@intlayer/engine 9.5.9 → 9.5.11

package/dist/esm/scan/detection/localeCode.mjs

@@ -0,0 +1,108 @@
+//#region src/scan/detection/localeCode.ts
+/**
+* Locale-code helpers shared by the scan checks and the routing detection.
+*
+* Dependency-free and browser-safe: consumed by the CLI, the backend audit and
+* the Chrome extension popup.
+*/
+/** Syntactic shape of a locale code: `en`, `en-US`, `pt_BR`, `zh-Hant-TW`… */
+const LOCALE_CODE_PATTERN = /^[a-z]{2,3}([_-][a-z]{4})?([_-]([a-z]{2}|[0-9]{3}))?$/i;
+/** Languages written right-to-left. */
+const RIGHT_TO_LEFT_LANGUAGES = /* @__PURE__ */ new Set([
+	"ar",
+	"arc",
+	"ckb",
+	"dv",
+	"fa",
+	"he",
+	"iw",
+	"ku",
+	"ps",
+	"sd",
+	"syr",
+	"ug",
+	"ur",
+	"yi"
+]);
+/** Lower-case a locale code and use `-` as separator (`pt_BR` → `pt-br`). */
+const normalizeLocaleCode = (localeCode) => localeCode.trim().replace(/_/g, "-").toLowerCase();
+/** Primary language subtag of a locale code (`en-US` → `en`). */
+const getLanguageCode = (localeCode) => normalizeLocaleCode(localeCode).split("-")[0] ?? "";
+/** Whether two locale codes designate the same locale, ignoring case/separator. */
+const isSameLocale = (firstLocale, secondLocale) => Boolean(firstLocale) && Boolean(secondLocale) && normalizeLocaleCode(firstLocale) === normalizeLocaleCode(secondLocale);
+/** Whether two locale codes share the same language (`fr` vs `fr-CA`). */
+const isSameLanguage = (firstLocale, secondLocale) => Boolean(firstLocale && secondLocale) && getLanguageCode(firstLocale) === getLanguageCode(secondLocale);
+/** Whether a string looks like a locale code (used on URL segments). */
+const looksLikeLocaleCode = (value) => LOCALE_CODE_PATTERN.test(value);
+/**
+* Whether a locale code is a valid BCP 47 tag (`en_US` and `english` are not).
+* `x-default` is accepted since it is a valid hreflang value.
+*/
+const isValidLocaleCode = (localeCode) => {
+	if (localeCode.toLowerCase() === "x-default") return true;
+	if (!LOCALE_CODE_PATTERN.test(localeCode) || localeCode.includes("_")) return false;
+	try {
+		return Intl.getCanonicalLocales(localeCode).length === 1;
+	} catch {
+		return false;
+	}
+};
+/** Whether the locale is written right-to-left (`ar`, `he-IL`, `fa`…). */
+const isRightToLeftLocale = (localeCode) => RIGHT_TO_LEFT_LANGUAGES.has(getLanguageCode(localeCode));
+/**
+* Find, among `candidateLocales`, the one matching `value` (a URL segment, a
+* subdomain label or a query-string value). Falls back to a language-only match
+* when no exact match exists (`/fr/` for hreflang `fr-FR`).
+*/
+const findMatchingLocale = (value, candidateLocales) => {
+	if (!value) return void 0;
+	const normalizedValue = normalizeLocaleCode(value);
+	const exactMatch = candidateLocales.find((locale) => normalizeLocaleCode(locale) === normalizedValue);
+	if (exactMatch) return exactMatch;
+	const languageMatches = candidateLocales.filter((locale) => getLanguageCode(locale) === normalizedValue);
+	return languageMatches.length === 1 ? languageMatches[0] : void 0;
+};
+/** Open Graph locale shape: `language_TERRITORY` (`en_GB`, `pt_BR`). */
+const OPEN_GRAPH_LOCALE_PATTERN = /^[a-z]{2,3}_[A-Z]{2}$/;
+/**
+* Whether a value is a valid `og:locale` (`en_GB`). Open Graph does not accept
+* a bare language (`en`) nor BCP 47 separators (`en-GB`): crawlers then fall
+* back to `en_US`.
+*/
+const isValidOpenGraphLocale = (localeCode) => OPEN_GRAPH_LOCALE_PATTERN.test(localeCode.trim());
+/** Region subtag of a locale code, if any (`en-GB` → `GB`, `zh-Hant-TW` → `TW`). */
+const getRegionCode = (localeCode) => {
+	try {
+		return new Intl.Locale(normalizeLocaleCode(localeCode)).region;
+	} catch {
+		return;
+	}
+};
+/** Most likely region of a language (`en` → `US`, `fr` → `FR`, `ja` → `JP`). */
+const getLikelyRegionCode = (languageCode) => {
+	try {
+		return new Intl.Locale(languageCode).maximize().region;
+	} catch {
+		return;
+	}
+};
+/**
+* Suggest the `og:locale` value for a locale code. The region comes from the
+* locale itself, then from `regionHint` (typically `<html lang>`) when it
+* shares the language, then from the language's most likely region.
+*
+* @example
+* toOpenGraphLocale('en', 'en-GB'); // 'en_GB'
+* toOpenGraphLocale('pt-br'); // 'pt_BR'
+* toOpenGraphLocale('fr'); // 'fr_FR'
+*/
+const toOpenGraphLocale = (localeCode, regionHint) => {
+	const languageCode = getLanguageCode(localeCode);
+	if (!/^[a-z]{2,3}$/.test(languageCode)) return void 0;
+	const regionCode = getRegionCode(localeCode) ?? (regionHint && isSameLanguage(localeCode, regionHint) ? getRegionCode(regionHint) : void 0) ?? getLikelyRegionCode(languageCode);
+	return regionCode && /^[A-Z]{2}$/.test(regionCode) ? `${languageCode}_${regionCode}` : void 0;
+};
+
+//#endregion
+export { findMatchingLocale, getLanguageCode, isRightToLeftLocale, isSameLanguage, isSameLocale, isValidLocaleCode, isValidOpenGraphLocale, looksLikeLocaleCode, normalizeLocaleCode, toOpenGraphLocale };
+//# sourceMappingURL=localeCode.mjs.map

package/dist/esm/scan/detection/localeCode.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"localeCode.mjs","names":[],"sources":["../../../../src/scan/detection/localeCode.ts"],"sourcesContent":["/**\n * Locale-code helpers shared by the scan checks and the routing detection.\n *\n * Dependency-free and browser-safe: consumed by the CLI, the backend audit and\n * the Chrome extension popup.\n */\n\n/** Syntactic shape of a locale code: `en`, `en-US`, `pt_BR`, `zh-Hant-TW`… */\nconst LOCALE_CODE_PATTERN =\n  /^[a-z]{2,3}([_-][a-z]{4})?([_-]([a-z]{2}|[0-9]{3}))?$/i;\n\n/** Languages written right-to-left. */\nconst RIGHT_TO_LEFT_LANGUAGES = new Set([\n  'ar',\n  'arc',\n  'ckb',\n  'dv',\n  'fa',\n  'he',\n  'iw',\n  'ku',\n  'ps',\n  'sd',\n  'syr',\n  'ug',\n  'ur',\n  'yi',\n]);\n\n/** Lower-case a locale code and use `-` as separator (`pt_BR` → `pt-br`). */\nexport const normalizeLocaleCode = (localeCode: string): string =>\n  localeCode.trim().replace(/_/g, '-').toLowerCase();\n\n/** Primary language subtag of a locale code (`en-US` → `en`). */\nexport const getLanguageCode = (localeCode: string): string =>\n  normalizeLocaleCode(localeCode).split('-')[0] ?? '';\n\n/** Whether two locale codes designate the same locale, ignoring case/separator. */\nexport const isSameLocale = (\n  firstLocale: string | undefined,\n  secondLocale: string | undefined\n): boolean =>\n  Boolean(firstLocale) &&\n  Boolean(secondLocale) &&\n  normalizeLocaleCode(firstLocale as string) ===\n    normalizeLocaleCode(secondLocale as string);\n\n/** Whether two locale codes share the same language (`fr` vs `fr-CA`). */\nexport const isSameLanguage = (\n  firstLocale: string | undefined,\n  secondLocale: string | undefined\n): boolean =>\n  Boolean(firstLocale && secondLocale) &&\n  getLanguageCode(firstLocale as string) ===\n    getLanguageCode(secondLocale as string);\n\n/** Whether a string looks like a locale code (used on URL segments). */\nexport const looksLikeLocaleCode = (value: string): boolean =>\n  LOCALE_CODE_PATTERN.test(value);\n\n/**\n * Whether a locale code is a valid BCP 47 tag (`en_US` and `english` are not).\n * `x-default` is accepted since it is a valid hreflang value.\n */\nexport const isValidLocaleCode = (localeCode: string): boolean => {\n  if (localeCode.toLowerCase() === 'x-default') return true;\n  if (!LOCALE_CODE_PATTERN.test(localeCode) || localeCode.includes('_')) {\n    return false;\n  }\n  try {\n    return Intl.getCanonicalLocales(localeCode).length === 1;\n  } catch {\n    return false;\n  }\n};\n\n/** Whether the locale is written right-to-left (`ar`, `he-IL`, `fa`…). */\nexport const isRightToLeftLocale = (localeCode: string): boolean =>\n  RIGHT_TO_LEFT_LANGUAGES.has(getLanguageCode(localeCode));\n\n/**\n * Find, among `candidateLocales`, the one matching `value` (a URL segment, a\n * subdomain label or a query-string value). Falls back to a language-only match\n * when no exact match exists (`/fr/` for hreflang `fr-FR`).\n */\nexport const findMatchingLocale = (\n  value: string | undefined | null,\n  candidateLocales: readonly string[]\n): string | undefined => {\n  if (!value) return undefined;\n  const normalizedValue = normalizeLocaleCode(value);\n\n  const exactMatch = candidateLocales.find(\n    (locale) => normalizeLocaleCode(locale) === normalizedValue\n  );\n  if (exactMatch) return exactMatch;\n\n  const languageMatches = candidateLocales.filter(\n    (locale) => getLanguageCode(locale) === normalizedValue\n  );\n  return languageMatches.length === 1 ? languageMatches[0] : undefined;\n};\n\n/** Open Graph locale shape: `language_TERRITORY` (`en_GB`, `pt_BR`). */\nconst OPEN_GRAPH_LOCALE_PATTERN = /^[a-z]{2,3}_[A-Z]{2}$/;\n\n/**\n * Whether a value is a valid `og:locale` (`en_GB`). Open Graph does not accept\n * a bare language (`en`) nor BCP 47 separators (`en-GB`): crawlers then fall\n * back to `en_US`.\n */\nexport const isValidOpenGraphLocale = (localeCode: string): boolean =>\n  OPEN_GRAPH_LOCALE_PATTERN.test(localeCode.trim());\n\n/** Region subtag of a locale code, if any (`en-GB` → `GB`, `zh-Hant-TW` → `TW`). */\nconst getRegionCode = (localeCode: string): string | undefined => {\n  try {\n    return new Intl.Locale(normalizeLocaleCode(localeCode)).region;\n  } catch {\n    return undefined;\n  }\n};\n\n/** Most likely region of a language (`en` → `US`, `fr` → `FR`, `ja` → `JP`). */\nconst getLikelyRegionCode = (languageCode: string): string | undefined => {\n  try {\n    return new Intl.Locale(languageCode).maximize().region;\n  } catch {\n    return undefined;\n  }\n};\n\n/**\n * Suggest the `og:locale` value for a locale code. The region comes from the\n * locale itself, then from `regionHint` (typically `<html lang>`) when it\n * shares the language, then from the language's most likely region.\n *\n * @example\n * toOpenGraphLocale('en', 'en-GB'); // 'en_GB'\n * toOpenGraphLocale('pt-br'); // 'pt_BR'\n * toOpenGraphLocale('fr'); // 'fr_FR'\n */\nexport const toOpenGraphLocale = (\n  localeCode: string,\n  regionHint?: string\n): string | undefined => {\n  const languageCode = getLanguageCode(localeCode);\n  if (!/^[a-z]{2,3}$/.test(languageCode)) return undefined;\n\n  const regionCode =\n    getRegionCode(localeCode) ??\n    (regionHint && isSameLanguage(localeCode, regionHint)\n      ? getRegionCode(regionHint)\n      : undefined) ??\n    getLikelyRegionCode(languageCode);\n\n  return regionCode && /^[A-Z]{2}$/.test(regionCode)\n    ? `${languageCode}_${regionCode}`\n    : undefined;\n};\n"],"mappings":";;;;;;;;AAQA,MAAM,sBACJ;;AAGF,MAAM,0CAA0B,IAAI,IAAI;CACtC;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;CACA;AACF,CAAC;;AAGD,MAAa,uBAAuB,eAClC,WAAW,KAAK,CAAC,CAAC,QAAQ,MAAM,GAAG,CAAC,CAAC,YAAY;;AAGnD,MAAa,mBAAmB,eAC9B,oBAAoB,UAAU,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,MAAM;;AAGnD,MAAa,gBACX,aACA,iBAEA,QAAQ,WAAW,KACnB,QAAQ,YAAY,KACpB,oBAAoB,WAAqB,MACvC,oBAAoB,YAAsB;;AAG9C,MAAa,kBACX,aACA,iBAEA,QAAQ,eAAe,YAAY,KACnC,gBAAgB,WAAqB,MACnC,gBAAgB,YAAsB;;AAG1C,MAAa,uBAAuB,UAClC,oBAAoB,KAAK,KAAK;;;;;AAMhC,MAAa,qBAAqB,eAAgC;CAChE,IAAI,WAAW,YAAY,MAAM,aAAa,OAAO;CACrD,IAAI,CAAC,oBAAoB,KAAK,UAAU,KAAK,WAAW,SAAS,GAAG,GAClE,OAAO;CAET,IAAI;EACF,OAAO,KAAK,oBAAoB,UAAU,CAAC,CAAC,WAAW;CACzD,QAAQ;EACN,OAAO;CACT;AACF;;AAGA,MAAa,uBAAuB,eAClC,wBAAwB,IAAI,gBAAgB,UAAU,CAAC;;;;;;AAOzD,MAAa,sBACX,OACA,qBACuB;CACvB,IAAI,CAAC,OAAO,OAAO;CACnB,MAAM,kBAAkB,oBAAoB,KAAK;CAEjD,MAAM,aAAa,iBAAiB,MACjC,WAAW,oBAAoB,MAAM,MAAM,eAC9C;CACA,IAAI,YAAY,OAAO;CAEvB,MAAM,kBAAkB,iBAAiB,QACtC,WAAW,gBAAgB,MAAM,MAAM,eAC1C;CACA,OAAO,gBAAgB,WAAW,IAAI,gBAAgB,KAAK;AAC7D;;AAGA,MAAM,4BAA4B;;;;;;AAOlC,MAAa,0BAA0B,eACrC,0BAA0B,KAAK,WAAW,KAAK,CAAC;;AAGlD,MAAM,iBAAiB,eAA2C;CAChE,IAAI;EACF,OAAO,IAAI,KAAK,OAAO,oBAAoB,UAAU,CAAC,CAAC,CAAC;CAC1D,QAAQ;EACN;CACF;AACF;;AAGA,MAAM,uBAAuB,iBAA6C;CACxE,IAAI;EACF,OAAO,IAAI,KAAK,OAAO,YAAY,CAAC,CAAC,SAAS,CAAC,CAAC;CAClD,QAAQ;EACN;CACF;AACF;;;;;;;;;;;AAYA,MAAa,qBACX,YACA,eACuB;CACvB,MAAM,eAAe,gBAAgB,UAAU;CAC/C,IAAI,CAAC,eAAe,KAAK,YAAY,GAAG,OAAO;CAE/C,MAAM,aACJ,cAAc,UAAU,MACvB,cAAc,eAAe,YAAY,UAAU,IAChD,cAAc,UAAU,IACxB,WACJ,oBAAoB,YAAY;CAElC,OAAO,cAAc,aAAa,KAAK,UAAU,IAC7C,GAAG,aAAa,GAAG,eACnB;AACN"}

package/dist/esm/scan/detection/localizedPages.mjs

@@ -0,0 +1,46 @@
+import { normalizeUrl, parseUrl } from "./url.mjs";
+import { isSameLocale } from "./localeCode.mjs";
+
+//#region src/scan/detection/localizedPages.ts
+const X_DEFAULT = "x-default";
+/**
+* Turns the hreflang alternates of a page into a navigable list: absolute,
+* deduped by URL (a real locale wins over `x-default`), flagging the current
+* page, `x-default` last.
+*
+* @param hreflangs - `<link rel="alternate" hreflang>` entries of the page.
+* @param currentUrl - Absolute URL of the page, used to resolve relative hrefs.
+*/
+const getLocalizedPages = (hreflangs, currentUrl) => {
+	const normalizedCurrentUrl = normalizeUrl(currentUrl);
+	const pagesByUrl = /* @__PURE__ */ new Map();
+	for (const { hreflang, href } of hreflangs) {
+		const url = hreflang ? parseUrl(href, currentUrl)?.href : void 0;
+		if (!url) continue;
+		const existingPage = pagesByUrl.get(url);
+		if (existingPage && existingPage.hreflang !== X_DEFAULT) continue;
+		pagesByUrl.set(url, {
+			hreflang,
+			url,
+			isCurrent: normalizeUrl(url) === normalizedCurrentUrl
+		});
+	}
+	return Array.from(pagesByUrl.values()).sort((firstPage, secondPage) => Number(firstPage.hreflang === X_DEFAULT) - Number(secondPage.hreflang === X_DEFAULT));
+};
+/**
+* Whether the page is the base (unlocalized / default-locale entry) version of
+* a multilingual page: the `x-default` target, the root of a `prefix-all`
+* site, or the unmarked default-locale URL. i18n issues tend to hide on the
+* localized versions rather than on this one.
+*/
+const isBaseLocalePage = ({ pageUrl, hreflangs, routing }) => {
+	if (routing.locales.length < 2) return false;
+	const xDefaultHref = hreflangs.find(({ hreflang }) => hreflang.toLowerCase() === X_DEFAULT)?.href;
+	if (xDefaultHref && normalizeUrl(xDefaultHref, pageUrl) === normalizeUrl(pageUrl)) return true;
+	if (routing.strategy === "prefix-all") return !routing.urlLocale;
+	return (routing.strategy === "prefix-no-default" || routing.strategy === "search-params" || routing.strategy === "subdomain") && isSameLocale(routing.urlLocale, routing.defaultLocale);
+};
+
+//#endregion
+export { getLocalizedPages, isBaseLocalePage };
+//# sourceMappingURL=localizedPages.mjs.map

package/dist/esm/scan/detection/localizedPages.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"localizedPages.mjs","names":[],"sources":["../../../../src/scan/detection/localizedPages.ts"],"sourcesContent":["import type { RoutingDetection } from './detectRoutingStrategy';\nimport { isSameLocale } from './localeCode';\nimport { normalizeUrl, parseUrl } from './url';\n\n/** A localized version of a page, read from its hreflang tags. */\nexport type LocalizedPage = {\n  hreflang: string;\n  /** Absolute URL of the localized version. */\n  url: string;\n  /** True when this alternate points to the page currently inspected. */\n  isCurrent: boolean;\n};\n\nconst X_DEFAULT = 'x-default';\n\n/**\n * Turns the hreflang alternates of a page into a navigable list: absolute,\n * deduped by URL (a real locale wins over `x-default`), flagging the current\n * page, `x-default` last.\n *\n * @param hreflangs - `<link rel=\"alternate\" hreflang>` entries of the page.\n * @param currentUrl - Absolute URL of the page, used to resolve relative hrefs.\n */\nexport const getLocalizedPages = (\n  hreflangs: { hreflang: string; href: string }[],\n  currentUrl: string\n): LocalizedPage[] => {\n  const normalizedCurrentUrl = normalizeUrl(currentUrl);\n  const pagesByUrl = new Map<string, LocalizedPage>();\n\n  for (const { hreflang, href } of hreflangs) {\n    const url = hreflang ? parseUrl(href, currentUrl)?.href : undefined;\n    if (!url) continue;\n\n    const existingPage = pagesByUrl.get(url);\n\n    // Keep a real locale over `x-default` when both point to the same URL.\n    if (existingPage && existingPage.hreflang !== X_DEFAULT) continue;\n\n    pagesByUrl.set(url, {\n      hreflang,\n      url,\n      isCurrent: normalizeUrl(url) === normalizedCurrentUrl,\n    });\n  }\n\n  return Array.from(pagesByUrl.values()).sort(\n    (firstPage, secondPage) =>\n      Number(firstPage.hreflang === X_DEFAULT) -\n      Number(secondPage.hreflang === X_DEFAULT)\n  );\n};\n\n/** Input of {@link isBaseLocalePage}. */\nexport type BaseLocalePageInput = {\n  pageUrl: string;\n  hreflangs: { hreflang: string; href: string }[];\n  routing: Pick<\n    RoutingDetection,\n    'strategy' | 'locales' | 'urlLocale' | 'defaultLocale'\n  >;\n};\n\n/**\n * Whether the page is the base (unlocalized / default-locale entry) version of\n * a multilingual page: the `x-default` target, the root of a `prefix-all`\n * site, or the unmarked default-locale URL. i18n issues tend to hide on the\n * localized versions rather than on this one.\n */\nexport const isBaseLocalePage = ({\n  pageUrl,\n  hreflangs,\n  routing,\n}: BaseLocalePageInput): boolean => {\n  if (routing.locales.length < 2) return false;\n\n  const xDefaultHref = hreflangs.find(\n    ({ hreflang }) => hreflang.toLowerCase() === X_DEFAULT\n  )?.href;\n  if (\n    xDefaultHref &&\n    normalizeUrl(xDefaultHref, pageUrl) === normalizeUrl(pageUrl)\n  ) {\n    return true;\n  }\n\n  if (routing.strategy === 'prefix-all') return !routing.urlLocale;\n\n  const hasUnmarkedDefaultLocale =\n    routing.strategy === 'prefix-no-default' ||\n    routing.strategy === 'search-params' ||\n    routing.strategy === 'subdomain';\n\n  return (\n    hasUnmarkedDefaultLocale &&\n    isSameLocale(routing.urlLocale, routing.defaultLocale)\n  );\n};\n"],"mappings":";;;;AAaA,MAAM,YAAY;;;;;;;;;AAUlB,MAAa,qBACX,WACA,eACoB;CACpB,MAAM,uBAAuB,aAAa,UAAU;CACpD,MAAM,6BAAa,IAAI,IAA2B;CAElD,KAAK,MAAM,EAAE,UAAU,UAAU,WAAW;EAC1C,MAAM,MAAM,WAAW,SAAS,MAAM,UAAU,CAAC,EAAE,OAAO;EAC1D,IAAI,CAAC,KAAK;EAEV,MAAM,eAAe,WAAW,IAAI,GAAG;EAGvC,IAAI,gBAAgB,aAAa,aAAa,WAAW;EAEzD,WAAW,IAAI,KAAK;GAClB;GACA;GACA,WAAW,aAAa,GAAG,MAAM;EACnC,CAAC;CACH;CAEA,OAAO,MAAM,KAAK,WAAW,OAAO,CAAC,CAAC,CAAC,MACpC,WAAW,eACV,OAAO,UAAU,aAAa,SAAS,IACvC,OAAO,WAAW,aAAa,SAAS,CAC5C;AACF;;;;;;;AAkBA,MAAa,oBAAoB,EAC/B,SACA,WACA,cACkC;CAClC,IAAI,QAAQ,QAAQ,SAAS,GAAG,OAAO;CAEvC,MAAM,eAAe,UAAU,MAC5B,EAAE,eAAe,SAAS,YAAY,MAAM,SAC/C,CAAC,EAAE;CACH,IACE,gBACA,aAAa,cAAc,OAAO,MAAM,aAAa,OAAO,GAE5D,OAAO;CAGT,IAAI,QAAQ,aAAa,cAAc,OAAO,CAAC,QAAQ;CAOvD,QAJE,QAAQ,aAAa,uBACrB,QAAQ,aAAa,mBACrB,QAAQ,aAAa,gBAIrB,aAAa,QAAQ,WAAW,QAAQ,aAAa;AAEzD"}