@mailwoman/corpus 9.2.0 → 9.3.0

package/out/recipes/scaffold.d.ts

@@ -0,0 +1,324 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Shared scaffolding for the synthetic-corpus SLICE RECIPES — the common bits the 16
+ *   `build-*-slice.mjs` scripts each re-implemented: the seeded LCG PRNG, the tuple reader, and the
+ *   canonical → `alignRow` → `LabeledRow` JSONL emit step. A recipe ({@link CorpusRecipe}) supplies
+ *   only its synthesis + filter; the `mailwoman corpus slice <recipe>` command supplies the I/O.
+ */
+import type { PathBuilderLike } from "path-ts";
+import type { AsyncChunkIterator, AsyncDataResource } from "spliterator";
+import { AsyncSequence } from "spliterator";
+/**
+ * {@link stableSourceIDFromParts} under the name the recipes use: arbitrary disambiguator keys (e.g. a variant index
+ * `v`) that aren't `ComponentTag`s, which is how the legacy builders kept per-variant ids unique.
+ */
+export declare function sliceSourceID(adapterID: string, parts: Record<string, string | undefined>): string;
+/**
+ * Where a country's convention writes the postcode inside the `«locality», «region»[, «country»]` admin tail.
+ *
+ * Position changes the tag assigned to the same digits. On the shipped model, `Heladería Frappé Manía, Avenida Country
+ * Club, Barcelona 6001, Anzoátegui, Venezuela` tags `6001` as `house_number` and loses the locality into the street,
+ * while the identical row written `… 6001 Barcelona, Anzoátegui, Venezuela` tags it `postcode` and recovers `locality:
+ * Barcelona`. So a slice emitting one placement teaches one family of countries.
+ *
+ * Each is attested by a gauntlet board row, which is the bar for adding another:
+ *
+ * - `leading` — `«postcode» «locality», «region»`. `Rua da Praia, 15, 8600-315 Lagos, Algarve, Portugal`
+ *   (`pt_structured`). The default, and what every tuple written before this field existed means.
+ * - `after_locality` — `«locality» «postcode», «region»`. `…, Barcelona 6001, Anzoátegui, Venezuela`
+ *   (`ve_city_postcode_trailing_state`).
+ * - `after_region` — `«locality», «region» «postcode»`. `12 MG Road, Indiranagar, Bengaluru, Karnataka 560038, India`
+ *   (`in_structured`).
+ */
+export type PostcodePlacement = "leading" | "after_locality" | "after_region";
+/**
+ * A (locality, region, postcode, country) source tuple — the input to tuples-mode recipes.
+ */
+export interface SliceTuple {
+    locality?: string;
+    /**
+     * The segment before the locality, when the source has one. A slice whose every row BEGINS with the locality teaches
+     * that the first named segment is the locality, which flips the model's default — measured on the v4.8.0 candidate,
+     * where `Ye Three Lords, 27 Minories, London EC3N 1DE` came back `locality: "Ye Three Lords"`.
+     */
+    dependentLocality?: string;
+    region?: string;
+    postcode?: string;
+    country?: string;
+    /**
+     * Defaults to `leading` when absent, which is what every tuple file written before this field existed means — so an
+     * old tuples file produces byte-identical rows.
+     */
+    postcodePlacement?: PostcodePlacement;
+    [k: string]: unknown;
+}
+/**
+ * The two seeded generators the legacy `build-*-slice` scripts used, re-exported from their home in
+ * `@mailwoman/core/utils` so a recipe keeps importing everything it needs from this one scaffold module.
+ *
+ * - `makeLcg` (`s = s*1664525 + 1013904223 mod 2^32`) — what the street/po-box/anchor builders seeded.
+ * - `makeMulberry32` — what the MAJORITY of them used (german, locale, boundary-stress, unit, fr-order, country-balanced,
+ *   intersection, fr-admin-split, street-affix, street-bare, po-box-cedex).
+ *
+ * A recipe must seed the same one its `.mjs` did, the same way it did (usually `seed`, but some derive a per-stream
+ * seed), or `--seed N` stops being byte-reproducible.
+ */
+export { makeLcg, mulberry32 as makeMulberry32, makeLcg as makeRandom } from "@mailwoman/core/utils";
+/**
+ * One CSV record, keyed by its lower-cased header name, every value trimmed. A column the header does not declare reads
+ * as `undefined`; a column the header declares but this record does not reach reads as `""`.
+ */
+export type CSVRecord = Record<string, string | undefined>;
+/**
+ * Read a CSV as header-keyed records.
+ *
+ * Returns the spliterator's own {@linkcode AsyncSequence}, so a caller composes `take`/`drop`/`filter` onto it — those
+ * ops fuse into one pull loop, and a `take` that is satisfied closes the source's file handle on the way out. Wrapping
+ * this in an `async function*` costs an async frame per row and takes those ops away; don't.
+ *
+ * A source at or below the spliterator's 128 KiB bulk threshold is read whole and parsed by the synchronous engine, so
+ * this is also the right reader for small sources — there is no buffered variant to reach for.
+ *
+ * @category CSV
+ */
+export declare function readCSVRecords(source: AsyncDataResource | AsyncChunkIterator): AsyncSequence<CSVRecord>;
+/**
+ * {@link readCSVRecords} over one member of a zip archive — what every recipe reading a cached OA source wants.
+ *
+ * A source a checkout has not cached yields nothing, after saying so. A lab holds the archives for the countries it has
+ * built, so a recipe naming ten sources routinely finds three, and the `unzip -p` subprocesses these replaced behaved
+ * the same way by accident — a non-zero exit warned and returned no rows. A recipe that ends up with NO tuples at all
+ * still throws; that is the case where the cache, not the recipe, is the problem.
+ *
+ * @category CSV
+ */
+export declare function readZippedCSVRecords(archivePath: PathBuilderLike, entryName: string): AsyncSequence<CSVRecord>;
+/**
+ * License stamped on the synthetic tuple-derived slices (`po-box`, `no-street`, `house-venue`): the output is
+ * generated, but it inherits the terms of the real tuples it is derived from, so the attribution travels with it.
+ */
+export declare const SYNTHETIC_TUPLE_LICENSE = "Synthetic \u2014 derived from CC-BY / public-domain input tuples";
+/**
+ * The surface key shared by the Norwegian slices (`no-fragment`, `no-street-led`).
+ *
+ * MUST match the NO digit board's `norm_surface`: NFC, lowercase, collapse whitespace — and KEEP diacritics.
+ * fr-fragment's norm strips them (NFD + combining-mark removal), which is right for French but would collapse
+ * `Tømmerlien` → `tommerlien` here, so a slice's exclusion check would never match the board's reserved `tømmerlien`
+ * and the train/eval split would leak silently. Diacritic street heads (…vegen/…veien with ø/å/æ) are the whole point
+ * of those slices' boundary; folding them away is not an option.
+ */
+export declare const foldNOSurface: (value: string) => string;
+/**
+ * A cached OpenAddresses extract: the zip and the CSV member.
+ */
+export interface OATupleSource {
+    zip: PathBuilderLike;
+    csv: string;
+}
+/**
+ * The four base fields every OA tuple reader extracts. `postcode` is `""` when the row carries none and
+ * {@link ReadOATuplesOptions.requirePostcode} is unset.
+ */
+export interface OATupleFields {
+    house_number: string;
+    street: string;
+    locality: string;
+    postcode: string;
+}
+export interface ReadOATuplesOptions<T> {
+    /**
+     * Stop after this many distinct tuples — the `break` closes the reader and releases the archive, which is what the
+     * GB-scale countrywide extracts need.
+     */
+    limit?: number;
+    /**
+     * Drop rows without a postcode (reversed-order and balance readers, where the postcode drives the rendering).
+     */
+    requirePostcode?: boolean;
+    /**
+     * Fold the postcode into the dedup key (`fr-order`'s key). The default key is
+     * `${house_number}|${street}|${locality}`, lower-cased.
+     */
+    dedupIncludesPostcode?: boolean;
+    /**
+     * Shape the recipe's tuple from the base fields, the raw record, and the (lower-cased) dedup key.
+     */
+    extra: (fields: OATupleFields, row: CSVRecord, key: string) => T;
+}
+/**
+ * Stream distinct tuples out of a cached OA zip — the reader the OA-skeleton recipes (`street-affix`, `unit`,
+ * `country-balanced`, `fr-order`) share. Field reads, filters, dedup keys and row order are exactly what each recipe's
+ * local copy did, so a recipe's slice stays byte-identical.
+ *
+ * @category CSV
+ */
+export declare function readOATuples<T>(source: OATupleSource, options: ReadOATuplesOptions<T>): Promise<T[]>;
+/**
+ * Stream-parse a tuples JSONL file, yielding each parsed object (blank/invalid lines skipped).
+ */
+export declare function readTuples(input: PathBuilderLike): AsyncSequence<SliceTuple>;
+/**
+ * A canonical row as the recipes assemble it, before `alignRow` turns it into a `LabeledRow`.
+ */
+export interface CanonicalSliceRow {
+    raw: string;
+    components: Record<string, string>;
+    country: string;
+    locale?: string;
+    source: string;
+    source_id: string;
+    corpus_version?: string;
+    license?: string;
+}
+/**
+ * Run a canonical row through `alignRow` and, on success, write the `LabeledRow` (+ `synth_method` / `synth_base_id`)
+ * as one JSONL line. Returns true if emitted, false if alignment quarantined it.
+ */
+export declare function alignAndWrite(write: (line: string) => void, canonical: CanonicalSliceRow, synthMethod: string, synthBaseID?: string | null): boolean;
+/**
+ * Parsed options a recipe's `run` receives. Common fields + the union of recipe-specific flags.
+ */
+export interface SliceRecipeOpts {
+    output: string;
+    seed: number;
+    variants: number;
+    input?: string;
+    count?: number;
+    golden?: boolean;
+    sourceName?: string;
+    houseNumberProb?: number;
+    pmbRatio?: number;
+    militaryRatio?: number;
+    reversedFraction?: number;
+    edgesDir?: string;
+    country?: string;
+    intlFraction?: number;
+    /**
+     * `german`: fraction of NATIVE-order rows rendered with no commas at all (`Neusser Str. 12 Nippes 50733 Köln`), the
+     * single-line register #1946 found the model loses the district on. Default 0.3.
+     */
+    commaFreeFraction?: number;
+    /**
+     * `german`: fraction of rows that carry a WOF Ortsteil of the tuple's locality as `dependent_locality`. Default 0.3;
+     * 0 when no admin database is readable.
+     */
+    ortsteilFraction?: number;
+    /**
+     * `german`: the WOF admin database the Ortsteil pool is read from. Default
+     * `$MAILWOMAN_DATA_ROOT/wof/admin-global-priority-importance.db`.
+     */
+    adminDB?: string;
+    /**
+     * `locale`: fraction of rows that append an explicit country surface form + a `country` component. Default 0.
+     */
+    countryFraction?: number;
+    /**
+     * `locale`: tri-state override of the per-part `districtAsLocality` mapping for this invocation. `undefined` (flag
+     * absent) leaves each `COUNTRY_SOURCES` part's own value untouched — every existing locale build stays
+     * byte-identical. `true`/`false` forces that value on every part read this run. ES's pedanía slice
+     * (`synth-es-pedania`) additionally uses `true` to select {@link LocaleCountrySource.pedaniaParts} instead of the
+     * default `parts` — see `locale.ts`.
+     */
+    districtAsLocality?: boolean;
+    bareProb?: number;
+    hnProb?: number;
+    communes?: string;
+    /**
+     * `fr-lieudit`: BAN `adresses-<dept>.csv` directory. Default `$MAILWOMAN_DATA_ROOT/corpus/sources/ban`.
+     */
+    banDir?: string;
+    multilocaleCount?: number;
+    /**
+     * `fr-fragment` / `no-fragment` / `no-street-led`: the eval board's reserved street-surface list. REQUIRED for those
+     * recipes — a slice that trains on its own eval set measures memorization. See their docstrings.
+     */
+    excludeSurfaces?: string;
+    /**
+     * `no-fragment`: share of rows that are counter-distribution (bare locality OR bare postcode).
+     */
+    counterProb?: number;
+    /**
+     * `no-fragment` knob 3: emit N copies of each street+number row whose number has >= longNumberMinDigits digits
+     * (oversample the failing long-number class). Default 1 = no boost.
+     */
+    longNumberBoost?: number;
+    /**
+     * `no-fragment` knob 3: minimum digit count for a number to count as "long" and be boosted. Default 3.
+     */
+    longNumberMinDigits?: number;
+    /**
+     * `sub-venue`: the sub-venue lexicon JSON. Default = the committed `corpus/data/sub-venue-lexicon.json`, resolved
+     * through the package manifest so it works from the source tree and from `out/`.
+     */
+    lexicon?: string;
+    /**
+     * `sub-venue`: directory of `sub-venue-extract` JSONLs, one per region. Default
+     * `$MAILWOMAN_DATA_ROOT/sub-venue/extracts`.
+     */
+    extractsDir?: string;
+    /**
+     * `sub-venue`: the `poi.db` spatial layer, read for the en-US and fr-FR venue + confound pools (the two of poi.db's
+     * four countries this slice has legs for). Default `$MAILWOMAN_DATA_ROOT/poi/poi.db`.
+     */
+    poiDB?: string;
+    /**
+     * `sub-venue`: GB/US/FR address-context tuples JSONL. Default the house-venue v3 tuples
+     * (`$MAILWOMAN_DATA_ROOT/corpus/intermediate/house-venue-tuples-v3.jsonl`); DE and ES read OpenAddresses directly.
+     */
+    subVenueTuples?: string;
+    /**
+     * `sub-venue`: share of emitted rows that are confound NEGATIVES. Default 0.3.
+     */
+    negativeFraction?: number;
+}
+/**
+ * Tally a recipe returns.
+ */
+export interface SliceStats {
+    read?: number;
+    emitted: number;
+    skipped: number;
+    /**
+     * Rows dropped because their street SURFACE is reserved by an eval board (`--exclude-surfaces`). Separate from
+     * `skipped` on purpose: a nonzero value is the audit trail that the train/eval split actually fired. Zero when a
+     * recipe has no board split.
+     */
+    contaminated?: number;
+}
+/**
+ * A single declared recipe-specific option flag (for the command's --help).
+ */
+export interface SliceRecipeOption {
+    flag: string;
+    description: string;
+}
+/**
+ * A slice recipe: its identity, input mode, and its synthesis `run`.
+ */
+export interface CorpusRecipe {
+    /**
+     * Recipe id, e.g. "street", "po-box" — the `<recipe>` positional.
+     */
+    name: string;
+    /**
+     * One-line description for `--list` / help.
+     */
+    description: string;
+    /**
+     * `tuples` reads `--input` JSONL; `generate` self-generates `--count` rows.
+     */
+    mode: "tuples" | "generate";
+    /**
+     * Recipe-specific flags this recipe honors (documentation only).
+     */
+    options?: SliceRecipeOption[];
+    /**
+     * Do the build: create the recipe's PRNG from `opts.seed` (its LEGACY generator — `makeLcg` or `makeMulberry32` — for
+     * byte-reproducibility), synthesize, and emit each row via `write`.
+     */
+    run(opts: SliceRecipeOpts, write: (line: string) => void): Promise<SliceStats>;
+}
+//# sourceMappingURL=scaffold.d.ts.map

package/out/recipes/scaffold.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"scaffold.d.ts","sourceRoot":"","sources":["../../lib/recipes/scaffold.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAKH,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,SAAS,CAAA;AAC9C,OAAO,KAAK,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,MAAM,aAAa,CAAA;AACxE,OAAO,EAAE,aAAa,EAAmC,MAAM,aAAa,CAAA;AAK5E;;;GAGG;AACH,wBAAgB,aAAa,CAAC,SAAS,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC,GAAG,MAAM,CAElG;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,MAAM,MAAM,iBAAiB,GAAG,SAAS,GAAG,gBAAgB,GAAG,cAAc,CAAA;AAE7E;;GAEG;AACH,MAAM,WAAW,UAAU;IAC1B,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB;;;;OAIG;IACH,iBAAiB,CAAC,EAAE,MAAM,CAAA;IAC1B,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB;;;OAGG;IACH,iBAAiB,CAAC,EAAE,iBAAiB,CAAA;IACrC,CAAC,CAAC,EAAE,MAAM,GAAG,OAAO,CAAA;CACpB;AAED;;;;;;;;;;GAUG;AACH,OAAO,EAAE,OAAO,EAAE,UAAU,IAAI,cAAc,EAAE,OAAO,IAAI,UAAU,EAAE,MAAM,uBAAuB,CAAA;AAEpG;;;GAGG;AACH,MAAM,MAAM,SAAS,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC,CAAA;AA4C1D;;;;;;;;;;;GAWG;AACH,wBAAgB,cAAc,CAAC,MAAM,EAAE,iBAAiB,GAAG,kBAAkB,GAAG,aAAa,CAAC,SAAS,CAAC,CAYvG;AAED;;;;;;;;;GASG;AACH,wBAAgB,oBAAoB,CAAC,WAAW,EAAE,eAAe,EAAE,SAAS,EAAE,MAAM,GAAG,aAAa,CAAC,SAAS,CAAC,CAU9G;AAED;;;GAGG;AACH,eAAO,MAAM,uBAAuB,qEAAgE,CAAA;AAEpG;;;;;;;;GAQG;AACH,eAAO,MAAM,aAAa,GAAI,OAAO,MAAM,KAAG,MACsB,CAAA;AAEpE;;GAEG;AACH,MAAM,WAAW,aAAa;IAC7B,GAAG,EAAE,eAAe,CAAA;IACpB,GAAG,EAAE,MAAM,CAAA;CACX;AAED;;;GAGG;AACH,MAAM,WAAW,aAAa;IAC7B,YAAY,EAAE,MAAM,CAAA;IACpB,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,QAAQ,EAAE,MAAM,CAAA;CAChB;AAED,MAAM,WAAW,mBAAmB,CAAC,CAAC;IACrC;;;OAGG;IACH,KAAK,CAAC,EAAE,MAAM,CAAA;IACd;;OAEG;IACH,eAAe,CAAC,EAAE,OAAO,CAAA;IACzB;;;OAGG;IACH,qBAAqB,CAAC,EAAE,OAAO,CAAA;IAC/B;;OAEG;IACH,KAAK,EAAE,CAAC,MAAM,EAAE,aAAa,EAAE,GAAG,EAAE,SAAS,EAAE,GAAG,EAAE,MAAM,KAAK,CAAC,CAAA;CAChE;AAED;;;;;;GAMG;AACH,wBAAsB,YAAY,CAAC,CAAC,EAAE,MAAM,EAAE,aAAa,EAAE,OAAO,EAAE,mBAAmB,CAAC,CAAC,CAAC,GAAG,OAAO,CAAC,CAAC,EAAE,CAAC,CA6B1G;AAED;;GAEG;AACH,wBAAgB,UAAU,CAAC,KAAK,EAAE,eAAe,GAAG,aAAa,CAAC,UAAU,CAAC,CAU5E;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC,GAAG,EAAE,MAAM,CAAA;IACX,UAAU,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAA;IAClC,OAAO,EAAE,MAAM,CAAA;IACf,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,MAAM,EAAE,MAAM,CAAA;IACd,SAAS,EAAE,MAAM,CAAA;IACjB,cAAc,CAAC,EAAE,MAAM,CAAA;IACvB,OAAO,CAAC,EAAE,MAAM,CAAA;CAChB;AAED;;;GAGG;AACH,wBAAgB,aAAa,CAC5B,KAAK,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,EAC7B,SAAS,EAAE,iBAAiB,EAC5B,WAAW,EAAE,MAAM,EACnB,WAAW,GAAE,MAAM,GAAG,IAAW,GAC/B,OAAO,CAOT;AAED;;GAEG;AACH,MAAM,WAAW,eAAe;IAC/B,MAAM,EAAE,MAAM,CAAA;IACd,IAAI,EAAE,MAAM,CAAA;IACZ,QAAQ,EAAE,MAAM,CAAA;IAChB,KAAK,CAAC,EAAE,MAAM,CAAA;IACd,KAAK,CAAC,EAAE,MAAM,CAAA;IACd,MAAM,CAAC,EAAE,OAAO,CAAA;IAChB,UAAU,CAAC,EAAE,MAAM,CAAA;IAEnB,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,aAAa,CAAC,EAAE,MAAM,CAAA;IACtB,gBAAgB,CAAC,EAAE,MAAM,CAAA;IACzB,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB,YAAY,CAAC,EAAE,MAAM,CAAA;IACrB;;;OAGG;IACH,iBAAiB,CAAC,EAAE,MAAM,CAAA;IAC1B;;;OAGG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAA;IACzB;;;OAGG;IACH,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB;;OAEG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB;;;;;;OAMG;IACH,kBAAkB,CAAC,EAAE,OAAO,CAAA;IAC5B,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB;;OAEG;IACH,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,gBAAgB,CAAC,EAAE,MAAM,CAAA;IACzB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB;;OAEG;IACH,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB;;OAEG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAA;IAC5B;;;OAGG;IACH,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB;;;OAGG;IACH,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB;;;OAGG;IACH,KAAK,CAAC,EAAE,MAAM,CAAA;IACd;;;OAGG;IACH,cAAc,CAAC,EAAE,MAAM,CAAA;IACvB;;OAEG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAA;CACzB;AAED;;GAEG;AACH,MAAM,WAAW,UAAU;IAC1B,IAAI,CAAC,EAAE,MAAM,CAAA;IACb,OAAO,EAAE,MAAM,CAAA;IACf,OAAO,EAAE,MAAM,CAAA;IACf;;;;OAIG;IACH,YAAY,CAAC,EAAE,MAAM,CAAA;CACrB;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC,IAAI,EAAE,MAAM,CAAA;IACZ,WAAW,EAAE,MAAM,CAAA;CACnB;AAED;;GAEG;AACH,MAAM,WAAW,YAAY;IAC5B;;OAEG;IACH,IAAI,EAAE,MAAM,CAAA;IACZ;;OAEG;IACH,WAAW,EAAE,MAAM,CAAA;IACnB;;OAEG;IACH,IAAI,EAAE,QAAQ,GAAG,UAAU,CAAA;IAC3B;;OAEG;IACH,OAAO,CAAC,EAAE,iBAAiB,EAAE,CAAA;IAC7B;;;OAGG;IACH,GAAG,CAAC,IAAI,EAAE,eAAe,EAAE,KAAK,EAAE,CAAC,IAAI,EAAE,MAAM,KAAK,IAAI,GAAG,OAAO,CAAC,UAAU,CAAC,CAAA;CAC9E"}

package/out/recipes/scaffold.js

@@ -0,0 +1,186 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   Shared scaffolding for the synthetic-corpus SLICE RECIPES — the common bits the 16
+ *   `build-*-slice.mjs` scripts each re-implemented: the seeded LCG PRNG, the tuple reader, and the
+ *   canonical → `alignRow` → `LabeledRow` JSONL emit step. A recipe ({@link CorpusRecipe}) supplies
+ *   only its synthesis + filter; the `mailwoman corpus slice <recipe>` command supplies the I/O.
+ */
+import { pathExists } from "@mailwoman/core/fs/readers";
+import { readZipEntry } from "@mailwoman/core/fs/zip";
+import { tryParsingJSON } from "@mailwoman/core/json";
+import { AsyncSequence, CSVSpliterator, TextSpliterator } from "spliterator";
+import { stableSourceIDFromParts } from "#adapters/utils";
+import { alignRow } from "#utils";
+/**
+ * {@link stableSourceIDFromParts} under the name the recipes use: arbitrary disambiguator keys (e.g. a variant index
+ * `v`) that aren't `ComponentTag`s, which is how the legacy builders kept per-variant ids unique.
+ */
+export function sliceSourceID(adapterID, parts) {
+    return stableSourceIDFromParts(adapterID, parts);
+}
+/**
+ * The two seeded generators the legacy `build-*-slice` scripts used, re-exported from their home in
+ * `@mailwoman/core/utils` so a recipe keeps importing everything it needs from this one scaffold module.
+ *
+ * - `makeLcg` (`s = s*1664525 + 1013904223 mod 2^32`) — what the street/po-box/anchor builders seeded.
+ * - `makeMulberry32` — what the MAJORITY of them used (german, locale, boundary-stress, unit, fr-order, country-balanced,
+ *   intersection, fr-admin-split, street-affix, street-bare, po-box-cedex).
+ *
+ * A recipe must seed the same one its `.mjs` did, the same way it did (usually `seed`, but some derive a per-stream
+ * seed), or `--seed N` stops being byte-reproducible.
+ */
+export { makeLcg, mulberry32 as makeMulberry32, makeLcg as makeRandom } from "@mailwoman/core/utils";
+/**
+ * Quote handling spans the ROW split, not only the column split: a newline inside a quoted field belongs to a single
+ * record, so the record boundaries can only be found by a scanner that already knows where the quotes are. Find the
+ * lines first and unquote afterwards and such a record splits in two — the first half short by however many columns
+ * followed the newline, which is how a `street` value comes to be read as `city`.
+ *
+ * `header: false` keeps the first record in the stream, so this module owns the lower-casing rather than
+ * `normalizeKeys` — see {@link toHeader}.
+ */
+const CSV_INIT = { header: false, enableQuoteHandling: true };
+/**
+ * Header names are lower-cased on the way in. `normalizeKeys` will not do it: it leaves an ALL CAPS header alone, and
+ * OpenAddresses ships `LON,LAT,NUMBER,STREET` while other extracts ship the same names lower-case. A recipe names its
+ * columns in lower case either way.
+ */
+function toHeader(cells) {
+    return cells.map((name) => name.trim().toLowerCase());
+}
+/**
+ * Line breaks inside a value become single spaces. A quote-aware parse is the first thing here able to return a value
+ * CONTAINING one — `us/ia/statewide.csv` has 12, all unit designators like `"#2\n#2"` — and every consumer synthesizes
+ * one-line address text from these cells with no guard, because until that parse landed no value could carry one.
+ * Collapsing keeps the record (the address is fine; the source's line break is not part of it) without emitting a
+ * training row with a newline inside it.
+ *
+ * Only `\r` and `\n`, deliberately — NOT `\s`. Runs of spaces and tabs pass through exactly as the source wrote them
+ * (OA's IA extract writes `NORTH`, three spaces, `MAIN STREET`), because those could always appear and every slice
+ * built to date contains them. Widening this to `\s+` silently rewrites values on rows with no line break at all.
+ * `scaffold.test.ts` pins both halves, against both readers.
+ */
+function toRecord(header, cells) {
+    const record = {};
+    for (let i = 0; i < header.length; i++) {
+        record[header[i]] = (cells[i] ?? "").replaceAll(/[\r\n]+/g, " ").trim();
+    }
+    return record;
+}
+/**
+ * Read a CSV as header-keyed records.
+ *
+ * Returns the spliterator's own {@linkcode AsyncSequence}, so a caller composes `take`/`drop`/`filter` onto it — those
+ * ops fuse into one pull loop, and a `take` that is satisfied closes the source's file handle on the way out. Wrapping
+ * this in an `async function*` costs an async frame per row and takes those ops away; don't.
+ *
+ * A source at or below the spliterator's 128 KiB bulk threshold is read whole and parsed by the synchronous engine, so
+ * this is also the right reader for small sources — there is no buffered variant to reach for.
+ *
+ * @category CSV
+ */
+export function readCSVRecords(source) {
+    let header = null;
+    return CSVSpliterator.fromAsync(source, CSV_INIT)
+        .filter((cells) => {
+        if (header)
+            return true;
+        header = toHeader(cells);
+        return false;
+    })
+        .map((cells) => toRecord(header, cells));
+}
+/**
+ * {@link readCSVRecords} over one member of a zip archive — what every recipe reading a cached OA source wants.
+ *
+ * A source a checkout has not cached yields nothing, after saying so. A lab holds the archives for the countries it has
+ * built, so a recipe naming ten sources routinely finds three, and the `unzip -p` subprocesses these replaced behaved
+ * the same way by accident — a non-zero exit warned and returned no rows. A recipe that ends up with NO tuples at all
+ * still throws; that is the case where the cache, not the recipe, is the problem.
+ *
+ * @category CSV
+ */
+export function readZippedCSVRecords(archivePath, entryName) {
+    return AsyncSequence.from(async () => {
+        if (!(await pathExists(String(archivePath)))) {
+            console.error(`  WARN: ${archivePath} is not cached — skipping ${entryName}`);
+            return [];
+        }
+        return readCSVRecords(readZipEntry(archivePath, entryName));
+    });
+}
+/**
+ * License stamped on the synthetic tuple-derived slices (`po-box`, `no-street`, `house-venue`): the output is
+ * generated, but it inherits the terms of the real tuples it is derived from, so the attribution travels with it.
+ */
+export const SYNTHETIC_TUPLE_LICENSE = "Synthetic — derived from CC-BY / public-domain input tuples";
+/**
+ * The surface key shared by the Norwegian slices (`no-fragment`, `no-street-led`).
+ *
+ * MUST match the NO digit board's `norm_surface`: NFC, lowercase, collapse whitespace — and KEEP diacritics.
+ * fr-fragment's norm strips them (NFD + combining-mark removal), which is right for French but would collapse
+ * `Tømmerlien` → `tommerlien` here, so a slice's exclusion check would never match the board's reserved `tømmerlien`
+ * and the train/eval split would leak silently. Diacritic street heads (…vegen/…veien with ø/å/æ) are the whole point
+ * of those slices' boundary; folding them away is not an option.
+ */
+export const foldNOSurface = (value) => value.normalize("NFC").toLowerCase().replaceAll(/\s+/g, " ").trim();
+/**
+ * Stream distinct tuples out of a cached OA zip — the reader the OA-skeleton recipes (`street-affix`, `unit`,
+ * `country-balanced`, `fr-order`) share. Field reads, filters, dedup keys and row order are exactly what each recipe's
+ * local copy did, so a recipe's slice stays byte-identical.
+ *
+ * @category CSV
+ */
+export async function readOATuples(source, options) {
+    const tuples = [];
+    const seen = new Set();
+    for await (const row of readZippedCSVRecords(source.zip, source.csv)) {
+        if (options.limit !== undefined && tuples.length >= options.limit)
+            break;
+        const street = row.street ?? "";
+        const locality = row.city ?? "";
+        const house_number = row.number ?? "";
+        const postcode = row.postcode ?? "";
+        if (!street || !locality || !house_number)
+            continue;
+        if (options.requirePostcode && !postcode)
+            continue;
+        const key = (options.dedupIncludesPostcode
+            ? `${house_number}|${street}|${locality}|${postcode}`
+            : `${house_number}|${street}|${locality}`).toLowerCase();
+        if (seen.has(key))
+            continue;
+        seen.add(key);
+        tuples.push(options.extra({ house_number, street, locality, postcode }, row, key));
+    }
+    return tuples;
+}
+/**
+ * Stream-parse a tuples JSONL file, yielding each parsed object (blank/invalid lines skipped).
+ */
+export function readTuples(input) {
+    // TextSpliterator (not JSONSpliterator) keeps the reader's established tolerance: malformed lines are skipped rather
+    // than rejecting the sequence. These operators fuse into the source's pull loop instead of adding an async-generator
+    // frame per tuple.
+    return TextSpliterator.fromAsync(input)
+        .map((line) => line.trim())
+        .filter((line) => Boolean(line))
+        .map((line) => tryParsingJSON(line))
+        .filter((tuple) => tuple !== null)
+        .map((tuple) => tuple);
+}
+/**
+ * Run a canonical row through `alignRow` and, on success, write the `LabeledRow` (+ `synth_method` / `synth_base_id`)
+ * as one JSONL line. Returns true if emitted, false if alignment quarantined it.
+ */
+export function alignAndWrite(write, canonical, synthMethod, synthBaseID = null) {
+    const aligned = alignRow(canonical);
+    if (!aligned.row)
+        return false;
+    write(JSON.stringify({ ...aligned.row, synth_method: synthMethod, synth_base_id: synthBaseID }) + "\n");
+    return true;
+}
+//# sourceMappingURL=scaffold.js.map

package/out/recipes/scaffold.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"scaffold.js","sourceRoot":"","sources":["../../lib/recipes/scaffold.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,EAAE,UAAU,EAAE,MAAM,4BAA4B,CAAA;AACvD,OAAO,EAAE,YAAY,EAAE,MAAM,wBAAwB,CAAA;AACrD,OAAO,EAAE,cAAc,EAAE,MAAM,sBAAsB,CAAA;AAGrD,OAAO,EAAE,aAAa,EAAE,cAAc,EAAE,eAAe,EAAE,MAAM,aAAa,CAAA;AAE5E,OAAO,EAAE,uBAAuB,EAAE,MAAM,iBAAiB,CAAA;AACzD,OAAO,EAAE,QAAQ,EAAE,MAAM,QAAQ,CAAA;AAEjC;;;GAGG;AACH,MAAM,UAAU,aAAa,CAAC,SAAiB,EAAE,KAAyC;IACzF,OAAO,uBAAuB,CAAC,SAAS,EAAE,KAAK,CAAC,CAAA;AACjD,CAAC;AA2CD;;;;;;;;;;GAUG;AACH,OAAO,EAAE,OAAO,EAAE,UAAU,IAAI,cAAc,EAAE,OAAO,IAAI,UAAU,EAAE,MAAM,uBAAuB,CAAA;AAQpG;;;;;;;;GAQG;AACH,MAAM,QAAQ,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,mBAAmB,EAAE,IAAI,EAAW,CAAA;AAEtE;;;;GAIG;AACH,SAAS,QAAQ,CAAC,KAAwB;IACzC,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC,CAAA;AACtD,CAAC;AAED;;;;;;;;;;;GAWG;AACH,SAAS,QAAQ,CAAC,MAAyB,EAAE,KAAwB;IACpE,MAAM,MAAM,GAAc,EAAE,CAAA;IAE5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACxC,MAAM,CAAC,MAAM,CAAC,CAAC,CAAE,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,UAAU,CAAC,UAAU,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;IACzE,CAAC;IAED,OAAO,MAAM,CAAA;AACd,CAAC;AAED;;;;;;;;;;;GAWG;AACH,MAAM,UAAU,cAAc,CAAC,MAA8C;IAC5E,IAAI,MAAM,GAAoB,IAAI,CAAA;IAElC,OAAO,cAAc,CAAC,SAAS,CAAW,MAAM,EAAE,QAAQ,CAAC;SACzD,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE;QACjB,IAAI,MAAM;YAAE,OAAO,IAAI,CAAA;QAEvB,MAAM,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAA;QAExB,OAAO,KAAK,CAAA;IACb,CAAC,CAAC;SACD,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,QAAQ,CAAC,MAAO,EAAE,KAAK,CAAC,CAAC,CAAA;AAC3C,CAAC;AAED;;;;;;;;;GASG;AACH,MAAM,UAAU,oBAAoB,CAAC,WAA4B,EAAE,SAAiB;IACnF,OAAO,aAAa,CAAC,IAAI,CAAY,KAAK,IAAI,EAAE;QAC/C,IAAI,CAAC,CAAC,MAAM,UAAU,CAAC,MAAM,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,CAAC;YAC9C,OAAO,CAAC,KAAK,CAAC,WAAW,WAAW,6BAA6B,SAAS,EAAE,CAAC,CAAA;YAE7E,OAAO,EAAE,CAAA;QACV,CAAC;QAED,OAAO,cAAc,CAAC,YAAY,CAAC,WAAW,EAAE,SAAS,CAAC,CAAC,CAAA;IAC5D,CAAC,CAAC,CAAA;AACH,CAAC;AAED;;;GAGG;AACH,MAAM,CAAC,MAAM,uBAAuB,GAAG,6DAA6D,CAAA;AAEpG;;;;;;;;GAQG;AACH,MAAM,CAAC,MAAM,aAAa,GAAG,CAAC,KAAa,EAAU,EAAE,CACtD,KAAK,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,WAAW,EAAE,CAAC,UAAU,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AA0CpE;;;;;;GAMG;AACH,MAAM,CAAC,KAAK,UAAU,YAAY,CAAI,MAAqB,EAAE,OAA+B;IAC3F,MAAM,MAAM,GAAQ,EAAE,CAAA;IACtB,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAA;IAE9B,IAAI,KAAK,EAAE,MAAM,GAAG,IAAI,oBAAoB,CAAC,MAAM,CAAC,GAAG,EAAE,MAAM,CAAC,GAAG,CAAC,EAAE,CAAC;QACtE,IAAI,OAAO,CAAC,KAAK,KAAK,SAAS,IAAI,MAAM,CAAC,MAAM,IAAI,OAAO,CAAC,KAAK;YAAE,MAAK;QAExE,MAAM,MAAM,GAAG,GAAG,CAAC,MAAM,IAAI,EAAE,CAAA;QAC/B,MAAM,QAAQ,GAAG,GAAG,CAAC,IAAI,IAAI,EAAE,CAAA;QAC/B,MAAM,YAAY,GAAG,GAAG,CAAC,MAAM,IAAI,EAAE,CAAA;QACrC,MAAM,QAAQ,GAAG,GAAG,CAAC,QAAQ,IAAI,EAAE,CAAA;QAEnC,IAAI,CAAC,MAAM,IAAI,CAAC,QAAQ,IAAI,CAAC,YAAY;YAAE,SAAQ;QAEnD,IAAI,OAAO,CAAC,eAAe,IAAI,CAAC,QAAQ;YAAE,SAAQ;QAElD,MAAM,GAAG,GAAG,CACX,OAAO,CAAC,qBAAqB;YAC5B,CAAC,CAAC,GAAG,YAAY,IAAI,MAAM,IAAI,QAAQ,IAAI,QAAQ,EAAE;YACrD,CAAC,CAAC,GAAG,YAAY,IAAI,MAAM,IAAI,QAAQ,EAAE,CAC1C,CAAC,WAAW,EAAE,CAAA;QAEf,IAAI,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC;YAAE,SAAQ;QAC3B,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,CAAA;QAEb,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,EAAE,YAAY,EAAE,MAAM,EAAE,QAAQ,EAAE,QAAQ,EAAE,EAAE,GAAG,EAAE,GAAG,CAAC,CAAC,CAAA;IACnF,CAAC;IAED,OAAO,MAAM,CAAA;AACd,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,UAAU,CAAC,KAAsB;IAChD,qHAAqH;IACrH,qHAAqH;IACrH,mBAAmB;IACnB,OAAO,eAAe,CAAC,SAAS,CAAC,KAAK,CAAC;SACrC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;SAC1B,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;SAC/B,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,cAAc,CAAa,IAAI,CAAC,CAAC;SAC/C,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,KAAK,IAAI,CAAC;SACjC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAM,CAAC,CAAA;AACzB,CAAC;AAgBD;;;GAGG;AACH,MAAM,UAAU,aAAa,CAC5B,KAA6B,EAC7B,SAA4B,EAC5B,WAAmB,EACnB,cAA6B,IAAI;IAEjC,MAAM,OAAO,GAAG,QAAQ,CAAC,SAA2C,CAAC,CAAA;IAErE,IAAI,CAAC,OAAO,CAAC,GAAG;QAAE,OAAO,KAAK,CAAA;IAC9B,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,EAAE,GAAG,OAAO,CAAC,GAAG,EAAE,YAAY,EAAE,WAAW,EAAE,aAAa,EAAE,WAAW,EAAE,CAAC,GAAG,IAAI,CAAC,CAAA;IAEvG,OAAO,IAAI,CAAA;AACZ,CAAC"}

package/out/recipes/si-bare-village.d.ts

@@ -0,0 +1,32 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `si-bare-village` — the Slovenian no-street counter-slice (#901 run-2). Slovenia's rural
+ *   addressing has NO street line: the village name is the street-level token and repeats as the
+ *   locality ("Zabiče 8, 6250 Zabiče"). The v1.9.8 check FALSIFIED the fr-bare-street slice alone on
+ *   exactly this class (SI resolve −3.4pp; "Apače 108" split into street "Apače 10" + house "8") —
+ *   the bare-street boundary lesson generalizes onto a form where the leading name must keep its
+ *   number whole and the trailing mention must stay locality-bound. This slice is the paired
+ *   counter-distribution: same lesson ("name before number, comma, then admin"), opposite polarity
+ *   on the trailing mention.
+ *
+ *   Three real-order templates cycle per tuple (mirrors the coord-golden orders so the eval and the
+ *   training distribution agree):
+ *
+ *   1. canonical  "«V» «n», «pc» «V»"
+ *   2. bare       "«V» «n», «V»"        (no postcode — the anchor-free form, the fr-bare lesson)
+ *   3. pc-first   "«pc» «V», «V» «n»"
+ *
+ *   Gold spans: leading «V» = street (matches OA ground truth — the village IS the address line),
+ *   «n» = house_number (NEVER split mid-digits), «pc» = postcode (never swallowing the neighbor),
+ *   trailing «V» = locality (the binding the resolver needs).
+ */
+import { type CorpusRecipe } from "#recipes/scaffold";
+/**
+ * Slice recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
+ * and `description` below for the surface form it generates.
+ */
+export declare const siBareVillageRecipe: CorpusRecipe;
+//# sourceMappingURL=si-bare-village.d.ts.map

package/out/recipes/si-bare-village.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"si-bare-village.d.ts","sourceRoot":"","sources":["../../lib/recipes/si-bare-village.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAEH,OAAO,EAA6C,KAAK,YAAY,EAAiB,MAAM,mBAAmB,CAAA;AAE/G;;;GAGG;AACH,eAAO,MAAM,mBAAmB,EAAE,YAgEjC,CAAA"}

package/out/recipes/si-bare-village.js

@@ -0,0 +1,88 @@
+/**
+ * @copyright Sister Software
+ * @license AGPL-3.0
+ * @author Teffen Ellis, et al.
+ *
+ *   `si-bare-village` — the Slovenian no-street counter-slice (#901 run-2). Slovenia's rural
+ *   addressing has NO street line: the village name is the street-level token and repeats as the
+ *   locality ("Zabiče 8, 6250 Zabiče"). The v1.9.8 check FALSIFIED the fr-bare-street slice alone on
+ *   exactly this class (SI resolve −3.4pp; "Apače 108" split into street "Apače 10" + house "8") —
+ *   the bare-street boundary lesson generalizes onto a form where the leading name must keep its
+ *   number whole and the trailing mention must stay locality-bound. This slice is the paired
+ *   counter-distribution: same lesson ("name before number, comma, then admin"), opposite polarity
+ *   on the trailing mention.
+ *
+ *   Three real-order templates cycle per tuple (mirrors the coord-golden orders so the eval and the
+ *   training distribution agree):
+ *
+ *   1. canonical  "«V» «n», «pc» «V»"
+ *   2. bare       "«V» «n», «V»"        (no postcode — the anchor-free form, the fr-bare lesson)
+ *   3. pc-first   "«pc» «V», «V» «n»"
+ *
+ *   Gold spans: leading «V» = street (matches OA ground truth — the village IS the address line),
+ *   «n» = house_number (NEVER split mid-digits), «pc» = postcode (never swallowing the neighbor),
+ *   trailing «V» = locality (the binding the resolver needs).
+ */
+import { alignAndWrite, makeMulberry32, readTuples, sliceSourceID } from "#recipes/scaffold";
+/**
+ * Slice recipe registered with the corpus builder — see the file header for the parse behaviour it exists to exercise,
+ * and `description` below for the surface form it generates.
+ */
+export const siBareVillageRecipe = {
+    name: "si-bare-village",
+    description: "SI no-street village form (#901 run-2): '«V» «n», «pc» «V»' — the fr-bare-street counter-distribution",
+    mode: "tuples",
+    async run(opts, write) {
+        makeMulberry32(opts.seed);
+        let read = 0;
+        let emitted = 0;
+        let skipped = 0;
+        for await (const t of readTuples(opts.input)) {
+            read++;
+            const village = String(t.locality ?? "").trim();
+            const number = String(t.number ?? "").trim();
+            const postcode = String(t.postcode ?? "").trim();
+            if (!village || !number || !postcode) {
+                skipped++;
+                continue;
+            }
+            const order = read % 3;
+            let raw;
+            const components = {
+                street: village,
+                house_number: number,
+                locality: village,
+            };
+            if (order === 0) {
+                components.postcode = postcode;
+                raw = `${village} ${number}, ${postcode} ${village}`;
+            }
+            else if (order === 1) {
+                raw = `${village} ${number}, ${village}`;
+            }
+            else {
+                components.postcode = postcode;
+                raw = `${postcode} ${village}, ${village} ${number}`;
+            }
+            const source_id = sliceSourceID("synth-si-bare-village", { ...components, o: String(order), v: String(read) });
+            const canonical = {
+                raw,
+                components,
+                country: "SI",
+                locale: "sl-SI",
+                source: "synth-si-bare-village",
+                source_id,
+                corpus_version: "0.9.9",
+                license: "Synthetic — si-bare-village; (village, number, postcode) from OpenAddresses SI (per-source attribution in the model card)",
+            };
+            if (alignAndWrite(write, canonical, "si-bare-village")) {
+                emitted++;
+            }
+            else {
+                skipped++;
+            }
+        }
+        return { read, emitted, skipped };
+    },
+};
+//# sourceMappingURL=si-bare-village.js.map

package/out/recipes/si-bare-village.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"si-bare-village.js","sourceRoot":"","sources":["../../lib/recipes/si-bare-village.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;AAEH,OAAO,EAAE,aAAa,EAAE,cAAc,EAAE,UAAU,EAAqB,aAAa,EAAE,MAAM,mBAAmB,CAAA;AAE/G;;;GAGG;AACH,MAAM,CAAC,MAAM,mBAAmB,GAAiB;IAChD,IAAI,EAAE,iBAAiB;IACvB,WAAW,EAAE,uGAAuG;IACpH,IAAI,EAAE,QAAQ;IACd,KAAK,CAAC,GAAG,CAAC,IAAI,EAAE,KAAK;QACpB,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACzB,IAAI,IAAI,GAAG,CAAC,CAAA;QACZ,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,OAAO,GAAG,CAAC,CAAA;QAEf,IAAI,KAAK,EAAE,MAAM,CAAC,IAAI,UAAU,CAAC,IAAI,CAAC,KAAM,CAAC,EAAE,CAAC;YAC/C,IAAI,EAAE,CAAA;YACN,MAAM,OAAO,GAAG,MAAM,CAAC,CAAC,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAC/C,MAAM,MAAM,GAAG,MAAM,CAAC,CAAC,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAC5C,MAAM,QAAQ,GAAG,MAAM,CAAC,CAAC,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAEhD,IAAI,CAAC,OAAO,IAAI,CAAC,MAAM,IAAI,CAAC,QAAQ,EAAE,CAAC;gBACtC,OAAO,EAAE,CAAA;gBAET,SAAQ;YACT,CAAC;YAED,MAAM,KAAK,GAAG,IAAI,GAAG,CAAC,CAAA;YACtB,IAAI,GAAW,CAAA;YAEf,MAAM,UAAU,GAA2B;gBAC1C,MAAM,EAAE,OAAO;gBACf,YAAY,EAAE,MAAM;gBACpB,QAAQ,EAAE,OAAO;aACjB,CAAA;YAED,IAAI,KAAK,KAAK,CAAC,EAAE,CAAC;gBACjB,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;gBAC9B,GAAG,GAAG,GAAG,OAAO,IAAI,MAAM,KAAK,QAAQ,IAAI,OAAO,EAAE,CAAA;YACrD,CAAC;iBAAM,IAAI,KAAK,KAAK,CAAC,EAAE,CAAC;gBACxB,GAAG,GAAG,GAAG,OAAO,IAAI,MAAM,KAAK,OAAO,EAAE,CAAA;YACzC,CAAC;iBAAM,CAAC;gBACP,UAAU,CAAC,QAAQ,GAAG,QAAQ,CAAA;gBAC9B,GAAG,GAAG,GAAG,QAAQ,IAAI,OAAO,KAAK,OAAO,IAAI,MAAM,EAAE,CAAA;YACrD,CAAC;YAED,MAAM,SAAS,GAAG,aAAa,CAAC,uBAAuB,EAAE,EAAE,GAAG,UAAU,EAAE,CAAC,EAAE,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;YAE9G,MAAM,SAAS,GAAG;gBACjB,GAAG;gBACH,UAAU;gBACV,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,OAAO;gBACf,MAAM,EAAE,uBAAuB;gBAC/B,SAAS;gBACT,cAAc,EAAE,OAAO;gBACvB,OAAO,EACN,2HAA2H;aAC5H,CAAA;YAED,IAAI,aAAa,CAAC,KAAK,EAAE,SAAS,EAAE,iBAAiB,CAAC,EAAE,CAAC;gBACxD,OAAO,EAAE,CAAA;YACV,CAAC;iBAAM,CAAC;gBACP,OAAO,EAAE,CAAA;YACV,CAAC;QACF,CAAC;QAED,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,CAAA;IAClC,CAAC;CACD,CAAA"}