@intlayer/engine 9.5.10 → 9.5.12

package/dist/esm/scan/checks.mjs.map

@@ -1 +1 @@
-{"version":3,"file":"checks.mjs","names":[],"sources":["../../../src/scan/checks.ts"],"sourcesContent":["import { ALL_LOCALES } from '@intlayer/types/allLocales';\nimport { analyzeBundleContent } from './analyzeBundleContent';\nimport {\n  type Anchor,\n  extractHreflangs,\n  extractHtmlDir,\n  extractHtmlLang,\n  hasCanonical,\n} from './parseHtml';\nimport type {\n  BundleChunkInput,\n  BundleContentAnalysis,\n  ScanEvent,\n} from './types';\n\n/** Format a byte count as a human-readable size. */\nexport const formatSize = (bytes: number): string => {\n  if (bytes >= 1024 * 1024) return `${(bytes / (1024 * 1024)).toFixed(2)} MB`;\n  if (bytes >= 1024) return `${(bytes / 1024).toFixed(2)} KB`;\n  return `${bytes} B`;\n};\n\n/**\n * Check the `<html>` element attributes (`lang`, `dir`) and the resulting\n * current-locale signal. Returns the detected language tag.\n */\nexport const checkHtmlAttributes = (\n  html: string,\n  targetUrl: string,\n  events: ScanEvent[]\n): { langTag: string | undefined } => {\n  const langTag = extractHtmlLang(html);\n  const dirTag = extractHtmlDir(html);\n\n  events.push({\n    type: `url_htmlLang\\\\${targetUrl}`,\n    status: langTag ? 'success' : 'error',\n    details: {\n      success: langTag,\n      error: langTag ? undefined : 'Missing html lang attribute',\n    },\n  });\n\n  events.push({\n    type: `url_currentLocale\\\\${targetUrl}`,\n    status: langTag ? 'success' : 'warning',\n    details: {\n      success: langTag,\n      warning: langTag ? undefined : 'No locale detected',\n    },\n  });\n\n  events.push({\n    type: `url_htmlDir\\\\${targetUrl}`,\n    status: dirTag ? 'success' : 'warning',\n    details: {\n      success: dirTag,\n      warning: dirTag ? undefined : 'Missing html dir attribute',\n    },\n  });\n\n  return { langTag };\n};\n\n/** Check the presence of a canonical link. */\nexport const checkCanonical = (\n  html: string,\n  targetUrl: string,\n  events: ScanEvent[]\n): void => {\n  const present = hasCanonical(html);\n  events.push({\n    type: `url_hasCanonical\\\\${targetUrl}`,\n    status: present ? 'success' : 'warning',\n    details: { warning: present ? undefined : 'Missing canonical link' },\n  });\n};\n\n/**\n * Check the page's hreflang structure and collect discovered locales into\n * `localesSet`.\n */\nexport const checkLinguisticStructure = (\n  html: string,\n  targetUrl: string,\n  localesSet: Set<string>,\n  events: ScanEvent[]\n): void => {\n  const langTag = extractHtmlLang(html);\n  if (langTag) localesSet.add(langTag);\n\n  const hreflangs = extractHreflangs(html);\n  for (const { hreflang } of hreflangs) {\n    if (hreflang !== 'x-default') localesSet.add(hreflang);\n  }\n\n  const hasXDefault = hreflangs.some((h) => h.hreflang === 'x-default');\n\n  events.push({\n    type: `url_hreflang\\\\${targetUrl}`,\n    status: hreflangs.length > 0 ? 'success' : 'warning',\n    details: {\n      success: hreflangs.length > 0 ? hreflangs : undefined,\n      warning: hreflangs.length === 0 ? 'No hreflang tags found' : undefined,\n    },\n  });\n\n  events.push({\n    type: `url_hasXDefault\\\\${targetUrl}`,\n    status: hasXDefault ? 'success' : 'error',\n    details: {\n      error: hasXDefault ? undefined : 'Missing x-default hreflang link',\n    },\n  });\n};\n\nconst normalizeHost = (host: string): string =>\n  host.startsWith('www.') ? host.slice(4) : host;\n\nconst localeValues = Object.values(ALL_LOCALES) as string[];\n\n/**\n * Check whether internal links carry a locale segment, mirroring the hosted\n * audit's URL-structure analysis but operating on parsed anchors instead of a\n * live DOM.\n */\nexport const checkUrlStructure = (\n  anchors: Anchor[],\n  origin: string,\n  targetUrl: string,\n  events: ScanEvent[]\n): void => {\n  const targetHostname = normalizeHost(new URL(origin).hostname);\n\n  let localizedCount = 0;\n  let totalInternalCount = 0;\n  const nonLocalizedLinks: string[] = [];\n\n  for (const { href } of anchors) {\n    if (!href || href.startsWith('#') || href.startsWith('javascript:'))\n      continue;\n\n    try {\n      const url = new URL(href, origin);\n      if (normalizeHost(url.hostname) !== targetHostname) continue;\n\n      totalInternalCount++;\n      const path = url.pathname.toLowerCase();\n      const hostname = url.hostname.toLowerCase();\n\n      const hasLocaleInPath = localeValues.some((locale) => {\n        const l = locale.toLowerCase();\n        return path === `/${l}` || path.includes(`/${l}/`);\n      });\n      const hasLocaleInSubdomain = localeValues.some((locale) => {\n        const l = locale.toLowerCase();\n        return hostname.startsWith(`${l}.`) || hostname.includes(`.${l}.`);\n      });\n\n      if (hasLocaleInPath || hasLocaleInSubdomain) {\n        localizedCount++;\n      } else {\n        nonLocalizedLinks.push(href);\n      }\n    } catch {\n      /* ignore malformed URLs */\n    }\n  }\n\n  const hasLocalizedLinks = localizedCount > 0;\n  const allAnchorsLocalized =\n    totalInternalCount === 0 || localizedCount === totalInternalCount;\n\n  events.push({\n    type: `url_hasLocalizedLinks\\\\${targetUrl}`,\n    status: hasLocalizedLinks ? 'success' : 'warning',\n    details: {\n      success: hasLocalizedLinks\n        ? `${localizedCount} localized links found out of ${totalInternalCount} internal links`\n        : undefined,\n      warning: hasLocalizedLinks ? undefined : 'No localized links found',\n    },\n  });\n\n  events.push({\n    type: `url_allAnchorsLocalized\\\\${targetUrl}`,\n    status: allAnchorsLocalized ? 'success' : 'warning',\n    details: {\n      warning: allAnchorsLocalized\n        ? undefined\n        : {\n            message: 'Some internal links are not localized',\n            links: nonLocalizedLinks,\n          },\n    },\n  });\n};\n\n/**\n * Analyze the JavaScript bundles for unused locale content and emit the\n * corresponding event. Returns the analysis so callers can report on it.\n */\nexport const checkBundleContent = (\n  chunks: BundleChunkInput[],\n  html: string,\n  currentLocale: string | undefined,\n  targetUrl: string,\n  totalPageSize: number,\n  events: ScanEvent[]\n): BundleContentAnalysis | undefined => {\n  if (!currentLocale) {\n    events.push({\n      type: `url_unusedBundleContent\\\\${targetUrl}`,\n      status: 'warning',\n      details: {\n        warning: 'Cannot analyse bundle content: page locale not detected',\n      },\n    });\n    return undefined;\n  }\n\n  const analysis = analyzeBundleContent(\n    chunks,\n    html,\n    currentLocale,\n    totalPageSize\n  );\n\n  // Status is driven by the main bundle — lazy chunks with unused content are expected.\n  const mainBundleMaxUnused = analysis.mainBundleChunks.reduce(\n    (max, c) => Math.max(max, c.unusedPercent),\n    0\n  );\n\n  const status =\n    mainBundleMaxUnused === 0\n      ? 'success'\n      : mainBundleMaxUnused <= 30\n        ? 'warning'\n        : 'error';\n\n  events.push({\n    type: `url_unusedBundleContent\\\\${targetUrl}`,\n    status,\n    details: { [status]: analysis },\n  });\n\n  return analysis;\n};\n\n/** Fetch and check `robots.txt`, emitting robots-related events. */\nexport const checkRobots = async (\n  origin: string,\n  discoveredLocales: Set<string>,\n  userAgent: string,\n  events: ScanEvent[]\n): Promise<void> => {\n  let robotsPresent = false;\n  let noLocalizedUrlsForgotten = true;\n  const errors: string[] = [];\n\n  try {\n    const response = await fetch(`${origin}/robots.txt`, {\n      headers: { 'User-Agent': userAgent },\n    });\n\n    if (response.ok) {\n      robotsPresent = true;\n      const content = await response.text();\n\n      if (content && discoveredLocales.size > 0) {\n        const disallowedPaths = content\n          .split('\\n')\n          .map((line) => line.trim().toLowerCase())\n          .filter((line) => line.startsWith('disallow:'))\n          .map((line) => line.slice('disallow:'.length).trim());\n\n        for (const locale of discoveredLocales) {\n          for (const path of disallowedPaths) {\n            if (path === `/${locale}` || path === `/${locale}/`) {\n              noLocalizedUrlsForgotten = false;\n              errors.push(\n                `Locale path \"${locale}\" appears to be blocked in robots.txt: ${path}`\n              );\n            }\n          }\n        }\n      }\n    }\n  } catch (error) {\n    errors.push(\n      `Failed to fetch robots.txt: ${error instanceof Error ? error.message : 'Unknown error'}`\n    );\n  }\n\n  events.push({\n    type: 'robots_robotsPresent',\n    status: robotsPresent ? 'success' : 'warning',\n    details: {\n      warning: robotsPresent ? undefined : 'No robots.txt found',\n      error: errors.length > 0 ? errors : undefined,\n    },\n  });\n\n  if (robotsPresent) {\n    events.push({\n      type: 'robots_noLocalizedUrlsForgotten',\n      status: noLocalizedUrlsForgotten ? 'success' : 'error',\n      details: { error: noLocalizedUrlsForgotten ? undefined : errors },\n    });\n  }\n};\n\n/** Fetch and check `sitemap.xml`, emitting sitemap-related events. */\nexport const checkSitemap = async (\n  origin: string,\n  discoveredLocales: Set<string>,\n  userAgent: string,\n  events: ScanEvent[]\n): Promise<void> => {\n  let sitemapPresent = false;\n  let hasXDefault = false;\n  let hasAlternates = false;\n  let noLocalizedUrlsForgotten = true;\n  const errors: string[] = [];\n\n  try {\n    const response = await fetch(`${origin}/sitemap.xml`, {\n      headers: { 'User-Agent': userAgent },\n    });\n\n    if (response.ok) {\n      sitemapPresent = true;\n      const content = await response.text();\n\n      const hreflangMatches = content.match(/hreflang\\s*=\\s*\"([^\"]+)\"/gi) ?? [];\n      const hreflangs = hreflangMatches.map((m) =>\n        m.replace(/hreflang\\s*=\\s*\"/i, '').replace(/\"$/, '')\n      );\n      hasAlternates = hreflangs.length > 0;\n      hasXDefault = hreflangs.includes('x-default');\n\n      if (discoveredLocales.size > 0) {\n        const urlBlocks = content.match(/<url\\b[\\s\\S]*?<\\/url>/gi) ?? [];\n        const allFoundLocales = new Set<string>();\n        let anyUrlMissingLocale = false;\n\n        for (const block of urlBlocks) {\n          const localesInUrl = new Set<string>();\n\n          for (const hreflang of block.match(/hreflang\\s*=\\s*\"([^\"]+)\"/gi) ??\n            []) {\n            const value = hreflang\n              .replace(/hreflang\\s*=\\s*\"/i, '')\n              .replace(/\"$/, '');\n            if (value !== 'x-default') {\n              localesInUrl.add(value);\n              allFoundLocales.add(value);\n            }\n          }\n\n          const loc = block.match(/<loc>([\\s\\S]*?)<\\/loc>/i)?.[1]?.trim();\n          if (loc) {\n            try {\n              const firstSegment = new URL(loc).pathname\n                .split('/')\n                .filter(Boolean)[0];\n              if (firstSegment && discoveredLocales.has(firstSegment)) {\n                localesInUrl.add(firstSegment);\n                allFoundLocales.add(firstSegment);\n              }\n            } catch {\n              /* invalid loc URL, skip */\n            }\n          }\n\n          const missing = [...discoveredLocales].filter(\n            (locale) => !localesInUrl.has(locale)\n          );\n          if (missing.length > 0 && missing.length < discoveredLocales.size) {\n            anyUrlMissingLocale = true;\n          }\n        }\n\n        const completelyMissing = [...discoveredLocales].filter(\n          (locale) => !allFoundLocales.has(locale)\n        );\n        if (anyUrlMissingLocale || completelyMissing.length > 0) {\n          noLocalizedUrlsForgotten = false;\n          if (completelyMissing.length > 0) {\n            errors.push(\n              `The following locales are completely missing from the sitemap: ${completelyMissing.join(', ')}`\n            );\n          }\n        }\n      }\n    }\n  } catch (error) {\n    errors.push(\n      `Failed to fetch sitemap.xml: ${error instanceof Error ? error.message : 'Unknown error'}`\n    );\n  }\n\n  events.push({\n    type: 'sitemap_sitemapPresent',\n    status: sitemapPresent ? 'success' : 'warning',\n    details: {\n      warning: sitemapPresent ? undefined : 'No sitemap.xml found',\n      error: errors.length > 0 ? errors : undefined,\n    },\n  });\n\n  if (sitemapPresent) {\n    events.push({\n      type: 'sitemap_noLocalizedUrlsForgotten',\n      status: noLocalizedUrlsForgotten ? 'success' : 'warning',\n      details: { warning: noLocalizedUrlsForgotten ? undefined : errors },\n    });\n\n    events.push({\n      type: 'sitemap_hasXDefault',\n      status: hasXDefault ? 'success' : 'warning',\n      details: {\n        warning: hasXDefault ? undefined : 'No x-default hreflang in sitemap',\n      },\n    });\n\n    events.push({\n      type: 'sitemap_hasAlternates',\n      status: hasAlternates ? 'success' : 'warning',\n      details: {\n        warning: hasAlternates\n          ? undefined\n          : 'No alternate language links found in sitemap',\n      },\n    });\n  }\n};\n"],"mappings":";;;;;;AAgBA,MAAa,cAAc,UAA0B;CACnD,IAAI,SAAS,SAAa,OAAO,IAAI,QAAS,QAAY,CAAE,QAAQ,CAAC,EAAE;CACvE,IAAI,SAAS,MAAM,OAAO,IAAI,QAAQ,KAAI,CAAE,QAAQ,CAAC,EAAE;CACvD,OAAO,GAAG,MAAM;AAClB;;;;;AAMA,MAAa,uBACX,MACA,WACA,WACoC;CACpC,MAAM,UAAU,gBAAgB,IAAI;CACpC,MAAM,SAAS,eAAe,IAAI;CAElC,OAAO,KAAK;EACV,MAAM,iBAAiB;EACvB,QAAQ,UAAU,YAAY;EAC9B,SAAS;GACP,SAAS;GACT,OAAO,UAAU,SAAY;EAC/B;CACF,CAAC;CAED,OAAO,KAAK;EACV,MAAM,sBAAsB;EAC5B,QAAQ,UAAU,YAAY;EAC9B,SAAS;GACP,SAAS;GACT,SAAS,UAAU,SAAY;EACjC;CACF,CAAC;CAED,OAAO,KAAK;EACV,MAAM,gBAAgB;EACtB,QAAQ,SAAS,YAAY;EAC7B,SAAS;GACP,SAAS;GACT,SAAS,SAAS,SAAY;EAChC;CACF,CAAC;CAED,OAAO,EAAE,QAAQ;AACnB;;AAGA,MAAa,kBACX,MACA,WACA,WACS;CACT,MAAM,UAAU,aAAa,IAAI;CACjC,OAAO,KAAK;EACV,MAAM,qBAAqB;EAC3B,QAAQ,UAAU,YAAY;EAC9B,SAAS,EAAE,SAAS,UAAU,SAAY,yBAAyB;CACrE,CAAC;AACH;;;;;AAMA,MAAa,4BACX,MACA,WACA,YACA,WACS;CACT,MAAM,UAAU,gBAAgB,IAAI;CACpC,IAAI,SAAS,WAAW,IAAI,OAAO;CAEnC,MAAM,YAAY,iBAAiB,IAAI;CACvC,KAAK,MAAM,EAAE,cAAc,WACzB,IAAI,aAAa,aAAa,WAAW,IAAI,QAAQ;CAGvD,MAAM,cAAc,UAAU,MAAM,MAAM,EAAE,aAAa,WAAW;CAEpE,OAAO,KAAK;EACV,MAAM,iBAAiB;EACvB,QAAQ,UAAU,SAAS,IAAI,YAAY;EAC3C,SAAS;GACP,SAAS,UAAU,SAAS,IAAI,YAAY;GAC5C,SAAS,UAAU,WAAW,IAAI,2BAA2B;EAC/D;CACF,CAAC;CAED,OAAO,KAAK;EACV,MAAM,oBAAoB;EAC1B,QAAQ,cAAc,YAAY;EAClC,SAAS,EACP,OAAO,cAAc,SAAY,kCACnC;CACF,CAAC;AACH;AAEA,MAAM,iBAAiB,SACrB,KAAK,WAAW,MAAM,IAAI,KAAK,MAAM,CAAC,IAAI;AAE5C,MAAM,eAAe,OAAO,OAAO,WAAW;;;;;;AAO9C,MAAa,qBACX,SACA,QACA,WACA,WACS;CACT,MAAM,iBAAiB,cAAc,IAAI,IAAI,MAAM,CAAC,CAAC,QAAQ;CAE7D,IAAI,iBAAiB;CACrB,IAAI,qBAAqB;CACzB,MAAM,oBAA8B,CAAC;CAErC,KAAK,MAAM,EAAE,UAAU,SAAS;EAC9B,IAAI,CAAC,QAAQ,KAAK,WAAW,GAAG,KAAK,KAAK,WAAW,aAAa,GAChE;EAEF,IAAI;GACF,MAAM,MAAM,IAAI,IAAI,MAAM,MAAM;GAChC,IAAI,cAAc,IAAI,QAAQ,MAAM,gBAAgB;GAEpD;GACA,MAAM,OAAO,IAAI,SAAS,YAAY;GACtC,MAAM,WAAW,IAAI,SAAS,YAAY;GAE1C,MAAM,kBAAkB,aAAa,MAAM,WAAW;IACpD,MAAM,IAAI,OAAO,YAAY;IAC7B,OAAO,SAAS,IAAI,OAAO,KAAK,SAAS,IAAI,EAAE,EAAE;GACnD,CAAC;GACD,MAAM,uBAAuB,aAAa,MAAM,WAAW;IACzD,MAAM,IAAI,OAAO,YAAY;IAC7B,OAAO,SAAS,WAAW,GAAG,EAAE,EAAE,KAAK,SAAS,SAAS,IAAI,EAAE,EAAE;GACnE,CAAC;GAED,IAAI,mBAAmB,sBACrB;QAEA,kBAAkB,KAAK,IAAI;EAE/B,QAAQ,CAER;CACF;CAEA,MAAM,oBAAoB,iBAAiB;CAC3C,MAAM,sBACJ,uBAAuB,KAAK,mBAAmB;CAEjD,OAAO,KAAK;EACV,MAAM,0BAA0B;EAChC,QAAQ,oBAAoB,YAAY;EACxC,SAAS;GACP,SAAS,oBACL,GAAG,eAAe,gCAAgC,mBAAmB,mBACrE;GACJ,SAAS,oBAAoB,SAAY;EAC3C;CACF,CAAC;CAED,OAAO,KAAK;EACV,MAAM,4BAA4B;EAClC,QAAQ,sBAAsB,YAAY;EAC1C,SAAS,EACP,SAAS,sBACL,SACA;GACE,SAAS;GACT,OAAO;EACT,EACN;CACF,CAAC;AACH;;;;;AAMA,MAAa,sBACX,QACA,MACA,eACA,WACA,eACA,WACsC;CACtC,IAAI,CAAC,eAAe;EAClB,OAAO,KAAK;GACV,MAAM,4BAA4B;GAClC,QAAQ;GACR,SAAS,EACP,SAAS,0DACX;EACF,CAAC;EACD;CACF;CAEA,MAAM,WAAW,qBACf,QACA,MACA,eACA,aACF;CAGA,MAAM,sBAAsB,SAAS,iBAAiB,QACnD,KAAK,MAAM,KAAK,IAAI,KAAK,EAAE,aAAa,GACzC,CACF;CAEA,MAAM,SACJ,wBAAwB,IACpB,YACA,uBAAuB,KACrB,YACA;CAER,OAAO,KAAK;EACV,MAAM,4BAA4B;EAClC;EACA,SAAS,GAAG,SAAS,SAAS;CAChC,CAAC;CAED,OAAO;AACT;;AAGA,MAAa,cAAc,OACzB,QACA,mBACA,WACA,WACkB;CAClB,IAAI,gBAAgB;CACpB,IAAI,2BAA2B;CAC/B,MAAM,SAAmB,CAAC;CAE1B,IAAI;EACF,MAAM,WAAW,MAAM,MAAM,GAAG,OAAO,cAAc,EACnD,SAAS,EAAE,cAAc,UAAU,EACrC,CAAC;EAED,IAAI,SAAS,IAAI;GACf,gBAAgB;GAChB,MAAM,UAAU,MAAM,SAAS,KAAK;GAEpC,IAAI,WAAW,kBAAkB,OAAO,GAAG;IACzC,MAAM,kBAAkB,QACrB,MAAM,IAAI,CAAC,CACX,KAAK,SAAS,KAAK,KAAK,CAAC,CAAC,YAAY,CAAC,CAAC,CACxC,QAAQ,SAAS,KAAK,WAAW,WAAW,CAAC,CAAC,CAC9C,KAAK,SAAS,KAAK,MAAM,CAAkB,CAAC,CAAC,KAAK,CAAC;IAEtD,KAAK,MAAM,UAAU,mBACnB,KAAK,MAAM,QAAQ,iBACjB,IAAI,SAAS,IAAI,YAAY,SAAS,IAAI,OAAO,IAAI;KACnD,2BAA2B;KAC3B,OAAO,KACL,gBAAgB,OAAO,yCAAyC,MAClE;IACF;GAGN;EACF;CACF,SAAS,OAAO;EACd,OAAO,KACL,+BAA+B,iBAAiB,QAAQ,MAAM,UAAU,iBAC1E;CACF;CAEA,OAAO,KAAK;EACV,MAAM;EACN,QAAQ,gBAAgB,YAAY;EACpC,SAAS;GACP,SAAS,gBAAgB,SAAY;GACrC,OAAO,OAAO,SAAS,IAAI,SAAS;EACtC;CACF,CAAC;CAED,IAAI,eACF,OAAO,KAAK;EACV,MAAM;EACN,QAAQ,2BAA2B,YAAY;EAC/C,SAAS,EAAE,OAAO,2BAA2B,SAAY,OAAO;CAClE,CAAC;AAEL;;AAGA,MAAa,eAAe,OAC1B,QACA,mBACA,WACA,WACkB;CAClB,IAAI,iBAAiB;CACrB,IAAI,cAAc;CAClB,IAAI,gBAAgB;CACpB,IAAI,2BAA2B;CAC/B,MAAM,SAAmB,CAAC;CAE1B,IAAI;EACF,MAAM,WAAW,MAAM,MAAM,GAAG,OAAO,eAAe,EACpD,SAAS,EAAE,cAAc,UAAU,EACrC,CAAC;EAED,IAAI,SAAS,IAAI;GACf,iBAAiB;GACjB,MAAM,UAAU,MAAM,SAAS,KAAK;GAGpC,MAAM,aADkB,QAAQ,MAAM,4BAA4B,KAAK,CAAC,EACvC,CAAC,KAAK,MACrC,EAAE,QAAQ,qBAAqB,EAAE,CAAC,CAAC,QAAQ,MAAM,EAAE,CACrD;GACA,gBAAgB,UAAU,SAAS;GACnC,cAAc,UAAU,SAAS,WAAW;GAE5C,IAAI,kBAAkB,OAAO,GAAG;IAC9B,MAAM,YAAY,QAAQ,MAAM,yBAAyB,KAAK,CAAC;IAC/D,MAAM,kCAAkB,IAAI,IAAY;IACxC,IAAI,sBAAsB;IAE1B,KAAK,MAAM,SAAS,WAAW;KAC7B,MAAM,+BAAe,IAAI,IAAY;KAErC,KAAK,MAAM,YAAY,MAAM,MAAM,4BAA4B,KAC7D,CAAC,GAAG;MACJ,MAAM,QAAQ,SACX,QAAQ,qBAAqB,EAAE,CAAC,CAChC,QAAQ,MAAM,EAAE;MACnB,IAAI,UAAU,aAAa;OACzB,aAAa,IAAI,KAAK;OACtB,gBAAgB,IAAI,KAAK;MAC3B;KACF;KAEA,MAAM,MAAM,MAAM,MAAM,yBAAyB,CAAC,GAAG,EAAE,EAAE,KAAK;KAC9D,IAAI,KACF,IAAI;MACF,MAAM,eAAe,IAAI,IAAI,GAAG,CAAC,CAAC,SAC/B,MAAM,GAAG,CAAC,CACV,OAAO,OAAO,CAAC,CAAC;MACnB,IAAI,gBAAgB,kBAAkB,IAAI,YAAY,GAAG;OACvD,aAAa,IAAI,YAAY;OAC7B,gBAAgB,IAAI,YAAY;MAClC;KACF,QAAQ,CAER;KAGF,MAAM,UAAU,CAAC,GAAG,iBAAiB,CAAC,CAAC,QACpC,WAAW,CAAC,aAAa,IAAI,MAAM,CACtC;KACA,IAAI,QAAQ,SAAS,KAAK,QAAQ,SAAS,kBAAkB,MAC3D,sBAAsB;IAE1B;IAEA,MAAM,oBAAoB,CAAC,GAAG,iBAAiB,CAAC,CAAC,QAC9C,WAAW,CAAC,gBAAgB,IAAI,MAAM,CACzC;IACA,IAAI,uBAAuB,kBAAkB,SAAS,GAAG;KACvD,2BAA2B;KAC3B,IAAI,kBAAkB,SAAS,GAC7B,OAAO,KACL,kEAAkE,kBAAkB,KAAK,IAAI,GAC/F;IAEJ;GACF;EACF;CACF,SAAS,OAAO;EACd,OAAO,KACL,gCAAgC,iBAAiB,QAAQ,MAAM,UAAU,iBAC3E;CACF;CAEA,OAAO,KAAK;EACV,MAAM;EACN,QAAQ,iBAAiB,YAAY;EACrC,SAAS;GACP,SAAS,iBAAiB,SAAY;GACtC,OAAO,OAAO,SAAS,IAAI,SAAS;EACtC;CACF,CAAC;CAED,IAAI,gBAAgB;EAClB,OAAO,KAAK;GACV,MAAM;GACN,QAAQ,2BAA2B,YAAY;GAC/C,SAAS,EAAE,SAAS,2BAA2B,SAAY,OAAO;EACpE,CAAC;EAED,OAAO,KAAK;GACV,MAAM;GACN,QAAQ,cAAc,YAAY;GAClC,SAAS,EACP,SAAS,cAAc,SAAY,mCACrC;EACF,CAAC;EAED,OAAO,KAAK;GACV,MAAM;GACN,QAAQ,gBAAgB,YAAY;GACpC,SAAS,EACP,SAAS,gBACL,SACA,+CACN;EACF,CAAC;CACH;AACF"}
+{"version":3,"file":"checks.mjs","names":[],"sources":["../../../src/scan/checks.ts"],"sourcesContent":["import { analyzeBundleContent } from './analyzeBundleContent';\nimport { classifyInternalLinks } from './detection/classifyInternalLinks';\nimport {\n  getUrlLocale,\n  type RoutingDetection,\n} from './detection/detectRoutingStrategy';\nimport {\n  getLanguageCode,\n  isRightToLeftLocale,\n  isSameLanguage,\n  isValidLocaleCode,\n  isValidOpenGraphLocale,\n  normalizeLocaleCode,\n  toOpenGraphLocale,\n} from './detection/localeCode';\nimport { isAbsoluteUrl, normalizeUrl, parseUrl } from './detection/url';\nimport { fetchText, type ScanFetchOptions } from './fetchText';\nimport {\n  type Anchor,\n  extractHreflangs,\n  extractHtmlLang,\n  type HreflangLink,\n} from './parseHtml';\nimport { isPathAllowedByRobots, parseRobots } from './robots';\nimport { collectSitemapEntries, getDefaultSitemapUrls } from './sitemap';\nimport type {\n  BundleChunkInput,\n  BundleContentAnalysis,\n  ChunkAnalysis,\n  ScanCheckStatus,\n  ScanEvent,\n} from './types';\n\n/** Receives every check result as soon as it is produced. */\nexport type EmitScanEvent = (event: ScanEvent) => void;\n\n/** Signals of the scanned page shared by the page-level checks. */\nexport type PageSignals = {\n  targetUrl: string;\n  langTag?: string;\n  dirTag?: string;\n  ogLocale?: string;\n  /** `og:locale:alternate` values. */\n  ogLocaleAlternates?: string[];\n  /** Raw `href` of the canonical link (`''` when present but empty). */\n  canonicalHref?: string;\n  hreflangs: HreflangLink[];\n  anchors: Anchor[];\n  routing: RoutingDetection;\n  /** Locale of the page: from its URL when possible, else `<html lang>`. */\n  pageLocale?: string;\n};\n\n/** Maximum number of links listed in a check's details. */\nconst MAX_LISTED_LINKS = 30;\n/** Maximum number of hreflang alternates fetched for the reciprocity check. */\nconst MAX_FETCHED_ALTERNATES = 8;\n/** Statuses meaning the site refuses automated requests, not a broken page. */\nconst BOT_BLOCKING_STATUSES = new Set([401, 403, 429]);\n\n/** Format a byte count as a human-readable size. */\nexport const formatSize = (bytes: number): string => {\n  if (bytes >= 1024 * 1024) return `${(bytes / (1024 * 1024)).toFixed(2)} MB`;\n  if (bytes >= 1024) return `${(bytes / 1024).toFixed(2)} KB`;\n  return `${bytes} B`;\n};\n\n/** Build a URL-scoped check type (`url_htmlLang\\https://…`). */\nconst urlCheckType = (checkName: string, targetUrl: string): string =>\n  `${checkName}\\\\${targetUrl}`;\n\n/** Emit an event whose details are attached to its own status. */\nconst emitCheck = (\n  emit: EmitScanEvent,\n  type: string,\n  status: ScanCheckStatus,\n  details?: unknown\n): void =>\n  emit({\n    type,\n    status,\n    details: details === undefined ? undefined : { [status]: details },\n  });\n\n/** Render an anchor as an HTML line for the details popover. */\nconst formatAnchor = ({ href, text }: Anchor): string =>\n  `<a href=\"${href}\">${text.slice(0, 60) || 'Link'}</a>`;\n\n/* ------------------------------------------------------------------------ */\n/*                              <html> attributes                            */\n/* ------------------------------------------------------------------------ */\n\n/** `url_htmlLang`: `<html lang>` is present and a valid BCP 47 tag. */\nexport const checkHtmlLang = (\n  { targetUrl, langTag }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_htmlLang', targetUrl);\n  if (!langTag) {\n    emitCheck(emit, type, 'error', 'Missing html lang attribute');\n  } else if (!isValidLocaleCode(langTag)) {\n    emitCheck(\n      emit,\n      type,\n      'error',\n      `\"${langTag}\" is not a valid BCP 47 language tag (expected e.g. \"en\" or \"en-US\")`\n    );\n  } else {\n    emitCheck(emit, type, 'success', langTag);\n  }\n};\n\n/**\n * `url_htmlDir`: the text direction matches the language. `dir` is only\n * required for right-to-left languages, `ltr` being the browser default.\n */\nexport const checkHtmlDir = (\n  { targetUrl, langTag, dirTag }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_htmlDir', targetUrl);\n  const direction = dirTag?.toLowerCase();\n\n  if (!langTag) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      'Cannot verify the text direction: html lang is missing'\n    );\n    return;\n  }\n\n  const expectedDirection = isRightToLeftLocale(langTag) ? 'rtl' : 'ltr';\n\n  if (\n    direction === expectedDirection ||\n    (!direction && expectedDirection === 'ltr')\n  ) {\n    emitCheck(emit, type, 'success', direction ?? 'ltr (browser default)');\n    return;\n  }\n\n  emitCheck(\n    emit,\n    type,\n    'error',\n    direction\n      ? `dir=\"${direction}\" does not match the ${expectedDirection.toUpperCase()} language \"${langTag}\"`\n      : `Missing dir=\"rtl\" for the right-to-left language \"${langTag}\"`\n  );\n};\n\n/**\n * `url_currentLocale`: the locale announced by `<html lang>`, the URL and the\n * self hreflang entry agree.\n */\nexport const checkLocaleConsistency = (\n  { targetUrl, langTag, hreflangs, routing }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_currentLocale', targetUrl);\n  const normalizedTargetUrl = normalizeUrl(targetUrl);\n  const selfHreflang = hreflangs.find(\n    ({ hreflang, href }) =>\n      hreflang.toLowerCase() !== 'x-default' &&\n      normalizeUrl(href, targetUrl) === normalizedTargetUrl\n  )?.hreflang;\n\n  const signals = {\n    htmlLang: langTag,\n    urlLocale: routing.urlLocale,\n    selfHreflang,\n  };\n  const definedLocales = Object.values(signals).filter(\n    (locale): locale is string => Boolean(locale)\n  );\n\n  if (definedLocales.length === 0) {\n    emitCheck(emit, type, 'warning', 'No locale detected on the page');\n    return;\n  }\n\n  const firstLocale = definedLocales[0];\n  const isConsistent = definedLocales.every((locale) =>\n    isSameLanguage(locale, firstLocale)\n  );\n\n  emitCheck(\n    emit,\n    type,\n    isConsistent ? 'success' : 'error',\n    isConsistent\n      ? signals\n      : {\n          message:\n            'The page announces different locales in html lang, its URL and its hreflang entry',\n          ...signals,\n        }\n  );\n};\n\n/** Describe an invalid Open Graph locale, with the value to use instead. */\nconst formatInvalidOpenGraphLocale = (\n  localeCode: string,\n  langTag: string | undefined,\n  property: string\n): string => {\n  const suggestion = toOpenGraphLocale(localeCode, langTag);\n  return `${property} \"${localeCode}\" is not a valid Open Graph locale (expected language_TERRITORY${\n    suggestion ? `, e.g. \"${suggestion}\"` : ''\n  }): crawlers ignore it and fall back to en_US`;\n};\n\n/**\n * `url_ogLocale`: `og:locale` is set, uses the Open Graph `language_TERRITORY`\n * format (`en_GB`, not `en` nor `en-GB`) and matches `<html lang>`.\n * `og:locale:alternate` values must use the same format.\n */\nexport const checkOgLocale = (\n  { targetUrl, ogLocale, ogLocaleAlternates = [], langTag }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_ogLocale', targetUrl);\n\n  if (!ogLocale) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      'Missing <meta property=\"og:locale\">: social previews default to en_US'\n    );\n    return;\n  }\n\n  if (!isValidOpenGraphLocale(ogLocale)) {\n    emitCheck(\n      emit,\n      type,\n      'error',\n      formatInvalidOpenGraphLocale(ogLocale, langTag, 'og:locale')\n    );\n    return;\n  }\n\n  if (langTag && !isSameLanguage(ogLocale, langTag)) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      `og:locale \"${ogLocale}\" does not match html lang \"${langTag}\"`\n    );\n    return;\n  }\n\n  const invalidAlternates = ogLocaleAlternates.filter(\n    (alternate) => !isValidOpenGraphLocale(alternate)\n  );\n\n  if (invalidAlternates.length > 0) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      invalidAlternates\n        .map((alternate) =>\n          formatInvalidOpenGraphLocale(\n            alternate,\n            undefined,\n            'og:locale:alternate'\n          )\n        )\n        .join('\\n')\n    );\n    return;\n  }\n\n  emitCheck(emit, type, 'success', ogLocale);\n};\n\n/* ------------------------------------------------------------------------ */\n/*                         Canonical & hreflang tags                         */\n/* ------------------------------------------------------------------------ */\n\n/**\n * `url_hasCanonical`: a canonical link exists and does not point to another\n * locale version (which would de-index the current one).\n */\nexport const checkCanonical = (\n  { targetUrl, canonicalHref, hreflangs, routing, pageLocale }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_hasCanonical', targetUrl);\n\n  if (canonicalHref === undefined || canonicalHref === '') {\n    emitCheck(emit, type, 'warning', 'Missing canonical link');\n    return;\n  }\n\n  const canonicalUrl = normalizeUrl(canonicalHref, targetUrl);\n  const otherLocaleAlternate = hreflangs.find(\n    ({ hreflang, href }) =>\n      hreflang.toLowerCase() !== 'x-default' &&\n      normalizeUrl(href, targetUrl) === canonicalUrl &&\n      !isSameLanguage(hreflang, pageLocale)\n  );\n  const canonicalLocale = getUrlLocale(canonicalUrl, routing);\n\n  if (\n    otherLocaleAlternate ||\n    (canonicalLocale &&\n      pageLocale &&\n      !isSameLanguage(canonicalLocale, pageLocale))\n  ) {\n    emitCheck(emit, type, 'error', {\n      message: `The canonical URL points to the \"${otherLocaleAlternate?.hreflang ?? canonicalLocale}\" version instead of this \"${pageLocale}\" page: search engines will drop this page from their index`,\n      canonical: canonicalUrl,\n    });\n    return;\n  }\n\n  emitCheck(\n    emit,\n    type,\n    'success',\n    canonicalUrl === normalizeUrl(targetUrl)\n      ? canonicalUrl\n      : {\n          message: 'Canonical points to another URL of the same locale',\n          canonical: canonicalUrl,\n        }\n  );\n};\n\n/**\n * `url_hreflang`: hreflang tags exist, use valid codes and absolute URLs, have\n * no duplicates, and include a self reference.\n */\nexport const checkHreflang = (\n  { targetUrl, hreflangs, canonicalHref }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_hreflang', targetUrl);\n\n  if (hreflangs.length === 0) {\n    emitCheck(emit, type, 'warning', 'No hreflang tags found');\n    return;\n  }\n\n  const issues: string[] = [];\n\n  const invalidCodes = hreflangs\n    .map(({ hreflang }) => hreflang)\n    .filter((hreflang) => !isValidLocaleCode(hreflang));\n  if (invalidCodes.length > 0) {\n    issues.push(`Invalid hreflang codes: ${invalidCodes.join(', ')}`);\n  }\n\n  const ukCodes = hreflangs\n    .map(({ hreflang }) => hreflang)\n    .filter((hreflang) => /-uk$/i.test(hreflang));\n  if (ukCodes.length > 0) {\n    issues.push(\n      `\"${ukCodes.join(', ')}\": the United Kingdom region code is \"GB\" (e.g. en-GB)`\n    );\n  }\n\n  // Search engines ignore alternates that are not fully qualified.\n  const relativeHrefs = hreflangs.filter(({ href }) => !isAbsoluteUrl(href));\n  if (relativeHrefs.length > 0) {\n    issues.push(\n      `hreflang URLs must be absolute (https://…), search engines ignore: ${relativeHrefs\n        .map(({ hreflang, href }) => `${hreflang} → ${href}`)\n        .join(', ')}`\n    );\n  }\n\n  const urlsByCode = new Map<string, Set<string>>();\n  for (const { hreflang, href } of hreflangs) {\n    const code = normalizeLocaleCode(hreflang);\n    const urls = urlsByCode.get(code) ?? new Set<string>();\n    urls.add(normalizeUrl(href, targetUrl));\n    urlsByCode.set(code, urls);\n  }\n  const conflictingCodes = [...urlsByCode.entries()]\n    .filter(([, urls]) => urls.size > 1)\n    .map(([code]) => code);\n  if (conflictingCodes.length > 0) {\n    issues.push(\n      `Same hreflang declared with different URLs: ${conflictingCodes.join(', ')}`\n    );\n  }\n\n  const selfUrls = new Set([normalizeUrl(targetUrl)]);\n  if (canonicalHref) selfUrls.add(normalizeUrl(canonicalHref, targetUrl));\n  const hasSelfReference = hreflangs.some(\n    ({ hreflang, href }) =>\n      hreflang.toLowerCase() !== 'x-default' &&\n      selfUrls.has(normalizeUrl(href, targetUrl))\n  );\n  if (!hasSelfReference) {\n    issues.push('No hreflang entry points to the page itself (self reference)');\n  }\n\n  const status: ScanCheckStatus =\n    relativeHrefs.length > 0\n      ? 'error'\n      : issues.length > 0\n        ? 'warning'\n        : 'success';\n\n  emitCheck(\n    emit,\n    type,\n    status,\n    issues.length > 0 ? { issues, hreflangs } : hreflangs\n  );\n};\n\n/** `url_hasXDefault`: an `x-default` hreflang is declared. */\nexport const checkXDefault = (\n  { targetUrl, hreflangs }: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const type = urlCheckType('url_hasXDefault', targetUrl);\n  const xDefault = hreflangs.find(\n    ({ hreflang }) => hreflang.toLowerCase() === 'x-default'\n  );\n\n  if (xDefault) {\n    emitCheck(emit, type, 'success', xDefault.href);\n  } else if (hreflangs.length === 0) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      'No hreflang tags, so no x-default either'\n    );\n  } else {\n    emitCheck(emit, type, 'error', 'Missing x-default hreflang link');\n  }\n};\n\n/** Outcome of fetching one hreflang alternate. */\ntype AlternateVerification = {\n  url: string;\n  hreflang: string;\n  /** `broken`: 4xx/5xx; `unverifiable`: unreachable or blocking bots. */\n  outcome: 'valid' | 'invalid' | 'broken' | 'unverifiable';\n  issues: string[];\n};\n\n/** Fetch an alternate and check it forms a valid hreflang pair with the page. */\nconst verifyAlternate = async (\n  alternateUrl: string,\n  hreflang: string,\n  selfUrls: Set<string>,\n  fetchOptions: ScanFetchOptions\n): Promise<AlternateVerification> => {\n  const response = await fetchText(alternateUrl, fetchOptions);\n  const verification = { url: alternateUrl, hreflang };\n\n  if (!response || BOT_BLOCKING_STATUSES.has(response.status)) {\n    return { ...verification, outcome: 'unverifiable', issues: [] };\n  }\n  if (!response.ok) {\n    return {\n      ...verification,\n      outcome: 'broken',\n      issues: [`HTTP ${response.status}`],\n    };\n  }\n\n  const issues: string[] = [];\n  if (response.redirected) issues.push(`redirects to ${response.finalUrl}`);\n\n  const hasReturnLink = extractHreflangs(response.text).some(({ href }) =>\n    selfUrls.has(normalizeUrl(href, response.finalUrl))\n  );\n  if (!hasReturnLink) issues.push('no hreflang link back to this page');\n\n  const alternateLang = extractHtmlLang(response.text);\n  if (alternateLang && !isSameLanguage(alternateLang, hreflang)) {\n    issues.push(`html lang is \"${alternateLang}\"`);\n  }\n\n  return {\n    ...verification,\n    outcome: issues.length > 0 ? 'invalid' : 'valid',\n    issues,\n  };\n};\n\n/**\n * `url_hreflangReciprocal`: every alternate answers with a 200, is not\n * redirected, links back to this page and declares the expected language.\n * Search engines ignore hreflang pairs without a return link.\n */\nexport const checkHreflangReciprocity = async (\n  { targetUrl, hreflangs, canonicalHref }: PageSignals,\n  fetchOptions: ScanFetchOptions,\n  emit: EmitScanEvent\n): Promise<void> => {\n  const type = urlCheckType('url_hreflangReciprocal', targetUrl);\n  const selfUrls = new Set([normalizeUrl(targetUrl)]);\n  if (canonicalHref) selfUrls.add(normalizeUrl(canonicalHref, targetUrl));\n\n  const alternatesByUrl = new Map<string, string>();\n  for (const { hreflang, href } of hreflangs) {\n    if (hreflang.toLowerCase() === 'x-default') continue;\n    const absoluteUrl = parseUrl(href, targetUrl)?.href;\n    if (!absoluteUrl || selfUrls.has(normalizeUrl(absoluteUrl))) continue;\n    if (!alternatesByUrl.has(absoluteUrl)) {\n      alternatesByUrl.set(absoluteUrl, hreflang);\n    }\n  }\n\n  if (alternatesByUrl.size === 0) return;\n\n  const verifications = await Promise.all(\n    [...alternatesByUrl.entries()]\n      .slice(0, MAX_FETCHED_ALTERNATES)\n      .map(([alternateUrl, hreflang]) =>\n        verifyAlternate(alternateUrl, hreflang, selfUrls, fetchOptions)\n      )\n  );\n\n  const unverifiableUrls = verifications\n    .filter(({ outcome }) => outcome === 'unverifiable')\n    .map(({ url }) => url);\n  const failingVerifications = verifications.filter(\n    ({ outcome }) => outcome === 'broken' || outcome === 'invalid'\n  );\n  const verifiedCount = verifications.length - unverifiableUrls.length;\n\n  if (verifiedCount === 0) {\n    emitCheck(emit, type, 'warning', {\n      message:\n        'The alternates could not be verified: they are unreachable or block automated requests',\n      urls: unverifiableUrls,\n    });\n    return;\n  }\n\n  if (failingVerifications.length === 0) {\n    emitCheck(\n      emit,\n      type,\n      'success',\n      `${verifiedCount} alternates checked, all reciprocal`\n    );\n    return;\n  }\n\n  emitCheck(\n    emit,\n    type,\n    failingVerifications.some(({ outcome }) => outcome === 'broken')\n      ? 'error'\n      : 'warning',\n    {\n      message: `${failingVerifications.length} of ${verifiedCount} checked hreflang alternates are not valid return pairs`,\n      alternates: failingVerifications.map(\n        ({ hreflang, url, issues }) =>\n          `${hreflang} ${url}: ${issues.join(', ')}`\n      ),\n      unverifiableUrls,\n    }\n  );\n};\n\n/* ------------------------------------------------------------------------ */\n/*                              Internal links                               */\n/* ------------------------------------------------------------------------ */\n\n/**\n * Link checks, interpreted through the detected routing strategy (so an\n * unprefixed link is correct on the default locale of a `prefix-no-default`\n * site, and `?lang=` links are understood on `search-params` sites):\n *\n * - `url_hasLocalizedLinks`: internal links keep the page locale\n * - `url_allAnchorsLocalized`: no internal link loses or switches the locale\n * - `url_hasLangSelector`: crawlable links to the other locale versions exist\n */\nexport const checkInternalLinks = (\n  signals: PageSignals,\n  emit: EmitScanEvent\n): void => {\n  const { targetUrl, routing, pageLocale } = signals;\n  const {\n    internalLinks,\n    sameLocaleLinks,\n    otherLocaleLinks,\n    unlocalizedLinks,\n    switcherLinks,\n  } = classifyInternalLinks(signals);\n  const listLinks = (links: Anchor[]): string[] =>\n    links.slice(0, MAX_LISTED_LINKS).map(formatAnchor);\n\n  const localizedLinksType = urlCheckType('url_hasLocalizedLinks', targetUrl);\n  const isLocaleInUrl =\n    routing.strategy !== 'no-prefix' && routing.strategy !== 'unknown';\n\n  if (internalLinks.length === 0) {\n    emitCheck(emit, localizedLinksType, 'warning', {\n      message: 'No internal links found on the page',\n      links: listLinks(switcherLinks),\n    });\n  } else if (!isLocaleInUrl || !pageLocale) {\n    emitCheck(emit, localizedLinksType, 'warning', {\n      message: `The locale is not reflected in the URLs (routing: ${routing.strategy}). Each locale needs its own crawlable URL to be indexed.`,\n      links: listLinks(internalLinks),\n    });\n  } else if (sameLocaleLinks.length === 0) {\n    emitCheck(emit, localizedLinksType, 'warning', {\n      message: `None of the ${internalLinks.length} internal links points to the \"${pageLocale}\" version (routing: ${routing.strategy})`,\n      links: listLinks(internalLinks),\n    });\n  } else {\n    emitCheck(emit, localizedLinksType, 'success', {\n      message: `${sameLocaleLinks.length} of ${internalLinks.length} internal links point to the \"${pageLocale}\" version (routing: ${routing.strategy})`,\n      links: listLinks(sameLocaleLinks),\n    });\n  }\n\n  if (isLocaleInUrl && pageLocale && internalLinks.length > 0) {\n    const brokenLinks = [...otherLocaleLinks, ...unlocalizedLinks];\n    emitCheck(\n      emit,\n      urlCheckType('url_allAnchorsLocalized', targetUrl),\n      brokenLinks.length === 0 ? 'success' : 'warning',\n      brokenLinks.length === 0\n        ? `All ${internalLinks.length} internal links keep the \"${pageLocale}\" locale`\n        : {\n            message: `${brokenLinks.length} internal links leave the \"${pageLocale}\" locale (${otherLocaleLinks.length} to another locale, ${unlocalizedLinks.length} without locale)`,\n            links: listLinks(brokenLinks),\n          }\n    );\n  }\n\n  if (routing.locales.length > 1) {\n    emitCheck(\n      emit,\n      urlCheckType('url_hasLangSelector', targetUrl),\n      switcherLinks.length > 0 ? 'success' : 'warning',\n      switcherLinks.length > 0\n        ? {\n            message: `${switcherLinks.length} crawlable links to other locale versions`,\n            links: listLinks(switcherLinks),\n          }\n        : 'No crawlable <a href> to the other locale versions: search engines only discover them through hreflang'\n    );\n  }\n};\n\n/* ------------------------------------------------------------------------ */\n/*                              Bundle content                               */\n/* ------------------------------------------------------------------------ */\n\nconst getFilename = (url: string): string =>\n  url.split('/').pop()?.split('?')[0] ?? url;\n\nconst formatChunk = (chunk: ChunkAnalysis) => ({\n  filename: getFilename(chunk.url),\n  url: chunk.url,\n  fileSize: formatSize(chunk.fileSize),\n  totalLocaleSize: formatSize(chunk.totalLocaleSize),\n  usedLocaleSize: formatSize(chunk.usedLocaleSize),\n  unusedLocaleSize: formatSize(chunk.unusedLocaleSize),\n  dictionariesFound: chunk.dictionariesFound,\n  unusedPercent: `${chunk.unusedPercent}%`,\n});\n\n/** Human-readable version of a {@link BundleContentAnalysis}, for display. */\nexport const formatBundleAnalysis = (analysis: BundleContentAnalysis) => ({\n  currentLocale: analysis.currentLocale,\n  totalPageSize: formatSize(analysis.totalPageSize),\n  renderedContentSize: formatSize(analysis.renderedContentSize),\n  contentSize: formatSize(analysis.contentSize),\n  totalLocaleSize: formatSize(analysis.totalLocaleSize),\n  totalUnusedLocaleSize: formatSize(analysis.totalUnusedLocaleSize),\n  unusedPercentOfLocale: `${analysis.unusedPercentOfLocale}%`,\n  mainBundleChunks: analysis.mainBundleChunks.map(formatChunk),\n  lazyBundleChunks: analysis.lazyBundleChunks.map(formatChunk),\n});\n\n/**\n * `url_unusedBundleContent`: share of other-locale translations shipped in the\n * eagerly loaded bundles. Returns the raw analysis for reporting.\n */\nexport const checkBundleContent = (\n  chunks: BundleChunkInput[],\n  html: string,\n  currentLocale: string | undefined,\n  targetUrl: string,\n  totalPageSize: number,\n  emit: EmitScanEvent\n): BundleContentAnalysis | undefined => {\n  const type = urlCheckType('url_unusedBundleContent', targetUrl);\n\n  if (!currentLocale) {\n    emitCheck(\n      emit,\n      type,\n      'warning',\n      'Cannot analyse bundle content: page locale not detected'\n    );\n    return undefined;\n  }\n\n  const analysis = analyzeBundleContent(\n    chunks,\n    html,\n    currentLocale,\n    totalPageSize\n  );\n\n  // Status is driven by the main bundle — lazy chunks with unused content are expected.\n  const mainBundleMaxUnused = analysis.mainBundleChunks.reduce(\n    (max, chunk) => Math.max(max, chunk.unusedPercent),\n    0\n  );\n\n  const status: ScanCheckStatus =\n    mainBundleMaxUnused === 0\n      ? 'success'\n      : mainBundleMaxUnused <= 30\n        ? 'warning'\n        : 'error';\n\n  emitCheck(emit, type, status, formatBundleAnalysis(analysis));\n\n  return analysis;\n};\n\n/* ------------------------------------------------------------------------ */\n/*                            robots.txt & sitemap                           */\n/* ------------------------------------------------------------------------ */\n\n/**\n * Fetch `robots.txt` and check that neither the site nor its localized URLs\n * are blocked for Googlebot.\n *\n * - `robots_robotsPresent`\n * - `robots_noLocalizedUrlsForgotten`\n *\n * @param localizedUrls - URLs of every locale version (hreflang alternates).\n * @returns The sitemaps declared in `robots.txt`.\n */\nexport const checkRobots = async (\n  origin: string,\n  localizedUrls: string[],\n  fetchOptions: ScanFetchOptions,\n  emit: EmitScanEvent\n): Promise<string[]> => {\n  const response = await fetchText(`${origin}/robots.txt`, fetchOptions);\n\n  if (!response?.ok) {\n    emitCheck(\n      emit,\n      'robots_robotsPresent',\n      'warning',\n      response\n        ? `No robots.txt found (HTTP ${response.status})`\n        : 'robots.txt is unreachable'\n    );\n    return [];\n  }\n\n  const robots = parseRobots(response.text);\n  emitCheck(emit, 'robots_robotsPresent', 'success', {\n    disallow: robots.disallowedPaths,\n    sitemaps: robots.sitemapUrls,\n  });\n\n  const blockedUrls = [origin, ...localizedUrls].filter((url) => {\n    const parsedUrl = parseUrl(url);\n    return (\n      parsedUrl?.origin === origin &&\n      !isPathAllowedByRobots(`${parsedUrl.pathname}${parsedUrl.search}`, robots)\n    );\n  });\n\n  const isSiteBlocked = !isPathAllowedByRobots('/', robots);\n  emitCheck(\n    emit,\n    'robots_noLocalizedUrlsForgotten',\n    blockedUrls.length === 0 ? 'success' : 'error',\n    blockedUrls.length === 0\n      ? true\n      : isSiteBlocked\n        ? 'robots.txt blocks the whole site for Googlebot (Disallow: /)'\n        : {\n            message: 'Localized URLs blocked by robots.txt',\n            urls: [...new Set(blockedUrls)],\n          }\n  );\n\n  return robots.sitemapUrls;\n};\n\n/**\n * Read the sitemaps (declared in `robots.txt`, else the default locations,\n * following sitemap indexes) and check their locale coverage.\n *\n * - `sitemap_sitemapPresent`\n * - `sitemap_noLocalizedUrlsForgotten`: every locale is listed, and entries\n *   declaring alternates declare all of them\n * - `sitemap_hasAlternates` / `sitemap_hasXDefault`\n */\nexport const checkSitemap = async (\n  origin: string,\n  declaredSitemapUrls: string[],\n  routing: RoutingDetection,\n  fetchOptions: ScanFetchOptions,\n  emit: EmitScanEvent\n): Promise<void> => {\n  const { fetchedSitemapUrls, entries, isTruncated } =\n    await collectSitemapEntries(\n      declaredSitemapUrls.length > 0\n        ? declaredSitemapUrls\n        : getDefaultSitemapUrls(origin),\n      fetchOptions\n    );\n\n  if (fetchedSitemapUrls.length === 0) {\n    emitCheck(\n      emit,\n      'sitemap_sitemapPresent',\n      'warning',\n      declaredSitemapUrls.length > 0\n        ? `Sitemaps declared in robots.txt are unreachable: ${declaredSitemapUrls.join(', ')}`\n        : 'No sitemap found (robots.txt Sitemap directive, /sitemap.xml, /sitemap_index.xml)'\n    );\n    return;\n  }\n\n  emitCheck(emit, 'sitemap_sitemapPresent', 'success', {\n    sitemaps: fetchedSitemapUrls,\n    urlCount: entries.length,\n    isTruncated,\n  });\n\n  const knownLocales = routing.locales;\n  const foundLanguages = new Set<string>();\n  const entriesWithoutSelfReference: string[] = [];\n  const relativeAlternates: string[] = [];\n\n  let partiallyTranslatedCount = 0;\n  let hasAlternates = false;\n  let hasXDefault = false;\n\n  for (const { loc, alternates } of entries) {\n    const locLocale = getUrlLocale(loc, routing);\n    if (locLocale) foundLanguages.add(getLanguageCode(locLocale));\n    if (alternates.length === 0) continue;\n\n    hasAlternates = true;\n    const entryLanguages = new Set<string>();\n    for (const { hreflang, href } of alternates) {\n      if (!isAbsoluteUrl(href)) {\n        relativeAlternates.push(`${loc}: ${hreflang} → ${href}`);\n      }\n      if (hreflang.toLowerCase() === 'x-default') {\n        hasXDefault = true;\n        continue;\n      }\n      entryLanguages.add(getLanguageCode(hreflang));\n      foundLanguages.add(getLanguageCode(hreflang));\n    }\n\n    // Google requires each <url> to list itself among its alternates.\n    const normalizedLoc = normalizeUrl(loc);\n    if (!alternates.some(({ href }) => normalizeUrl(href) === normalizedLoc)) {\n      entriesWithoutSelfReference.push(loc);\n    }\n    if (entryLanguages.size < knownLocales.length) partiallyTranslatedCount++;\n  }\n\n  // Subdomain / domain sites usually serve one sitemap per host.\n  const isSingleOriginRouting =\n    routing.strategy !== 'subdomain' && routing.strategy !== 'domain';\n\n  if (knownLocales.length > 1) {\n    const missingLocales = isSingleOriginRouting\n      ? knownLocales.filter(\n          (locale) => !foundLanguages.has(getLanguageCode(locale))\n        )\n      : [];\n    const issues = [\n      ...(missingLocales.length > 0\n        ? [`Locales missing from the sitemap: ${missingLocales.join(', ')}`]\n        : []),\n      ...entriesWithoutSelfReference\n        .slice(0, MAX_LISTED_LINKS)\n        .map((loc) => `${loc} is not listed among its own alternates`),\n    ];\n    emitCheck(\n      emit,\n      'sitemap_noLocalizedUrlsForgotten',\n      issues.length === 0 ? 'success' : 'warning',\n      issues.length === 0\n        ? {\n            message: `Every locale (${knownLocales.join(', ')}) is listed`,\n            partiallyTranslatedUrlCount: partiallyTranslatedCount,\n          }\n        : issues\n    );\n  }\n\n  if (!hasAlternates) {\n    emitCheck(\n      emit,\n      'sitemap_hasAlternates',\n      'warning',\n      'No alternate language links found in the sitemap'\n    );\n  } else if (relativeAlternates.length > 0) {\n    emitCheck(emit, 'sitemap_hasAlternates', 'error', {\n      message: `${relativeAlternates.length} sitemap hreflang alternates are not absolute URLs (https://…): search engines ignore them`,\n      alternates: relativeAlternates.slice(0, MAX_LISTED_LINKS),\n    });\n  } else {\n    emitCheck(emit, 'sitemap_hasAlternates', 'success', true);\n  }\n\n  emitCheck(\n    emit,\n    'sitemap_hasXDefault',\n    hasXDefault ? 'success' : 'warning',\n    hasXDefault ? true : 'No x-default hreflang in the sitemap'\n  );\n};\n"],"mappings":";;;;;;;;;;;;AAsDA,MAAM,mBAAmB;;AAEzB,MAAM,yBAAyB;;AAE/B,MAAM,wCAAwB,IAAI,IAAI;CAAC;CAAK;CAAK;AAAG,CAAC;;AAGrD,MAAa,cAAc,UAA0B;CACnD,IAAI,SAAS,SAAa,OAAO,IAAI,QAAS,QAAY,CAAE,QAAQ,CAAC,EAAE;CACvE,IAAI,SAAS,MAAM,OAAO,IAAI,QAAQ,KAAI,CAAE,QAAQ,CAAC,EAAE;CACvD,OAAO,GAAG,MAAM;AAClB;;AAGA,MAAM,gBAAgB,WAAmB,cACvC,GAAG,UAAU,IAAI;;AAGnB,MAAM,aACJ,MACA,MACA,QACA,YAEA,KAAK;CACH;CACA;CACA,SAAS,YAAY,SAAY,SAAY,GAAG,SAAS,QAAQ;AACnE,CAAC;;AAGH,MAAM,gBAAgB,EAAE,MAAM,WAC5B,YAAY,KAAK,IAAI,KAAK,MAAM,GAAG,EAAE,KAAK,OAAO;;AAOnD,MAAa,iBACX,EAAE,WAAW,WACb,SACS;CACT,MAAM,OAAO,aAAa,gBAAgB,SAAS;CACnD,IAAI,CAAC,SACH,UAAU,MAAM,MAAM,SAAS,6BAA6B;MACvD,IAAI,CAAC,kBAAkB,OAAO,GACnC,UACE,MACA,MACA,SACA,IAAI,QAAQ,qEACd;MAEA,UAAU,MAAM,MAAM,WAAW,OAAO;AAE5C;;;;;AAMA,MAAa,gBACX,EAAE,WAAW,SAAS,UACtB,SACS;CACT,MAAM,OAAO,aAAa,eAAe,SAAS;CAClD,MAAM,YAAY,QAAQ,YAAY;CAEtC,IAAI,CAAC,SAAS;EACZ,UACE,MACA,MACA,WACA,wDACF;EACA;CACF;CAEA,MAAM,oBAAoB,oBAAoB,OAAO,IAAI,QAAQ;CAEjE,IACE,cAAc,qBACb,CAAC,aAAa,sBAAsB,OACrC;EACA,UAAU,MAAM,MAAM,WAAW,aAAa,uBAAuB;EACrE;CACF;CAEA,UACE,MACA,MACA,SACA,YACI,QAAQ,UAAU,uBAAuB,kBAAkB,YAAY,EAAE,aAAa,QAAQ,KAC9F,qDAAqD,QAAQ,EACnE;AACF;;;;;AAMA,MAAa,0BACX,EAAE,WAAW,SAAS,WAAW,WACjC,SACS;CACT,MAAM,OAAO,aAAa,qBAAqB,SAAS;CACxD,MAAM,sBAAsB,aAAa,SAAS;CAClD,MAAM,eAAe,UAAU,MAC5B,EAAE,UAAU,WACX,SAAS,YAAY,MAAM,eAC3B,aAAa,MAAM,SAAS,MAAM,mBACtC,CAAC,EAAE;CAEH,MAAM,UAAU;EACd,UAAU;EACV,WAAW,QAAQ;EACnB;CACF;CACA,MAAM,iBAAiB,OAAO,OAAO,OAAO,CAAC,CAAC,QAC3C,WAA6B,QAAQ,MAAM,CAC9C;CAEA,IAAI,eAAe,WAAW,GAAG;EAC/B,UAAU,MAAM,MAAM,WAAW,gCAAgC;EACjE;CACF;CAEA,MAAM,cAAc,eAAe;CACnC,MAAM,eAAe,eAAe,OAAO,WACzC,eAAe,QAAQ,WAAW,CACpC;CAEA,UACE,MACA,MACA,eAAe,YAAY,SAC3B,eACI,UACA;EACE,SACE;EACF,GAAG;CACL,CACN;AACF;;AAGA,MAAM,gCACJ,YACA,SACA,aACW;CACX,MAAM,aAAa,kBAAkB,YAAY,OAAO;CACxD,OAAO,GAAG,SAAS,IAAI,WAAW,iEAChC,aAAa,WAAW,WAAW,KAAK,GACzC;AACH;;;;;;AAOA,MAAa,iBACX,EAAE,WAAW,UAAU,qBAAqB,CAAC,GAAG,WAChD,SACS;CACT,MAAM,OAAO,aAAa,gBAAgB,SAAS;CAEnD,IAAI,CAAC,UAAU;EACb,UACE,MACA,MACA,WACA,yEACF;EACA;CACF;CAEA,IAAI,CAAC,uBAAuB,QAAQ,GAAG;EACrC,UACE,MACA,MACA,SACA,6BAA6B,UAAU,SAAS,WAAW,CAC7D;EACA;CACF;CAEA,IAAI,WAAW,CAAC,eAAe,UAAU,OAAO,GAAG;EACjD,UACE,MACA,MACA,WACA,cAAc,SAAS,8BAA8B,QAAQ,EAC/D;EACA;CACF;CAEA,MAAM,oBAAoB,mBAAmB,QAC1C,cAAc,CAAC,uBAAuB,SAAS,CAClD;CAEA,IAAI,kBAAkB,SAAS,GAAG;EAChC,UACE,MACA,MACA,WACA,kBACG,KAAK,cACJ,6BACE,WACA,QACA,qBACF,CACF,CAAC,CACA,KAAK,IAAI,CACd;EACA;CACF;CAEA,UAAU,MAAM,MAAM,WAAW,QAAQ;AAC3C;;;;;AAUA,MAAa,kBACX,EAAE,WAAW,eAAe,WAAW,SAAS,cAChD,SACS;CACT,MAAM,OAAO,aAAa,oBAAoB,SAAS;CAEvD,IAAI,kBAAkB,UAAa,kBAAkB,IAAI;EACvD,UAAU,MAAM,MAAM,WAAW,wBAAwB;EACzD;CACF;CAEA,MAAM,eAAe,aAAa,eAAe,SAAS;CAC1D,MAAM,uBAAuB,UAAU,MACpC,EAAE,UAAU,WACX,SAAS,YAAY,MAAM,eAC3B,aAAa,MAAM,SAAS,MAAM,gBAClC,CAAC,eAAe,UAAU,UAAU,CACxC;CACA,MAAM,kBAAkB,aAAa,cAAc,OAAO;CAE1D,IACE,wBACC,mBACC,cACA,CAAC,eAAe,iBAAiB,UAAU,GAC7C;EACA,UAAU,MAAM,MAAM,SAAS;GAC7B,SAAS,oCAAoC,sBAAsB,YAAY,gBAAgB,6BAA6B,WAAW;GACvI,WAAW;EACb,CAAC;EACD;CACF;CAEA,UACE,MACA,MACA,WACA,iBAAiB,aAAa,SAAS,IACnC,eACA;EACE,SAAS;EACT,WAAW;CACb,CACN;AACF;;;;;AAMA,MAAa,iBACX,EAAE,WAAW,WAAW,iBACxB,SACS;CACT,MAAM,OAAO,aAAa,gBAAgB,SAAS;CAEnD,IAAI,UAAU,WAAW,GAAG;EAC1B,UAAU,MAAM,MAAM,WAAW,wBAAwB;EACzD;CACF;CAEA,MAAM,SAAmB,CAAC;CAE1B,MAAM,eAAe,UAClB,KAAK,EAAE,eAAe,QAAQ,CAAC,CAC/B,QAAQ,aAAa,CAAC,kBAAkB,QAAQ,CAAC;CACpD,IAAI,aAAa,SAAS,GACxB,OAAO,KAAK,2BAA2B,aAAa,KAAK,IAAI,GAAG;CAGlE,MAAM,UAAU,UACb,KAAK,EAAE,eAAe,QAAQ,CAAC,CAC/B,QAAQ,aAAa,QAAQ,KAAK,QAAQ,CAAC;CAC9C,IAAI,QAAQ,SAAS,GACnB,OAAO,KACL,IAAI,QAAQ,KAAK,IAAI,EAAE,uDACzB;CAIF,MAAM,gBAAgB,UAAU,QAAQ,EAAE,WAAW,CAAC,cAAc,IAAI,CAAC;CACzE,IAAI,cAAc,SAAS,GACzB,OAAO,KACL,sEAAsE,cACnE,KAAK,EAAE,UAAU,WAAW,GAAG,SAAS,KAAK,MAAM,CAAC,CACpD,KAAK,IAAI,GACd;CAGF,MAAM,6BAAa,IAAI,IAAyB;CAChD,KAAK,MAAM,EAAE,UAAU,UAAU,WAAW;EAC1C,MAAM,OAAO,oBAAoB,QAAQ;EACzC,MAAM,OAAO,WAAW,IAAI,IAAI,qBAAK,IAAI,IAAY;EACrD,KAAK,IAAI,aAAa,MAAM,SAAS,CAAC;EACtC,WAAW,IAAI,MAAM,IAAI;CAC3B;CACA,MAAM,mBAAmB,CAAC,GAAG,WAAW,QAAQ,CAAC,CAAC,CAC/C,QAAQ,GAAG,UAAU,KAAK,OAAO,CAAC,CAAC,CACnC,KAAK,CAAC,UAAU,IAAI;CACvB,IAAI,iBAAiB,SAAS,GAC5B,OAAO,KACL,+CAA+C,iBAAiB,KAAK,IAAI,GAC3E;CAGF,MAAM,2BAAW,IAAI,IAAI,CAAC,aAAa,SAAS,CAAC,CAAC;CAClD,IAAI,eAAe,SAAS,IAAI,aAAa,eAAe,SAAS,CAAC;CAMtE,IAAI,CALqB,UAAU,MAChC,EAAE,UAAU,WACX,SAAS,YAAY,MAAM,eAC3B,SAAS,IAAI,aAAa,MAAM,SAAS,CAAC,CAE1B,GAClB,OAAO,KAAK,8DAA8D;CAG5E,MAAM,SACJ,cAAc,SAAS,IACnB,UACA,OAAO,SAAS,IACd,YACA;CAER,UACE,MACA,MACA,QACA,OAAO,SAAS,IAAI;EAAE;EAAQ;CAAU,IAAI,SAC9C;AACF;;AAGA,MAAa,iBACX,EAAE,WAAW,aACb,SACS;CACT,MAAM,OAAO,aAAa,mBAAmB,SAAS;CACtD,MAAM,WAAW,UAAU,MACxB,EAAE,eAAe,SAAS,YAAY,MAAM,WAC/C;CAEA,IAAI,UACF,UAAU,MAAM,MAAM,WAAW,SAAS,IAAI;MACzC,IAAI,UAAU,WAAW,GAC9B,UACE,MACA,MACA,WACA,0CACF;MAEA,UAAU,MAAM,MAAM,SAAS,iCAAiC;AAEpE;;AAYA,MAAM,kBAAkB,OACtB,cACA,UACA,UACA,iBACmC;CACnC,MAAM,WAAW,MAAM,UAAU,cAAc,YAAY;CAC3D,MAAM,eAAe;EAAE,KAAK;EAAc;CAAS;CAEnD,IAAI,CAAC,YAAY,sBAAsB,IAAI,SAAS,MAAM,GACxD,OAAO;EAAE,GAAG;EAAc,SAAS;EAAgB,QAAQ,CAAC;CAAE;CAEhE,IAAI,CAAC,SAAS,IACZ,OAAO;EACL,GAAG;EACH,SAAS;EACT,QAAQ,CAAC,QAAQ,SAAS,QAAQ;CACpC;CAGF,MAAM,SAAmB,CAAC;CAC1B,IAAI,SAAS,YAAY,OAAO,KAAK,gBAAgB,SAAS,UAAU;CAKxE,IAAI,CAHkB,iBAAiB,SAAS,IAAI,CAAC,CAAC,MAAM,EAAE,WAC5D,SAAS,IAAI,aAAa,MAAM,SAAS,QAAQ,CAAC,CAEnC,GAAG,OAAO,KAAK,oCAAoC;CAEpE,MAAM,gBAAgB,gBAAgB,SAAS,IAAI;CACnD,IAAI,iBAAiB,CAAC,eAAe,eAAe,QAAQ,GAC1D,OAAO,KAAK,iBAAiB,cAAc,EAAE;CAG/C,OAAO;EACL,GAAG;EACH,SAAS,OAAO,SAAS,IAAI,YAAY;EACzC;CACF;AACF;;;;;;AAOA,MAAa,2BAA2B,OACtC,EAAE,WAAW,WAAW,iBACxB,cACA,SACkB;CAClB,MAAM,OAAO,aAAa,0BAA0B,SAAS;CAC7D,MAAM,2BAAW,IAAI,IAAI,CAAC,aAAa,SAAS,CAAC,CAAC;CAClD,IAAI,eAAe,SAAS,IAAI,aAAa,eAAe,SAAS,CAAC;CAEtE,MAAM,kCAAkB,IAAI,IAAoB;CAChD,KAAK,MAAM,EAAE,UAAU,UAAU,WAAW;EAC1C,IAAI,SAAS,YAAY,MAAM,aAAa;EAC5C,MAAM,cAAc,SAAS,MAAM,SAAS,CAAC,EAAE;EAC/C,IAAI,CAAC,eAAe,SAAS,IAAI,aAAa,WAAW,CAAC,GAAG;EAC7D,IAAI,CAAC,gBAAgB,IAAI,WAAW,GAClC,gBAAgB,IAAI,aAAa,QAAQ;CAE7C;CAEA,IAAI,gBAAgB,SAAS,GAAG;CAEhC,MAAM,gBAAgB,MAAM,QAAQ,IAClC,CAAC,GAAG,gBAAgB,QAAQ,CAAC,CAAC,CAC3B,MAAM,GAAG,sBAAsB,CAAC,CAChC,KAAK,CAAC,cAAc,cACnB,gBAAgB,cAAc,UAAU,UAAU,YAAY,CAChE,CACJ;CAEA,MAAM,mBAAmB,cACtB,QAAQ,EAAE,cAAc,YAAY,cAAc,CAAC,CACnD,KAAK,EAAE,UAAU,GAAG;CACvB,MAAM,uBAAuB,cAAc,QACxC,EAAE,cAAc,YAAY,YAAY,YAAY,SACvD;CACA,MAAM,gBAAgB,cAAc,SAAS,iBAAiB;CAE9D,IAAI,kBAAkB,GAAG;EACvB,UAAU,MAAM,MAAM,WAAW;GAC/B,SACE;GACF,MAAM;EACR,CAAC;EACD;CACF;CAEA,IAAI,qBAAqB,WAAW,GAAG;EACrC,UACE,MACA,MACA,WACA,GAAG,cAAc,oCACnB;EACA;CACF;CAEA,UACE,MACA,MACA,qBAAqB,MAAM,EAAE,cAAc,YAAY,QAAQ,IAC3D,UACA,WACJ;EACE,SAAS,GAAG,qBAAqB,OAAO,MAAM,cAAc;EAC5D,YAAY,qBAAqB,KAC9B,EAAE,UAAU,KAAK,aAChB,GAAG,SAAS,GAAG,IAAI,IAAI,OAAO,KAAK,IAAI,GAC3C;EACA;CACF,CACF;AACF;;;;;;;;;;AAeA,MAAa,sBACX,SACA,SACS;CACT,MAAM,EAAE,WAAW,SAAS,eAAe;CAC3C,MAAM,EACJ,eACA,iBACA,kBACA,kBACA,kBACE,sBAAsB,OAAO;CACjC,MAAM,aAAa,UACjB,MAAM,MAAM,GAAG,gBAAgB,CAAC,CAAC,IAAI,YAAY;CAEnD,MAAM,qBAAqB,aAAa,yBAAyB,SAAS;CAC1E,MAAM,gBACJ,QAAQ,aAAa,eAAe,QAAQ,aAAa;CAE3D,IAAI,cAAc,WAAW,GAC3B,UAAU,MAAM,oBAAoB,WAAW;EAC7C,SAAS;EACT,OAAO,UAAU,aAAa;CAChC,CAAC;MACI,IAAI,CAAC,iBAAiB,CAAC,YAC5B,UAAU,MAAM,oBAAoB,WAAW;EAC7C,SAAS,qDAAqD,QAAQ,SAAS;EAC/E,OAAO,UAAU,aAAa;CAChC,CAAC;MACI,IAAI,gBAAgB,WAAW,GACpC,UAAU,MAAM,oBAAoB,WAAW;EAC7C,SAAS,eAAe,cAAc,OAAO,iCAAiC,WAAW,sBAAsB,QAAQ,SAAS;EAChI,OAAO,UAAU,aAAa;CAChC,CAAC;MAED,UAAU,MAAM,oBAAoB,WAAW;EAC7C,SAAS,GAAG,gBAAgB,OAAO,MAAM,cAAc,OAAO,gCAAgC,WAAW,sBAAsB,QAAQ,SAAS;EAChJ,OAAO,UAAU,eAAe;CAClC,CAAC;CAGH,IAAI,iBAAiB,cAAc,cAAc,SAAS,GAAG;EAC3D,MAAM,cAAc,CAAC,GAAG,kBAAkB,GAAG,gBAAgB;EAC7D,UACE,MACA,aAAa,2BAA2B,SAAS,GACjD,YAAY,WAAW,IAAI,YAAY,WACvC,YAAY,WAAW,IACnB,OAAO,cAAc,OAAO,4BAA4B,WAAW,YACnE;GACE,SAAS,GAAG,YAAY,OAAO,6BAA6B,WAAW,YAAY,iBAAiB,OAAO,sBAAsB,iBAAiB,OAAO;GACzJ,OAAO,UAAU,WAAW;EAC9B,CACN;CACF;CAEA,IAAI,QAAQ,QAAQ,SAAS,GAC3B,UACE,MACA,aAAa,uBAAuB,SAAS,GAC7C,cAAc,SAAS,IAAI,YAAY,WACvC,cAAc,SAAS,IACnB;EACE,SAAS,GAAG,cAAc,OAAO;EACjC,OAAO,UAAU,aAAa;CAChC,IACA,wGACN;AAEJ;AAMA,MAAM,eAAe,QACnB,IAAI,MAAM,GAAG,CAAC,CAAC,IAAI,CAAC,EAAE,MAAM,GAAG,CAAC,CAAC,MAAM;AAEzC,MAAM,eAAe,WAA0B;CAC7C,UAAU,YAAY,MAAM,GAAG;CAC/B,KAAK,MAAM;CACX,UAAU,WAAW,MAAM,QAAQ;CACnC,iBAAiB,WAAW,MAAM,eAAe;CACjD,gBAAgB,WAAW,MAAM,cAAc;CAC/C,kBAAkB,WAAW,MAAM,gBAAgB;CACnD,mBAAmB,MAAM;CACzB,eAAe,GAAG,MAAM,cAAc;AACxC;;AAGA,MAAa,wBAAwB,cAAqC;CACxE,eAAe,SAAS;CACxB,eAAe,WAAW,SAAS,aAAa;CAChD,qBAAqB,WAAW,SAAS,mBAAmB;CAC5D,aAAa,WAAW,SAAS,WAAW;CAC5C,iBAAiB,WAAW,SAAS,eAAe;CACpD,uBAAuB,WAAW,SAAS,qBAAqB;CAChE,uBAAuB,GAAG,SAAS,sBAAsB;CACzD,kBAAkB,SAAS,iBAAiB,IAAI,WAAW;CAC3D,kBAAkB,SAAS,iBAAiB,IAAI,WAAW;AAC7D;;;;;AAMA,MAAa,sBACX,QACA,MACA,eACA,WACA,eACA,SACsC;CACtC,MAAM,OAAO,aAAa,2BAA2B,SAAS;CAE9D,IAAI,CAAC,eAAe;EAClB,UACE,MACA,MACA,WACA,yDACF;EACA;CACF;CAEA,MAAM,WAAW,qBACf,QACA,MACA,eACA,aACF;CAGA,MAAM,sBAAsB,SAAS,iBAAiB,QACnD,KAAK,UAAU,KAAK,IAAI,KAAK,MAAM,aAAa,GACjD,CACF;CASA,UAAU,MAAM,MANd,wBAAwB,IACpB,YACA,uBAAuB,KACrB,YACA,SAEsB,qBAAqB,QAAQ,CAAC;CAE5D,OAAO;AACT;;;;;;;;;;;AAgBA,MAAa,cAAc,OACzB,QACA,eACA,cACA,SACsB;CACtB,MAAM,WAAW,MAAM,UAAU,GAAG,OAAO,cAAc,YAAY;CAErE,IAAI,CAAC,UAAU,IAAI;EACjB,UACE,MACA,wBACA,WACA,WACI,6BAA6B,SAAS,OAAO,KAC7C,2BACN;EACA,OAAO,CAAC;CACV;CAEA,MAAM,SAAS,YAAY,SAAS,IAAI;CACxC,UAAU,MAAM,wBAAwB,WAAW;EACjD,UAAU,OAAO;EACjB,UAAU,OAAO;CACnB,CAAC;CAED,MAAM,cAAc,CAAC,QAAQ,GAAG,aAAa,CAAC,CAAC,QAAQ,QAAQ;EAC7D,MAAM,YAAY,SAAS,GAAG;EAC9B,OACE,WAAW,WAAW,UACtB,CAAC,sBAAsB,GAAG,UAAU,WAAW,UAAU,UAAU,MAAM;CAE7E,CAAC;CAED,MAAM,gBAAgB,CAAC,sBAAsB,KAAK,MAAM;CACxD,UACE,MACA,mCACA,YAAY,WAAW,IAAI,YAAY,SACvC,YAAY,WAAW,IACnB,OACA,gBACE,iEACA;EACE,SAAS;EACT,MAAM,CAAC,GAAG,IAAI,IAAI,WAAW,CAAC;CAChC,CACR;CAEA,OAAO,OAAO;AAChB;;;;;;;;;;AAWA,MAAa,eAAe,OAC1B,QACA,qBACA,SACA,cACA,SACkB;CAClB,MAAM,EAAE,oBAAoB,SAAS,gBACnC,MAAM,sBACJ,oBAAoB,SAAS,IACzB,sBACA,sBAAsB,MAAM,GAChC,YACF;CAEF,IAAI,mBAAmB,WAAW,GAAG;EACnC,UACE,MACA,0BACA,WACA,oBAAoB,SAAS,IACzB,oDAAoD,oBAAoB,KAAK,IAAI,MACjF,mFACN;EACA;CACF;CAEA,UAAU,MAAM,0BAA0B,WAAW;EACnD,UAAU;EACV,UAAU,QAAQ;EAClB;CACF,CAAC;CAED,MAAM,eAAe,QAAQ;CAC7B,MAAM,iCAAiB,IAAI,IAAY;CACvC,MAAM,8BAAwC,CAAC;CAC/C,MAAM,qBAA+B,CAAC;CAEtC,IAAI,2BAA2B;CAC/B,IAAI,gBAAgB;CACpB,IAAI,cAAc;CAElB,KAAK,MAAM,EAAE,KAAK,gBAAgB,SAAS;EACzC,MAAM,YAAY,aAAa,KAAK,OAAO;EAC3C,IAAI,WAAW,eAAe,IAAI,gBAAgB,SAAS,CAAC;EAC5D,IAAI,WAAW,WAAW,GAAG;EAE7B,gBAAgB;EAChB,MAAM,iCAAiB,IAAI,IAAY;EACvC,KAAK,MAAM,EAAE,UAAU,UAAU,YAAY;GAC3C,IAAI,CAAC,cAAc,IAAI,GACrB,mBAAmB,KAAK,GAAG,IAAI,IAAI,SAAS,KAAK,MAAM;GAEzD,IAAI,SAAS,YAAY,MAAM,aAAa;IAC1C,cAAc;IACd;GACF;GACA,eAAe,IAAI,gBAAgB,QAAQ,CAAC;GAC5C,eAAe,IAAI,gBAAgB,QAAQ,CAAC;EAC9C;EAGA,MAAM,gBAAgB,aAAa,GAAG;EACtC,IAAI,CAAC,WAAW,MAAM,EAAE,WAAW,aAAa,IAAI,MAAM,aAAa,GACrE,4BAA4B,KAAK,GAAG;EAEtC,IAAI,eAAe,OAAO,aAAa,QAAQ;CACjD;CAGA,MAAM,wBACJ,QAAQ,aAAa,eAAe,QAAQ,aAAa;CAE3D,IAAI,aAAa,SAAS,GAAG;EAC3B,MAAM,iBAAiB,wBACnB,aAAa,QACV,WAAW,CAAC,eAAe,IAAI,gBAAgB,MAAM,CAAC,CACzD,IACA,CAAC;EACL,MAAM,SAAS,CACb,GAAI,eAAe,SAAS,IACxB,CAAC,qCAAqC,eAAe,KAAK,IAAI,GAAG,IACjE,CAAC,GACL,GAAG,4BACA,MAAM,GAAG,gBAAgB,CAAC,CAC1B,KAAK,QAAQ,GAAG,IAAI,wCAAwC,CACjE;EACA,UACE,MACA,oCACA,OAAO,WAAW,IAAI,YAAY,WAClC,OAAO,WAAW,IACd;GACE,SAAS,iBAAiB,aAAa,KAAK,IAAI,EAAE;GAClD,6BAA6B;EAC/B,IACA,MACN;CACF;CAEA,IAAI,CAAC,eACH,UACE,MACA,yBACA,WACA,kDACF;MACK,IAAI,mBAAmB,SAAS,GACrC,UAAU,MAAM,yBAAyB,SAAS;EAChD,SAAS,GAAG,mBAAmB,OAAO;EACtC,YAAY,mBAAmB,MAAM,GAAG,gBAAgB;CAC1D,CAAC;MAED,UAAU,MAAM,yBAAyB,WAAW,IAAI;CAG1D,UACE,MACA,uBACA,cAAc,YAAY,WAC1B,cAAc,OAAO,sCACvB;AACF"}

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

@@ -0,0 +1,54 @@
+//#region src/scan/detection/checkDetails.ts
+/**
+* Split the details attached to a scan check into its message, the listed
+* links (HTML source) and the other listed items. Shapes without any of them
+* (bundle summaries, hreflang lists…) return empty lists.
+*
+* @param details - The `success` / `warning` / `error` payload of a check.
+*/
+const splitCheckDetails = (details) => {
+	if (typeof details === "string") return {
+		message: details,
+		links: [],
+		items: []
+	};
+	if (Array.isArray(details)) return {
+		links: [],
+		items: details.map(String)
+	};
+	if (!details || typeof details !== "object") return {
+		links: [],
+		items: []
+	};
+	const { message, issues, links, alternates, urls } = details;
+	const listedItems = [
+		issues,
+		alternates,
+		urls
+	].find(Array.isArray);
+	return {
+		message: typeof message === "string" ? message : void 0,
+		links: Array.isArray(links) ? links.map((link) => String(link).replace(/`{3,4}(?:html)?\n?/g, "")) : [],
+		items: (listedItems ?? []).map(String)
+	};
+};
+/**
+* Turn the details attached to a scan check into short printable lines: its
+* message, then the listed issues / links / URLs. Shared by the CLI report and
+* the Chrome extension popup.
+*
+* @param details - The `success` / `warning` / `error` payload of a check.
+* @returns The message first, then one line per listed item.
+*/
+const getCheckDetailLines = (details) => {
+	const { message, links, items } = splitCheckDetails(details);
+	return [
+		...message ? [message] : [],
+		...links,
+		...items
+	];
+};
+
+//#endregion
+export { getCheckDetailLines, splitCheckDetails };
+//# sourceMappingURL=checkDetails.mjs.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"checkDetails.mjs","names":[],"sources":["../../../../src/scan/detection/checkDetails.ts"],"sourcesContent":["/** Details of a scan check, split for display. */\nexport type CheckDetailParts = {\n  /** Explanation of the check result. */\n  message?: string;\n  /** Offending `<a>` elements, as HTML source. */\n  links: string[];\n  /** Other listed items: issues, hreflang alternates, URLs… */\n  items: string[];\n};\n\n/**\n * Split the details attached to a scan check into its message, the listed\n * links (HTML source) and the other listed items. Shapes without any of them\n * (bundle summaries, hreflang lists…) return empty lists.\n *\n * @param details - The `success` / `warning` / `error` payload of a check.\n */\nexport const splitCheckDetails = (details: unknown): CheckDetailParts => {\n  if (typeof details === 'string') {\n    return { message: details, links: [], items: [] };\n  }\n  if (Array.isArray(details)) {\n    return { links: [], items: details.map(String) };\n  }\n  if (!details || typeof details !== 'object') return { links: [], items: [] };\n\n  const { message, issues, links, alternates, urls } = details as Record<\n    string,\n    unknown\n  >;\n  const listedItems = [issues, alternates, urls].find(Array.isArray) as\n    | unknown[]\n    | undefined;\n\n  return {\n    message: typeof message === 'string' ? message : undefined,\n    links: Array.isArray(links)\n      ? links.map((link) => String(link).replace(/`{3,4}(?:html)?\\n?/g, ''))\n      : [],\n    items: (listedItems ?? []).map(String),\n  };\n};\n\n/**\n * Turn the details attached to a scan check into short printable lines: its\n * message, then the listed issues / links / URLs. Shared by the CLI report and\n * the Chrome extension popup.\n *\n * @param details - The `success` / `warning` / `error` payload of a check.\n * @returns The message first, then one line per listed item.\n */\nexport const getCheckDetailLines = (details: unknown): string[] => {\n  const { message, links, items } = splitCheckDetails(details);\n\n  return [...(message ? [message] : []), ...links, ...items];\n};\n"],"mappings":";;;;;;;;AAiBA,MAAa,qBAAqB,YAAuC;CACvE,IAAI,OAAO,YAAY,UACrB,OAAO;EAAE,SAAS;EAAS,OAAO,CAAC;EAAG,OAAO,CAAC;CAAE;CAElD,IAAI,MAAM,QAAQ,OAAO,GACvB,OAAO;EAAE,OAAO,CAAC;EAAG,OAAO,QAAQ,IAAI,MAAM;CAAE;CAEjD,IAAI,CAAC,WAAW,OAAO,YAAY,UAAU,OAAO;EAAE,OAAO,CAAC;EAAG,OAAO,CAAC;CAAE;CAE3E,MAAM,EAAE,SAAS,QAAQ,OAAO,YAAY,SAAS;CAIrD,MAAM,cAAc;EAAC;EAAQ;EAAY;CAAI,CAAC,CAAC,KAAK,MAAM,OAAO;CAIjE,OAAO;EACL,SAAS,OAAO,YAAY,WAAW,UAAU;EACjD,OAAO,MAAM,QAAQ,KAAK,IACtB,MAAM,KAAK,SAAS,OAAO,IAAI,CAAC,CAAC,QAAQ,uBAAuB,EAAE,CAAC,IACnE,CAAC;EACL,QAAQ,eAAe,CAAC,EAAC,CAAE,IAAI,MAAM;CACvC;AACF;;;;;;;;;AAUA,MAAa,uBAAuB,YAA+B;CACjE,MAAM,EAAE,SAAS,OAAO,UAAU,kBAAkB,OAAO;CAE3D,OAAO;EAAC,GAAI,UAAU,CAAC,OAAO,IAAI,CAAC;EAAI,GAAG;EAAO,GAAG;CAAK;AAC3D"}

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

@@ -0,0 +1,46 @@
+import { normalizeHostname, normalizeUrl, parseUrl } from "./url.mjs";
+import { isSameLanguage } from "./localeCode.mjs";
+import { getUrlLocale } from "./detectRoutingStrategy.mjs";
+
+//#region src/scan/detection/classifyInternalLinks.ts
+const NON_PAGE_EXTENSION_PATTERN = /\.(?:pdf|png|jpe?g|gif|svg|webp|avif|ico|zip|xml|txt|json|css|js|mp4|webm|mp3)$/i;
+/**
+* Group the internal anchors of a page by the locale they lead to, reading
+* each URL through the detected routing strategy (so unprefixed links count as
+* the default locale on a `prefix-no-default` site, and `?lang=` is understood
+* on a `search-params` site).
+*/
+const classifyInternalLinks = ({ targetUrl, anchors, hreflangs, routing, pageLocale }) => {
+	const targetHostname = normalizeHostname(new URL(targetUrl).hostname);
+	const localeHostnames = new Set(Object.keys(routing.hostLocales ?? {}));
+	const otherLocaleAlternateUrls = new Set(hreflangs.filter(({ hreflang }) => hreflang.toLowerCase() !== "x-default" && !isSameLanguage(hreflang, pageLocale)).map(({ href }) => normalizeUrl(href, targetUrl)));
+	const classification = {
+		internalLinks: [],
+		sameLocaleLinks: [],
+		otherLocaleLinks: [],
+		unlocalizedLinks: [],
+		switcherLinks: []
+	};
+	for (const anchor of anchors) {
+		if (/^(?:#|javascript:|mailto:|tel:)/i.test(anchor.href)) continue;
+		const url = parseUrl(anchor.href, targetUrl);
+		if (!url || !/^https?:$/.test(url.protocol)) continue;
+		const hostname = normalizeHostname(url.hostname);
+		if (hostname !== targetHostname && !localeHostnames.has(hostname)) continue;
+		if (NON_PAGE_EXTENSION_PATTERN.test(url.pathname)) continue;
+		if (anchor.hreflang && !isSameLanguage(anchor.hreflang, pageLocale) || otherLocaleAlternateUrls.has(normalizeUrl(url.href))) {
+			classification.switcherLinks.push(anchor);
+			continue;
+		}
+		classification.internalLinks.push(anchor);
+		const linkLocale = getUrlLocale(url.href, routing);
+		if (!linkLocale) classification.unlocalizedLinks.push(anchor);
+		else if (isSameLanguage(linkLocale, pageLocale)) classification.sameLocaleLinks.push(anchor);
+		else classification.otherLocaleLinks.push(anchor);
+	}
+	return classification;
+};
+
+//#endregion
+export { classifyInternalLinks };
+//# sourceMappingURL=classifyInternalLinks.mjs.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"classifyInternalLinks.mjs","names":[],"sources":["../../../../src/scan/detection/classifyInternalLinks.ts"],"sourcesContent":["import { getUrlLocale, type RoutingDetection } from './detectRoutingStrategy';\nimport { isSameLanguage } from './localeCode';\nimport { normalizeHostname, normalizeUrl, parseUrl } from './url';\n\n/** A page anchor, as parsed from HTML or read from a live DOM. */\nexport type LinkAnchor = {\n  href: string;\n  text: string;\n  /** `hreflang` attribute, set on language-switcher links. */\n  hreflang?: string;\n};\n\n/** Input of {@link classifyInternalLinks}. */\nexport type LinkClassificationInput<Anchor extends LinkAnchor> = {\n  /** Absolute URL of the page holding the anchors. */\n  targetUrl: string;\n  anchors: Anchor[];\n  hreflangs: { hreflang: string; href: string }[];\n  routing: RoutingDetection;\n  /** Locale of the page. */\n  pageLocale?: string;\n};\n\n/** Internal anchors of a page, grouped by the locale they lead to. */\nexport type LinkClassification<Anchor extends LinkAnchor> = {\n  /** Internal page links, language-switcher links excluded. */\n  internalLinks: Anchor[];\n  /** Internal links keeping the page locale. */\n  sameLocaleLinks: Anchor[];\n  /** Internal links leading to another locale. */\n  otherLocaleLinks: Anchor[];\n  /** Internal links whose URL carries no locale. */\n  unlocalizedLinks: Anchor[];\n  /** Links to the other locale versions (hreflang alternates). */\n  switcherLinks: Anchor[];\n};\n\nconst NON_PAGE_EXTENSION_PATTERN =\n  /\\.(?:pdf|png|jpe?g|gif|svg|webp|avif|ico|zip|xml|txt|json|css|js|mp4|webm|mp3)$/i;\n\n/**\n * Group the internal anchors of a page by the locale they lead to, reading\n * each URL through the detected routing strategy (so unprefixed links count as\n * the default locale on a `prefix-no-default` site, and `?lang=` is understood\n * on a `search-params` site).\n */\nexport const classifyInternalLinks = <Anchor extends LinkAnchor>({\n  targetUrl,\n  anchors,\n  hreflangs,\n  routing,\n  pageLocale,\n}: LinkClassificationInput<Anchor>): LinkClassification<Anchor> => {\n  const targetHostname = normalizeHostname(new URL(targetUrl).hostname);\n  const localeHostnames = new Set(Object.keys(routing.hostLocales ?? {}));\n  const otherLocaleAlternateUrls = new Set(\n    hreflangs\n      .filter(\n        ({ hreflang }) =>\n          hreflang.toLowerCase() !== 'x-default' &&\n          !isSameLanguage(hreflang, pageLocale)\n      )\n      .map(({ href }) => normalizeUrl(href, targetUrl))\n  );\n\n  const classification: LinkClassification<Anchor> = {\n    internalLinks: [],\n    sameLocaleLinks: [],\n    otherLocaleLinks: [],\n    unlocalizedLinks: [],\n    switcherLinks: [],\n  };\n\n  for (const anchor of anchors) {\n    if (/^(?:#|javascript:|mailto:|tel:)/i.test(anchor.href)) continue;\n    const url = parseUrl(anchor.href, targetUrl);\n    if (!url || !/^https?:$/.test(url.protocol)) continue;\n\n    const hostname = normalizeHostname(url.hostname);\n    if (hostname !== targetHostname && !localeHostnames.has(hostname)) continue;\n    if (NON_PAGE_EXTENSION_PATTERN.test(url.pathname)) continue;\n\n    if (\n      (anchor.hreflang && !isSameLanguage(anchor.hreflang, pageLocale)) ||\n      otherLocaleAlternateUrls.has(normalizeUrl(url.href))\n    ) {\n      classification.switcherLinks.push(anchor);\n      continue;\n    }\n\n    classification.internalLinks.push(anchor);\n    const linkLocale = getUrlLocale(url.href, routing);\n    if (!linkLocale) {\n      classification.unlocalizedLinks.push(anchor);\n    } else if (isSameLanguage(linkLocale, pageLocale)) {\n      classification.sameLocaleLinks.push(anchor);\n    } else {\n      classification.otherLocaleLinks.push(anchor);\n    }\n  }\n\n  return classification;\n};\n"],"mappings":";;;;;AAqCA,MAAM,6BACJ;;;;;;;AAQF,MAAa,yBAAoD,EAC/D,WACA,SACA,WACA,SACA,iBACiE;CACjE,MAAM,iBAAiB,kBAAkB,IAAI,IAAI,SAAS,CAAC,CAAC,QAAQ;CACpE,MAAM,kBAAkB,IAAI,IAAI,OAAO,KAAK,QAAQ,eAAe,CAAC,CAAC,CAAC;CACtE,MAAM,2BAA2B,IAAI,IACnC,UACG,QACE,EAAE,eACD,SAAS,YAAY,MAAM,eAC3B,CAAC,eAAe,UAAU,UAAU,CACxC,CAAC,CACA,KAAK,EAAE,WAAW,aAAa,MAAM,SAAS,CAAC,CACpD;CAEA,MAAM,iBAA6C;EACjD,eAAe,CAAC;EAChB,iBAAiB,CAAC;EAClB,kBAAkB,CAAC;EACnB,kBAAkB,CAAC;EACnB,eAAe,CAAC;CAClB;CAEA,KAAK,MAAM,UAAU,SAAS;EAC5B,IAAI,mCAAmC,KAAK,OAAO,IAAI,GAAG;EAC1D,MAAM,MAAM,SAAS,OAAO,MAAM,SAAS;EAC3C,IAAI,CAAC,OAAO,CAAC,YAAY,KAAK,IAAI,QAAQ,GAAG;EAE7C,MAAM,WAAW,kBAAkB,IAAI,QAAQ;EAC/C,IAAI,aAAa,kBAAkB,CAAC,gBAAgB,IAAI,QAAQ,GAAG;EACnE,IAAI,2BAA2B,KAAK,IAAI,QAAQ,GAAG;EAEnD,IACG,OAAO,YAAY,CAAC,eAAe,OAAO,UAAU,UAAU,KAC/D,yBAAyB,IAAI,aAAa,IAAI,IAAI,CAAC,GACnD;GACA,eAAe,cAAc,KAAK,MAAM;GACxC;EACF;EAEA,eAAe,cAAc,KAAK,MAAM;EACxC,MAAM,aAAa,aAAa,IAAI,MAAM,OAAO;EACjD,IAAI,CAAC,YACH,eAAe,iBAAiB,KAAK,MAAM;OACtC,IAAI,eAAe,YAAY,UAAU,GAC9C,eAAe,gBAAgB,KAAK,MAAM;OAE1C,eAAe,iBAAiB,KAAK,MAAM;CAE/C;CAEA,OAAO;AACT"}

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

@@ -0,0 +1,141 @@
+import { getFirstPathSegment, normalizeHostname, normalizeUrl, parseUrl } from "./url.mjs";
+import { findMatchingLocale, looksLikeLocaleCode } from "./localeCode.mjs";
+
+//#region src/scan/detection/detectRoutingStrategy.ts
+/** Locale of a URL according to an already-detected routing strategy. */
+const getUrlLocale = (url, routing) => {
+	const parsedUrl = parseUrl(url);
+	if (!parsedUrl) return void 0;
+	switch (routing.strategy) {
+		case "prefix-all":
+		case "prefix-no-default": return findMatchingLocale(getFirstPathSegment(parsedUrl.pathname), routing.locales) ?? (routing.strategy === "prefix-no-default" ? routing.defaultLocale : void 0);
+		case "search-params": return findMatchingLocale(parsedUrl.searchParams.get(routing.searchParamName ?? ""), routing.locales) ?? routing.defaultLocale;
+		case "subdomain":
+		case "domain": return routing.hostLocales?.[normalizeHostname(parsedUrl.hostname)] ?? findMatchingLocale(parsedUrl.hostname.split(".")[0], routing.locales);
+		default: return;
+	}
+};
+/** Detect the strategy from at least two hreflang alternates. */
+const detectFromAlternates = (localizedUrls, locales, xDefaultUrl) => {
+	const alternateCount = localizedUrls.length;
+	const findDefaultLocale = (unmarked) => unmarked.length === 1 ? unmarked[0]?.locale : localizedUrls.find(({ url }) => xDefaultUrl && normalizeUrl(url.href) === normalizeUrl(xDefaultUrl))?.locale;
+	const hostnames = new Set(localizedUrls.map(({ url }) => normalizeHostname(url.hostname)));
+	if (hostnames.size > 1) {
+		const hostLocales = Object.fromEntries(localizedUrls.map(({ url, locale }) => [normalizeHostname(url.hostname), locale]));
+		const subdomainPrefixed = localizedUrls.filter(({ url, locale }) => findMatchingLocale(url.hostname.split(".")[0], [locale]) !== void 0);
+		const isSubdomain = subdomainPrefixed.length * 2 >= localizedUrls.length;
+		return {
+			strategy: isSubdomain ? "subdomain" : "domain",
+			confidence: "high",
+			hostLocales,
+			defaultLocale: isSubdomain ? findDefaultLocale(localizedUrls.filter((entry) => !subdomainPrefixed.includes(entry))) : void 0,
+			evidence: `hreflang alternates use ${hostnames.size} different hosts`
+		};
+	}
+	const parameterMatchCount = /* @__PURE__ */ new Map();
+	for (const { url, locale } of localizedUrls) for (const [parameterName, parameterValue] of url.searchParams) if (findMatchingLocale(parameterValue, [locale])) parameterMatchCount.set(parameterName, (parameterMatchCount.get(parameterName) ?? 0) + 1);
+	const [searchParamName, searchParamCount] = [...parameterMatchCount.entries()].sort((first, second) => second[1] - first[1])[0] ?? ["", 0];
+	if (searchParamCount >= Math.max(alternateCount - 1, 1)) return {
+		strategy: "search-params",
+		confidence: "high",
+		searchParamName,
+		defaultLocale: findDefaultLocale(localizedUrls.filter(({ url }) => !url.searchParams.has(searchParamName))),
+		evidence: `hreflang alternates carry the locale in the "?${searchParamName}=" parameter`
+	};
+	const unprefixed = localizedUrls.filter(({ url, locale }) => !findMatchingLocale(getFirstPathSegment(url.pathname), [locale]));
+	if (unprefixed.length === 0) return {
+		strategy: "prefix-all",
+		confidence: "high",
+		defaultLocale: findDefaultLocale([]),
+		evidence: "every hreflang alternate starts with a locale segment"
+	};
+	if (unprefixed.length === 1) return {
+		strategy: "prefix-no-default",
+		confidence: "high",
+		defaultLocale: unprefixed[0]?.locale,
+		evidence: `every hreflang alternate but "${unprefixed[0]?.locale}" starts with a locale segment`
+	};
+	if (new Set(localizedUrls.map(({ url }) => normalizeUrl(url.href))).size === 1) return {
+		strategy: "no-prefix",
+		confidence: "high",
+		evidence: `all ${locales.length} hreflang alternates point to the same URL`
+	};
+	return {
+		strategy: "unknown",
+		confidence: "high",
+		evidence: "hreflang alternates use different URLs without a locale marker (localized slugs?)"
+	};
+};
+/** Guess the strategy from the scanned URL when no alternates are declared. */
+const detectFromPageUrl = (pageUrl, htmlLang) => {
+	const firstSegment = getFirstPathSegment(pageUrl.pathname);
+	const pageLocales = htmlLang ? [htmlLang] : [];
+	if (firstSegment && (findMatchingLocale(firstSegment, pageLocales) || !htmlLang && looksLikeLocaleCode(firstSegment))) return {
+		strategy: "prefix-all",
+		confidence: "low",
+		evidence: `the page URL starts with the locale segment "/${firstSegment}"`
+	};
+	for (const [parameterName, parameterValue] of pageUrl.searchParams) if (findMatchingLocale(parameterValue, pageLocales)) return {
+		strategy: "search-params",
+		confidence: "low",
+		searchParamName: parameterName,
+		evidence: `the page URL carries the locale in "?${parameterName}="`
+	};
+	if (findMatchingLocale(pageUrl.hostname.split(".")[0], pageLocales)) return {
+		strategy: "subdomain",
+		confidence: "low",
+		hostLocales: { [normalizeHostname(pageUrl.hostname)]: htmlLang ?? "" },
+		evidence: "the page hostname starts with the locale"
+	};
+	return {
+		strategy: htmlLang ? "no-prefix" : "unknown",
+		confidence: "low",
+		evidence: htmlLang ? "no hreflang alternates and no locale marker in the page URL" : "no hreflang alternates, no lang attribute and no locale marker in the URL"
+	};
+};
+/**
+* Detect how the site encodes the locale in its URLs (path prefix, query
+* parameter, subdomain, domain or nothing).
+*
+* The hreflang alternates are the most reliable signal: each one pairs a
+* locale with its URL, so the shared pattern reveals the strategy. Without
+* alternates the scanned URL alone is used, with `confidence: 'low'`.
+*/
+const detectRoutingStrategy = ({ pageUrl, htmlLang, hreflangs }) => {
+	const parsedPageUrl = parseUrl(pageUrl);
+	const localizedUrls = [];
+	for (const { hreflang, href } of hreflangs) {
+		if (hreflang.toLowerCase() === "x-default") continue;
+		const url = parseUrl(href, pageUrl);
+		if (url) localizedUrls.push({
+			locale: hreflang,
+			url
+		});
+	}
+	const xDefaultHref = hreflangs.find(({ hreflang }) => hreflang.toLowerCase() === "x-default")?.href;
+	const xDefaultUrl = xDefaultHref ? parseUrl(xDefaultHref, pageUrl)?.href : void 0;
+	const locales = [...new Set(localizedUrls.map(({ locale }) => locale))];
+	const detection = localizedUrls.length >= 2 ? {
+		...detectFromAlternates(localizedUrls, locales, xDefaultUrl),
+		locales
+	} : {
+		...parsedPageUrl ? detectFromPageUrl(parsedPageUrl, htmlLang) : {
+			strategy: "unknown",
+			confidence: "low",
+			evidence: "invalid page URL"
+		},
+		locales: htmlLang ? [htmlLang] : locales
+	};
+	if (detection.strategy === "prefix-all" && detection.locales.length === 0) {
+		const firstSegment = parsedPageUrl ? getFirstPathSegment(parsedPageUrl.pathname) : void 0;
+		if (firstSegment) detection.locales = [firstSegment];
+	}
+	return {
+		...detection,
+		urlLocale: getUrlLocale(pageUrl, detection)
+	};
+};
+
+//#endregion
+export { detectRoutingStrategy, getUrlLocale };
+//# sourceMappingURL=detectRoutingStrategy.mjs.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"detectRoutingStrategy.mjs","names":[],"sources":["../../../../src/scan/detection/detectRoutingStrategy.ts"],"sourcesContent":["import { findMatchingLocale, looksLikeLocaleCode } from './localeCode';\nimport {\n  getFirstPathSegment,\n  normalizeHostname,\n  normalizeUrl,\n  parseUrl,\n} from './url';\n\n/**\n * How a website exposes its locales in URLs. Mirrors Intlayer's\n * `routing.mode` values, plus the domain-based strategies.\n *\n * - `prefix-all`: every locale is prefixed (`/en/about`, `/fr/about`)\n * - `prefix-no-default`: the default locale is not prefixed (`/about`, `/fr/about`)\n * - `search-params`: the locale is a query parameter (`/about?lang=fr`)\n * - `subdomain`: the locale is a subdomain (`fr.example.com`)\n * - `domain`: one domain per locale (`example.fr`, `example.de`)\n * - `no-prefix`: a single URL serves every locale (cookie / header based)\n * - `unknown`: not enough signals to decide\n */\nexport type RoutingStrategy =\n  | 'prefix-all'\n  | 'prefix-no-default'\n  | 'search-params'\n  | 'subdomain'\n  | 'domain'\n  | 'no-prefix'\n  | 'unknown';\n\n/** Result of {@link detectRoutingStrategy}. */\nexport type RoutingDetection = {\n  strategy: RoutingStrategy;\n  /**\n   * `high` when inferred from the hreflang alternates, `low` when guessed from\n   * the scanned URL alone.\n   */\n  confidence: 'high' | 'low';\n  /** Locales known to the site (hreflang values, else the page `lang`). */\n  locales: string[];\n  /** Locale served without any marker (`prefix-no-default`, `search-params`…). */\n  defaultLocale?: string;\n  /** Query parameter carrying the locale, for `search-params`. */\n  searchParamName?: string;\n  /** Hostname → locale map, for `subdomain` / `domain`. */\n  hostLocales?: Record<string, string>;\n  /** Locale of the scanned page according to its URL alone. */\n  urlLocale?: string;\n  /** Human-readable explanation of the decision. */\n  evidence: string;\n};\n\n/** Input of {@link detectRoutingStrategy}. */\nexport type RoutingDetectionInput = {\n  /** Absolute URL of the scanned page. */\n  pageUrl: string;\n  /** Value of `<html lang>`, when present. */\n  htmlLang?: string;\n  /** `<link rel=\"alternate\" hreflang>` entries of the page. */\n  hreflangs: { hreflang: string; href: string }[];\n};\n\ntype LocalizedUrl = { locale: string; url: URL };\n\n/** Locale of a URL according to an already-detected routing strategy. */\nexport const getUrlLocale = (\n  url: string,\n  routing: RoutingDetection\n): string | undefined => {\n  const parsedUrl = parseUrl(url);\n  if (!parsedUrl) return undefined;\n\n  switch (routing.strategy) {\n    case 'prefix-all':\n    case 'prefix-no-default':\n      return (\n        findMatchingLocale(\n          getFirstPathSegment(parsedUrl.pathname),\n          routing.locales\n        ) ??\n        (routing.strategy === 'prefix-no-default'\n          ? routing.defaultLocale\n          : undefined)\n      );\n    case 'search-params':\n      return (\n        findMatchingLocale(\n          parsedUrl.searchParams.get(routing.searchParamName ?? ''),\n          routing.locales\n        ) ?? routing.defaultLocale\n      );\n    case 'subdomain':\n    case 'domain':\n      return (\n        routing.hostLocales?.[normalizeHostname(parsedUrl.hostname)] ??\n        findMatchingLocale(parsedUrl.hostname.split('.')[0], routing.locales)\n      );\n    default:\n      return undefined;\n  }\n};\n\n/** Detect the strategy from at least two hreflang alternates. */\nconst detectFromAlternates = (\n  localizedUrls: LocalizedUrl[],\n  locales: string[],\n  xDefaultUrl: string | undefined\n): Omit<RoutingDetection, 'urlLocale' | 'locales'> => {\n  const alternateCount = localizedUrls.length;\n  const findDefaultLocale = (unmarked: LocalizedUrl[]): string | undefined =>\n    unmarked.length === 1\n      ? unmarked[0]?.locale\n      : localizedUrls.find(\n          ({ url }) =>\n            xDefaultUrl && normalizeUrl(url.href) === normalizeUrl(xDefaultUrl)\n        )?.locale;\n\n  // One host per locale → subdomain or domain based routing.\n  const hostnames = new Set(\n    localizedUrls.map(({ url }) => normalizeHostname(url.hostname))\n  );\n  if (hostnames.size > 1) {\n    const hostLocales = Object.fromEntries(\n      localizedUrls.map(({ url, locale }) => [\n        normalizeHostname(url.hostname),\n        locale,\n      ])\n    );\n    const subdomainPrefixed = localizedUrls.filter(\n      ({ url, locale }) =>\n        findMatchingLocale(url.hostname.split('.')[0], [locale]) !== undefined\n    );\n    // Most hosts must start with their locale: `www.example.de` is a domain.\n    const isSubdomain = subdomainPrefixed.length * 2 >= localizedUrls.length;\n    return {\n      strategy: isSubdomain ? 'subdomain' : 'domain',\n      confidence: 'high',\n      hostLocales,\n      defaultLocale: isSubdomain\n        ? findDefaultLocale(\n            localizedUrls.filter((entry) => !subdomainPrefixed.includes(entry))\n          )\n        : undefined,\n      evidence: `hreflang alternates use ${hostnames.size} different hosts`,\n    };\n  }\n\n  // A query parameter whose value is the locale on (almost) every alternate.\n  const parameterMatchCount = new Map<string, number>();\n  for (const { url, locale } of localizedUrls) {\n    for (const [parameterName, parameterValue] of url.searchParams) {\n      if (findMatchingLocale(parameterValue, [locale])) {\n        parameterMatchCount.set(\n          parameterName,\n          (parameterMatchCount.get(parameterName) ?? 0) + 1\n        );\n      }\n    }\n  }\n  const [searchParamName, searchParamCount] = [\n    ...parameterMatchCount.entries(),\n  ].sort((first, second) => second[1] - first[1])[0] ?? ['', 0];\n  if (searchParamCount >= Math.max(alternateCount - 1, 1)) {\n    return {\n      strategy: 'search-params',\n      confidence: 'high',\n      searchParamName,\n      defaultLocale: findDefaultLocale(\n        localizedUrls.filter(\n          ({ url }) => !url.searchParams.has(searchParamName)\n        )\n      ),\n      evidence: `hreflang alternates carry the locale in the \"?${searchParamName}=\" parameter`,\n    };\n  }\n\n  // A locale path prefix on every alternate, or on every one but the default.\n  const unprefixed = localizedUrls.filter(\n    ({ url, locale }) =>\n      !findMatchingLocale(getFirstPathSegment(url.pathname), [locale])\n  );\n  if (unprefixed.length === 0) {\n    return {\n      strategy: 'prefix-all',\n      confidence: 'high',\n      defaultLocale: findDefaultLocale([]),\n      evidence: 'every hreflang alternate starts with a locale segment',\n    };\n  }\n  if (unprefixed.length === 1) {\n    return {\n      strategy: 'prefix-no-default',\n      confidence: 'high',\n      defaultLocale: unprefixed[0]?.locale,\n      evidence: `every hreflang alternate but \"${unprefixed[0]?.locale}\" starts with a locale segment`,\n    };\n  }\n\n  const distinctUrls = new Set(\n    localizedUrls.map(({ url }) => normalizeUrl(url.href))\n  );\n  if (distinctUrls.size === 1) {\n    return {\n      strategy: 'no-prefix',\n      confidence: 'high',\n      evidence: `all ${locales.length} hreflang alternates point to the same URL`,\n    };\n  }\n\n  return {\n    strategy: 'unknown',\n    confidence: 'high',\n    evidence:\n      'hreflang alternates use different URLs without a locale marker (localized slugs?)',\n  };\n};\n\n/** Guess the strategy from the scanned URL when no alternates are declared. */\nconst detectFromPageUrl = (\n  pageUrl: URL,\n  htmlLang: string | undefined\n): Omit<RoutingDetection, 'urlLocale' | 'locales'> => {\n  const firstSegment = getFirstPathSegment(pageUrl.pathname);\n  const pageLocales = htmlLang ? [htmlLang] : [];\n\n  if (\n    firstSegment &&\n    (findMatchingLocale(firstSegment, pageLocales) ||\n      (!htmlLang && looksLikeLocaleCode(firstSegment)))\n  ) {\n    return {\n      strategy: 'prefix-all',\n      confidence: 'low',\n      evidence: `the page URL starts with the locale segment \"/${firstSegment}\"`,\n    };\n  }\n\n  for (const [parameterName, parameterValue] of pageUrl.searchParams) {\n    if (findMatchingLocale(parameterValue, pageLocales)) {\n      return {\n        strategy: 'search-params',\n        confidence: 'low',\n        searchParamName: parameterName,\n        evidence: `the page URL carries the locale in \"?${parameterName}=\"`,\n      };\n    }\n  }\n\n  if (findMatchingLocale(pageUrl.hostname.split('.')[0], pageLocales)) {\n    return {\n      strategy: 'subdomain',\n      confidence: 'low',\n      hostLocales: { [normalizeHostname(pageUrl.hostname)]: htmlLang ?? '' },\n      evidence: 'the page hostname starts with the locale',\n    };\n  }\n\n  return {\n    strategy: htmlLang ? 'no-prefix' : 'unknown',\n    confidence: 'low',\n    evidence: htmlLang\n      ? 'no hreflang alternates and no locale marker in the page URL'\n      : 'no hreflang alternates, no lang attribute and no locale marker in the URL',\n  };\n};\n\n/**\n * Detect how the site encodes the locale in its URLs (path prefix, query\n * parameter, subdomain, domain or nothing).\n *\n * The hreflang alternates are the most reliable signal: each one pairs a\n * locale with its URL, so the shared pattern reveals the strategy. Without\n * alternates the scanned URL alone is used, with `confidence: 'low'`.\n */\nexport const detectRoutingStrategy = ({\n  pageUrl,\n  htmlLang,\n  hreflangs,\n}: RoutingDetectionInput): RoutingDetection => {\n  const parsedPageUrl = parseUrl(pageUrl);\n\n  const localizedUrls: LocalizedUrl[] = [];\n  for (const { hreflang, href } of hreflangs) {\n    if (hreflang.toLowerCase() === 'x-default') continue;\n    const url = parseUrl(href, pageUrl);\n    if (url) localizedUrls.push({ locale: hreflang, url });\n  }\n  const xDefaultHref = hreflangs.find(\n    ({ hreflang }) => hreflang.toLowerCase() === 'x-default'\n  )?.href;\n  const xDefaultUrl = xDefaultHref\n    ? parseUrl(xDefaultHref, pageUrl)?.href\n    : undefined;\n\n  const locales = [...new Set(localizedUrls.map(({ locale }) => locale))];\n\n  const detection: RoutingDetection =\n    localizedUrls.length >= 2\n      ? {\n          ...detectFromAlternates(localizedUrls, locales, xDefaultUrl),\n          locales,\n        }\n      : {\n          ...(parsedPageUrl\n            ? detectFromPageUrl(parsedPageUrl, htmlLang)\n            : {\n                strategy: 'unknown',\n                confidence: 'low',\n                evidence: 'invalid page URL',\n              }),\n          locales: htmlLang ? [htmlLang] : locales,\n        };\n\n  if (detection.strategy === 'prefix-all' && detection.locales.length === 0) {\n    const firstSegment = parsedPageUrl\n      ? getFirstPathSegment(parsedPageUrl.pathname)\n      : undefined;\n    if (firstSegment) detection.locales = [firstSegment];\n  }\n\n  return { ...detection, urlLocale: getUrlLocale(pageUrl, detection) };\n};\n"],"mappings":";;;;;AAgEA,MAAa,gBACX,KACA,YACuB;CACvB,MAAM,YAAY,SAAS,GAAG;CAC9B,IAAI,CAAC,WAAW,OAAO;CAEvB,QAAQ,QAAQ,UAAhB;EACE,KAAK;EACL,KAAK,qBACH,OACE,mBACE,oBAAoB,UAAU,QAAQ,GACtC,QAAQ,OACV,MACC,QAAQ,aAAa,sBAClB,QAAQ,gBACR;EAER,KAAK,iBACH,OACE,mBACE,UAAU,aAAa,IAAI,QAAQ,mBAAmB,EAAE,GACxD,QAAQ,OACV,KAAK,QAAQ;EAEjB,KAAK;EACL,KAAK,UACH,OACE,QAAQ,cAAc,kBAAkB,UAAU,QAAQ,MAC1D,mBAAmB,UAAU,SAAS,MAAM,GAAG,CAAC,CAAC,IAAI,QAAQ,OAAO;EAExE,SACE;CACJ;AACF;;AAGA,MAAM,wBACJ,eACA,SACA,gBACoD;CACpD,MAAM,iBAAiB,cAAc;CACrC,MAAM,qBAAqB,aACzB,SAAS,WAAW,IAChB,SAAS,EAAE,EAAE,SACb,cAAc,MACX,EAAE,UACD,eAAe,aAAa,IAAI,IAAI,MAAM,aAAa,WAAW,CACtE,CAAC,EAAE;CAGT,MAAM,YAAY,IAAI,IACpB,cAAc,KAAK,EAAE,UAAU,kBAAkB,IAAI,QAAQ,CAAC,CAChE;CACA,IAAI,UAAU,OAAO,GAAG;EACtB,MAAM,cAAc,OAAO,YACzB,cAAc,KAAK,EAAE,KAAK,aAAa,CACrC,kBAAkB,IAAI,QAAQ,GAC9B,MACF,CAAC,CACH;EACA,MAAM,oBAAoB,cAAc,QACrC,EAAE,KAAK,aACN,mBAAmB,IAAI,SAAS,MAAM,GAAG,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,MAAM,MACjE;EAEA,MAAM,cAAc,kBAAkB,SAAS,KAAK,cAAc;EAClE,OAAO;GACL,UAAU,cAAc,cAAc;GACtC,YAAY;GACZ;GACA,eAAe,cACX,kBACE,cAAc,QAAQ,UAAU,CAAC,kBAAkB,SAAS,KAAK,CAAC,CACpE,IACA;GACJ,UAAU,2BAA2B,UAAU,KAAK;EACtD;CACF;CAGA,MAAM,sCAAsB,IAAI,IAAoB;CACpD,KAAK,MAAM,EAAE,KAAK,YAAY,eAC5B,KAAK,MAAM,CAAC,eAAe,mBAAmB,IAAI,cAChD,IAAI,mBAAmB,gBAAgB,CAAC,MAAM,CAAC,GAC7C,oBAAoB,IAClB,gBACC,oBAAoB,IAAI,aAAa,KAAK,KAAK,CAClD;CAIN,MAAM,CAAC,iBAAiB,oBAAoB,CAC1C,GAAG,oBAAoB,QAAQ,CACjC,CAAC,CAAC,MAAM,OAAO,WAAW,OAAO,KAAK,MAAM,EAAE,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC;CAC5D,IAAI,oBAAoB,KAAK,IAAI,iBAAiB,GAAG,CAAC,GACpD,OAAO;EACL,UAAU;EACV,YAAY;EACZ;EACA,eAAe,kBACb,cAAc,QACX,EAAE,UAAU,CAAC,IAAI,aAAa,IAAI,eAAe,CACpD,CACF;EACA,UAAU,iDAAiD,gBAAgB;CAC7E;CAIF,MAAM,aAAa,cAAc,QAC9B,EAAE,KAAK,aACN,CAAC,mBAAmB,oBAAoB,IAAI,QAAQ,GAAG,CAAC,MAAM,CAAC,CACnE;CACA,IAAI,WAAW,WAAW,GACxB,OAAO;EACL,UAAU;EACV,YAAY;EACZ,eAAe,kBAAkB,CAAC,CAAC;EACnC,UAAU;CACZ;CAEF,IAAI,WAAW,WAAW,GACxB,OAAO;EACL,UAAU;EACV,YAAY;EACZ,eAAe,WAAW,EAAE,EAAE;EAC9B,UAAU,iCAAiC,WAAW,EAAE,EAAE,OAAO;CACnE;CAMF,IAAI,IAHqB,IACvB,cAAc,KAAK,EAAE,UAAU,aAAa,IAAI,IAAI,CAAC,CAExC,CAAC,CAAC,SAAS,GACxB,OAAO;EACL,UAAU;EACV,YAAY;EACZ,UAAU,OAAO,QAAQ,OAAO;CAClC;CAGF,OAAO;EACL,UAAU;EACV,YAAY;EACZ,UACE;CACJ;AACF;;AAGA,MAAM,qBACJ,SACA,aACoD;CACpD,MAAM,eAAe,oBAAoB,QAAQ,QAAQ;CACzD,MAAM,cAAc,WAAW,CAAC,QAAQ,IAAI,CAAC;CAE7C,IACE,iBACC,mBAAmB,cAAc,WAAW,KAC1C,CAAC,YAAY,oBAAoB,YAAY,IAEhD,OAAO;EACL,UAAU;EACV,YAAY;EACZ,UAAU,iDAAiD,aAAa;CAC1E;CAGF,KAAK,MAAM,CAAC,eAAe,mBAAmB,QAAQ,cACpD,IAAI,mBAAmB,gBAAgB,WAAW,GAChD,OAAO;EACL,UAAU;EACV,YAAY;EACZ,iBAAiB;EACjB,UAAU,wCAAwC,cAAc;CAClE;CAIJ,IAAI,mBAAmB,QAAQ,SAAS,MAAM,GAAG,CAAC,CAAC,IAAI,WAAW,GAChE,OAAO;EACL,UAAU;EACV,YAAY;EACZ,aAAa,GAAG,kBAAkB,QAAQ,QAAQ,IAAI,YAAY,GAAG;EACrE,UAAU;CACZ;CAGF,OAAO;EACL,UAAU,WAAW,cAAc;EACnC,YAAY;EACZ,UAAU,WACN,gEACA;CACN;AACF;;;;;;;;;AAUA,MAAa,yBAAyB,EACpC,SACA,UACA,gBAC6C;CAC7C,MAAM,gBAAgB,SAAS,OAAO;CAEtC,MAAM,gBAAgC,CAAC;CACvC,KAAK,MAAM,EAAE,UAAU,UAAU,WAAW;EAC1C,IAAI,SAAS,YAAY,MAAM,aAAa;EAC5C,MAAM,MAAM,SAAS,MAAM,OAAO;EAClC,IAAI,KAAK,cAAc,KAAK;GAAE,QAAQ;GAAU;EAAI,CAAC;CACvD;CACA,MAAM,eAAe,UAAU,MAC5B,EAAE,eAAe,SAAS,YAAY,MAAM,WAC/C,CAAC,EAAE;CACH,MAAM,cAAc,eAChB,SAAS,cAAc,OAAO,CAAC,EAAE,OACjC;CAEJ,MAAM,UAAU,CAAC,GAAG,IAAI,IAAI,cAAc,KAAK,EAAE,aAAa,MAAM,CAAC,CAAC;CAEtE,MAAM,YACJ,cAAc,UAAU,IACpB;EACE,GAAG,qBAAqB,eAAe,SAAS,WAAW;EAC3D;CACF,IACA;EACE,GAAI,gBACA,kBAAkB,eAAe,QAAQ,IACzC;GACE,UAAU;GACV,YAAY;GACZ,UAAU;EACZ;EACJ,SAAS,WAAW,CAAC,QAAQ,IAAI;CACnC;CAEN,IAAI,UAAU,aAAa,gBAAgB,UAAU,QAAQ,WAAW,GAAG;EACzE,MAAM,eAAe,gBACjB,oBAAoB,cAAc,QAAQ,IAC1C;EACJ,IAAI,cAAc,UAAU,UAAU,CAAC,YAAY;CACrD;CAEA,OAAO;EAAE,GAAG;EAAW,WAAW,aAAa,SAAS,SAAS;CAAE;AACrE"}

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

@@ -0,0 +1,87 @@
+import { technologySignatures } from "./technologySignatures.mjs";
+import { getSiteDomain, parseUrl } from "./url.mjs";
+
+//#region src/scan/detection/detectTechnologies.ts
+/** Largest script prefix scanned per chunk, to bound regex cost. */
+const MAX_SCRIPT_LENGTH = 5e6;
+/** Shorten a matched snippet for display in the evidence. */
+const truncate = (text, maxLength = 60) => text.length > maxLength ? `${text.slice(0, maxLength)}…` : text;
+/** Return the first signal of `signature` found in `input`, if any. */
+const matchSignature = (signature, input, lowerCaseStorageKeys) => {
+	const versionFromGlobal = signature.versionGlobal ? input.globalVersions?.[signature.versionGlobal] : void 0;
+	for (const globalName of signature.globals ?? []) if (input.globals?.includes(globalName)) return {
+		evidence: `window.${globalName}`,
+		version: versionFromGlobal
+	};
+	if (versionFromGlobal) return {
+		evidence: `window.${signature.versionGlobal}`,
+		version: versionFromGlobal
+	};
+	for (const domMarker of signature.domMarkers ?? []) if (input.domMarkers?.includes(domMarker)) return { evidence: `${domMarker} DOM marker` };
+	for (const storageKey of signature.storageKeys ?? []) if (lowerCaseStorageKeys.has(storageKey.toLowerCase())) return { evidence: `"${storageKey}" cookie / storage key` };
+	for (const pattern of signature.html ?? []) {
+		const match = input.html?.match(pattern);
+		if (match) return {
+			evidence: `HTML contains "${truncate(match[0])}"`,
+			version: match[1]
+		};
+	}
+	const isThirdPartyService = signature.category === "tms" || signature.category === "translation-proxy";
+	const pageHostname = input.pageUrl ? parseUrl(input.pageUrl)?.hostname : void 0;
+	const isOwnSite = (url) => {
+		const hostname = parseUrl(url)?.hostname;
+		return Boolean(pageHostname && hostname) && getSiteDomain(hostname) === getSiteDomain(pageHostname);
+	};
+	for (const pattern of signature.resourceUrls ?? []) {
+		const resourceUrl = input.resourceUrls?.find((url) => pattern.test(url) && !(isThirdPartyService && isOwnSite(url)));
+		if (resourceUrl) return { evidence: `loads ${truncate(resourceUrl, 90)}` };
+	}
+	for (const pattern of signature.bundle ?? []) for (const script of input.scripts ?? []) {
+		const match = script.slice(0, MAX_SCRIPT_LENGTH).match(pattern);
+		if (match) return {
+			evidence: `JavaScript bundle contains "${truncate(match[0])}"`,
+			version: match[1]
+		};
+	}
+};
+/**
+* Every window global and version path referenced by the signatures. Live-page
+* collectors (extension, puppeteer) read these and pass back the ones present.
+*/
+const getTechnologyGlobalNames = () => ({
+	globals: [...new Set(technologySignatures.flatMap((signature) => signature.globals ?? []))],
+	versionGlobals: technologySignatures.flatMap((signature) => signature.versionGlobal ? [signature.versionGlobal] : [])
+});
+/**
+* Detect the frameworks, i18n libraries, TMS and translation proxies used by a
+* page from the raw signals collected on it.
+*
+* @param input - The {@link TechnologyDetectionInput} signals.
+* @param signatures - Fingerprints to test, defaults to {@link technologySignatures}.
+* @returns Detected technologies, in signature order.
+*/
+const detectTechnologies = (input, signatures = technologySignatures) => {
+	const lowerCaseStorageKeys = new Set((input.storageKeys ?? []).map((key) => key.toLowerCase()));
+	const matchedTechnologies = [];
+	for (const signature of signatures) {
+		const match = matchSignature(signature, input, lowerCaseStorageKeys);
+		if (!match) continue;
+		matchedTechnologies.push({
+			id: signature.id,
+			name: signature.name,
+			category: signature.category,
+			version: match.version || void 0,
+			evidence: match.evidence
+		});
+	}
+	const signatureById = new Map(signatures.map((signature) => [signature.id, signature]));
+	const getIds = (technologies) => new Set(technologies.map(({ id }) => id));
+	const matchedIds = getIds(matchedTechnologies);
+	const supportedTechnologies = matchedTechnologies.filter(({ id }) => !signatureById.get(id)?.requires?.some((requiredId) => !matchedIds.has(requiredId)));
+	const supportedIds = getIds(supportedTechnologies);
+	return supportedTechnologies.filter(({ id }) => !signatureById.get(id)?.hiddenBy?.some((hidingId) => supportedIds.has(hidingId)));
+};
+
+//#endregion
+export { detectTechnologies, getTechnologyGlobalNames };
+//# sourceMappingURL=detectTechnologies.mjs.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"detectTechnologies.mjs","names":[],"sources":["../../../../src/scan/detection/detectTechnologies.ts"],"sourcesContent":["import {\n  type TechnologyCategory,\n  type TechnologySignature,\n  technologySignatures,\n} from './technologySignatures';\nimport { getSiteDomain, parseUrl } from './url';\n\n/** A technology detected on the scanned page. */\nexport type DetectedTechnology = {\n  /** Stable identifier, e.g. `nextjs`, `crowdin`. */\n  id: string;\n  /** Human-readable name, e.g. `Next.js`. */\n  name: string;\n  category: TechnologyCategory;\n  /** Version when it can be read from the page, e.g. `14.2.3`. */\n  version?: string;\n  /** Short explanation of the signal that triggered the detection. */\n  evidence: string;\n};\n\n/**\n * Raw signals collected from a page. Every field is optional so each consumer\n * provides what it can: the CLI basic mode only has HTML + scripts, the\n * backend and the extension also read window globals and storage keys.\n */\nexport type TechnologyDetectionInput = {\n  /**\n   * URL of the scanned page. Resource URLs on its own site are ignored for\n   * third-party services, so the Crowdin website is not reported as using\n   * Crowdin because it loads its own favicon.\n   */\n  pageUrl?: string;\n  /** HTML document, inline scripts included. */\n  html?: string;\n  /** Script / stylesheet / iframe / network request URLs. */\n  resourceUrls?: string[];\n  /** Contents of the same-origin JavaScript chunks. */\n  scripts?: string[];\n  /** Names of the window globals that exist on the page. */\n  globals?: string[];\n  /** Cookie / web-storage keys found on the page. */\n  storageKeys?: string[];\n  /** Live-DOM markers, see {@link TechnologySignature.domMarkers}. */\n  domMarkers?: string[];\n  /** Values read from {@link TechnologySignature.versionGlobal} paths. */\n  globalVersions?: Record<string, string>;\n};\n\n/** Largest script prefix scanned per chunk, to bound regex cost. */\nconst MAX_SCRIPT_LENGTH = 5_000_000;\n\n/** Shorten a matched snippet for display in the evidence. */\nconst truncate = (text: string, maxLength = 60): string =>\n  text.length > maxLength ? `${text.slice(0, maxLength)}…` : text;\n\ntype SignatureMatch = { evidence: string; version?: string };\n\n/** Return the first signal of `signature` found in `input`, if any. */\nconst matchSignature = (\n  signature: TechnologySignature,\n  input: TechnologyDetectionInput,\n  lowerCaseStorageKeys: Set<string>\n): SignatureMatch | undefined => {\n  const versionFromGlobal = signature.versionGlobal\n    ? input.globalVersions?.[signature.versionGlobal]\n    : undefined;\n\n  for (const globalName of signature.globals ?? []) {\n    if (input.globals?.includes(globalName)) {\n      return { evidence: `window.${globalName}`, version: versionFromGlobal };\n    }\n  }\n\n  if (versionFromGlobal) {\n    return {\n      evidence: `window.${signature.versionGlobal}`,\n      version: versionFromGlobal,\n    };\n  }\n\n  for (const domMarker of signature.domMarkers ?? []) {\n    if (input.domMarkers?.includes(domMarker)) {\n      return { evidence: `${domMarker} DOM marker` };\n    }\n  }\n\n  for (const storageKey of signature.storageKeys ?? []) {\n    if (lowerCaseStorageKeys.has(storageKey.toLowerCase())) {\n      return { evidence: `\"${storageKey}\" cookie / storage key` };\n    }\n  }\n\n  for (const pattern of signature.html ?? []) {\n    const match = input.html?.match(pattern);\n    if (match) {\n      return {\n        evidence: `HTML contains \"${truncate(match[0])}\"`,\n        version: match[1],\n      };\n    }\n  }\n\n  const isThirdPartyService =\n    signature.category === 'tms' || signature.category === 'translation-proxy';\n  const pageHostname = input.pageUrl\n    ? parseUrl(input.pageUrl)?.hostname\n    : undefined;\n  const isOwnSite = (url: string): boolean => {\n    const hostname = parseUrl(url)?.hostname;\n    return (\n      Boolean(pageHostname && hostname) &&\n      getSiteDomain(hostname as string) ===\n        getSiteDomain(pageHostname as string)\n    );\n  };\n\n  for (const pattern of signature.resourceUrls ?? []) {\n    const resourceUrl = input.resourceUrls?.find(\n      (url) => pattern.test(url) && !(isThirdPartyService && isOwnSite(url))\n    );\n    if (resourceUrl) {\n      return { evidence: `loads ${truncate(resourceUrl, 90)}` };\n    }\n  }\n\n  for (const pattern of signature.bundle ?? []) {\n    for (const script of input.scripts ?? []) {\n      const match = script.slice(0, MAX_SCRIPT_LENGTH).match(pattern);\n      if (match) {\n        return {\n          evidence: `JavaScript bundle contains \"${truncate(match[0])}\"`,\n          version: match[1],\n        };\n      }\n    }\n  }\n\n  return undefined;\n};\n\n/**\n * Every window global and version path referenced by the signatures. Live-page\n * collectors (extension, puppeteer) read these and pass back the ones present.\n */\nexport const getTechnologyGlobalNames = (): {\n  globals: string[];\n  versionGlobals: string[];\n} => ({\n  globals: [\n    ...new Set(\n      technologySignatures.flatMap((signature) => signature.globals ?? [])\n    ),\n  ],\n  versionGlobals: technologySignatures.flatMap((signature) =>\n    signature.versionGlobal ? [signature.versionGlobal] : []\n  ),\n});\n\n/**\n * Detect the frameworks, i18n libraries, TMS and translation proxies used by a\n * page from the raw signals collected on it.\n *\n * @param input - The {@link TechnologyDetectionInput} signals.\n * @param signatures - Fingerprints to test, defaults to {@link technologySignatures}.\n * @returns Detected technologies, in signature order.\n */\nexport const detectTechnologies = (\n  input: TechnologyDetectionInput,\n  signatures: readonly TechnologySignature[] = technologySignatures\n): DetectedTechnology[] => {\n  const lowerCaseStorageKeys = new Set(\n    (input.storageKeys ?? []).map((key) => key.toLowerCase())\n  );\n\n  const matchedTechnologies: DetectedTechnology[] = [];\n  for (const signature of signatures) {\n    const match = matchSignature(signature, input, lowerCaseStorageKeys);\n    if (!match) continue;\n    matchedTechnologies.push({\n      id: signature.id,\n      name: signature.name,\n      category: signature.category,\n      version: match.version || undefined,\n      evidence: match.evidence,\n    });\n  }\n\n  const signatureById = new Map(\n    signatures.map((signature) => [signature.id, signature])\n  );\n  const getIds = (technologies: DetectedTechnology[]): Set<string> =>\n    new Set(technologies.map(({ id }) => id));\n\n  // Requirements first, so a dropped technology cannot hide another one.\n  const matchedIds = getIds(matchedTechnologies);\n  const supportedTechnologies = matchedTechnologies.filter(\n    ({ id }) =>\n      !signatureById\n        .get(id)\n        ?.requires?.some((requiredId) => !matchedIds.has(requiredId))\n  );\n\n  const supportedIds = getIds(supportedTechnologies);\n  return supportedTechnologies.filter(\n    ({ id }) =>\n      !signatureById\n        .get(id)\n        ?.hiddenBy?.some((hidingId) => supportedIds.has(hidingId))\n  );\n};\n"],"mappings":";;;;;AAiDA,MAAM,oBAAoB;;AAG1B,MAAM,YAAY,MAAc,YAAY,OAC1C,KAAK,SAAS,YAAY,GAAG,KAAK,MAAM,GAAG,SAAS,EAAE,KAAK;;AAK7D,MAAM,kBACJ,WACA,OACA,yBAC+B;CAC/B,MAAM,oBAAoB,UAAU,gBAChC,MAAM,iBAAiB,UAAU,iBACjC;CAEJ,KAAK,MAAM,cAAc,UAAU,WAAW,CAAC,GAC7C,IAAI,MAAM,SAAS,SAAS,UAAU,GACpC,OAAO;EAAE,UAAU,UAAU;EAAc,SAAS;CAAkB;CAI1E,IAAI,mBACF,OAAO;EACL,UAAU,UAAU,UAAU;EAC9B,SAAS;CACX;CAGF,KAAK,MAAM,aAAa,UAAU,cAAc,CAAC,GAC/C,IAAI,MAAM,YAAY,SAAS,SAAS,GACtC,OAAO,EAAE,UAAU,GAAG,UAAU,aAAa;CAIjD,KAAK,MAAM,cAAc,UAAU,eAAe,CAAC,GACjD,IAAI,qBAAqB,IAAI,WAAW,YAAY,CAAC,GACnD,OAAO,EAAE,UAAU,IAAI,WAAW,wBAAwB;CAI9D,KAAK,MAAM,WAAW,UAAU,QAAQ,CAAC,GAAG;EAC1C,MAAM,QAAQ,MAAM,MAAM,MAAM,OAAO;EACvC,IAAI,OACF,OAAO;GACL,UAAU,kBAAkB,SAAS,MAAM,EAAE,EAAE;GAC/C,SAAS,MAAM;EACjB;CAEJ;CAEA,MAAM,sBACJ,UAAU,aAAa,SAAS,UAAU,aAAa;CACzD,MAAM,eAAe,MAAM,UACvB,SAAS,MAAM,OAAO,CAAC,EAAE,WACzB;CACJ,MAAM,aAAa,QAAyB;EAC1C,MAAM,WAAW,SAAS,GAAG,CAAC,EAAE;EAChC,OACE,QAAQ,gBAAgB,QAAQ,KAChC,cAAc,QAAkB,MAC9B,cAAc,YAAsB;CAE1C;CAEA,KAAK,MAAM,WAAW,UAAU,gBAAgB,CAAC,GAAG;EAClD,MAAM,cAAc,MAAM,cAAc,MACrC,QAAQ,QAAQ,KAAK,GAAG,KAAK,EAAE,uBAAuB,UAAU,GAAG,EACtE;EACA,IAAI,aACF,OAAO,EAAE,UAAU,SAAS,SAAS,aAAa,EAAE,IAAI;CAE5D;CAEA,KAAK,MAAM,WAAW,UAAU,UAAU,CAAC,GACzC,KAAK,MAAM,UAAU,MAAM,WAAW,CAAC,GAAG;EACxC,MAAM,QAAQ,OAAO,MAAM,GAAG,iBAAiB,CAAC,CAAC,MAAM,OAAO;EAC9D,IAAI,OACF,OAAO;GACL,UAAU,+BAA+B,SAAS,MAAM,EAAE,EAAE;GAC5D,SAAS,MAAM;EACjB;CAEJ;AAIJ;;;;;AAMA,MAAa,kCAGP;CACJ,SAAS,CACP,GAAG,IAAI,IACL,qBAAqB,SAAS,cAAc,UAAU,WAAW,CAAC,CAAC,CACrE,CACF;CACA,gBAAgB,qBAAqB,SAAS,cAC5C,UAAU,gBAAgB,CAAC,UAAU,aAAa,IAAI,CAAC,CACzD;AACF;;;;;;;;;AAUA,MAAa,sBACX,OACA,aAA6C,yBACpB;CACzB,MAAM,uBAAuB,IAAI,KAC9B,MAAM,eAAe,CAAC,EAAC,CAAE,KAAK,QAAQ,IAAI,YAAY,CAAC,CAC1D;CAEA,MAAM,sBAA4C,CAAC;CACnD,KAAK,MAAM,aAAa,YAAY;EAClC,MAAM,QAAQ,eAAe,WAAW,OAAO,oBAAoB;EACnE,IAAI,CAAC,OAAO;EACZ,oBAAoB,KAAK;GACvB,IAAI,UAAU;GACd,MAAM,UAAU;GAChB,UAAU,UAAU;GACpB,SAAS,MAAM,WAAW;GAC1B,UAAU,MAAM;EAClB,CAAC;CACH;CAEA,MAAM,gBAAgB,IAAI,IACxB,WAAW,KAAK,cAAc,CAAC,UAAU,IAAI,SAAS,CAAC,CACzD;CACA,MAAM,UAAU,iBACd,IAAI,IAAI,aAAa,KAAK,EAAE,SAAS,EAAE,CAAC;CAG1C,MAAM,aAAa,OAAO,mBAAmB;CAC7C,MAAM,wBAAwB,oBAAoB,QAC/C,EAAE,SACD,CAAC,cACE,IAAI,EAAE,CAAC,EACN,UAAU,MAAM,eAAe,CAAC,WAAW,IAAI,UAAU,CAAC,CAClE;CAEA,MAAM,eAAe,OAAO,qBAAqB;CACjD,OAAO,sBAAsB,QAC1B,EAAE,SACD,CAAC,cACE,IAAI,EAAE,CAAC,EACN,UAAU,MAAM,aAAa,aAAa,IAAI,QAAQ,CAAC,CAC/D;AACF"}

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

@@ -0,0 +1,10 @@
+import { technologySignatures } from "./technologySignatures.mjs";
+import { getFirstPathSegment, getSiteDomain, isAbsoluteUrl, normalizeHostname, normalizeUrl, parseUrl } from "./url.mjs";
+import { detectTechnologies, getTechnologyGlobalNames } from "./detectTechnologies.mjs";
+import { findMatchingLocale, getLanguageCode, isRightToLeftLocale, isSameLanguage, isSameLocale, isValidLocaleCode, isValidOpenGraphLocale, looksLikeLocaleCode, normalizeLocaleCode, toOpenGraphLocale } from "./localeCode.mjs";
+import { detectRoutingStrategy, getUrlLocale } from "./detectRoutingStrategy.mjs";
+import { classifyInternalLinks } from "./classifyInternalLinks.mjs";
+import { getCheckDetailLines, splitCheckDetails } from "./checkDetails.mjs";
+import { getLocalizedPages, isBaseLocalePage } from "./localizedPages.mjs";
+
+export { classifyInternalLinks, detectRoutingStrategy, detectTechnologies, findMatchingLocale, getCheckDetailLines, getFirstPathSegment, getLanguageCode, getLocalizedPages, getSiteDomain, getTechnologyGlobalNames, getUrlLocale, isAbsoluteUrl, isBaseLocalePage, isRightToLeftLocale, isSameLanguage, isSameLocale, isValidLocaleCode, isValidOpenGraphLocale, looksLikeLocaleCode, normalizeHostname, normalizeLocaleCode, normalizeUrl, parseUrl, splitCheckDetails, technologySignatures, toOpenGraphLocale };