAAoB;IAC9C,sFAAsF;IACtF,MAAM,CAAC,GAAG,MAAM,EAAE,CAAA;IAClB,IAAI,CAAC,GAAG,GAAG;QAAE,OAAO,MAAM,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,EAAE,CAAC,CAAC,CAAA,CAAC,OAAO;IACjE,IAAI,CAAC,GAAG,GAAG;QAAE,OAAO,MAAM,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAA,CAAC,UAAU;IACvE,IAAI,CAAC,GAAG,IAAI;QAAE,OAAO,MAAM,CAAC,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,CAAC,CAAA,CAAC,YAAY;IAC5E,OAAO,MAAM,CAAC,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,KAAK,CAAC,CAAC,CAAA,CAAC,cAAc;AACnE,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,kBAAkB,CACjC,IAAgE,EAChE,OAA2B,EAAE;IAE7B,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,UAAU,GAAG,IAAI,CAAC,UAAU,IAAI,iBAAiB,CAAA;IACvD,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,IAAI,GAAG,CAAA;IAErC,MAAM,MAAM,GAAG,eAAe,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;IAC5C,MAAM,GAAG,GAAG,iBAAiB,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;IACzC,IAAI,CAAC,GAAG;QAAE,OAAO,IAAI,CAAA;IAErB,MAAM,MAAM,GAAG,qBAAqB,CAAC,UAAU,CAAC,MAAM,CAAC,EAAE,MAAM,CAAC,CAAA;IAChE,MAAM,MAAM,GAAG,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,OAAO,CAAC,MAAM,CAAC,CAAE,CAAA;IACtE,MAAM,WAAW,GAAG,kBAAkB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IAEtD,mEAAmE;IACnE,MAAM,OAAO,GAAG,IAAI,CAAC,MAAM,IAAI,GAAG,CAAC,GAAG,IAAI,MAAM,EAAE,GAAG,QAAQ,CAAA;IAC7D,IAAI,OAAO,EAAE,CAAC;QACb,MAAM,SAAS,GAAG,GAAG,CAAC,GAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,GAAI,CAAC,MAAM,CAAC,CAAE,CAAA;QACnE,MAAM,SAAS,GAAG,kBAAkB,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;QACvD,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,WAAW,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAO,CAAA;QACzF,MAAM,GAAG,GAAG,GAAG,UAAU,KAAK,SAAS,KAAK,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAA;QAC5F,OAAO;YACN,GAAG;YACH,UAAU,EAAE;gBACX,GAAG,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,IAAI,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;gBAC/D,MAAM,EAAE,IAAI,CAAC,MAAO;gBACpB,MAAM,EAAE,SAAS;gBACjB,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,MAAM,EAAE,IAAI,CAAC,MAAM;gBACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,OAAO,EAAE,IAAI,CAAC,OAAO;aACrB;YACD,MAAM;YACN,QAAQ,EAAE,iBAAiB;SAC3B,CAAA;IACF,CAAC;IAED,sDAAsD;IACtD,MAAM,GAAG,GAAG,GAAG,WAAW,KAAK,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAA;IAC/E,OAAO;QACN,GAAG;QACH,UAAU,EAAE;YACX,MAAM,EAAE,WAAW;YACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,MAAM,EAAE,IAAI,CAAC,MAAM;YACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,OAAO,EAAE,IAAI,CAAC,OAAO;SACrB;QACD,MAAM;QACN,QAAQ,EAAE,QAAQ;KAClB,CAAA;AACF,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,eAAe,CAAC,OAAe;IAC9C,MAAM,CAAC,GAAG,OAAO,CAAC,IAAI,EAAE,CAAC,WAAW,EAAE,CAAA;IACtC,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,eAAe;QAAE,OAAO,OAAO,CAAA;IACtE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,gBAAgB;QAAE,OAAO,OAAO,CAAA;IACrF,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,WAAW;QAAE,OAAO,OAAO,CAAA;IAClE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,OAAO;QAAE,OAAO,OAAO,CAAA;IAC9D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,WAAW;QAAE,OAAO,OAAO,CAAA;IAClE,OAAO,OAAO,CAAA;AACf,CAAC;AAED,qFAAqF;AACrF,MAAM,UAAU,gBAAgB;IAC/B,OAAO,uBAAuB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,CAAC,CAAA;AACpD,CAAC"}
+{"version":3,"file":"synthesize-po-box.js","sourceRoot":"","sources":["../../src/synthesize-po-box.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAkBH;;;;;GAKG;AACH,MAAM,CAAC,MAAM,uBAAuB,GAAkC;IACrE;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,SAAS,EAAE,QAAQ,EAAE,KAAK,EAAE,iBAAiB,EAAE,KAAK,CAAC;QACrF,GAAG,EAAE,CAAC,KAAK,EAAE,GAAG,CAAC;KACjB;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,KAAK,EAAE,iBAAiB,CAAC;QACzD,GAAG,EAAE,CAAC,KAAK,EAAE,GAAG,CAAC;KACjB;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,iBAAiB,CAAC;KAClD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,iBAAiB,EAAE,SAAS,EAAE,YAAY,CAAC;KAC3E;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,QAAQ,EAAE,UAAU,EAAE,iBAAiB,EAAE,aAAa,EAAE,aAAa,CAAC;KAChF;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,IAAI,EAAE,MAAM,EAAE,eAAe,EAAE,KAAK,CAAC;KAC/C;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,IAAI,EAAE,MAAM,EAAE,cAAc,EAAE,IAAI,EAAE,MAAM,CAAC;KACrD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,OAAO,EAAE,MAAM,EAAE,UAAU,EAAE,qBAAqB,CAAC;KAC7D;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,OAAO,EAAE,UAAU,EAAE,iBAAiB,EAAE,IAAI,CAAC;KACvD;IACD;QACC,MAAM,EAAE,OAAO;QACf,OAAO,EAAE,CAAC,SAAS,EAAE,mBAAmB,EAAE,IAAI,CAAC;KAC/C;CACD,CAAA;AAED,MAAM,iBAAiB,GAAG,IAAI,GAAG,CAAyB,uBAAuB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,CAAC,CAAA;AAE5G;;;GAGG;AACH,MAAM,UAAU,qBAAqB,CAAC,GAAW,EAAE,MAAoB;IACtE,IAAI,MAAM,EAAE,GAAG,GAAG;QAAE,OAAO,GAAG,CAAA;IAC9B,MAAM,QAAQ,GAAiC;QAC9C,qDAAqD;QACrD,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;QAC/D,wCAAwC;QACxC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;QAC/D,mDAAmD;QACnD,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,CAAC,EAAE,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC;KAC5B,CAAA;IACD,MAAM,CAAC,GAAG,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,QAAQ,CAAC,MAAM,CAAC,CAAE,CAAA;IAC3D,OAAO,CAAC,CAAC,GAAG,CAAC,CAAA;AACd,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,kBAAkB,CAAC,MAAc,EAAE,MAAc;IAChE,OAAO,GAAG,MAAM,IAAI,MAAM,EAAE,CAAA;AAC7B,CAAC;AAkBD,SAAS,iBAAiB,CAAC,MAAoB;IAC9C,sFAAsF;IACtF,MAAM,CAAC,GAAG,MAAM,EAAE,CAAA;IAClB,IAAI,CAAC,GAAG,GAAG;QAAE,OAAO,MAAM,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,EAAE,CAAC,CAAC,CAAA,CAAC,OAAO;IACjE,IAAI,CAAC,GAAG,GAAG;QAAE,OAAO,MAAM,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC,CAAA,CAAC,UAAU;IACvE,IAAI,CAAC,GAAG,IAAI;QAAE,OAAO,MAAM,CAAC,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,CAAC,CAAA,CAAC,YAAY;IAC5E,OAAO,MAAM,CAAC,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,KAAK,CAAC,CAAC,CAAA,CAAC,cAAc;AACnE,CAAC;AAED;;;;GAIG;AACH,MAAM,UAAU,kBAAkB,CACjC,IAAgE,EAChE,OAA2B,EAAE;IAE7B,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,UAAU,GAAG,IAAI,CAAC,UAAU,IAAI,iBAAiB,CAAA;IACvD,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,IAAI,GAAG,CAAA;IAErC,MAAM,MAAM,GAAG,eAAe,CAAC,IAAI,CAAC,OAAO,CAAC,CAAA;IAC5C,MAAM,GAAG,GAAG,iBAAiB,CAAC,GAAG,CAAC,MAAM,CAAC,CAAA;IACzC,IAAI,CAAC,GAAG;QAAE,OAAO,IAAI,CAAA;IAErB,MAAM,MAAM,GAAG,qBAAqB,CAAC,UAAU,CAAC,MAAM,CAAC,EAAE,MAAM,CAAC,CAAA;IAChE,MAAM,MAAM,GAAG,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,OAAO,CAAC,MAAM,CAAC,CAAE,CAAA;IACtE,MAAM,WAAW,GAAG,kBAAkB,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IAEtD,mEAAmE;IACnE,MAAM,OAAO,GAAG,IAAI,CAAC,MAAM,IAAI,GAAG,CAAC,GAAG,IAAI,MAAM,EAAE,GAAG,QAAQ,CAAA;IAC7D,IAAI,OAAO,EAAE,CAAC;QACb,MAAM,SAAS,GAAG,GAAG,CAAC,GAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,GAAI,CAAC,MAAM,CAAC,CAAE,CAAA;QACnE,MAAM,SAAS,GAAG,kBAAkB,CAAC,SAAS,EAAE,MAAM,CAAC,CAAA;QACvD,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,WAAW,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAO,CAAA;QACzF,MAAM,GAAG,GAAG,GAAG,UAAU,KAAK,SAAS,KAAK,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAA;QAC5F,OAAO;YACN,GAAG;YACH,UAAU,EAAE;gBACX,GAAG,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,IAAI,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;gBAC/D,MAAM,EAAE,IAAI,CAAC,MAAO;gBACpB,MAAM,EAAE,SAAS;gBACjB,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,MAAM,EAAE,IAAI,CAAC,MAAM;gBACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,OAAO,EAAE,IAAI,CAAC,OAAO;aACrB;YACD,MAAM;YACN,QAAQ,EAAE,iBAAiB;SAC3B,CAAA;IACF,CAAC;IAED,kGAAkG;IAClG,oGAAoG;IACpG,MAAM,SAAS,GAAG,OAAO,CAAC,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC,CAAA;IAC5D,MAAM,IAAI,GAAG,SAAS,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,KAAK,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,QAAQ,IAAI,IAAI,CAAC,QAAQ,EAAE,CAAA;IAClH,MAAM,GAAG,GAAG,GAAG,WAAW,KAAK,IAAI,EAAE,CAAA;IACrC,OAAO;QACN,GAAG;QACH,UAAU,EAAE;YACX,MAAM,EAAE,WAAW;YACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,GAAG,CAAC,SAAS,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YAC7C,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,OAAO,EAAE,IAAI,CAAC,OAAO;SACrB;QACD,MAAM;QACN,QAAQ,EAAE,QAAQ;KAClB,CAAA;AACF,CAAC;AAED;;;;;;;GAOG;AACH,MAAM,SAAS,GAA0D;IACxE,EAAE,IAAI,EAAE,KAAK,EAAE,WAAW,EAAE,IAAI,EAAE;IAClC,EAAE,IAAI,EAAE,KAAK,EAAE,WAAW,EAAE,IAAI,EAAE;IAClC,EAAE,IAAI,EAAE,MAAM,EAAE,WAAW,EAAE,KAAK,EAAE;CACpC,CAAA;AACD,MAAM,YAAY,GAAG,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAU,CAAA;AACnD,4FAA4F;AAC5F,MAAM,cAAc,GAAwE;IAC3F,EAAE,MAAM,EAAE,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,KAAK,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,IAAI,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,EAAE;IACpF,EAAE,MAAM,EAAE,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,KAAK,MAAM,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,EAAE;IACzF,EAAE,MAAM,EAAE,IAAI,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,EAAE,CAAC,MAAM,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,EAAE,GAAG,CAAC,EAAE,EAAE;CACpF,CAAA;AAED,mGAAmG;AACnG,MAAM,UAAU,0BAA0B,CAAC,OAA2B,EAAE;IACvE,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,MAAM,CAAA;IACzC,MAAM,IAAI,GAAG,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,SAAS,CAAC,MAAM,CAAC,CAAE,CAAA;IAChE,MAAM,MAAM,GAAG,MAAM,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,CAAC,CAAA;IACtD,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,cAAc,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,cAAc,CAAC,MAAM,CAAC,CAAE,CAAA;IACrF,MAAM,MAAM,GAAG,GAAG,CAAC,MAAM,CAAC,CAAA;IAC1B,MAAM,EAAE,GAAG,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,YAAY,CAAC,MAAM,CAAC,CAAE,CAAA;IACpE,MAAM,MAAM,GAAG,IAAI,CAAC,WAAW,IAAI,MAAM,EAAE,GAAG,GAAG,CAAA;IACjD,MAAM,QAAQ,GAAG,MAAM,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,IAAI,IAAI,MAAM,QAAQ,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,IAAI,IAAI,MAAM,EAAE,CAAA;IACpH,MAAM,GAAG,GAAG,GAAG,QAAQ,KAAK,EAAE,IAAI,MAAM,IAAI,MAAM,EAAE,CAAA;IACpD,OAAO;QACN,GAAG;QACH,UAAU,EAAE;YACX,MAAM,EAAE,QAAQ;YAChB,QAAQ,EAAE,EAAE;YACZ,MAAM;YACN,QAAQ,EAAE,MAAM;YAChB,OAAO,EAAE,IAAI;SACb;QACD,MAAM,EAAE,OAAO;QACf,QAAQ,EAAE,iBAAiB;KAC3B,CAAA;AACF,CAAC;AAED;;;GAGG;AACH,MAAM,UAAU,eAAe,CAAC,OAAe;IAC9C,MAAM,CAAC,GAAG,OAAO,CAAC,IAAI,EAAE,CAAC,WAAW,EAAE,CAAA;IACtC,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,eAAe;QAAE,OAAO,OAAO,CAAA;IACtE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,gBAAgB;QAAE,OAAO,OAAO,CAAA;IACrF,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,WAAW;QAAE,OAAO,OAAO,CAAA;IAClE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,aAAa;QAAE,OAAO,OAAO,CAAA;IACpE,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,OAAO;QAAE,OAAO,OAAO,CAAA;IAC9D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,QAAQ;QAAE,OAAO,OAAO,CAAA;IAC/D,IAAI,CAAC,KAAK,IAAI,IAAI,CAAC,KAAK,KAAK,IAAI,CAAC,KAAK,WAAW;QAAE,OAAO,OAAO,CAAA;IAClE,OAAO,OAAO,CAAA;AACf,CAAC;AAED,qFAAqF;AACrF,MAAM,UAAU,gBAAgB;IAC/B,OAAO,uBAAuB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,CAAC,CAAA;AACpD,CAAC"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mailwoman/corpus",
-  "version": "4.10.0",
+  "version": "4.11.0",
   "description": "Mailwoman corpus pipeline: BIO-labeled dataset builder for the neural classifier.",
   "license": "AGPL-3.0-only",
   "repository": {
@@ -17,9 +17,9 @@
   },
   "dependencies": {
     "@dsnp/parquetjs": "1.8.7",
-    "@mailwoman/codex": "4.10.0",
-    "@mailwoman/core": "4.10.0",
-    "@mailwoman/formatter": "4.10.0",
+    "@mailwoman/codex": "4.11.0",
+    "@mailwoman/core": "4.11.0",
+    "@mailwoman/formatter": "4.11.0",
     "csv-parse": "^5.6.0",
     "fast-glob": "^3.3.3",
     "fastest-levenshtein": "^1.0.16",