@mailwoman/corpus 6.4.0 → 6.5.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1 +1 @@
1
- {"version":3,"file":"no-fragment.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/no-fragment.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAIH,OAAO,EAA6C,KAAK,WAAW,EAAiB,MAAM,eAAe,CAAA;AAgB1G,eAAO,MAAM,gBAAgB,EAAE,WAyH9B,CAAA"}
1
+ {"version":3,"file":"no-fragment.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/no-fragment.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAIH,OAAO,EAA6C,KAAK,WAAW,EAAiB,MAAM,eAAe,CAAA;AAgB1G,eAAO,MAAM,gBAAgB,EAAE,WA8I9B,CAAA"}
@@ -64,6 +64,14 @@ export const noFragmentRecipe = {
64
64
  flag: "--counter-prob <n>",
65
65
  description: "Share of ALL rows that are counter-distribution (bare locality OR bare postcode) (default 0.30)",
66
66
  },
67
+ {
68
+ flag: "--long-number-boost <n>",
69
+ description: "knob 3: emit N copies of each street+number row whose number has >= --long-number-min-digits digits — oversample the failing long-number class the model calls a postcode (default 1 = no boost)",
70
+ },
71
+ {
72
+ flag: "--long-number-min-digits <n>",
73
+ description: "knob 3: minimum digit count for a number to be 'long' and boosted (default 3)",
74
+ },
67
75
  ],
68
76
  async run(opts, write) {
69
77
  const random = makeMulberry32(opts.seed);
@@ -81,6 +89,8 @@ export const noFragmentRecipe = {
81
89
  throw new Error(`no-fragment: --exclude-surfaces "${excludePath}" listed no surfaces`);
82
90
  const bareStreetProb = opts.bareProb ?? 0.3;
83
91
  const counterProb = opts.counterProb ?? 0.3;
92
+ const longNumberBoost = Math.max(1, Math.floor(opts.longNumberBoost ?? 1));
93
+ const longNumberMinDigits = opts.longNumberMinDigits ?? 3;
84
94
  // Harvested from the tuples — every NO row carries its locality and postcode, so the two
85
95
  // counter-classes need no second source.
86
96
  const localities = new Set();
@@ -89,8 +99,11 @@ export const noFragmentRecipe = {
89
99
  let emitted = 0;
90
100
  let skipped = 0;
91
101
  let contaminated = 0;
102
+ let emitSeq = 0;
92
103
  const emit = (raw, components, klass) => {
93
- const source_id = shardSourceID("synth-no-fragment", { ...components, k: klass, v: String(read) });
104
+ // emitSeq keeps every emit distinct knob 3 emits N copies of one long-number row, and
105
+ // (components, read) alone would collide their source_id and let downstream dedup drop the boost.
106
+ const source_id = shardSourceID("synth-no-fragment", { ...components, k: klass, v: `${read}:${emitSeq++}` });
94
107
  const canonical = {
95
108
  raw,
96
109
  components,
@@ -146,7 +159,14 @@ export const noFragmentRecipe = {
146
159
  }
147
160
  else {
148
161
  const klass = number.includes("/") ? "slash-hn" : "street-hn";
149
- emit(`${street} ${number}`, { street, house_number: number }, klass);
162
+ // knob 3: the failing class is street + LONG number (Leppdalsvegen 1285 -> postcode). The
163
+ // digit count, not the slash, is what tips the length prior toward postcode. Oversample
164
+ // those rows to fight the prior with volume and teach the street/number boundary directly.
165
+ const digits = (number.match(/\d/g) ?? []).length;
166
+ const copies = digits >= longNumberMinDigits ? longNumberBoost : 1;
167
+ for (let c = 0; c < copies; c++) {
168
+ emit(`${street} ${number}`, { street, house_number: number }, klass);
169
+ }
150
170
  }
151
171
  }
152
172
  return { read, emitted, skipped, contaminated };
@@ -1 +1 @@
1
- {"version":3,"file":"no-fragment.js","sourceRoot":"","sources":["../../../src/shard-recipes/no-fragment.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,SAAS,CAAA;AAEtC,OAAO,EAAE,aAAa,EAAE,cAAc,EAAE,UAAU,EAAoB,aAAa,EAAE,MAAM,eAAe,CAAA;AAE1G;;;;GAIG;AACH,MAAM,IAAI,GAAG,CAAC,KAAa,EAAU,EAAE,CAAC,KAAK,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,WAAW,EAAE,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AAExG,6FAA6F;AAC7F,MAAM,OAAO,GAAG,CAAC,KAAa,EAAU,EAAE,CACzC,KAAK;KACH,KAAK,CAAC,KAAK,CAAC;KACZ,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAE,CAAC,WAAW,EAAE,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;KACpE,IAAI,CAAC,GAAG,CAAC,CAAA;AAEZ,MAAM,CAAC,MAAM,gBAAgB,GAAgB;IAC5C,IAAI,EAAE,aAAa;IACnB,WAAW,EACV,2JAA2J;IAC5J,IAAI,EAAE,QAAQ;IACd,OAAO,EAAE;QACR;YACC,IAAI,EAAE,2BAA2B;YACjC,WAAW,EACV,gHAAgH;SACjH;QACD;YACC,IAAI,EAAE,wBAAwB;YAC9B,WAAW,EAAE,mGAAmG;SAChH;QACD;YACC,IAAI,EAAE,oBAAoB;YAC1B,WAAW,EAAE,iGAAiG;SAC9G;KACD;IACD,KAAK,CAAC,GAAG,CAAC,IAAI,EAAE,KAAK;QACpB,MAAM,MAAM,GAAG,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACxC,MAAM,WAAW,GAAG,IAAI,CAAC,eAAe,CAAA;QAExC,IAAI,CAAC,WAAW,EAAE,CAAC;YAClB,MAAM,IAAI,KAAK,CACd,uFAAuF;gBACtF,qGAAqG;gBACrG,sFAAsF,CACvF,CAAA;QACF,CAAC;QAED,MAAM,QAAQ,GAAG,IAAI,GAAG,CACvB,YAAY,CAAC,WAAW,EAAE,MAAM,CAAC;aAC/B,KAAK,CAAC,IAAI,CAAC;aACX,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;aAC1B,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CACjD,CAAA;QAED,IAAI,CAAC,QAAQ,CAAC,IAAI;YAAE,MAAM,IAAI,KAAK,CAAC,oCAAoC,WAAW,sBAAsB,CAAC,CAAA;QAE1G,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,IAAI,GAAG,CAAA;QAC3C,MAAM,WAAW,GAAG,IAAI,CAAC,WAAW,IAAI,GAAG,CAAA;QAE3C,yFAAyF;QACzF,yCAAyC;QACzC,MAAM,UAAU,GAAG,IAAI,GAAG,EAAU,CAAA;QACpC,MAAM,SAAS,GAAG,IAAI,GAAG,EAAU,CAAA;QAEnC,IAAI,IAAI,GAAG,CAAC,CAAA;QACZ,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,YAAY,GAAG,CAAC,CAAA;QAEpB,MAAM,IAAI,GAAG,CAAC,GAAW,EAAE,UAAkC,EAAE,KAAa,EAAQ,EAAE;YACrF,MAAM,SAAS,GAAG,aAAa,CAAC,mBAAmB,EAAE,EAAE,GAAG,UAAU,EAAE,CAAC,EAAE,KAAK,EAAE,CAAC,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;YAClG,MAAM,SAAS,GAAG;gBACjB,GAAG;gBACH,UAAU;gBACV,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,OAAO;gBACf,MAAM,EAAE,mBAAmB;gBAC3B,SAAS;gBACT,cAAc,EAAE,QAAQ;gBACxB,OAAO,EAAE,8FAA8F;aACvG,CAAA;YAED,IAAI,aAAa,CAAC,KAAK,EAAE,SAAS,EAAE,aAAa,CAAC;gBAAE,OAAO,EAAE,CAAA;;gBACxD,OAAO,EAAE,CAAA;QACf,CAAC,CAAA;QAED,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,UAAU,CAAC,IAAI,CAAC,KAAM,CAAC,EAAE,CAAC;YACnD,IAAI,EAAE,CAAA;YACN,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAChD,MAAM,QAAQ,GAAG,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC,CAAA;YAC7D,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAChD,MAAM,QAAQ,GAAG,MAAM,CAAC,KAAK,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAEpD,IAAI,QAAQ;gBAAE,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAA;YACtC,IAAI,QAAQ;gBAAE,SAAS,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAA;YAErC,IAAI,CAAC,MAAM,EAAE,CAAC;gBACb,OAAO,EAAE,CAAA;gBACT,SAAQ;YACT,CAAC;YAED,iEAAiE;YACjE,IAAI,QAAQ,CAAC,GAAG,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,EAAE,CAAC;gBAChC,YAAY,EAAE,CAAA;gBACd,SAAQ;YACT,CAAC;YAED,0FAA0F;YAC1F,2FAA2F;YAC3F,0EAA0E;YAC1E,IAAI,MAAM,EAAE,GAAG,WAAW,EAAE,CAAC;gBAC5B,IAAI,MAAM,EAAE,GAAG,GAAG,IAAI,UAAU,CAAC,IAAI,EAAE,CAAC;oBACvC,MAAM,GAAG,GAAG,CAAC,GAAG,UAAU,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,UAAU,CAAC,IAAI,CAAC,CAAE,CAAA;oBAEpE,IAAI,CAAC,GAAG,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,EAAE,uBAAuB,CAAC,CAAA;gBACtD,CAAC;qBAAM,IAAI,SAAS,CAAC,IAAI,EAAE,CAAC;oBAC3B,MAAM,EAAE,GAAG,CAAC,GAAG,SAAS,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,SAAS,CAAC,IAAI,CAAC,CAAE,CAAA;oBAEjE,IAAI,CAAC,EAAE,EAAE,EAAE,QAAQ,EAAE,EAAE,EAAE,EAAE,uBAAuB,CAAC,CAAA;gBACpD,CAAC;gBACD,SAAQ;YACT,CAAC;YAED,0FAA0F;YAC1F,8FAA8F;YAC9F,IAAI,CAAC,MAAM,IAAI,MAAM,EAAE,GAAG,cAAc,EAAE,CAAC;gBAC1C,IAAI,CAAC,MAAM,EAAE,EAAE,MAAM,EAAE,EAAE,aAAa,CAAC,CAAA;YACxC,CAAC;iBAAM,CAAC;gBACP,MAAM,KAAK,GAAG,MAAM,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,WAAW,CAAA;gBAE7D,IAAI,CAAC,GAAG,MAAM,IAAI,MAAM,EAAE,EAAE,EAAE,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,EAAE,KAAK,CAAC,CAAA;YACrE,CAAC;QACF,CAAC;QAED,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,CAAA;IAChD,CAAC;CACD,CAAA"}
1
+ {"version":3,"file":"no-fragment.js","sourceRoot":"","sources":["../../../src/shard-recipes/no-fragment.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,SAAS,CAAA;AAEtC,OAAO,EAAE,aAAa,EAAE,cAAc,EAAE,UAAU,EAAoB,aAAa,EAAE,MAAM,eAAe,CAAA;AAE1G;;;;GAIG;AACH,MAAM,IAAI,GAAG,CAAC,KAAa,EAAU,EAAE,CAAC,KAAK,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,WAAW,EAAE,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AAExG,6FAA6F;AAC7F,MAAM,OAAO,GAAG,CAAC,KAAa,EAAU,EAAE,CACzC,KAAK;KACH,KAAK,CAAC,KAAK,CAAC;KACZ,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAE,CAAC,WAAW,EAAE,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;KACpE,IAAI,CAAC,GAAG,CAAC,CAAA;AAEZ,MAAM,CAAC,MAAM,gBAAgB,GAAgB;IAC5C,IAAI,EAAE,aAAa;IACnB,WAAW,EACV,2JAA2J;IAC5J,IAAI,EAAE,QAAQ;IACd,OAAO,EAAE;QACR;YACC,IAAI,EAAE,2BAA2B;YACjC,WAAW,EACV,gHAAgH;SACjH;QACD;YACC,IAAI,EAAE,wBAAwB;YAC9B,WAAW,EAAE,mGAAmG;SAChH;QACD;YACC,IAAI,EAAE,oBAAoB;YAC1B,WAAW,EAAE,iGAAiG;SAC9G;QACD;YACC,IAAI,EAAE,yBAAyB;YAC/B,WAAW,EACV,kMAAkM;SACnM;QACD;YACC,IAAI,EAAE,8BAA8B;YACpC,WAAW,EAAE,+EAA+E;SAC5F;KACD;IACD,KAAK,CAAC,GAAG,CAAC,IAAI,EAAE,KAAK;QACpB,MAAM,MAAM,GAAG,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACxC,MAAM,WAAW,GAAG,IAAI,CAAC,eAAe,CAAA;QAExC,IAAI,CAAC,WAAW,EAAE,CAAC;YAClB,MAAM,IAAI,KAAK,CACd,uFAAuF;gBACtF,qGAAqG;gBACrG,sFAAsF,CACvF,CAAA;QACF,CAAC;QAED,MAAM,QAAQ,GAAG,IAAI,GAAG,CACvB,YAAY,CAAC,WAAW,EAAE,MAAM,CAAC;aAC/B,KAAK,CAAC,IAAI,CAAC;aACX,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;aAC1B,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CACjD,CAAA;QAED,IAAI,CAAC,QAAQ,CAAC,IAAI;YAAE,MAAM,IAAI,KAAK,CAAC,oCAAoC,WAAW,sBAAsB,CAAC,CAAA;QAE1G,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,IAAI,GAAG,CAAA;QAC3C,MAAM,WAAW,GAAG,IAAI,CAAC,WAAW,IAAI,GAAG,CAAA;QAC3C,MAAM,eAAe,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,eAAe,IAAI,CAAC,CAAC,CAAC,CAAA;QAC1E,MAAM,mBAAmB,GAAG,IAAI,CAAC,mBAAmB,IAAI,CAAC,CAAA;QAEzD,yFAAyF;QACzF,yCAAyC;QACzC,MAAM,UAAU,GAAG,IAAI,GAAG,EAAU,CAAA;QACpC,MAAM,SAAS,GAAG,IAAI,GAAG,EAAU,CAAA;QAEnC,IAAI,IAAI,GAAG,CAAC,CAAA;QACZ,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,OAAO,GAAG,CAAC,CAAA;QACf,IAAI,YAAY,GAAG,CAAC,CAAA;QACpB,IAAI,OAAO,GAAG,CAAC,CAAA;QAEf,MAAM,IAAI,GAAG,CAAC,GAAW,EAAE,UAAkC,EAAE,KAAa,EAAQ,EAAE;YACrF,wFAAwF;YACxF,kGAAkG;YAClG,MAAM,SAAS,GAAG,aAAa,CAAC,mBAAmB,EAAE,EAAE,GAAG,UAAU,EAAE,CAAC,EAAE,KAAK,EAAE,CAAC,EAAE,GAAG,IAAI,IAAI,OAAO,EAAE,EAAE,EAAE,CAAC,CAAA;YAC5G,MAAM,SAAS,GAAG;gBACjB,GAAG;gBACH,UAAU;gBACV,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,OAAO;gBACf,MAAM,EAAE,mBAAmB;gBAC3B,SAAS;gBACT,cAAc,EAAE,QAAQ;gBACxB,OAAO,EAAE,8FAA8F;aACvG,CAAA;YAED,IAAI,aAAa,CAAC,KAAK,EAAE,SAAS,EAAE,aAAa,CAAC;gBAAE,OAAO,EAAE,CAAA;;gBACxD,OAAO,EAAE,CAAA;QACf,CAAC,CAAA;QAED,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,UAAU,CAAC,IAAI,CAAC,KAAM,CAAC,EAAE,CAAC;YACnD,IAAI,EAAE,CAAA;YACN,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAChD,MAAM,QAAQ,GAAG,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC,CAAA;YAC7D,MAAM,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAChD,MAAM,QAAQ,GAAG,MAAM,CAAC,KAAK,CAAC,QAAQ,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAA;YAEpD,IAAI,QAAQ;gBAAE,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAA;YACtC,IAAI,QAAQ;gBAAE,SAAS,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAA;YAErC,IAAI,CAAC,MAAM,EAAE,CAAC;gBACb,OAAO,EAAE,CAAA;gBACT,SAAQ;YACT,CAAC;YAED,iEAAiE;YACjE,IAAI,QAAQ,CAAC,GAAG,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,EAAE,CAAC;gBAChC,YAAY,EAAE,CAAA;gBACd,SAAQ;YACT,CAAC;YAED,0FAA0F;YAC1F,2FAA2F;YAC3F,0EAA0E;YAC1E,IAAI,MAAM,EAAE,GAAG,WAAW,EAAE,CAAC;gBAC5B,IAAI,MAAM,EAAE,GAAG,GAAG,IAAI,UAAU,CAAC,IAAI,EAAE,CAAC;oBACvC,MAAM,GAAG,GAAG,CAAC,GAAG,UAAU,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,UAAU,CAAC,IAAI,CAAC,CAAE,CAAA;oBAEpE,IAAI,CAAC,GAAG,EAAE,EAAE,QAAQ,EAAE,GAAG,EAAE,EAAE,uBAAuB,CAAC,CAAA;gBACtD,CAAC;qBAAM,IAAI,SAAS,CAAC,IAAI,EAAE,CAAC;oBAC3B,MAAM,EAAE,GAAG,CAAC,GAAG,SAAS,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,GAAG,SAAS,CAAC,IAAI,CAAC,CAAE,CAAA;oBAEjE,IAAI,CAAC,EAAE,EAAE,EAAE,QAAQ,EAAE,EAAE,EAAE,EAAE,uBAAuB,CAAC,CAAA;gBACpD,CAAC;gBACD,SAAQ;YACT,CAAC;YAED,0FAA0F;YAC1F,8FAA8F;YAC9F,IAAI,CAAC,MAAM,IAAI,MAAM,EAAE,GAAG,cAAc,EAAE,CAAC;gBAC1C,IAAI,CAAC,MAAM,EAAE,EAAE,MAAM,EAAE,EAAE,aAAa,CAAC,CAAA;YACxC,CAAC;iBAAM,CAAC;gBACP,MAAM,KAAK,GAAG,MAAM,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,CAAC,CAAC,WAAW,CAAA;gBAC7D,0FAA0F;gBAC1F,wFAAwF;gBACxF,2FAA2F;gBAC3F,MAAM,MAAM,GAAG,CAAC,MAAM,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAA;gBACjD,MAAM,MAAM,GAAG,MAAM,IAAI,mBAAmB,CAAC,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC,CAAA;gBAElE,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;oBACjC,IAAI,CAAC,GAAG,MAAM,IAAI,MAAM,EAAE,EAAE,EAAE,MAAM,EAAE,YAAY,EAAE,MAAM,EAAE,EAAE,KAAK,CAAC,CAAA;gBACrE,CAAC;YACF,CAAC;QACF,CAAC;QAED,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,CAAA;IAChD,CAAC;CACD,CAAA"}
@@ -80,6 +80,13 @@ export interface ShardRecipeOpts {
80
80
  excludeSurfaces?: string;
81
81
  /** `no-fragment`: share of rows that are counter-distribution (bare locality OR bare postcode). */
82
82
  counterProb?: number;
83
+ /**
84
+ * `no-fragment` knob 3: emit N copies of each street+number row whose number has >= longNumberMinDigits digits
85
+ * (oversample the failing long-number class). Default 1 = no boost.
86
+ */
87
+ longNumberBoost?: number;
88
+ /** `no-fragment` knob 3: minimum digit count for a number to count as "long" and be boosted. Default 3. */
89
+ longNumberMinDigits?: number;
83
90
  }
84
91
  /** Tally a recipe returns. */
85
92
  export interface ShardStats {
@@ -1 +1 @@
1
- {"version":3,"file":"scaffold.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/scaffold.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAKH;;;;GAIG;AACH,wBAAgB,aAAa,CAAC,SAAS,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC,GAAG,MAAM,CAElG;AAED,+FAA+F;AAC/F,MAAM,WAAW,UAAU;IAC1B,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB,CAAC,CAAC,EAAE,MAAM,GAAG,OAAO,CAAA;CACpB;AAED;;;GAGG;AACH,wBAAgB,OAAO,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,MAAM,CAQlD;AAED,6CAA6C;AAC7C,eAAO,MAAM,UAAU,gBAAU,CAAA;AAEjC;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,MAAM,CAUzD;AAED,mGAAmG;AACnG,wBAAuB,UAAU,CAAC,KAAK,EAAE,MAAM,GAAG,cAAc,CAAC,UAAU,CAAC,CAgB3E;AAED,kGAAkG;AAClG,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,oGAAoG;AACpG,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,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,gBAAgB,CAAC,EAAE,MAAM,CAAA;IACzB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB,mGAAmG;IACnG,WAAW,CAAC,EAAE,MAAM,CAAA;CACpB;AAED,8BAA8B;AAC9B,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,gFAAgF;AAChF,MAAM,WAAW,iBAAiB;IACjC,IAAI,EAAE,MAAM,CAAA;IACZ,WAAW,EAAE,MAAM,CAAA;CACnB;AAED,yEAAyE;AACzE,MAAM,WAAW,WAAW;IAC3B,sEAAsE;IACtE,IAAI,EAAE,MAAM,CAAA;IACZ,gDAAgD;IAChD,WAAW,EAAE,MAAM,CAAA;IACnB,gFAAgF;IAChF,IAAI,EAAE,QAAQ,GAAG,UAAU,CAAA;IAC3B,qEAAqE;IACrE,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"}
1
+ {"version":3,"file":"scaffold.d.ts","sourceRoot":"","sources":["../../../src/shard-recipes/scaffold.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAKH;;;;GAIG;AACH,wBAAgB,aAAa,CAAC,SAAS,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC,GAAG,MAAM,CAElG;AAED,+FAA+F;AAC/F,MAAM,WAAW,UAAU;IAC1B,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,OAAO,CAAC,EAAE,MAAM,CAAA;IAChB,CAAC,CAAC,EAAE,MAAM,GAAG,OAAO,CAAA;CACpB;AAED;;;GAGG;AACH,wBAAgB,OAAO,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,MAAM,CAQlD;AAED,6CAA6C;AAC7C,eAAO,MAAM,UAAU,gBAAU,CAAA;AAEjC;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,MAAM,CAUzD;AAED,mGAAmG;AACnG,wBAAuB,UAAU,CAAC,KAAK,EAAE,MAAM,GAAG,cAAc,CAAC,UAAU,CAAC,CAgB3E;AAED,kGAAkG;AAClG,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,oGAAoG;AACpG,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,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,MAAM,CAAC,EAAE,MAAM,CAAA;IACf,QAAQ,CAAC,EAAE,MAAM,CAAA;IACjB,gBAAgB,CAAC,EAAE,MAAM,CAAA;IACzB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB,mGAAmG;IACnG,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAA;IACxB,2GAA2G;IAC3G,mBAAmB,CAAC,EAAE,MAAM,CAAA;CAC5B;AAED,8BAA8B;AAC9B,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,gFAAgF;AAChF,MAAM,WAAW,iBAAiB;IACjC,IAAI,EAAE,MAAM,CAAA;IACZ,WAAW,EAAE,MAAM,CAAA;CACnB;AAED,yEAAyE;AACzE,MAAM,WAAW,WAAW;IAC3B,sEAAsE;IACtE,IAAI,EAAE,MAAM,CAAA;IACZ,gDAAgD;IAChD,WAAW,EAAE,MAAM,CAAA;IACnB,gFAAgF;IAChF,IAAI,EAAE,QAAQ,GAAG,UAAU,CAAA;IAC3B,qEAAqE;IACrE,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/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@mailwoman/corpus",
3
- "version": "6.4.0",
3
+ "version": "6.5.0",
4
4
  "description": "Mailwoman corpus pipeline: BIO-labeled dataset builder for the neural classifier.",
5
5
  "license": "AGPL-3.0-only OR LicenseRef-Commercial",
6
6
  "repository": {
@@ -82,9 +82,9 @@
82
82
  },
83
83
  "dependencies": {
84
84
  "@dsnp/parquetjs": "1.8.8",
85
- "@mailwoman/codex": "6.4.0",
86
- "@mailwoman/core": "6.4.0",
87
- "@mailwoman/formatter": "6.4.0",
85
+ "@mailwoman/codex": "6.5.0",
86
+ "@mailwoman/core": "6.5.0",
87
+ "@mailwoman/formatter": "6.5.0",
88
88
  "csv-parse": "^7.0.1",
89
89
  "fast-glob": "^3.3.3",
90
90
  "fastest-levenshtein": "^1.0.16",