@empyria/name 0.1.2 → 0.1.4

package/lib/Name.js

@@ -92,6 +92,17 @@ export const TITLE_LIST = [
 ]
 export const TITLES = enumise(TITLE_LIST)
 
+// Honorifics as compact keys (transliterated, alphanumerics only, upper-case) — the form
+// `nameFp` compares each token against. `TITLES` (an enumise() object) can't be used directly:
+// it has no `.has`, its keys carry `_` separators, and its values keep original casing.
+const TITLE_KEYS = new Set(
+	TITLE_LIST.map((t) =>
+		transliterate(t)
+			.toUpperCase()
+			.replace(/[^A-Z0-9]/g, ''),
+	),
+)
+
 export function findTitle(_title) {
 	if (!_title) return undefined
 
@@ -99,4 +110,58 @@ export function findTitle(_title) {
 	return TITLE_LIST.find((t) => transliterate(t).toUpperCase() === title)
 }
 
+/**
+ * Base fold: transliterate to ASCII, collapse any run of whitespace to a single space, trim.
+ * @param {unknown} value
+ * @returns {string}
+ */
+export function fold(value) {
+	return transliterate(String(value ?? ''))
+		.replace(/\s+/g, ' ')
+		.trim()
+}
+
+/**
+ * Compact identifier key — transliterated, alphanumerics only, upper-case. Format-agnostic:
+ * `idKey('5299 00 T8BM49') === idKey('529900t8bm49')`, `idKey('DE 123.456.789') === 'DE123456789'`.
+ * Empty string for empty/nullish input (callers skip those).
+ * @param {unknown} value
+ * @returns {string}
+ */
+export function idKey(value) {
+	return fold(value)
+		.toUpperCase()
+		.replace(/[^A-Z0-9]/g, '')
+}
+
+/**
+ * Free-text name form for external full-text engines (BAILII, CourtListener, opencaselaw) —
+ * transliterated, lower-case, single-spaced, tokens intact so their tokenisers still work.
+ * @param {unknown} value
+ * @returns {string}
+ */
+export function nameText(value) {
+	return fold(value).toLowerCase()
+}
+
+/**
+ * Order-, case-, punctuation-, script-invariant name fingerprint: transliterate, upper-case,
+ * split on whitespace, strip non-alphanumerics per token, drop honorifics and 1-char tokens,
+ * dedupe, sort. `nameFp('Dr. Steve Charles') === nameFp('charles, steve') === 'CHARLES STEVE'`.
+ *
+ * Best-effort for organisations — legal-form / "&" vs "and" differences aren't reconciled here
+ * (`nameFp('J.P. Morgan & Co') !== nameFp('JPMorgan and Co')`); the `:Ident` exact match and the
+ * raw-name text search are the reliable paths for companies.
+ * @param {unknown} value
+ * @returns {string}
+ */
+export function nameFp(value) {
+	const tokens = fold(value)
+		.toUpperCase()
+		.split(' ')
+		.map((token) => token.replace(/[^A-Z0-9]/g, ''))
+		.filter((token) => token.length > 1 && !TITLE_KEYS.has(token))
+	return [...new Set(tokens)].sort().join(' ')
+}
+
 export { transliterate, slugify, TRANS, capitalise, shortify }

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@empyria/name",
-	"version": "0.1.2",
+	"version": "0.1.4",
 	"description": "Name and title manipulation library for Empyria",
 	"license": "MIT",
 	"author": "Imre Fazekas <imre.fazekas@icloud.com>",

package/test/Name.test.js

@@ -3,6 +3,10 @@ import {
 	TITLE_LIST,
 	TITLES,
 	findTitle,
+	fold,
+	idKey,
+	nameFp,
+	nameText,
 	TRANS,
 	capitalise,
 	shortify,
@@ -73,3 +77,22 @@ describe('re-exported string helpers (from @empyria/classification)', () => {
 		expect(slugify('Jean Pierre')).toBe('jean-pierre')
 	})
 })
+
+describe('fold / idKey / nameText / nameFp', () => {
+	test('idKey is format-agnostic', () => {
+		expect(idKey('5299 00 T8BM49')).toBe(idKey('529900t8bm49'))
+		expect(idKey('DE 123.456.789')).toBe('DE123456789')
+	})
+
+	test('nameText transliterates + lower-cases, keeps token spacing', () => {
+		expect(nameText('  José   MÜLLER ')).toBe('jose muller')
+	})
+
+	test('nameFp is order / case / punctuation / honorific / script invariant', () => {
+		expect(nameFp('Steve Charles')).toBe(nameFp('Charles Steve'))
+		expect(nameFp('Dr. Steve Charles')).toBe(nameFp('charles, steve'))
+		expect(nameFp('Professor Ada Lovelace')).toBe(nameFp('lovelace, ada'))
+		expect(nameFp('Владимир Путин')).toBe(nameFp('Putin Vladimir'))
+		expect(nameFp('Mr.')).toBe('')
+	})
+})