@intlayer/engine 9.0.0-canary.12

package/dist/cjs/scan/parseHtml.cjs

@@ -0,0 +1,127 @@
+Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
+
+//#region src/scan/parseHtml.ts
+/**
+* Tiny dependency-free HTML extraction helpers.
+*
+* The hosted backend audit relies on Cheerio + a real browser, but the CLI scan
+* must stay dependency-light. These regex-based helpers cover the handful of
+* head/anchor signals the score needs. They are intentionally forgiving: when a
+* tag can't be parsed it is simply skipped rather than throwing.
+*/
+/** Compute the UTF-8 byte length of a string in both Node and browser builds. */
+const byteLength = (text) => typeof Buffer !== "undefined" ? Buffer.byteLength(text, "utf-8") : new TextEncoder().encode(text).length;
+/** Read an attribute value off a single tag's attribute string. */
+const readAttribute = (attributes, attributeName) => {
+	const match = attributes.match(new RegExp(`${attributeName}\\s*=\\s*("([^"]*)"|'([^']*)'|([^\\s>]+))`, "i"));
+	if (!match) return void 0;
+	return match[2] ?? match[3] ?? match[4];
+};
+/** Extract the `lang` attribute of the `<html>` element, if present. */
+const extractHtmlLang = (html) => {
+	const htmlTag = html.match(/<html\b([^>]*)>/i);
+	return htmlTag ? readAttribute(htmlTag[1], "lang") : void 0;
+};
+/** Extract the `dir` attribute of the `<html>` element, if present. */
+const extractHtmlDir = (html) => {
+	const htmlTag = html.match(/<html\b([^>]*)>/i);
+	return htmlTag ? readAttribute(htmlTag[1], "dir") : void 0;
+};
+/** Extract the document `<title>` text. */
+const extractTitle = (html) => {
+	const match = html.match(/<title[^>]*>([\s\S]*?)<\/title>/i);
+	return match ? match[1].trim() : "";
+};
+/** Extract the `<meta name="description">` content. */
+const extractMetaDescription = (html) => {
+	const metas = html.match(/<meta\b[^>]*>/gi) ?? [];
+	for (const meta of metas) if (/name\s*=\s*("|')?description\1?/i.test(meta)) return readAttribute(meta, "content") ?? "";
+	return "";
+};
+/** Extract the `<meta property="og:image">` content. */
+const extractOgImage = (html) => {
+	const metas = html.match(/<meta\b[^>]*>/gi) ?? [];
+	for (const meta of metas) if (/property\s*=\s*("|')?og:image\1?/i.test(meta)) return readAttribute(meta, "content");
+};
+/** Whether a `<link rel="canonical">` element is present. */
+const hasCanonical = (html) => {
+	return (html.match(/<link\b[^>]*>/gi) ?? []).some((link) => /rel\s*=\s*("|')?canonical\1?/i.test(link));
+};
+/** Extract every `<link rel="alternate" hreflang="…" href="…">` element. */
+const extractHreflangs = (html) => {
+	const links = html.match(/<link\b[^>]*>/gi) ?? [];
+	const result = [];
+	for (const link of links) {
+		if (!/rel\s*=\s*("|')?alternate\1?/i.test(link)) continue;
+		const hreflang = readAttribute(link, "hreflang");
+		const href = readAttribute(link, "href");
+		if (hreflang && href) result.push({
+			hreflang,
+			href
+		});
+	}
+	return result;
+};
+/**
+* Extract every eagerly-loaded script URL: `<script src>`,
+* `<link rel="modulepreload">` and `<link rel="preload" as="script">`.
+*
+* @param html - The raw HTML document.
+* @param baseUrl - Base URL used to resolve relative script URLs.
+* @returns Absolute, de-duplicated script URLs.
+*/
+const extractScriptUrls = (html, baseUrl) => {
+	const urls = /* @__PURE__ */ new Set();
+	const add = (raw) => {
+		if (!raw) return;
+		try {
+			urls.add(new URL(raw, baseUrl).href);
+		} catch {}
+	};
+	for (const script of html.match(/<script\b[^>]*>/gi) ?? []) add(readAttribute(script, "src"));
+	for (const link of html.match(/<link\b[^>]*>/gi) ?? []) {
+		const rel = readAttribute(link, "rel")?.toLowerCase();
+		const as = readAttribute(link, "as")?.toLowerCase();
+		if (rel === "modulepreload" || rel === "preload" && as === "script") add(readAttribute(link, "href"));
+	}
+	return Array.from(urls);
+};
+/** Extract every `<a href="…">text</a>` anchor from the document. */
+const extractAnchors = (html) => {
+	const anchors = [];
+	const anchorPattern = /<a\b([^>]*)>([\s\S]*?)<\/a>/gi;
+	let match = anchorPattern.exec(html);
+	while (match !== null) {
+		const href = readAttribute(match[1], "href");
+		if (href) {
+			const text = match[2].replace(/<[^>]+>/g, " ").replace(/\s+/g, " ").trim();
+			anchors.push({
+				href,
+				text
+			});
+		}
+		match = anchorPattern.exec(html);
+	}
+	return anchors;
+};
+/**
+* Extract visible text snippets from an HTML document (scripts, styles and
+* tags stripped). Used to approximate the rendered content size without a DOM.
+*/
+const extractVisibleTextStrings = (html) => {
+	return html.replace(/<script[\s\S]*?<\/script>/gi, " ").replace(/<style[\s\S]*?<\/style>/gi, " ").replace(/<noscript[\s\S]*?<\/noscript>/gi, " ").replace(/<!--[\s\S]*?-->/g, " ").replace(/<[^>]+>/g, "\n").split("\n").map((line) => line.replace(/\s+/g, " ").trim()).filter((line) => line.length > 1);
+};
+
+//#endregion
+exports.byteLength = byteLength;
+exports.extractAnchors = extractAnchors;
+exports.extractHreflangs = extractHreflangs;
+exports.extractHtmlDir = extractHtmlDir;
+exports.extractHtmlLang = extractHtmlLang;
+exports.extractMetaDescription = extractMetaDescription;
+exports.extractOgImage = extractOgImage;
+exports.extractScriptUrls = extractScriptUrls;
+exports.extractTitle = extractTitle;
+exports.extractVisibleTextStrings = extractVisibleTextStrings;
+exports.hasCanonical = hasCanonical;
+//# sourceMappingURL=parseHtml.cjs.map

package/dist/cjs/scan/parseHtml.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"parseHtml.cjs","names":[],"sources":["../../../src/scan/parseHtml.ts"],"sourcesContent":["/**\n * Tiny dependency-free HTML extraction helpers.\n *\n * The hosted backend audit relies on Cheerio + a real browser, but the CLI scan\n * must stay dependency-light. These regex-based helpers cover the handful of\n * head/anchor signals the score needs. They are intentionally forgiving: when a\n * tag can't be parsed it is simply skipped rather than throwing.\n */\n\n/** Compute the UTF-8 byte length of a string in both Node and browser builds. */\nexport const byteLength = (text: string): number =>\n  typeof Buffer !== 'undefined'\n    ? Buffer.byteLength(text, 'utf-8')\n    : new TextEncoder().encode(text).length;\n\n/** Read an attribute value off a single tag's attribute string. */\nconst readAttribute = (\n  attributes: string,\n  attributeName: string\n): string | undefined => {\n  const match = attributes.match(\n    new RegExp(`${attributeName}\\\\s*=\\\\s*(\"([^\"]*)\"|'([^']*)'|([^\\\\s>]+))`, 'i')\n  );\n  if (!match) return undefined;\n  return match[2] ?? match[3] ?? match[4];\n};\n\n/** Extract the `lang` attribute of the `<html>` element, if present. */\nexport const extractHtmlLang = (html: string): string | undefined => {\n  const htmlTag = html.match(/<html\\b([^>]*)>/i);\n  return htmlTag ? readAttribute(htmlTag[1], 'lang') : undefined;\n};\n\n/** Extract the `dir` attribute of the `<html>` element, if present. */\nexport const extractHtmlDir = (html: string): string | undefined => {\n  const htmlTag = html.match(/<html\\b([^>]*)>/i);\n  return htmlTag ? readAttribute(htmlTag[1], 'dir') : undefined;\n};\n\n/** Extract the document `<title>` text. */\nexport const extractTitle = (html: string): string => {\n  const match = html.match(/<title[^>]*>([\\s\\S]*?)<\\/title>/i);\n  return match ? match[1].trim() : '';\n};\n\n/** Extract the `<meta name=\"description\">` content. */\nexport const extractMetaDescription = (html: string): string => {\n  const metas = html.match(/<meta\\b[^>]*>/gi) ?? [];\n  for (const meta of metas) {\n    if (/name\\s*=\\s*(\"|')?description\\1?/i.test(meta)) {\n      return readAttribute(meta, 'content') ?? '';\n    }\n  }\n  return '';\n};\n\n/** Extract the `<meta property=\"og:image\">` content. */\nexport const extractOgImage = (html: string): string | undefined => {\n  const metas = html.match(/<meta\\b[^>]*>/gi) ?? [];\n  for (const meta of metas) {\n    if (/property\\s*=\\s*(\"|')?og:image\\1?/i.test(meta)) {\n      return readAttribute(meta, 'content');\n    }\n  }\n  return undefined;\n};\n\n/** Whether a `<link rel=\"canonical\">` element is present. */\nexport const hasCanonical = (html: string): boolean => {\n  const links = html.match(/<link\\b[^>]*>/gi) ?? [];\n  return links.some((link) => /rel\\s*=\\s*(\"|')?canonical\\1?/i.test(link));\n};\n\n/** A parsed `<link rel=\"alternate\" hreflang>` element. */\nexport type HreflangLink = { hreflang: string; href: string };\n\n/** Extract every `<link rel=\"alternate\" hreflang=\"…\" href=\"…\">` element. */\nexport const extractHreflangs = (html: string): HreflangLink[] => {\n  const links = html.match(/<link\\b[^>]*>/gi) ?? [];\n  const result: HreflangLink[] = [];\n  for (const link of links) {\n    if (!/rel\\s*=\\s*(\"|')?alternate\\1?/i.test(link)) continue;\n    const hreflang = readAttribute(link, 'hreflang');\n    const href = readAttribute(link, 'href');\n    if (hreflang && href) result.push({ hreflang, href });\n  }\n  return result;\n};\n\n/**\n * Extract every eagerly-loaded script URL: `<script src>`,\n * `<link rel=\"modulepreload\">` and `<link rel=\"preload\" as=\"script\">`.\n *\n * @param html - The raw HTML document.\n * @param baseUrl - Base URL used to resolve relative script URLs.\n * @returns Absolute, de-duplicated script URLs.\n */\nexport const extractScriptUrls = (html: string, baseUrl: string): string[] => {\n  const urls = new Set<string>();\n\n  const add = (raw: string | undefined) => {\n    if (!raw) return;\n    try {\n      urls.add(new URL(raw, baseUrl).href);\n    } catch {\n      /* ignore malformed URLs */\n    }\n  };\n\n  for (const script of html.match(/<script\\b[^>]*>/gi) ?? []) {\n    add(readAttribute(script, 'src'));\n  }\n\n  for (const link of html.match(/<link\\b[^>]*>/gi) ?? []) {\n    const rel = readAttribute(link, 'rel')?.toLowerCase();\n    const as = readAttribute(link, 'as')?.toLowerCase();\n    if (rel === 'modulepreload' || (rel === 'preload' && as === 'script')) {\n      add(readAttribute(link, 'href'));\n    }\n  }\n\n  return Array.from(urls);\n};\n\n/** A parsed `<a href>` anchor. */\nexport type Anchor = { href: string; text: string };\n\n/** Extract every `<a href=\"…\">text</a>` anchor from the document. */\nexport const extractAnchors = (html: string): Anchor[] => {\n  const anchors: Anchor[] = [];\n  const anchorPattern = /<a\\b([^>]*)>([\\s\\S]*?)<\\/a>/gi;\n  let match = anchorPattern.exec(html);\n  while (match !== null) {\n    const href = readAttribute(match[1], 'href');\n    if (href) {\n      const text = match[2]\n        .replace(/<[^>]+>/g, ' ')\n        .replace(/\\s+/g, ' ')\n        .trim();\n      anchors.push({ href, text });\n    }\n    match = anchorPattern.exec(html);\n  }\n  return anchors;\n};\n\n/**\n * Extract visible text snippets from an HTML document (scripts, styles and\n * tags stripped). Used to approximate the rendered content size without a DOM.\n */\nexport const extractVisibleTextStrings = (html: string): string[] => {\n  const withoutNonVisible = html\n    .replace(/<script[\\s\\S]*?<\\/script>/gi, ' ')\n    .replace(/<style[\\s\\S]*?<\\/style>/gi, ' ')\n    .replace(/<noscript[\\s\\S]*?<\\/noscript>/gi, ' ')\n    .replace(/<!--[\\s\\S]*?-->/g, ' ');\n\n  return withoutNonVisible\n    .replace(/<[^>]+>/g, '\\n')\n    .split('\\n')\n    .map((line) => line.replace(/\\s+/g, ' ').trim())\n    .filter((line) => line.length > 1);\n};\n"],"mappings":";;;;;;;;;;;;AAUA,MAAa,cAAc,SACzB,OAAO,WAAW,cACd,OAAO,WAAW,MAAM,QAAQ,GAChC,IAAI,aAAa,CAAC,OAAO,KAAK,CAAC;;AAGrC,MAAM,iBACJ,YACA,kBACuB;CACvB,MAAM,QAAQ,WAAW,MACvB,IAAI,OAAO,GAAG,cAAc,4CAA4C,IAAI,CAC7E;AACD,KAAI,CAAC,MAAO,QAAO;AACnB,QAAO,MAAM,MAAM,MAAM,MAAM,MAAM;;;AAIvC,MAAa,mBAAmB,SAAqC;CACnE,MAAM,UAAU,KAAK,MAAM,mBAAmB;AAC9C,QAAO,UAAU,cAAc,QAAQ,IAAI,OAAO,GAAG;;;AAIvD,MAAa,kBAAkB,SAAqC;CAClE,MAAM,UAAU,KAAK,MAAM,mBAAmB;AAC9C,QAAO,UAAU,cAAc,QAAQ,IAAI,MAAM,GAAG;;;AAItD,MAAa,gBAAgB,SAAyB;CACpD,MAAM,QAAQ,KAAK,MAAM,mCAAmC;AAC5D,QAAO,QAAQ,MAAM,GAAG,MAAM,GAAG;;;AAInC,MAAa,0BAA0B,SAAyB;CAC9D,MAAM,QAAQ,KAAK,MAAM,kBAAkB,IAAI,EAAE;AACjD,MAAK,MAAM,QAAQ,MACjB,KAAI,mCAAmC,KAAK,KAAK,CAC/C,QAAO,cAAc,MAAM,UAAU,IAAI;AAG7C,QAAO;;;AAIT,MAAa,kBAAkB,SAAqC;CAClE,MAAM,QAAQ,KAAK,MAAM,kBAAkB,IAAI,EAAE;AACjD,MAAK,MAAM,QAAQ,MACjB,KAAI,oCAAoC,KAAK,KAAK,CAChD,QAAO,cAAc,MAAM,UAAU;;;AAO3C,MAAa,gBAAgB,SAA0B;AAErD,SADc,KAAK,MAAM,kBAAkB,IAAI,EAAE,EACpC,MAAM,SAAS,gCAAgC,KAAK,KAAK,CAAC;;;AAOzE,MAAa,oBAAoB,SAAiC;CAChE,MAAM,QAAQ,KAAK,MAAM,kBAAkB,IAAI,EAAE;CACjD,MAAM,SAAyB,EAAE;AACjC,MAAK,MAAM,QAAQ,OAAO;AACxB,MAAI,CAAC,gCAAgC,KAAK,KAAK,CAAE;EACjD,MAAM,WAAW,cAAc,MAAM,WAAW;EAChD,MAAM,OAAO,cAAc,MAAM,OAAO;AACxC,MAAI,YAAY,KAAM,QAAO,KAAK;GAAE;GAAU;GAAM,CAAC;;AAEvD,QAAO;;;;;;;;;;AAWT,MAAa,qBAAqB,MAAc,YAA8B;CAC5E,MAAM,uBAAO,IAAI,KAAa;CAE9B,MAAM,OAAO,QAA4B;AACvC,MAAI,CAAC,IAAK;AACV,MAAI;AACF,QAAK,IAAI,IAAI,IAAI,KAAK,QAAQ,CAAC,KAAK;UAC9B;;AAKV,MAAK,MAAM,UAAU,KAAK,MAAM,oBAAoB,IAAI,EAAE,CACxD,KAAI,cAAc,QAAQ,MAAM,CAAC;AAGnC,MAAK,MAAM,QAAQ,KAAK,MAAM,kBAAkB,IAAI,EAAE,EAAE;EACtD,MAAM,MAAM,cAAc,MAAM,MAAM,EAAE,aAAa;EACrD,MAAM,KAAK,cAAc,MAAM,KAAK,EAAE,aAAa;AACnD,MAAI,QAAQ,mBAAoB,QAAQ,aAAa,OAAO,SAC1D,KAAI,cAAc,MAAM,OAAO,CAAC;;AAIpC,QAAO,MAAM,KAAK,KAAK;;;AAOzB,MAAa,kBAAkB,SAA2B;CACxD,MAAM,UAAoB,EAAE;CAC5B,MAAM,gBAAgB;CACtB,IAAI,QAAQ,cAAc,KAAK,KAAK;AACpC,QAAO,UAAU,MAAM;EACrB,MAAM,OAAO,cAAc,MAAM,IAAI,OAAO;AAC5C,MAAI,MAAM;GACR,MAAM,OAAO,MAAM,GAChB,QAAQ,YAAY,IAAI,CACxB,QAAQ,QAAQ,IAAI,CACpB,MAAM;AACT,WAAQ,KAAK;IAAE;IAAM;IAAM,CAAC;;AAE9B,UAAQ,cAAc,KAAK,KAAK;;AAElC,QAAO;;;;;;AAOT,MAAa,6BAA6B,SAA2B;AAOnE,QAN0B,KACvB,QAAQ,+BAA+B,IAAI,CAC3C,QAAQ,6BAA6B,IAAI,CACzC,QAAQ,mCAAmC,IAAI,CAC/C,QAAQ,oBAAoB,IAEP,CACrB,QAAQ,YAAY,KAAK,CACzB,MAAM,KAAK,CACX,KAAK,SAAS,KAAK,QAAQ,QAAQ,IAAI,CAAC,MAAM,CAAC,CAC/C,QAAQ,SAAS,KAAK,SAAS,EAAE"}

package/dist/cjs/scan/scanWebsite.cjs

@@ -0,0 +1,205 @@
+Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
+const require_runtime = require('../_virtual/_rolldown/runtime.cjs');
+const require_scan_calculateScore = require('./calculateScore.cjs');
+const require_scan_parseHtml = require('./parseHtml.cjs');
+const require_scan_checks = require('./checks.cjs');
+let _intlayer_config_logger = require("@intlayer/config/logger");
+let _intlayer_config_colors = require("@intlayer/config/colors");
+
+//#region src/scan/scanWebsite.ts
+const DEFAULT_USER_AGENT = "Mozilla/5.0 (compatible; IntlayerScanBot/1.0; +https://intlayer.org)";
+const DEFAULT_TIMEOUT_MS = 3e4;
+/**
+* Log a recommendation to install `puppeteer` for a deeper scan. Mirrors the
+* style used by other optional-dependency hints across the CLI.
+*/
+const logDeepScanRecommendation = () => {
+	(0, _intlayer_config_logger.logger)([
+		(0, _intlayer_config_logger.colorize)("Recommended: Install", _intlayer_config_colors.GREY),
+		(0, _intlayer_config_logger.colorize)("puppeteer", _intlayer_config_colors.GREY_LIGHT),
+		(0, _intlayer_config_logger.colorize)("package to enable a deeper scan (renders client-side content & lazy-loaded chunks). See documentation:", _intlayer_config_colors.GREY),
+		(0, _intlayer_config_logger.colorize)("https://intlayer.org/doc/concept/cli#scan", _intlayer_config_colors.GREY_LIGHT)
+	]);
+};
+/**
+* Render the page with a locally installed `puppeteer` to capture
+* client-rendered content, the accurate transfer size, and lazy-loaded chunks.
+*
+* `puppeteer` is imported dynamically through a non-literal specifier so it is
+* never bundled and stays an optional dependency: when it is absent the import
+* rejects and the caller falls back to the basic scan.
+*
+* @returns The deep-scan result, or `null` when `puppeteer` is unavailable.
+*/
+const runDeepScan = async (targetUrl, userAgent, timeoutMs) => {
+	const moduleName = "puppeteer";
+	let puppeteer;
+	try {
+		const mod = await import(moduleName);
+		puppeteer = mod.default ?? mod;
+	} catch {
+		return null;
+	}
+	let browser;
+	try {
+		browser = await puppeteer.launch({
+			headless: true,
+			executablePath: process.env.PUPPETEER_EXECUTABLE_PATH,
+			args: [
+				"--no-sandbox",
+				"--disable-setuid-sandbox",
+				"--disable-dev-shm-usage",
+				"--disable-gpu"
+			]
+		});
+		const page = await browser.newPage();
+		await page.setUserAgent(userAgent);
+		await page.setExtraHTTPHeaders({ "Accept-Language": "en-US,en;q=0.9" });
+		const origin = new URL(targetUrl).origin;
+		const jsResponseMap = /* @__PURE__ */ new Map();
+		let totalPageSize = 0;
+		const pendingResponses = [];
+		page.on("response", (response) => {
+			pendingResponses.push((async () => {
+				try {
+					if (response.status() !== 200) return;
+					const buffer = await response.buffer();
+					totalPageSize += buffer.length;
+					const responseUrl = response.url();
+					if (((response.headers()["content-type"] ?? "").includes("javascript") || /\.(js|mjs|cjs)(\?|$)/.test(responseUrl)) && responseUrl.startsWith(origin)) jsResponseMap.set(responseUrl, buffer.toString("utf-8"));
+				} catch {}
+			})());
+		});
+		await page.goto(targetUrl, {
+			waitUntil: "domcontentloaded",
+			timeout: timeoutMs
+		});
+		await page.waitForNetworkIdle({
+			idleTime: 1e3,
+			timeout: 1e4
+		}).catch(() => {});
+		await Promise.allSettled(pendingResponses);
+		const html = await page.content();
+		const mainBundleUrls = new Set(require_scan_parseHtml.extractScriptUrls(html, targetUrl));
+		const chunks = Array.from(jsResponseMap.entries()).map(([url, content]) => ({
+			url,
+			isMainBundle: mainBundleUrls.has(url),
+			content
+		}));
+		return {
+			html,
+			totalPageSize,
+			chunks
+		};
+	} finally {
+		if (browser) await browser.close();
+	}
+};
+/** Fetch the raw HTML document, measuring its byte size. */
+const fetchHtml = async (url, userAgent, timeoutMs) => {
+	const controller = new AbortController();
+	const timer = setTimeout(() => controller.abort(), timeoutMs);
+	try {
+		const response = await fetch(url, {
+			headers: {
+				"User-Agent": userAgent,
+				"Accept-Language": "en-US,en;q=0.9"
+			},
+			signal: controller.signal
+		});
+		return {
+			html: await response.text(),
+			finalUrl: response.url || url
+		};
+	} finally {
+		clearTimeout(timer);
+	}
+};
+/**
+* Fetch every eagerly-loaded script. Same-origin scripts keep their content so
+* their locale weight can be analyzed; third-party scripts only contribute to
+* the measured page size (their locale-like keys cause false positives).
+*/
+const fetchScripts = async (scriptUrls, origin, userAgent) => {
+	const chunks = [];
+	let scriptBytes = 0;
+	await Promise.all(scriptUrls.map(async (scriptUrl) => {
+		try {
+			const response = await fetch(scriptUrl, { headers: { "User-Agent": userAgent } });
+			if (!response.ok) return;
+			const content = await response.text();
+			scriptBytes += require_scan_parseHtml.byteLength(content);
+			if (scriptUrl.startsWith(origin)) chunks.push({
+				url: scriptUrl,
+				isMainBundle: true,
+				content
+			});
+		} catch {}
+	}));
+	return {
+		chunks,
+		scriptBytes
+	};
+};
+/**
+* Scan a single web page for i18n/SEO health and bundle weight.
+*
+* In `deep` mode (default) the page is rendered with a locally installed
+* `puppeteer`; when `puppeteer` is missing the scan transparently falls back to
+* a `basic` fetch-based pass and logs a recommendation to install it.
+*
+* @param targetUrl - The absolute URL to scan.
+* @param options - {@link ScanOptions} controlling depth, timeout and UA.
+* @returns The {@link ScanResult} including score, page size and per-check events.
+*/
+const scanWebsite = async (targetUrl, options = {}) => {
+	const { deep = true, timeoutMs = DEFAULT_TIMEOUT_MS, userAgent = DEFAULT_USER_AGENT } = options;
+	const origin = new URL(targetUrl).origin;
+	let mode = "basic";
+	let html;
+	let totalPageSize;
+	let chunks;
+	const deepResult = deep ? await runDeepScan(targetUrl, userAgent, timeoutMs) : null;
+	if (deepResult) {
+		mode = "deep";
+		html = deepResult.html;
+		chunks = deepResult.chunks;
+		totalPageSize = deepResult.totalPageSize;
+	} else {
+		if (deep) logDeepScanRecommendation();
+		const { html: fetchedHtml, finalUrl } = await fetchHtml(targetUrl, userAgent, timeoutMs);
+		html = fetchedHtml;
+		const { chunks: fetchedChunks, scriptBytes } = await fetchScripts(require_scan_parseHtml.extractScriptUrls(fetchedHtml, finalUrl), origin, userAgent);
+		chunks = fetchedChunks;
+		totalPageSize = require_scan_parseHtml.byteLength(fetchedHtml) + scriptBytes;
+	}
+	const htmlSize = require_scan_parseHtml.byteLength(html);
+	const events = [];
+	const localesSet = /* @__PURE__ */ new Set();
+	const { langTag } = require_scan_checks.checkHtmlAttributes(html, targetUrl, events);
+	require_scan_checks.checkCanonical(html, targetUrl, events);
+	require_scan_checks.checkLinguisticStructure(html, targetUrl, localesSet, events);
+	require_scan_checks.checkUrlStructure(require_scan_parseHtml.extractAnchors(html), origin, targetUrl, events);
+	const bundle = require_scan_checks.checkBundleContent(chunks, html, langTag, targetUrl, totalPageSize, events);
+	await require_scan_checks.checkRobots(origin, localesSet, userAgent, events);
+	await require_scan_checks.checkSitemap(origin, localesSet, userAgent, events);
+	const rawScore = events.reduce((score, event) => require_scan_calculateScore.mutateScore(score, event), {
+		score: 0,
+		totalScore: 0
+	});
+	return {
+		url: targetUrl,
+		mode,
+		totalPageSize,
+		htmlSize,
+		score: require_scan_calculateScore.toScorePercent(rawScore),
+		rawScore,
+		events,
+		locales: Array.from(localesSet),
+		bundle
+	};
+};
+
+//#endregion
+exports.scanWebsite = scanWebsite;
+//# sourceMappingURL=scanWebsite.cjs.map

package/dist/cjs/scan/scanWebsite.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"scanWebsite.cjs","names":["GREY","GREY_LIGHT","extractScriptUrls","byteLength","checkHtmlAttributes","extractAnchors","checkBundleContent","checkRobots","checkSitemap","mutateScore","toScorePercent"],"sources":["../../../src/scan/scanWebsite.ts"],"sourcesContent":["import { GREY, GREY_LIGHT } from '@intlayer/config/colors';\nimport { colorize, logger } from '@intlayer/config/logger';\nimport { mutateScore, type Score, toScorePercent } from './calculateScore';\nimport {\n  checkBundleContent,\n  checkCanonical,\n  checkHtmlAttributes,\n  checkLinguisticStructure,\n  checkRobots,\n  checkSitemap,\n  checkUrlStructure,\n} from './checks';\nimport { byteLength, extractAnchors, extractScriptUrls } from './parseHtml';\nimport type {\n  BundleChunkInput,\n  ScanEvent,\n  ScanOptions,\n  ScanResult,\n} from './types';\n\nconst DEFAULT_USER_AGENT =\n  'Mozilla/5.0 (compatible; IntlayerScanBot/1.0; +https://intlayer.org)';\nconst DEFAULT_TIMEOUT_MS = 30_000;\n\n/**\n * Log a recommendation to install `puppeteer` for a deeper scan. Mirrors the\n * style used by other optional-dependency hints across the CLI.\n */\nconst logDeepScanRecommendation = (): void => {\n  logger([\n    colorize('Recommended: Install', GREY),\n    colorize('puppeteer', GREY_LIGHT),\n    colorize(\n      'package to enable a deeper scan (renders client-side content & lazy-loaded chunks). See documentation:',\n      GREY\n    ),\n    colorize('https://intlayer.org/doc/concept/cli#scan', GREY_LIGHT),\n  ]);\n};\n\n/** Outcome of a render-based deep scan. */\ntype DeepScanResult = {\n  html: string;\n  totalPageSize: number;\n  chunks: BundleChunkInput[];\n};\n\n/**\n * Render the page with a locally installed `puppeteer` to capture\n * client-rendered content, the accurate transfer size, and lazy-loaded chunks.\n *\n * `puppeteer` is imported dynamically through a non-literal specifier so it is\n * never bundled and stays an optional dependency: when it is absent the import\n * rejects and the caller falls back to the basic scan.\n *\n * @returns The deep-scan result, or `null` when `puppeteer` is unavailable.\n */\nconst runDeepScan = async (\n  targetUrl: string,\n  userAgent: string,\n  timeoutMs: number\n): Promise<DeepScanResult | null> => {\n  const moduleName = 'puppeteer';\n  // Optional dependency resolved at runtime — typed as `any` because it is not\n  // a declared dependency of this package.\n  let puppeteer: any;\n  try {\n    const mod: any = await import(moduleName);\n    puppeteer = mod.default ?? mod;\n  } catch {\n    return null;\n  }\n\n  let browser: any;\n  try {\n    browser = await puppeteer.launch({\n      headless: true,\n      executablePath: process.env.PUPPETEER_EXECUTABLE_PATH,\n      args: [\n        '--no-sandbox',\n        '--disable-setuid-sandbox',\n        '--disable-dev-shm-usage',\n        '--disable-gpu',\n      ],\n    });\n\n    const page = await browser.newPage();\n    await page.setUserAgent(userAgent);\n    await page.setExtraHTTPHeaders({ 'Accept-Language': 'en-US,en;q=0.9' });\n\n    const origin = new URL(targetUrl).origin;\n    const jsResponseMap = new Map<string, string>();\n    let totalPageSize = 0;\n    const pendingResponses: Promise<void>[] = [];\n\n    page.on('response', (response: any) => {\n      pendingResponses.push(\n        (async () => {\n          try {\n            if (response.status() !== 200) return;\n            const buffer = await response.buffer();\n            totalPageSize += buffer.length;\n\n            const responseUrl: string = response.url();\n            const contentType: string =\n              response.headers()['content-type'] ?? '';\n            const isJavaScript =\n              contentType.includes('javascript') ||\n              /\\.(js|mjs|cjs)(\\?|$)/.test(responseUrl);\n            if (isJavaScript && responseUrl.startsWith(origin)) {\n              jsResponseMap.set(responseUrl, buffer.toString('utf-8'));\n            }\n          } catch {\n            /* response already consumed or aborted */\n          }\n        })()\n      );\n    });\n\n    await page.goto(targetUrl, {\n      waitUntil: 'domcontentloaded',\n      timeout: timeoutMs,\n    });\n    await page\n      .waitForNetworkIdle({ idleTime: 1000, timeout: 10_000 })\n      .catch(() => {\n        /* ok if it never fully idles */\n      });\n    await Promise.allSettled(pendingResponses);\n\n    const html: string = await page.content();\n    const mainBundleUrls = new Set(extractScriptUrls(html, targetUrl));\n    const chunks: BundleChunkInput[] = Array.from(jsResponseMap.entries()).map(\n      ([url, content]) => ({\n        url,\n        isMainBundle: mainBundleUrls.has(url),\n        content,\n      })\n    );\n\n    return { html, totalPageSize, chunks };\n  } finally {\n    if (browser) await browser.close();\n  }\n};\n\n/** Fetch the raw HTML document, measuring its byte size. */\nconst fetchHtml = async (\n  url: string,\n  userAgent: string,\n  timeoutMs: number\n): Promise<{ html: string; finalUrl: string }> => {\n  const controller = new AbortController();\n  const timer = setTimeout(() => controller.abort(), timeoutMs);\n  try {\n    const response = await fetch(url, {\n      headers: { 'User-Agent': userAgent, 'Accept-Language': 'en-US,en;q=0.9' },\n      signal: controller.signal,\n    });\n    const html = await response.text();\n    return { html, finalUrl: response.url || url };\n  } finally {\n    clearTimeout(timer);\n  }\n};\n\n/**\n * Fetch every eagerly-loaded script. Same-origin scripts keep their content so\n * their locale weight can be analyzed; third-party scripts only contribute to\n * the measured page size (their locale-like keys cause false positives).\n */\nconst fetchScripts = async (\n  scriptUrls: string[],\n  origin: string,\n  userAgent: string\n): Promise<{ chunks: BundleChunkInput[]; scriptBytes: number }> => {\n  const chunks: BundleChunkInput[] = [];\n  let scriptBytes = 0;\n\n  await Promise.all(\n    scriptUrls.map(async (scriptUrl) => {\n      try {\n        const response = await fetch(scriptUrl, {\n          headers: { 'User-Agent': userAgent },\n        });\n        if (!response.ok) return;\n        const content = await response.text();\n        scriptBytes += byteLength(content);\n        if (scriptUrl.startsWith(origin)) {\n          chunks.push({ url: scriptUrl, isMainBundle: true, content });\n        }\n      } catch {\n        /* ignore unreachable scripts */\n      }\n    })\n  );\n\n  return { chunks, scriptBytes };\n};\n\n/**\n * Scan a single web page for i18n/SEO health and bundle weight.\n *\n * In `deep` mode (default) the page is rendered with a locally installed\n * `puppeteer`; when `puppeteer` is missing the scan transparently falls back to\n * a `basic` fetch-based pass and logs a recommendation to install it.\n *\n * @param targetUrl - The absolute URL to scan.\n * @param options - {@link ScanOptions} controlling depth, timeout and UA.\n * @returns The {@link ScanResult} including score, page size and per-check events.\n */\nexport const scanWebsite = async (\n  targetUrl: string,\n  options: ScanOptions = {}\n): Promise<ScanResult> => {\n  const {\n    deep = true,\n    timeoutMs = DEFAULT_TIMEOUT_MS,\n    userAgent = DEFAULT_USER_AGENT,\n  } = options;\n\n  const origin = new URL(targetUrl).origin;\n\n  let mode: ScanResult['mode'] = 'basic';\n  let html: string;\n  let totalPageSize: number;\n  let chunks: BundleChunkInput[];\n\n  const deepResult = deep\n    ? await runDeepScan(targetUrl, userAgent, timeoutMs)\n    : null;\n\n  if (deepResult) {\n    mode = 'deep';\n    html = deepResult.html;\n    chunks = deepResult.chunks;\n    totalPageSize = deepResult.totalPageSize;\n  } else {\n    if (deep) logDeepScanRecommendation();\n    const { html: fetchedHtml, finalUrl } = await fetchHtml(\n      targetUrl,\n      userAgent,\n      timeoutMs\n    );\n    html = fetchedHtml;\n    const scriptUrls = extractScriptUrls(fetchedHtml, finalUrl);\n    const { chunks: fetchedChunks, scriptBytes } = await fetchScripts(\n      scriptUrls,\n      origin,\n      userAgent\n    );\n    chunks = fetchedChunks;\n    totalPageSize = byteLength(fetchedHtml) + scriptBytes;\n  }\n\n  const htmlSize = byteLength(html);\n\n  const events: ScanEvent[] = [];\n  const localesSet = new Set<string>();\n\n  const { langTag } = checkHtmlAttributes(html, targetUrl, events);\n  checkCanonical(html, targetUrl, events);\n  checkLinguisticStructure(html, targetUrl, localesSet, events);\n  checkUrlStructure(extractAnchors(html), origin, targetUrl, events);\n  const bundle = checkBundleContent(\n    chunks,\n    html,\n    langTag,\n    targetUrl,\n    totalPageSize,\n    events\n  );\n  await checkRobots(origin, localesSet, userAgent, events);\n  await checkSitemap(origin, localesSet, userAgent, events);\n\n  const rawScore = events.reduce<Score>(\n    (score, event) => mutateScore(score, event),\n    { score: 0, totalScore: 0 }\n  );\n\n  return {\n    url: targetUrl,\n    mode,\n    totalPageSize,\n    htmlSize,\n    score: toScorePercent(rawScore),\n    rawScore,\n    events,\n    locales: Array.from(localesSet),\n    bundle,\n  };\n};\n"],"mappings":";;;;;;;;;AAoBA,MAAM,qBACJ;AACF,MAAM,qBAAqB;;;;;AAM3B,MAAM,kCAAwC;AAC5C,qCAAO;wCACI,wBAAwBA,6BAAK;wCAC7B,aAAaC,mCAAW;wCAE/B,0GACAD,6BACD;wCACQ,6CAA6CC,mCAAW;EAClE,CAAC;;;;;;;;;;;;AAoBJ,MAAM,cAAc,OAClB,WACA,WACA,cACmC;CACnC,MAAM,aAAa;CAGnB,IAAI;AACJ,KAAI;EACF,MAAM,MAAW,MAAM,OAAO;AAC9B,cAAY,IAAI,WAAW;SACrB;AACN,SAAO;;CAGT,IAAI;AACJ,KAAI;AACF,YAAU,MAAM,UAAU,OAAO;GAC/B,UAAU;GACV,gBAAgB,QAAQ,IAAI;GAC5B,MAAM;IACJ;IACA;IACA;IACA;IACD;GACF,CAAC;EAEF,MAAM,OAAO,MAAM,QAAQ,SAAS;AACpC,QAAM,KAAK,aAAa,UAAU;AAClC,QAAM,KAAK,oBAAoB,EAAE,mBAAmB,kBAAkB,CAAC;EAEvE,MAAM,SAAS,IAAI,IAAI,UAAU,CAAC;EAClC,MAAM,gCAAgB,IAAI,KAAqB;EAC/C,IAAI,gBAAgB;EACpB,MAAM,mBAAoC,EAAE;AAE5C,OAAK,GAAG,aAAa,aAAkB;AACrC,oBAAiB,MACd,YAAY;AACX,QAAI;AACF,SAAI,SAAS,QAAQ,KAAK,IAAK;KAC/B,MAAM,SAAS,MAAM,SAAS,QAAQ;AACtC,sBAAiB,OAAO;KAExB,MAAM,cAAsB,SAAS,KAAK;AAM1C,WAJE,SAAS,SAAS,CAAC,mBAAmB,IAE1B,SAAS,aAAa,IAClC,uBAAuB,KAAK,YAAY,KACtB,YAAY,WAAW,OAAO,CAChD,eAAc,IAAI,aAAa,OAAO,SAAS,QAAQ,CAAC;YAEpD;OAGN,CACL;IACD;AAEF,QAAM,KAAK,KAAK,WAAW;GACzB,WAAW;GACX,SAAS;GACV,CAAC;AACF,QAAM,KACH,mBAAmB;GAAE,UAAU;GAAM,SAAS;GAAQ,CAAC,CACvD,YAAY,GAEX;AACJ,QAAM,QAAQ,WAAW,iBAAiB;EAE1C,MAAM,OAAe,MAAM,KAAK,SAAS;EACzC,MAAM,iBAAiB,IAAI,IAAIC,yCAAkB,MAAM,UAAU,CAAC;EAClE,MAAM,SAA6B,MAAM,KAAK,cAAc,SAAS,CAAC,CAAC,KACpE,CAAC,KAAK,cAAc;GACnB;GACA,cAAc,eAAe,IAAI,IAAI;GACrC;GACD,EACF;AAED,SAAO;GAAE;GAAM;GAAe;GAAQ;WAC9B;AACR,MAAI,QAAS,OAAM,QAAQ,OAAO;;;;AAKtC,MAAM,YAAY,OAChB,KACA,WACA,cACgD;CAChD,MAAM,aAAa,IAAI,iBAAiB;CACxC,MAAM,QAAQ,iBAAiB,WAAW,OAAO,EAAE,UAAU;AAC7D,KAAI;EACF,MAAM,WAAW,MAAM,MAAM,KAAK;GAChC,SAAS;IAAE,cAAc;IAAW,mBAAmB;IAAkB;GACzE,QAAQ,WAAW;GACpB,CAAC;AAEF,SAAO;GAAE,YADU,SAAS,MAAM;GACnB,UAAU,SAAS,OAAO;GAAK;WACtC;AACR,eAAa,MAAM;;;;;;;;AASvB,MAAM,eAAe,OACnB,YACA,QACA,cACiE;CACjE,MAAM,SAA6B,EAAE;CACrC,IAAI,cAAc;AAElB,OAAM,QAAQ,IACZ,WAAW,IAAI,OAAO,cAAc;AAClC,MAAI;GACF,MAAM,WAAW,MAAM,MAAM,WAAW,EACtC,SAAS,EAAE,cAAc,WAAW,EACrC,CAAC;AACF,OAAI,CAAC,SAAS,GAAI;GAClB,MAAM,UAAU,MAAM,SAAS,MAAM;AACrC,kBAAeC,kCAAW,QAAQ;AAClC,OAAI,UAAU,WAAW,OAAO,CAC9B,QAAO,KAAK;IAAE,KAAK;IAAW,cAAc;IAAM;IAAS,CAAC;UAExD;GAGR,CACH;AAED,QAAO;EAAE;EAAQ;EAAa;;;;;;;;;;;;;AAchC,MAAa,cAAc,OACzB,WACA,UAAuB,EAAE,KACD;CACxB,MAAM,EACJ,OAAO,MACP,YAAY,oBACZ,YAAY,uBACV;CAEJ,MAAM,SAAS,IAAI,IAAI,UAAU,CAAC;CAElC,IAAI,OAA2B;CAC/B,IAAI;CACJ,IAAI;CACJ,IAAI;CAEJ,MAAM,aAAa,OACf,MAAM,YAAY,WAAW,WAAW,UAAU,GAClD;AAEJ,KAAI,YAAY;AACd,SAAO;AACP,SAAO,WAAW;AAClB,WAAS,WAAW;AACpB,kBAAgB,WAAW;QACtB;AACL,MAAI,KAAM,4BAA2B;EACrC,MAAM,EAAE,MAAM,aAAa,aAAa,MAAM,UAC5C,WACA,WACA,UACD;AACD,SAAO;EAEP,MAAM,EAAE,QAAQ,eAAe,gBAAgB,MAAM,aADlCD,yCAAkB,aAAa,SAEtC,EACV,QACA,UACD;AACD,WAAS;AACT,kBAAgBC,kCAAW,YAAY,GAAG;;CAG5C,MAAM,WAAWA,kCAAW,KAAK;CAEjC,MAAM,SAAsB,EAAE;CAC9B,MAAM,6BAAa,IAAI,KAAa;CAEpC,MAAM,EAAE,YAAYC,wCAAoB,MAAM,WAAW,OAAO;AAChE,oCAAe,MAAM,WAAW,OAAO;AACvC,8CAAyB,MAAM,WAAW,YAAY,OAAO;AAC7D,uCAAkBC,sCAAe,KAAK,EAAE,QAAQ,WAAW,OAAO;CAClE,MAAM,SAASC,uCACb,QACA,MACA,SACA,WACA,eACA,OACD;AACD,OAAMC,gCAAY,QAAQ,YAAY,WAAW,OAAO;AACxD,OAAMC,iCAAa,QAAQ,YAAY,WAAW,OAAO;CAEzD,MAAM,WAAW,OAAO,QACrB,OAAO,UAAUC,wCAAY,OAAO,MAAM,EAC3C;EAAE,OAAO;EAAG,YAAY;EAAG,CAC5B;AAED,QAAO;EACL,KAAK;EACL;EACA;EACA;EACA,OAAOC,2CAAe,SAAS;EAC/B;EACA;EACA,SAAS,MAAM,KAAK,WAAW;EAC/B;EACD"}

package/dist/cjs/utils/autoDecorateContent.cjs

@@ -0,0 +1,90 @@
+Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
+const require_runtime = require('../_virtual/_rolldown/runtime.cjs');
+let _intlayer_core_markdown = require("@intlayer/core/markdown");
+let _intlayer_core_transpiler = require("@intlayer/core/transpiler");
+let _intlayer_types_nodeType = require("@intlayer/types/nodeType");
+_intlayer_types_nodeType = require_runtime.__toESM(_intlayer_types_nodeType);
+
+//#region src/utils/autoDecorateContent.ts
+/**
+* Check if a string is a markdown string
+*/
+const isMarkdown = (str) => {
+	return [
+		/^\s*---/m,
+		/^\s*#+\s/m,
+		/^\s*[-*+]\s/m,
+		/^\s*\d+\.\s/m,
+		/^\s*>\s/m,
+		/\[.+\]\(.+\)/,
+		/!\[.+\]\(.+\)/,
+		/`{1,3}.+`{1,3}/,
+		/\*\*.+\*\*/,
+		/__.+__/,
+		/<(https?:\/\/[^\s>]+)>/
+	].some((pattern) => pattern.test(str));
+};
+/**
+* Check if a string is an insertion string
+*/
+const isInsertion = (str) => (0, _intlayer_core_transpiler.getInsertionValues)(str).length > 0;
+/**
+* Check if a string is an HTML/JSX string
+* Matches:
+* - <Tag>
+* - </Tag>
+* - <Tag />
+* - <Tag attribute="value">
+* - <Component.SubComponent>
+*/
+const isHTML = (str) => {
+	return /<[a-zA-Z][a-zA-Z0-9\-.]*(\s+[^>]*)?\/?>/.test(str) || /<\/[a-zA-Z][a-zA-Z0-9\-.]*\s*>/.test(str);
+};
+const leafNodeTypes = [
+	_intlayer_types_nodeType.HTML,
+	_intlayer_types_nodeType.MARKDOWN,
+	_intlayer_types_nodeType.INSERTION,
+	_intlayer_types_nodeType.FILE,
+	_intlayer_types_nodeType.REACT_NODE,
+	_intlayer_types_nodeType.TEXT,
+	_intlayer_types_nodeType.NUMBER,
+	_intlayer_types_nodeType.BOOLEAN,
+	_intlayer_types_nodeType.NULL,
+	_intlayer_types_nodeType.UNKNOWN
+];
+/**
+* Automatically decorate content strings with md() or insert() if they match
+*/
+const autoDecorateContent = (content, options = true) => {
+	if (options === false) return content;
+	const { markdown = true, html: htmlOption = true, insertion = true } = typeof options === "object" ? options : {};
+	if (typeof content === "string") {
+		if (markdown && isMarkdown(content)) return {
+			...(0, _intlayer_core_transpiler.md)(content),
+			metadata: (0, _intlayer_core_markdown.getMarkdownMetadata)(content)
+		};
+		if (htmlOption && isHTML(content)) return (0, _intlayer_core_transpiler.html)(content);
+		if (insertion && isInsertion(content)) return (0, _intlayer_core_transpiler.insert)(content);
+		return content;
+	}
+	if (Array.isArray(content)) return content.map((item) => autoDecorateContent(item, options));
+	if (content && typeof content === "object") {
+		if ("nodeType" in content) {
+			const nodeType = content.nodeType;
+			if (leafNodeTypes.includes(nodeType)) return content;
+			if (nodeType in content) return {
+				...content,
+				[nodeType]: autoDecorateContent(content[nodeType], options)
+			};
+			return content;
+		}
+		const result = {};
+		for (const key of Object.keys(content)) result[key] = autoDecorateContent(content[key], options);
+		return result;
+	}
+	return content;
+};
+
+//#endregion
+exports.autoDecorateContent = autoDecorateContent;
+//# sourceMappingURL=autoDecorateContent.cjs.map

package/dist/cjs/utils/autoDecorateContent.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"autoDecorateContent.cjs","names":["NodeTypes"],"sources":["../../../src/utils/autoDecorateContent.ts"],"sourcesContent":["import { getMarkdownMetadata } from '@intlayer/core/markdown';\nimport {\n  getInsertionValues,\n  html,\n  insert,\n  md,\n} from '@intlayer/core/transpiler';\nimport * as NodeTypes from '@intlayer/types/nodeType';\n\n/**\n * Check if a string is a markdown string\n */\nconst isMarkdown = (str: string): boolean => {\n  // Check for common markdown indicators\n  const patterns = [\n    /^\\s*---/m, // Front Matter\n    /^\\s*#+\\s/m, // Headers: # Title\n    /^\\s*[-*+]\\s/m, // Unordered lists: - Item or * Item\n    /^\\s*\\d+\\.\\s/m, // Ordered lists: 1. Item\n    /^\\s*>\\s/m, // Blockquotes: > Quote\n    /\\[.+\\]\\(.+\\)/, // Links: [text](url)\n    /!\\[.+\\]\\(.+\\)/, // Images: ![alt](url)\n    /`{1,3}.+`{1,3}/, // Code blocks or inline code: `code` or ```code```\n    /\\*\\*.+\\*\\*/, // Bold: **text**\n    /__.+__/, // Bold: __text__\n    /<(https?:\\/\\/[^\\s>]+)>/, // Autolinks: <http://...>\n  ];\n\n  return patterns.some((pattern) => pattern.test(str));\n};\n\n/**\n * Check if a string is an insertion string\n */\nconst isInsertion = (str: string): boolean =>\n  getInsertionValues(str).length > 0;\n\n/**\n * Check if a string is an HTML/JSX string\n * Matches:\n * - <Tag>\n * - </Tag>\n * - <Tag />\n * - <Tag attribute=\"value\">\n * - <Component.SubComponent>\n */\nconst isHTML = (str: string): boolean => {\n  // 1. Matches opening or self-closing tags: <Tag ... > or <Tag ... />\n  //    - Must start with < followed by a letter (to avoid math comparisons like a < b)\n  //    - Allows alphanumeric, hyphens, and dots (for Namespaced components) in tag name\n  //    - Allows attributes until the closing >\n  const openTagRegex = /<[a-zA-Z][a-zA-Z0-9\\-.]*(\\s+[^>]*)?\\/?>/;\n\n  // 2. Matches closing tags: </Tag>\n  const closeTagRegex = /<\\/[a-zA-Z][a-zA-Z0-9\\-.]*\\s*>/;\n\n  return openTagRegex.test(str) || closeTagRegex.test(str);\n};\n\nconst leafNodeTypes: string[] = [\n  NodeTypes.HTML,\n  NodeTypes.MARKDOWN,\n  NodeTypes.INSERTION,\n  NodeTypes.FILE,\n  NodeTypes.REACT_NODE,\n  NodeTypes.TEXT,\n  NodeTypes.NUMBER,\n  NodeTypes.BOOLEAN,\n  NodeTypes.NULL,\n  NodeTypes.UNKNOWN,\n];\n\ntype AutoTransformationOptions = {\n  markdown?: boolean;\n  html?: boolean;\n  insertion?: boolean;\n};\n\n/**\n * Automatically decorate content strings with md() or insert() if they match\n */\nexport const autoDecorateContent = (\n  content: any,\n  options: boolean | AutoTransformationOptions = true\n): any => {\n  if (options === false) {\n    return content;\n  }\n\n  const {\n    markdown = true,\n    html: htmlOption = true,\n    insertion = true,\n  } = typeof options === 'object' ? options : {};\n\n  if (typeof content === 'string') {\n    if (markdown && isMarkdown(content)) {\n      const markdownNode = md(content);\n\n      return {\n        ...markdownNode,\n        metadata: getMarkdownMetadata(content),\n      };\n    }\n\n    if (htmlOption && isHTML(content)) {\n      return html(content);\n    }\n\n    if (insertion && isInsertion(content)) {\n      return insert(content);\n    }\n\n    return content;\n  }\n\n  if (Array.isArray(content)) {\n    return content.map((item) => autoDecorateContent(item, options));\n  }\n\n  if (content && typeof content === 'object') {\n    // If it's already a decorated node (has nodeType)\n    if ('nodeType' in content) {\n      const nodeType = content.nodeType;\n\n      // If it's a leaf node type, don't re-decorate its content\n      if (leafNodeTypes.includes(nodeType)) {\n        return content;\n      }\n\n      // If it's a container node type (like translation, enumeration, etc.), recurse into its content field\n      if (nodeType in content) {\n        return {\n          ...content,\n          [nodeType]: autoDecorateContent(content[nodeType], options),\n        };\n      }\n\n      return content;\n    }\n\n    // Plain object, recurse into all keys\n    const result: Record<string, any> = {};\n    for (const key of Object.keys(content)) {\n      result[key] = autoDecorateContent(content[key], options);\n    }\n    return result;\n  }\n\n  return content;\n};\n"],"mappings":";;;;;;;;;;;AAYA,MAAM,cAAc,QAAyB;AAgB3C,QAAO;EAbL;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EAGa,CAAC,MAAM,YAAY,QAAQ,KAAK,IAAI,CAAC;;;;;AAMtD,MAAM,eAAe,0DACA,IAAI,CAAC,SAAS;;;;;;;;;;AAWnC,MAAM,UAAU,QAAyB;AAUvC,QAAO,0CAAa,KAAK,IAAI,IAAI,iCAAc,KAAK,IAAI;;AAG1D,MAAM,gBAA0B;CAC9BA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACVA,yBAAU;CACX;;;;AAWD,MAAa,uBACX,SACA,UAA+C,SACvC;AACR,KAAI,YAAY,MACd,QAAO;CAGT,MAAM,EACJ,WAAW,MACX,MAAM,aAAa,MACnB,YAAY,SACV,OAAO,YAAY,WAAW,UAAU,EAAE;AAE9C,KAAI,OAAO,YAAY,UAAU;AAC/B,MAAI,YAAY,WAAW,QAAQ,CAGjC,QAAO;GACL,qCAHsB,QAGP;GACf,2DAA8B,QAAQ;GACvC;AAGH,MAAI,cAAc,OAAO,QAAQ,CAC/B,4CAAY,QAAQ;AAGtB,MAAI,aAAa,YAAY,QAAQ,CACnC,8CAAc,QAAQ;AAGxB,SAAO;;AAGT,KAAI,MAAM,QAAQ,QAAQ,CACxB,QAAO,QAAQ,KAAK,SAAS,oBAAoB,MAAM,QAAQ,CAAC;AAGlE,KAAI,WAAW,OAAO,YAAY,UAAU;AAE1C,MAAI,cAAc,SAAS;GACzB,MAAM,WAAW,QAAQ;AAGzB,OAAI,cAAc,SAAS,SAAS,CAClC,QAAO;AAIT,OAAI,YAAY,QACd,QAAO;IACL,GAAG;KACF,WAAW,oBAAoB,QAAQ,WAAW,QAAQ;IAC5D;AAGH,UAAO;;EAIT,MAAM,SAA8B,EAAE;AACtC,OAAK,MAAM,OAAO,OAAO,KAAK,QAAQ,CACpC,QAAO,OAAO,oBAAoB,QAAQ,MAAM,QAAQ;AAE1D,SAAO;;AAGT,QAAO"}

package/dist/cjs/utils/buildComponentFilesList.cjs

@@ -0,0 +1,94 @@
+Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
+const require_runtime = require('../_virtual/_rolldown/runtime.cjs');
+let node_path = require("node:path");
+let fast_glob = require("fast-glob");
+fast_glob = require_runtime.__toESM(fast_glob);
+let _intlayer_config_utils = require("@intlayer/config/utils");
+let _intlayer_config_defaultValues = require("@intlayer/config/defaultValues");
+
+//#region src/utils/buildComponentFilesList.ts
+/**
+* Normalizes a pattern value to an array
+*/
+const normalizeToArray = (value) => Array.isArray(value) ? value : [value];
+/**
+* Remove directories that are subdirectories of others in the list so files
+* are never scanned twice.
+* Example: ['/root', '/root/src'] → ['/root']
+*/
+const getDistinctRootDirs = (dirs) => {
+	const uniqueDirs = Array.from(new Set(dirs.map((dir) => (0, node_path.resolve)(dir))));
+	uniqueDirs.sort((a, b) => a.length - b.length);
+	return uniqueDirs.reduce((acc, dir) => {
+		if (!acc.some((parent) => {
+			const rel = (0, node_path.relative)(parent, dir);
+			return !rel.startsWith("..") && !(0, node_path.isAbsolute)(rel) && rel !== "";
+		})) acc.push(dir);
+		return acc;
+	}, []);
+};
+/**
+* Returns true when the resolved path passes through a `node_modules` segment.
+* Works on both POSIX and Windows paths.
+*/
+const isInsideNodeModules = (dir) => (0, node_path.resolve)(dir).split(node_path.sep).includes("node_modules");
+/**
+* Default exclude patterns derived from TRAVERSE_PATTERN.
+* Extracted once so the function body can reference them without re-computing.
+*/
+const DEFAULT_EXCLUDE_PATTERNS = _intlayer_config_defaultValues.TRAVERSE_PATTERN.filter((p) => typeof p === "string" && p.startsWith("!")).map((p) => p.slice(1));
+/**
+* Builds a deduplicated list of absolute file paths matching the given patterns.
+*
+* Handles multiple root directories (deduplicates overlapping roots), exclude
+* patterns, negation patterns embedded in `traversePattern`, and optional
+* dot-file inclusion.
+*
+* Special case: `codeDir` entries that live inside `node_modules` (e.g. a
+* design-system package installed as a workspace dependency) are scanned as
+* their own explicit roots. They are NOT collapsed into the project root so
+* the `*\/node_modules\/**` exclusion does not silently drop them.
+*
+* @example
+* // Single root with excludes
+* const files = buildComponentFilesList(config);
+*
+* @example
+* // Design-system package inside node_modules is still scanned
+* // intlayer.config.ts: { content: { codeDir: ['node_modules/my-ds/src'] } }
+* const files = buildComponentFilesList(config);
+*/
+const buildComponentFilesList = (config, excludePattern) => {
+	const traversePattern = config.build.traversePattern;
+	const compilerTransformPattern = config.compiler.transformPattern;
+	const contentDeclarationPattern = config.content.fileExtensions.map((ext) => `/**/*${ext}`);
+	const patterns = [...traversePattern, ...normalizeToArray(compilerTransformPattern)].filter((pattern) => typeof pattern === "string").filter((pattern) => !pattern.startsWith("!")).map(_intlayer_config_utils.normalizePath);
+	const userExcludes = traversePattern.filter((pattern) => typeof pattern === "string" && pattern.startsWith("!")).map((pattern) => pattern.slice(1));
+	const baseExcludePatterns = Array.from(new Set([
+		...DEFAULT_EXCLUDE_PATTERNS,
+		...userExcludes,
+		...excludePattern ?? [],
+		...contentDeclarationPattern
+	])).filter((pattern) => typeof pattern === "string").map(_intlayer_config_utils.normalizePath);
+	const resolvedCodeDirs = (config.content.codeDir ?? []).map((dir) => (0, node_path.resolve)(dir));
+	const inNodeModulesCodeDirs = resolvedCodeDirs.filter(isInsideNodeModules);
+	const normalCodeDirs = resolvedCodeDirs.filter((dir) => !isInsideNodeModules(dir));
+	const normalFiles = getDistinctRootDirs([config.system.baseDir, ...normalCodeDirs]).flatMap((root) => fast_glob.default.sync(patterns, {
+		cwd: root,
+		ignore: baseExcludePatterns,
+		absolute: true,
+		dot: true
+	}));
+	const nodeModulesExcludePatterns = baseExcludePatterns.filter((pattern) => !pattern.includes("node_modules"));
+	const nodeModulesFiles = inNodeModulesCodeDirs.flatMap((dir) => fast_glob.default.sync(patterns, {
+		cwd: dir,
+		ignore: nodeModulesExcludePatterns,
+		absolute: true,
+		dot: false
+	}));
+	return Array.from(new Set([...normalFiles, ...nodeModulesFiles]));
+};
+
+//#endregion
+exports.buildComponentFilesList = buildComponentFilesList;
+//# sourceMappingURL=buildComponentFilesList.cjs.map

package/dist/cjs/utils/buildComponentFilesList.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"buildComponentFilesList.cjs","names":["sep","TRAVERSE_PATTERN","normalizePath","fg"],"sources":["../../../src/utils/buildComponentFilesList.ts"],"sourcesContent":["import { isAbsolute, relative, resolve, sep } from 'node:path';\nimport { TRAVERSE_PATTERN } from '@intlayer/config/defaultValues';\nimport { normalizePath } from '@intlayer/config/utils';\nimport type { IntlayerConfig } from '@intlayer/types/config';\nimport fg from 'fast-glob';\n\n/**\n * Normalizes a pattern value to an array\n */\nconst normalizeToArray = <T>(value: T | T[]): T[] =>\n  Array.isArray(value) ? value : [value];\n\n/**\n * Remove directories that are subdirectories of others in the list so files\n * are never scanned twice.\n * Example: ['/root', '/root/src'] → ['/root']\n */\nconst getDistinctRootDirs = (dirs: string[]): string[] => {\n  const uniqueDirs = Array.from(new Set(dirs.map((dir) => resolve(dir))));\n  uniqueDirs.sort((a, b) => a.length - b.length);\n\n  return uniqueDirs.reduce((acc: string[], dir) => {\n    const isNested = acc.some((parent) => {\n      const rel = relative(parent, dir);\n\n      return !rel.startsWith('..') && !isAbsolute(rel) && rel !== '';\n    });\n    if (!isNested) acc.push(dir);\n\n    return acc;\n  }, []);\n};\n\n/**\n * Returns true when the resolved path passes through a `node_modules` segment.\n * Works on both POSIX and Windows paths.\n */\nconst isInsideNodeModules = (dir: string): boolean =>\n  resolve(dir).split(sep).includes('node_modules');\n\n/**\n * Default exclude patterns derived from TRAVERSE_PATTERN.\n * Extracted once so the function body can reference them without re-computing.\n */\nconst DEFAULT_EXCLUDE_PATTERNS: string[] = TRAVERSE_PATTERN.filter(\n  (p): p is string => typeof p === 'string' && p.startsWith('!')\n).map((p) => p.slice(1));\n\n/**\n * Builds a deduplicated list of absolute file paths matching the given patterns.\n *\n * Handles multiple root directories (deduplicates overlapping roots), exclude\n * patterns, negation patterns embedded in `traversePattern`, and optional\n * dot-file inclusion.\n *\n * Special case: `codeDir` entries that live inside `node_modules` (e.g. a\n * design-system package installed as a workspace dependency) are scanned as\n * their own explicit roots. They are NOT collapsed into the project root so\n * the `*\\/node_modules\\/**` exclusion does not silently drop them.\n *\n * @example\n * // Single root with excludes\n * const files = buildComponentFilesList(config);\n *\n * @example\n * // Design-system package inside node_modules is still scanned\n * // intlayer.config.ts: { content: { codeDir: ['node_modules/my-ds/src'] } }\n * const files = buildComponentFilesList(config);\n */\nexport const buildComponentFilesList = (\n  config: IntlayerConfig,\n  excludePattern?: string[]\n): string[] => {\n  const traversePattern = config.build.traversePattern;\n  const compilerTransformPattern = config.compiler.transformPattern;\n  const contentDeclarationPattern = config.content.fileExtensions.map(\n    (ext) => `/**/*${ext}`\n  );\n\n  const patterns = [\n    ...traversePattern,\n    ...normalizeToArray(compilerTransformPattern),\n  ]\n    .filter((pattern) => typeof pattern === 'string')\n    .filter((pattern) => !pattern.startsWith('!'))\n    .map(normalizePath);\n\n  // User-supplied negations from traversePattern\n  const userExcludes = traversePattern\n    .filter(\n      (pattern): pattern is string =>\n        typeof pattern === 'string' && pattern.startsWith('!')\n    )\n    .map((pattern) => pattern.slice(1));\n\n  // Full exclude list: defaults (from TRAVERSE_PATTERN) + user overrides + caller extras + content files.\n  // DEFAULT_EXCLUDE_PATTERNS acts as a safety floor — if the user provides a\n  // partial traversePattern that omits some defaults, they are still applied.\n  const baseExcludePatterns = Array.from(\n    new Set([\n      ...DEFAULT_EXCLUDE_PATTERNS,\n      ...userExcludes,\n      ...(excludePattern ?? []),\n      ...contentDeclarationPattern,\n    ])\n  )\n    .filter((pattern): pattern is string => typeof pattern === 'string')\n    .map(normalizePath);\n\n  // Separate codeDir entries that live inside node_modules.\n  // getDistinctRootDirs would collapse them into the project root, after which\n  // the **/node_modules/** ignore would silently exclude them.\n  const resolvedCodeDirs = (config.content.codeDir ?? []).map((dir) =>\n    resolve(dir)\n  );\n  const inNodeModulesCodeDirs = resolvedCodeDirs.filter(isInsideNodeModules);\n  const normalCodeDirs = resolvedCodeDirs.filter(\n    (dir) => !isInsideNodeModules(dir)\n  );\n\n  // Normal roots: project base + regular codeDir entries, deduplicated.\n  const normalRoots = getDistinctRootDirs([\n    config.system.baseDir,\n    ...normalCodeDirs,\n  ]);\n\n  const normalFiles = normalRoots.flatMap((root) =>\n    fg.sync(patterns, {\n      cwd: root,\n      ignore: baseExcludePatterns,\n      absolute: true,\n      dot: true, // needed for .intlayer and similar\n    })\n  );\n\n  // node_modules codeDir roots: scanned directly from the package root.\n  // **/node_modules/** is removed from the ignore list so it doesn't block\n  // the scan (it is still applied inside the package for nested node_modules).\n  const nodeModulesExcludePatterns = baseExcludePatterns.filter(\n    (pattern) => !pattern.includes('node_modules')\n  );\n\n  const nodeModulesFiles = inNodeModulesCodeDirs.flatMap((dir) =>\n    fg.sync(patterns, {\n      cwd: dir,\n      ignore: nodeModulesExcludePatterns,\n      absolute: true,\n      dot: false,\n    })\n  );\n\n  return Array.from(new Set([...normalFiles, ...nodeModulesFiles]));\n};\n"],"mappings":";;;;;;;;;;;;AASA,MAAM,oBAAuB,UAC3B,MAAM,QAAQ,MAAM,GAAG,QAAQ,CAAC,MAAM;;;;;;AAOxC,MAAM,uBAAuB,SAA6B;CACxD,MAAM,aAAa,MAAM,KAAK,IAAI,IAAI,KAAK,KAAK,+BAAgB,IAAI,CAAC,CAAC,CAAC;AACvE,YAAW,MAAM,GAAG,MAAM,EAAE,SAAS,EAAE,OAAO;AAE9C,QAAO,WAAW,QAAQ,KAAe,QAAQ;AAM/C,MAAI,CALa,IAAI,MAAM,WAAW;GACpC,MAAM,8BAAe,QAAQ,IAAI;AAEjC,UAAO,CAAC,IAAI,WAAW,KAAK,IAAI,2BAAY,IAAI,IAAI,QAAQ;IAEjD,CAAE,KAAI,KAAK,IAAI;AAE5B,SAAO;IACN,EAAE,CAAC;;;;;;AAOR,MAAM,uBAAuB,+BACnB,IAAI,CAAC,MAAMA,cAAI,CAAC,SAAS,eAAe;;;;;AAMlD,MAAM,2BAAqCC,gDAAiB,QACzD,MAAmB,OAAO,MAAM,YAAY,EAAE,WAAW,IAAI,CAC/D,CAAC,KAAK,MAAM,EAAE,MAAM,EAAE,CAAC;;;;;;;;;;;;;;;;;;;;;;AAuBxB,MAAa,2BACX,QACA,mBACa;CACb,MAAM,kBAAkB,OAAO,MAAM;CACrC,MAAM,2BAA2B,OAAO,SAAS;CACjD,MAAM,4BAA4B,OAAO,QAAQ,eAAe,KAC7D,QAAQ,QAAQ,MAClB;CAED,MAAM,WAAW,CACf,GAAG,iBACH,GAAG,iBAAiB,yBAAyB,CAC9C,CACE,QAAQ,YAAY,OAAO,YAAY,SAAS,CAChD,QAAQ,YAAY,CAAC,QAAQ,WAAW,IAAI,CAAC,CAC7C,IAAIC,qCAAc;CAGrB,MAAM,eAAe,gBAClB,QACE,YACC,OAAO,YAAY,YAAY,QAAQ,WAAW,IAAI,CACzD,CACA,KAAK,YAAY,QAAQ,MAAM,EAAE,CAAC;CAKrC,MAAM,sBAAsB,MAAM,KAChC,IAAI,IAAI;EACN,GAAG;EACH,GAAG;EACH,GAAI,kBAAkB,EAAE;EACxB,GAAG;EACJ,CAAC,CACH,CACE,QAAQ,YAA+B,OAAO,YAAY,SAAS,CACnE,IAAIA,qCAAc;CAKrB,MAAM,oBAAoB,OAAO,QAAQ,WAAW,EAAE,EAAE,KAAK,+BACnD,IAAI,CACb;CACD,MAAM,wBAAwB,iBAAiB,OAAO,oBAAoB;CAC1E,MAAM,iBAAiB,iBAAiB,QACrC,QAAQ,CAAC,oBAAoB,IAAI,CACnC;CAQD,MAAM,cALc,oBAAoB,CACtC,OAAO,OAAO,SACd,GAAG,eACJ,CAE8B,CAAC,SAAS,SACvCC,kBAAG,KAAK,UAAU;EAChB,KAAK;EACL,QAAQ;EACR,UAAU;EACV,KAAK;EACN,CAAC,CACH;CAKD,MAAM,6BAA6B,oBAAoB,QACpD,YAAY,CAAC,QAAQ,SAAS,eAAe,CAC/C;CAED,MAAM,mBAAmB,sBAAsB,SAAS,QACtDA,kBAAG,KAAK,UAAU;EAChB,KAAK;EACL,QAAQ;EACR,UAAU;EACV,KAAK;EACN,CAAC,CACH;AAED,QAAO,MAAM,KAAK,IAAI,IAAI,CAAC,GAAG,aAAa,GAAG,iBAAiB,CAAC,CAAC"}