@fedify/vocab-runtime 2.0.31-dev.2306 → 2.0.31-dev.2308

package/deno.json

@@ -1,6 +1,6 @@
 {
   "name": "@fedify/vocab-runtime",
-  "version": "2.0.31-dev.2306+e775cf8c",
+  "version": "2.0.31-dev.2308+34090b3a",
   "license": "MIT",
   "exports": {
     ".": "./src/mod.ts",

package/dist/mod.cjs

@@ -4504,7 +4504,7 @@ const preloadedContexts = {
 //#endregion
 //#region deno.json
 var name = "@fedify/vocab-runtime";
-var version = "2.0.31-dev.2306+e775cf8c";
+var version = "2.0.31-dev.2308+34090b3a";
 //#endregion
 //#region src/request.ts
 /**
@@ -5730,6 +5730,50 @@ async function exportMultibaseKey(key) {
 	return new TextDecoder().decode(encoded);
 }
 //#endregion
+//#region src/langtag.ts
+const EXTLANG_PAIRS = new Set(Object.entries({
+	ar: "aao abh abv acm acq acw acx acy adf aeb aec afb ajp apc apd arb arq ars ary arz auz avl ayh ayl ayn ayp bbz pga shu ssh",
+	kok: "gom knn",
+	lv: "ltg lvs",
+	ms: "bjn btj bve bvu coa dup hji jak jax kvb kvr kxd lce lcf liw max meo mfa mfb min mqg msi mui orn ors pel pse tmw urk vkk vkt xmm zlm zmi zsm",
+	sgn: "ads aed aen afg ajs ase asf asp asq asw bfi bfk bog bqn bqy bvl bzs cds csc csd cse csf csg csl csn csq csr csx doq dse dsl dsz dyl ecs ehs esl esn eso eth fcs fse fsl fss gds gse gsg gsm gss gus hab haf hds hks hos hps hsh hsl icl iks ils inl ins ise isg isr jcs jhs jks jls jos jsl jus kgi kvk lbs lgs lls lsb lsc lsg lsl lsn lso lsp lst lsv lsw lsy lws mdl mfs mre msd msr mzc mzg mzy nbs ncs nsi nsl nsp nsr nzs okl pgz pks prl prz psc psd psg psl pso psp psr pys rib rms rnb rsi rsl rsm rsn sdl sfb sfs sgg sgx slf sls sqk sqs sqx ssp ssr svk swl syy szs tse tsm tsq tss tsy tza ugn ugy ukl uks vgt vsi vsl vsv wbs xki xml xms yds ygs yhs ysl ysm zhk zib zsl",
+	sw: "swc swh",
+	uz: "uzn uzs",
+	zh: "cdo cjy cmn cnp cpx csp czh czo gan hak hnm hsn luh lzh mnp nan sjc wuu yue"
+}).flatMap(([prefix, extlangs]) => extlangs.split(" ").map((extlang) => `${prefix}-${extlang}`)));
+const PREFIX_PATTERN = /^[A-Za-z]{2,3}$/;
+const EXTLANG_PATTERN = /^[A-Za-z]{3}$/;
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+function normalizeLanguageTag(tag) {
+	const subtags = tag.split("-");
+	if (subtags.length < 2) return tag;
+	const [prefix, extlang, ...rest] = subtags;
+	if (!PREFIX_PATTERN.test(prefix) || !EXTLANG_PATTERN.test(extlang)) return tag;
+	const language = extlang.toLowerCase();
+	if (!EXTLANG_PAIRS.has(`${prefix.toLowerCase()}-${language}`)) return tag;
+	return [language, ...rest].join("-");
+}
+//#endregion
 //#region src/langstr.ts
 /**
 * A language-tagged string which corresponds to the `rdf:langString` type.
@@ -5744,11 +5788,13 @@ var LanguageString = class extends String {
 	* Constructs a new `LanguageString`.
 	* @param value A string value written in the given language.
 	* @param language The language of the string.  If a string is given, it will
-	*                 be parsed as a `Intl.Locale` object.
+	*                 be parsed as a `Intl.Locale` object.  Extended language
+	*                 subtags are normalized first, e.g., `zh-yue` is parsed as
+	*                 `yue`.
 	*/
 	constructor(value, language) {
 		super(value);
-		this.locale = typeof language === "string" ? new Intl.Locale(language) : language;
+		this.locale = typeof language === "string" ? new Intl.Locale(normalizeLanguageTag(language)) : language;
 	}
 };
 LanguageString.prototype[Symbol.for("Deno.customInspect")] = function(inspect, options) {
@@ -5778,5 +5824,6 @@ exports.importSpki = importSpki;
 exports.isValidPublicIPv4Address = isValidPublicIPv4Address;
 exports.isValidPublicIPv6Address = isValidPublicIPv6Address;
 exports.logRequest = logRequest;
+exports.normalizeLanguageTag = normalizeLanguageTag;
 exports.preloadedContexts = preloadedContexts;
 exports.validatePublicUrl = validatePublicUrl;

package/dist/mod.d.cts

@@ -265,11 +265,37 @@ declare class LanguageString extends String {
   * Constructs a new `LanguageString`.
   * @param value A string value written in the given language.
   * @param language The language of the string.  If a string is given, it will
-  *                 be parsed as a `Intl.Locale` object.
+  *                 be parsed as a `Intl.Locale` object.  Extended language
+  *                 subtags are normalized first, e.g., `zh-yue` is parsed as
+  *                 `yue`.
   */
   constructor(value: string, language: Intl.Locale | string);
 }
 //#endregion
+//#region src/langtag.d.ts
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+declare function normalizeLanguageTag(tag: string): string;
+//#endregion
 //#region src/multibase/types.d.ts
 type BaseCode = "\x00" | "0" | "7" | "9" | "f" | "F" | "v" | "V" | "t" | "T" | "b" | "B" | "c" | "C" | "h" | "k" | "K" | "z" | "Z" | "m" | "M" | "u" | "U";
 /**
@@ -344,4 +370,4 @@ declare function isValidPublicIPv4Address(address: string): boolean;
 declare function isValidPublicIPv6Address(address: string): boolean;
 declare function expandIPv6Address(address: string): string;
 //#endregion
-export { type AuthenticatedDocumentLoaderFactory, type CreateRequestOptions, type DocumentLoader, type DocumentLoaderFactory, type DocumentLoaderFactoryOptions, type DocumentLoaderOptions, FetchError, type GetDocumentLoaderOptions, type GetUserAgentOptions, LanguageString, type RemoteDocument, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, preloadedContexts, validatePublicUrl };
+export { type AuthenticatedDocumentLoaderFactory, type CreateRequestOptions, type DocumentLoader, type DocumentLoaderFactory, type DocumentLoaderFactoryOptions, type DocumentLoaderOptions, FetchError, type GetDocumentLoaderOptions, type GetUserAgentOptions, LanguageString, type RemoteDocument, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, normalizeLanguageTag, preloadedContexts, validatePublicUrl };

package/dist/mod.d.ts

@@ -264,11 +264,37 @@ declare class LanguageString extends String {
   * Constructs a new `LanguageString`.
   * @param value A string value written in the given language.
   * @param language The language of the string.  If a string is given, it will
-  *                 be parsed as a `Intl.Locale` object.
+  *                 be parsed as a `Intl.Locale` object.  Extended language
+  *                 subtags are normalized first, e.g., `zh-yue` is parsed as
+  *                 `yue`.
   */
   constructor(value: string, language: Intl.Locale | string);
 }
 //#endregion
+//#region src/langtag.d.ts
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+declare function normalizeLanguageTag(tag: string): string;
+//#endregion
 //#region src/multibase/types.d.ts
 type BaseCode = "\x00" | "0" | "7" | "9" | "f" | "F" | "v" | "V" | "t" | "T" | "b" | "B" | "c" | "C" | "h" | "k" | "K" | "z" | "Z" | "m" | "M" | "u" | "U";
 /**
@@ -343,4 +369,4 @@ declare function isValidPublicIPv4Address(address: string): boolean;
 declare function isValidPublicIPv6Address(address: string): boolean;
 declare function expandIPv6Address(address: string): string;
 //#endregion
-export { type AuthenticatedDocumentLoaderFactory, type CreateRequestOptions, type DocumentLoader, type DocumentLoaderFactory, type DocumentLoaderFactoryOptions, type DocumentLoaderOptions, FetchError, type GetDocumentLoaderOptions, type GetUserAgentOptions, LanguageString, type RemoteDocument, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, preloadedContexts, validatePublicUrl };
+export { type AuthenticatedDocumentLoaderFactory, type CreateRequestOptions, type DocumentLoader, type DocumentLoaderFactory, type DocumentLoaderFactoryOptions, type DocumentLoaderOptions, FetchError, type GetDocumentLoaderOptions, type GetUserAgentOptions, LanguageString, type RemoteDocument, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, normalizeLanguageTag, preloadedContexts, validatePublicUrl };

package/dist/mod.js

@@ -4499,7 +4499,7 @@ const preloadedContexts = {
 //#endregion
 //#region deno.json
 var name = "@fedify/vocab-runtime";
-var version = "2.0.31-dev.2306+e775cf8c";
+var version = "2.0.31-dev.2308+34090b3a";
 //#endregion
 //#region src/request.ts
 /**
@@ -5725,6 +5725,50 @@ async function exportMultibaseKey(key) {
 	return new TextDecoder().decode(encoded);
 }
 //#endregion
+//#region src/langtag.ts
+const EXTLANG_PAIRS = new Set(Object.entries({
+	ar: "aao abh abv acm acq acw acx acy adf aeb aec afb ajp apc apd arb arq ars ary arz auz avl ayh ayl ayn ayp bbz pga shu ssh",
+	kok: "gom knn",
+	lv: "ltg lvs",
+	ms: "bjn btj bve bvu coa dup hji jak jax kvb kvr kxd lce lcf liw max meo mfa mfb min mqg msi mui orn ors pel pse tmw urk vkk vkt xmm zlm zmi zsm",
+	sgn: "ads aed aen afg ajs ase asf asp asq asw bfi bfk bog bqn bqy bvl bzs cds csc csd cse csf csg csl csn csq csr csx doq dse dsl dsz dyl ecs ehs esl esn eso eth fcs fse fsl fss gds gse gsg gsm gss gus hab haf hds hks hos hps hsh hsl icl iks ils inl ins ise isg isr jcs jhs jks jls jos jsl jus kgi kvk lbs lgs lls lsb lsc lsg lsl lsn lso lsp lst lsv lsw lsy lws mdl mfs mre msd msr mzc mzg mzy nbs ncs nsi nsl nsp nsr nzs okl pgz pks prl prz psc psd psg psl pso psp psr pys rib rms rnb rsi rsl rsm rsn sdl sfb sfs sgg sgx slf sls sqk sqs sqx ssp ssr svk swl syy szs tse tsm tsq tss tsy tza ugn ugy ukl uks vgt vsi vsl vsv wbs xki xml xms yds ygs yhs ysl ysm zhk zib zsl",
+	sw: "swc swh",
+	uz: "uzn uzs",
+	zh: "cdo cjy cmn cnp cpx csp czh czo gan hak hnm hsn luh lzh mnp nan sjc wuu yue"
+}).flatMap(([prefix, extlangs]) => extlangs.split(" ").map((extlang) => `${prefix}-${extlang}`)));
+const PREFIX_PATTERN = /^[A-Za-z]{2,3}$/;
+const EXTLANG_PATTERN = /^[A-Za-z]{3}$/;
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+function normalizeLanguageTag(tag) {
+	const subtags = tag.split("-");
+	if (subtags.length < 2) return tag;
+	const [prefix, extlang, ...rest] = subtags;
+	if (!PREFIX_PATTERN.test(prefix) || !EXTLANG_PATTERN.test(extlang)) return tag;
+	const language = extlang.toLowerCase();
+	if (!EXTLANG_PAIRS.has(`${prefix.toLowerCase()}-${language}`)) return tag;
+	return [language, ...rest].join("-");
+}
+//#endregion
 //#region src/langstr.ts
 /**
 * A language-tagged string which corresponds to the `rdf:langString` type.
@@ -5739,11 +5783,13 @@ var LanguageString = class extends String {
 	* Constructs a new `LanguageString`.
 	* @param value A string value written in the given language.
 	* @param language The language of the string.  If a string is given, it will
-	*                 be parsed as a `Intl.Locale` object.
+	*                 be parsed as a `Intl.Locale` object.  Extended language
+	*                 subtags are normalized first, e.g., `zh-yue` is parsed as
+	*                 `yue`.
 	*/
 	constructor(value, language) {
 		super(value);
-		this.locale = typeof language === "string" ? new Intl.Locale(language) : language;
+		this.locale = typeof language === "string" ? new Intl.Locale(normalizeLanguageTag(language)) : language;
 	}
 };
 LanguageString.prototype[Symbol.for("Deno.customInspect")] = function(inspect, options) {
@@ -5753,4 +5799,4 @@ LanguageString.prototype[Symbol.for("nodejs.util.inspect.custom")] = function(_d
 	return `<${this.locale.baseName}> ${inspect(this.toString(), options)}`;
 };
 //#endregion
-export { FetchError, LanguageString, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, preloadedContexts, validatePublicUrl };
+export { FetchError, LanguageString, UrlError, createActivityPubRequest, decodeMultibase, encodeMultibase, encodingFromBaseData, expandIPv6Address, exportMultibaseKey, exportSpki, getDocumentLoader, getRemoteDocument, getUserAgent, importMultibaseKey, importPem, importPkcs1, importSpki, isValidPublicIPv4Address, isValidPublicIPv6Address, logRequest, normalizeLanguageTag, preloadedContexts, validatePublicUrl };

package/dist/tests/{body-CGeu_wlY.mjs → body-Bx6-vXo_.mjs}

@@ -1,4 +1,4 @@
-import { t as FetchError } from "./request-JVGlYdPG.mjs";
+import { t as FetchError } from "./request-BV7yaWdl.mjs";
 import { getLogger } from "@logtape/logtape";
 //#region src/body.ts
 /** Decoded JSON body limit: 16 MiB.  @internal */

package/dist/tests/{body-KqP1lgFZ.cjs → body-DGmcZ1Kz.cjs}

@@ -1,5 +1,5 @@
 require("./chunk-Do9eywBl.cjs");
-const require_request = require("./request-75RoXeN4.cjs");
+const require_request = require("./request-Ban3HXuI.cjs");
 let _logtape_logtape = require("@logtape/logtape");
 //#region src/body.ts
 /** Decoded JSON body limit: 16 MiB.  @internal */

package/dist/tests/body.test.cjs

@@ -1,5 +1,5 @@
 require("./chunk-Do9eywBl.cjs");
-const require_body = require("./body-KqP1lgFZ.cjs");
+const require_body = require("./body-DGmcZ1Kz.cjs");
 let node_assert_strict = require("node:assert/strict");
 let node_test = require("node:test");
 let node_zlib = require("node:zlib");

package/dist/tests/body.test.mjs

@@ -1,4 +1,4 @@
-import { r as readBoundedText, t as BodyTooLargeError } from "./body-CGeu_wlY.mjs";
+import { r as readBoundedText, t as BodyTooLargeError } from "./body-Bx6-vXo_.mjs";
 import { deepStrictEqual, ok, rejects } from "node:assert/strict";
 import { test } from "node:test";
 import { gunzipSync, gzipSync } from "node:zlib";

package/dist/tests/docloader.test.cjs

@@ -1,6 +1,6 @@
 const require_chunk = require("./chunk-Do9eywBl.cjs");
-const require_request = require("./request-75RoXeN4.cjs");
-const require_body = require("./body-KqP1lgFZ.cjs");
+const require_request = require("./request-Ban3HXuI.cjs");
+const require_body = require("./body-DGmcZ1Kz.cjs");
 const require_link = require("./link-FguCydMA.cjs");
 const require_url = require("./url-Qk1svKZU.cjs");
 let node_test = require("node:test");

package/dist/tests/docloader.test.mjs

@@ -1,5 +1,5 @@
-import { a as name, i as logRequest, n as createActivityPubRequest, o as version, t as FetchError } from "./request-JVGlYdPG.mjs";
-import { n as MAX_BODY_SIZE, r as readBoundedText } from "./body-CGeu_wlY.mjs";
+import { a as name, i as logRequest, n as createActivityPubRequest, o as version, t as FetchError } from "./request-BV7yaWdl.mjs";
+import { n as MAX_BODY_SIZE, r as readBoundedText } from "./body-Bx6-vXo_.mjs";
 import { t as HttpHeaderLink } from "./link-NUUWCdnK.mjs";
 import { o as validatePublicUrl, t as UrlError } from "./url-DzlMS01z.mjs";
 import { test } from "node:test";

package/dist/tests/langstr.test.cjs

@@ -1,4 +1,5 @@
 const require_chunk = require("./chunk-Do9eywBl.cjs");
+const require_langtag = require("./langtag-Bm8oKhIO.cjs");
 let node_test = require("node:test");
 let node_assert = require("node:assert");
 let node_util = require("node:util");
@@ -17,11 +18,13 @@ var LanguageString = class extends String {
 	* Constructs a new `LanguageString`.
 	* @param value A string value written in the given language.
 	* @param language The language of the string.  If a string is given, it will
-	*                 be parsed as a `Intl.Locale` object.
+	*                 be parsed as a `Intl.Locale` object.  Extended language
+	*                 subtags are normalized first, e.g., `zh-yue` is parsed as
+	*                 `yue`.
 	*/
 	constructor(value, language) {
 		super(value);
-		this.locale = typeof language === "string" ? new Intl.Locale(language) : language;
+		this.locale = typeof language === "string" ? new Intl.Locale(require_langtag.normalizeLanguageTag(language)) : language;
 	}
 };
 LanguageString.prototype[Symbol.for("Deno.customInspect")] = function(inspect, options) {
@@ -38,6 +41,13 @@ LanguageString.prototype[Symbol.for("nodejs.util.inspect.custom")] = function(_d
 	(0, node_assert.deepStrictEqual)(langStr.locale, new Intl.Locale("en"));
 	(0, node_assert.deepStrictEqual)(new LanguageString("Hello", new Intl.Locale("en")), langStr);
 });
+(0, node_test.test)("new LanguageString() with extlang tags", () => {
+	(0, node_assert.deepStrictEqual)(new LanguageString("你好", "zh-YUE").locale.baseName, "yue");
+	(0, node_assert.deepStrictEqual)(new LanguageString("你好", "zh-yue-HK").locale.baseName, "yue-HK");
+	(0, node_assert.deepStrictEqual)(new LanguageString("你好", "zh-cmn-Hant").locale.baseName, new Intl.Locale("cmn-Hant").baseName);
+	(0, node_assert.throws)(() => new LanguageString("Hello", "en-USA"), RangeError);
+	(0, node_assert.throws)(() => new LanguageString("你好", "zh-yue-"), RangeError);
+});
 (0, node_test.test)("Deno.inspect(LanguageString)", () => {
 	const langStr = new LanguageString("Hello, 'world'", "en");
 	(0, node_assert.deepStrictEqual)(node_util.default.inspect(langStr, { colors: false }), "<en> \"Hello, 'world'\"");

package/dist/tests/langstr.test.mjs

@@ -1,5 +1,6 @@
+import { t as normalizeLanguageTag } from "./langtag-QZZgy903.mjs";
 import { test } from "node:test";
-import { deepStrictEqual } from "node:assert";
+import { deepStrictEqual, throws } from "node:assert";
 import util from "node:util";
 //#region src/langstr.ts
 /**
@@ -15,11 +16,13 @@ var LanguageString = class extends String {
 	* Constructs a new `LanguageString`.
 	* @param value A string value written in the given language.
 	* @param language The language of the string.  If a string is given, it will
-	*                 be parsed as a `Intl.Locale` object.
+	*                 be parsed as a `Intl.Locale` object.  Extended language
+	*                 subtags are normalized first, e.g., `zh-yue` is parsed as
+	*                 `yue`.
 	*/
 	constructor(value, language) {
 		super(value);
-		this.locale = typeof language === "string" ? new Intl.Locale(language) : language;
+		this.locale = typeof language === "string" ? new Intl.Locale(normalizeLanguageTag(language)) : language;
 	}
 };
 LanguageString.prototype[Symbol.for("Deno.customInspect")] = function(inspect, options) {
@@ -36,6 +39,13 @@ test("new LanguageString()", () => {
 	deepStrictEqual(langStr.locale, new Intl.Locale("en"));
 	deepStrictEqual(new LanguageString("Hello", new Intl.Locale("en")), langStr);
 });
+test("new LanguageString() with extlang tags", () => {
+	deepStrictEqual(new LanguageString("你好", "zh-YUE").locale.baseName, "yue");
+	deepStrictEqual(new LanguageString("你好", "zh-yue-HK").locale.baseName, "yue-HK");
+	deepStrictEqual(new LanguageString("你好", "zh-cmn-Hant").locale.baseName, new Intl.Locale("cmn-Hant").baseName);
+	throws(() => new LanguageString("Hello", "en-USA"), RangeError);
+	throws(() => new LanguageString("你好", "zh-yue-"), RangeError);
+});
 test("Deno.inspect(LanguageString)", () => {
 	const langStr = new LanguageString("Hello, 'world'", "en");
 	deepStrictEqual(util.inspect(langStr, { colors: false }), "<en> \"Hello, 'world'\"");

package/dist/tests/langtag-Bm8oKhIO.cjs

@@ -0,0 +1,50 @@
+//#region src/langtag.ts
+const EXTLANG_PAIRS = new Set(Object.entries({
+	ar: "aao abh abv acm acq acw acx acy adf aeb aec afb ajp apc apd arb arq ars ary arz auz avl ayh ayl ayn ayp bbz pga shu ssh",
+	kok: "gom knn",
+	lv: "ltg lvs",
+	ms: "bjn btj bve bvu coa dup hji jak jax kvb kvr kxd lce lcf liw max meo mfa mfb min mqg msi mui orn ors pel pse tmw urk vkk vkt xmm zlm zmi zsm",
+	sgn: "ads aed aen afg ajs ase asf asp asq asw bfi bfk bog bqn bqy bvl bzs cds csc csd cse csf csg csl csn csq csr csx doq dse dsl dsz dyl ecs ehs esl esn eso eth fcs fse fsl fss gds gse gsg gsm gss gus hab haf hds hks hos hps hsh hsl icl iks ils inl ins ise isg isr jcs jhs jks jls jos jsl jus kgi kvk lbs lgs lls lsb lsc lsg lsl lsn lso lsp lst lsv lsw lsy lws mdl mfs mre msd msr mzc mzg mzy nbs ncs nsi nsl nsp nsr nzs okl pgz pks prl prz psc psd psg psl pso psp psr pys rib rms rnb rsi rsl rsm rsn sdl sfb sfs sgg sgx slf sls sqk sqs sqx ssp ssr svk swl syy szs tse tsm tsq tss tsy tza ugn ugy ukl uks vgt vsi vsl vsv wbs xki xml xms yds ygs yhs ysl ysm zhk zib zsl",
+	sw: "swc swh",
+	uz: "uzn uzs",
+	zh: "cdo cjy cmn cnp cpx csp czh czo gan hak hnm hsn luh lzh mnp nan sjc wuu yue"
+}).flatMap(([prefix, extlangs]) => extlangs.split(" ").map((extlang) => `${prefix}-${extlang}`)));
+const PREFIX_PATTERN = /^[A-Za-z]{2,3}$/;
+const EXTLANG_PATTERN = /^[A-Za-z]{3}$/;
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+function normalizeLanguageTag(tag) {
+	const subtags = tag.split("-");
+	if (subtags.length < 2) return tag;
+	const [prefix, extlang, ...rest] = subtags;
+	if (!PREFIX_PATTERN.test(prefix) || !EXTLANG_PATTERN.test(extlang)) return tag;
+	const language = extlang.toLowerCase();
+	if (!EXTLANG_PAIRS.has(`${prefix.toLowerCase()}-${language}`)) return tag;
+	return [language, ...rest].join("-");
+}
+//#endregion
+Object.defineProperty(exports, "normalizeLanguageTag", {
+	enumerable: true,
+	get: function() {
+		return normalizeLanguageTag;
+	}
+});

package/dist/tests/langtag-QZZgy903.mjs

@@ -0,0 +1,45 @@
+//#region src/langtag.ts
+const EXTLANG_PAIRS = new Set(Object.entries({
+	ar: "aao abh abv acm acq acw acx acy adf aeb aec afb ajp apc apd arb arq ars ary arz auz avl ayh ayl ayn ayp bbz pga shu ssh",
+	kok: "gom knn",
+	lv: "ltg lvs",
+	ms: "bjn btj bve bvu coa dup hji jak jax kvb kvr kxd lce lcf liw max meo mfa mfb min mqg msi mui orn ors pel pse tmw urk vkk vkt xmm zlm zmi zsm",
+	sgn: "ads aed aen afg ajs ase asf asp asq asw bfi bfk bog bqn bqy bvl bzs cds csc csd cse csf csg csl csn csq csr csx doq dse dsl dsz dyl ecs ehs esl esn eso eth fcs fse fsl fss gds gse gsg gsm gss gus hab haf hds hks hos hps hsh hsl icl iks ils inl ins ise isg isr jcs jhs jks jls jos jsl jus kgi kvk lbs lgs lls lsb lsc lsg lsl lsn lso lsp lst lsv lsw lsy lws mdl mfs mre msd msr mzc mzg mzy nbs ncs nsi nsl nsp nsr nzs okl pgz pks prl prz psc psd psg psl pso psp psr pys rib rms rnb rsi rsl rsm rsn sdl sfb sfs sgg sgx slf sls sqk sqs sqx ssp ssr svk swl syy szs tse tsm tsq tss tsy tza ugn ugy ukl uks vgt vsi vsl vsv wbs xki xml xms yds ygs yhs ysl ysm zhk zib zsl",
+	sw: "swc swh",
+	uz: "uzn uzs",
+	zh: "cdo cjy cmn cnp cpx csp czh czo gan hak hnm hsn luh lzh mnp nan sjc wuu yue"
+}).flatMap(([prefix, extlangs]) => extlangs.split(" ").map((extlang) => `${prefix}-${extlang}`)));
+const PREFIX_PATTERN = /^[A-Za-z]{2,3}$/;
+const EXTLANG_PATTERN = /^[A-Za-z]{3}$/;
+/**
+* Normalizes a BCP 47 language tag that has an extended language subtag
+* (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+* described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+* tags, so tags should be normalized before being passed to it.
+*
+* Only registered extlangs that follow their registered prefix are removed;
+* any other tag is returned unchanged, so the result still needs to be
+* validated, e.g., by `Intl.Locale`.
+*
+* NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+* releases.  When merging it into the *main* branch, remove the `@internal`
+* tag, formalize it as a public API (`@since 2.5.0`), and add it to
+* the changelog fragment.
+*
+* [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+*
+* @param tag The BCP 47 language tag to normalize.
+* @returns The normalized language tag.
+* @internal
+*/
+function normalizeLanguageTag(tag) {
+	const subtags = tag.split("-");
+	if (subtags.length < 2) return tag;
+	const [prefix, extlang, ...rest] = subtags;
+	if (!PREFIX_PATTERN.test(prefix) || !EXTLANG_PATTERN.test(extlang)) return tag;
+	const language = extlang.toLowerCase();
+	if (!EXTLANG_PAIRS.has(`${prefix.toLowerCase()}-${language}`)) return tag;
+	return [language, ...rest].join("-");
+}
+//#endregion
+export { normalizeLanguageTag as t };

package/dist/tests/langtag.test.cjs

@@ -0,0 +1,46 @@
+require("./chunk-Do9eywBl.cjs");
+const require_langtag = require("./langtag-Bm8oKhIO.cjs");
+let node_test = require("node:test");
+let node_assert = require("node:assert");
+//#region src/langtag.test.ts
+(0, node_test.test)("normalizeLanguageTag() removes extlang subtags", () => {
+	for (const [input, expected] of [
+		["zh-yue", "yue"],
+		["zh-YUE", "yue"],
+		["ZH-Yue-hk", "yue-hk"],
+		["zh-yue-HK", "yue-HK"],
+		["zh-cmn-Hant", "cmn-Hant"],
+		["zh-yue-u-nu-hanidec", "yue-u-nu-hanidec"],
+		["sgn-ase", "ase"],
+		["ar-arz", "arz"],
+		["ms-zlm", "zlm"],
+		["kok-gom", "gom"],
+		["koK-gom", "gom"],
+		["lv-ltg", "ltg"],
+		["sw-swc", "swc"],
+		["uz-uzs", "uzs"],
+		["zh-yue-cmn", "yue-cmn"],
+		["zh-yue-", "yue-"]
+	]) (0, node_assert.deepStrictEqual)(require_langtag.normalizeLanguageTag(input), expected, input);
+});
+(0, node_test.test)("normalizeLanguageTag() leaves other tags unchanged", () => {
+	for (const tag of [
+		"",
+		"en",
+		"en-US",
+		"en-USA",
+		"en-yue",
+		"zh",
+		"zh-Hant-TW",
+		"yue",
+		"yue-HK",
+		"x-private",
+		"i-klingon",
+		"zh-min-nan",
+		"koK-gom",
+		"zh-yuK",
+		"constructor-yue",
+		"toString-yue"
+	]) (0, node_assert.deepStrictEqual)(require_langtag.normalizeLanguageTag(tag), tag, tag);
+});
+//#endregion

package/dist/tests/langtag.test.d.cts

@@ -0,0 +1 @@
+export { };

package/dist/tests/langtag.test.d.mts

@@ -0,0 +1 @@
+export { };

package/dist/tests/langtag.test.mjs

@@ -0,0 +1,46 @@
+import { t as normalizeLanguageTag } from "./langtag-QZZgy903.mjs";
+import { test } from "node:test";
+import { deepStrictEqual } from "node:assert";
+//#region src/langtag.test.ts
+test("normalizeLanguageTag() removes extlang subtags", () => {
+	for (const [input, expected] of [
+		["zh-yue", "yue"],
+		["zh-YUE", "yue"],
+		["ZH-Yue-hk", "yue-hk"],
+		["zh-yue-HK", "yue-HK"],
+		["zh-cmn-Hant", "cmn-Hant"],
+		["zh-yue-u-nu-hanidec", "yue-u-nu-hanidec"],
+		["sgn-ase", "ase"],
+		["ar-arz", "arz"],
+		["ms-zlm", "zlm"],
+		["kok-gom", "gom"],
+		["koK-gom", "gom"],
+		["lv-ltg", "ltg"],
+		["sw-swc", "swc"],
+		["uz-uzs", "uzs"],
+		["zh-yue-cmn", "yue-cmn"],
+		["zh-yue-", "yue-"]
+	]) deepStrictEqual(normalizeLanguageTag(input), expected, input);
+});
+test("normalizeLanguageTag() leaves other tags unchanged", () => {
+	for (const tag of [
+		"",
+		"en",
+		"en-US",
+		"en-USA",
+		"en-yue",
+		"zh",
+		"zh-Hant-TW",
+		"yue",
+		"yue-HK",
+		"x-private",
+		"i-klingon",
+		"zh-min-nan",
+		"koK-gom",
+		"zh-yuK",
+		"constructor-yue",
+		"toString-yue"
+	]) deepStrictEqual(normalizeLanguageTag(tag), tag, tag);
+});
+//#endregion
+export {};

package/dist/tests/{request-JVGlYdPG.mjs → request-BV7yaWdl.mjs}

@@ -1,7 +1,7 @@
 import process from "node:process";
 //#region deno.json
 var name = "@fedify/vocab-runtime";
-var version = "2.0.31-dev.2306+e775cf8c";
+var version = "2.0.31-dev.2308+34090b3a";
 //#endregion
 //#region src/request.ts
 /**

package/dist/tests/{request-75RoXeN4.cjs → request-Ban3HXuI.cjs}

@@ -3,7 +3,7 @@ let node_process = require("node:process");
 node_process = require_chunk.__toESM(node_process);
 //#region deno.json
 var name = "@fedify/vocab-runtime";
-var version = "2.0.31-dev.2306+e775cf8c";
+var version = "2.0.31-dev.2308+34090b3a";
 //#endregion
 //#region src/request.ts
 /**

package/dist/tests/request.test.cjs

@@ -1,5 +1,5 @@
 const require_chunk = require("./chunk-Do9eywBl.cjs");
-const require_request = require("./request-75RoXeN4.cjs");
+const require_request = require("./request-Ban3HXuI.cjs");
 let node_test = require("node:test");
 let node_process = require("node:process");
 node_process = require_chunk.__toESM(node_process);

package/dist/tests/request.test.mjs

@@ -1,4 +1,4 @@
-import { o as version, r as getUserAgent } from "./request-JVGlYdPG.mjs";
+import { o as version, r as getUserAgent } from "./request-BV7yaWdl.mjs";
 import { test } from "node:test";
 import process from "node:process";
 import { deepStrictEqual } from "node:assert";

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@fedify/vocab-runtime",
-  "version": "2.0.31-dev.2306+e775cf8c",
+  "version": "2.0.31-dev.2308+34090b3a",
   "homepage": "https://fedify.dev/",
   "repository": {
     "type": "git",

package/src/langstr.test.ts

@@ -1,4 +1,4 @@
-import { deepStrictEqual } from "node:assert";
+import { deepStrictEqual, throws } from "node:assert";
 import { test } from "node:test";
 import util from "node:util";
 import { LanguageString } from "./langstr.ts";
@@ -11,6 +11,23 @@ test("new LanguageString()", () => {
   deepStrictEqual(new LanguageString("Hello", new Intl.Locale("en")), langStr);
 });
 
+test("new LanguageString() with extlang tags", () => {
+  // Intl.Locale objects have no own enumerable properties, so their base names
+  // are compared instead:
+  deepStrictEqual(new LanguageString("你好", "zh-YUE").locale.baseName, "yue");
+  deepStrictEqual(
+    new LanguageString("你好", "zh-yue-HK").locale.baseName,
+    "yue-HK",
+  );
+  // Whether cmn is canonicalized to zh depends on the runtime:
+  deepStrictEqual(
+    new LanguageString("你好", "zh-cmn-Hant").locale.baseName,
+    new Intl.Locale("cmn-Hant").baseName,
+  );
+  throws(() => new LanguageString("Hello", "en-USA"), RangeError);
+  throws(() => new LanguageString("你好", "zh-yue-"), RangeError);
+});
+
 test("Deno.inspect(LanguageString)", () => {
   const langStr = new LanguageString("Hello, 'world'", "en");
   deepStrictEqual(

package/src/langstr.ts

@@ -1,3 +1,5 @@
+import { normalizeLanguageTag } from "./langtag.ts";
+
 /**
  * A language-tagged string which corresponds to the `rdf:langString` type.
  */
@@ -12,12 +14,14 @@ export class LanguageString extends String {
    * Constructs a new `LanguageString`.
    * @param value A string value written in the given language.
    * @param language The language of the string.  If a string is given, it will
-   *                 be parsed as a `Intl.Locale` object.
+   *                 be parsed as a `Intl.Locale` object.  Extended language
+   *                 subtags are normalized first, e.g., `zh-yue` is parsed as
+   *                 `yue`.
    */
   constructor(value: string, language: Intl.Locale | string) {
     super(value);
     this.locale = typeof language === "string"
-      ? new Intl.Locale(language)
+      ? new Intl.Locale(normalizeLanguageTag(language))
       : language;
   }
 }

package/src/langtag.test.ts

@@ -0,0 +1,53 @@
+import { deepStrictEqual } from "node:assert";
+import { test } from "node:test";
+import { normalizeLanguageTag } from "./langtag.ts";
+
+test("normalizeLanguageTag() removes extlang subtags", () => {
+  const cases: [string, string][] = [
+    ["zh-yue", "yue"],
+    ["zh-YUE", "yue"],
+    ["ZH-Yue-hk", "yue-hk"],
+    ["zh-yue-HK", "yue-HK"],
+    ["zh-cmn-Hant", "cmn-Hant"],
+    ["zh-yue-u-nu-hanidec", "yue-u-nu-hanidec"],
+    ["sgn-ase", "ase"],
+    ["ar-arz", "arz"],
+    ["ms-zlm", "zlm"],
+    ["kok-gom", "gom"],
+    ["koK-gom", "gom"],
+    ["lv-ltg", "ltg"],
+    ["sw-swc", "swc"],
+    ["uz-uzs", "uzs"],
+    // Valid tags never have more than one extlang, so only the first one is
+    // removed and the result is still rejected by Intl.Locale:
+    ["zh-yue-cmn", "yue-cmn"],
+    ["zh-yue-", "yue-"],
+  ];
+  for (const [input, expected] of cases) {
+    deepStrictEqual(normalizeLanguageTag(input), expected, input);
+  }
+});
+
+test("normalizeLanguageTag() leaves other tags unchanged", () => {
+  const tags = [
+    "",
+    "en",
+    "en-US",
+    "en-USA",
+    "en-yue",
+    "zh",
+    "zh-Hant-TW",
+    "yue",
+    "yue-HK",
+    "x-private",
+    "i-klingon",
+    "zh-min-nan",
+    "koK-gom",
+    "zh-yuK",
+    "constructor-yue",
+    "toString-yue",
+  ];
+  for (const tag of tags) {
+    deepStrictEqual(normalizeLanguageTag(tag), tag, tag);
+  }
+});

package/src/langtag.ts

@@ -0,0 +1,76 @@
+/**
+ * Extended language subtags (extlangs) registered in the IANA Language Subtag
+ * Registry, grouped by their prefixes.  Every extlang record has exactly one
+ * prefix and a `Preferred-Value` equal to the extlang subtag itself, so
+ * replacing the prefix and the extlang with the extlang alone yields
+ * the canonical form.
+ *
+ * Generated from the registry with File-Date: 2026-09-17.
+ */
+const EXTLANGS: Record<string, string> = {
+  ar: "aao abh abv acm acq acw acx acy adf aeb aec afb ajp apc apd arb arq " +
+    "ars ary arz auz avl ayh ayl ayn ayp bbz pga shu ssh",
+  kok: "gom knn",
+  lv: "ltg lvs",
+  ms: "bjn btj bve bvu coa dup hji jak jax kvb kvr kxd lce lcf liw max meo " +
+    "mfa mfb min mqg msi mui orn ors pel pse tmw urk vkk vkt xmm zlm zmi " +
+    "zsm",
+  sgn: "ads aed aen afg ajs ase asf asp asq asw bfi bfk bog bqn bqy bvl bzs " +
+    "cds csc csd cse csf csg csl csn csq csr csx doq dse dsl dsz dyl ecs " +
+    "ehs esl esn eso eth fcs fse fsl fss gds gse gsg gsm gss gus hab haf " +
+    "hds hks hos hps hsh hsl icl iks ils inl ins ise isg isr jcs jhs jks " +
+    "jls jos jsl jus kgi kvk lbs lgs lls lsb lsc lsg lsl lsn lso lsp lst " +
+    "lsv lsw lsy lws mdl mfs mre msd msr mzc mzg mzy nbs ncs nsi nsl nsp " +
+    "nsr nzs okl pgz pks prl prz psc psd psg psl pso psp psr pys rib rms " +
+    "rnb rsi rsl rsm rsn sdl sfb sfs sgg sgx slf sls sqk sqs sqx ssp ssr " +
+    "svk swl syy szs tse tsm tsq tss tsy tza ugn ugy ukl uks vgt vsi vsl " +
+    "vsv wbs xki xml xms yds ygs yhs ysl ysm zhk zib zsl",
+  sw: "swc swh",
+  uz: "uzn uzs",
+  zh: "cdo cjy cmn cnp cpx csp czh czo gan hak hnm hsn luh lzh mnp nan sjc " +
+    "wuu yue",
+};
+
+const EXTLANG_PAIRS: ReadonlySet<string> = new Set(
+  Object.entries(EXTLANGS).flatMap(([prefix, extlangs]) =>
+    extlangs.split(" ").map((extlang) => `${prefix}-${extlang}`)
+  ),
+);
+
+const PREFIX_PATTERN = /^[A-Za-z]{2,3}$/;
+const EXTLANG_PATTERN = /^[A-Za-z]{3}$/;
+
+/**
+ * Normalizes a BCP 47 language tag that has an extended language subtag
+ * (extlang), such as `zh-yue`, into its canonical form, such as `yue`, as
+ * described in [RFC 5646, Section 4.5][1].  `Intl.Locale` rejects extlang
+ * tags, so tags should be normalized before being passed to it.
+ *
+ * Only registered extlangs that follow their registered prefix are removed;
+ * any other tag is returned unchanged, so the result still needs to be
+ * validated, e.g., by `Intl.Locale`.
+ *
+ * NOTE: This function is marked as internal in the 2.0.x–2.4.x patch
+ * releases.  When merging it into the *main* branch, remove the `@internal`
+ * tag, formalize it as a public API (`@since 2.5.0`), and add it to
+ * the changelog fragment.
+ *
+ * [1]: https://www.rfc-editor.org/rfc/rfc5646.html#section-4.5
+ *
+ * @param tag The BCP 47 language tag to normalize.
+ * @returns The normalized language tag.
+ * @internal
+ */
+export function normalizeLanguageTag(tag: string): string {
+  const subtags = tag.split("-");
+  if (subtags.length < 2) return tag;
+  const [prefix, extlang, ...rest] = subtags;
+  // Checked before case folding, since `toLowerCase()` maps some non-ASCII
+  // characters (e.g., U+212A KELVIN SIGN) to ASCII letters:
+  if (!PREFIX_PATTERN.test(prefix) || !EXTLANG_PATTERN.test(extlang)) {
+    return tag;
+  }
+  const language = extlang.toLowerCase();
+  if (!EXTLANG_PAIRS.has(`${prefix.toLowerCase()}-${language}`)) return tag;
+  return [language, ...rest].join("-");
+}

package/src/mod.ts

@@ -25,6 +25,7 @@ export {
   importSpki,
 } from "./key.ts";
 export { LanguageString } from "./langstr.ts";
+export { normalizeLanguageTag } from "./langtag.ts";
 export {
   decodeMultibase,
   encodeMultibase,