@verbaly/compiler 0.13.1 → 0.14.5

package/README.md

@@ -53,6 +53,8 @@ render: { links: { docs: { href: '/docs', target: '_blank', rel: 'noopener' } }
 
 Per-element `data-verbaly-links='{"repo":"https://…"}'` merges over the config map.
 
+**Multi-locale SEO** — set `render.baseUrl` (or `--base-url`) and every page gets reciprocal `<link rel="alternate" hreflang>` (plus `x-default`) for the whole locale set; `--sitemap` writes a locale-aware `sitemap-i18n.xml`. `--clean` drops stale `dist/<locale>/` pages before mirroring. Injection is idempotent.
+
 ## 🔍 Pseudo-localization
 
 `verbaly pseudo` fills a QA catalog (`en-XA` by default, `--locale <id>` to change) from the source: accented letters, `⟦…⟧` markers and ~33% length padding reveal hardcoded strings, clipped layouts and concatenation bugs. Params, variant blocks and tags survive verbatim — the same structural validation as `translate`.

package/dist/cli.js

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
 import { parseArgs } from "node:util";
-import { existsSync, mkdirSync, readFileSync, readdirSync, writeFileSync } from "node:fs";
+import { existsSync, mkdirSync, readFileSync, readdirSync, rmSync, writeFileSync } from "node:fs";
 import { dirname, join, relative, resolve } from "node:path";
 import { RICH_TAGS, createVerbaly, parse, parseTags, safeHref } from "verbaly";
 import { pathToFileURL } from "node:url";
@@ -879,6 +879,8 @@ function pseudoCatalogs(cfg, catalogs, locale = PSEUDO_LOCALE) {
 }
 //#endregion
 //#region src/render.ts
+const HREFLANG_OPEN = "<!--verbaly:hreflang-->";
+const HREFLANG_CLOSE = "<!--/verbaly:hreflang-->";
 const VOID_TAGS = /* @__PURE__ */ new Set([
 	"area",
 	"base",
@@ -906,22 +908,19 @@ function renderHtml(html, options) {
 	const richTags = new Set(options.richTags ?? RICH_TAGS);
 	const globalLinks = options.richLinks;
 	const sourceLocale = options.sourceLocale ?? "en";
-	const messages = {};
-	for (const [locale, catalog] of Object.entries(options.catalogs)) {
-		const clean = {};
-		for (const [key, msg] of Object.entries(catalog)) if (msg) clean[key] = msg;
-		messages[locale] = clean;
-	}
 	const v = createVerbaly({
 		locale: options.locale,
 		fallback: sourceLocale,
-		messages
+		messages: options.catalogs
 	});
 	const t = v.t;
 	const ms = new MagicString(html);
 	const missing = /* @__PURE__ */ new Set();
 	const skip = protectedRanges(html);
 	const inSkip = (index) => skip.some(([from, to]) => index >= from && index < to);
+	const selfUrl = options.alternates?.find((a) => a.hreflang === options.locale)?.href;
+	const sourceUrl = options.alternates?.find((a) => a.hreflang === "x-default")?.href;
+	const rewriteUrl = selfUrl !== void 0 && selfUrl !== sourceUrl;
 	START_TAG.lastIndex = 0;
 	let m;
 	while ((m = START_TAG.exec(html)) !== null) {
@@ -935,6 +934,8 @@ function renderHtml(html, options) {
 			continue;
 		}
 		const attrs = parseAttrs(attrChunk);
+		if (rewriteUrl && tagName === "link" && attrs.get("rel")?.toLowerCase() === "canonical" && decodeEntities(attrs.get("href") ?? "") === sourceUrl) setAttribute(ms, html, chunkStart, openEnd, attrChunk, "href", selfUrl);
+		if (rewriteUrl && tagName === "meta" && attrs.get("property") === "og:url" && decodeEntities(attrs.get("content") ?? "") === sourceUrl) setAttribute(ms, html, chunkStart, openEnd, attrChunk, "content", selfUrl);
 		const key = attrs.get(attr);
 		const attrMapRaw = attrs.get(attrsAttr);
 		if (key === void 0 && attrMapRaw === void 0) continue;
@@ -968,32 +969,66 @@ function renderHtml(html, options) {
 			}
 		}
 	}
+	if (options.alternates?.length) injectAlternates(ms, html, options.alternates);
 	return {
 		html: ms.toString(),
 		missing: [...missing]
 	};
 }
+function injectAlternates(ms, html, alternates) {
+	const links = alternates.map((a) => `<link rel="alternate" hreflang="${escapeAttr(a.hreflang)}" href="${escapeAttr(a.href)}">`).join("");
+	const block = `${HREFLANG_OPEN}${links}${HREFLANG_CLOSE}`;
+	const from = html.indexOf(HREFLANG_OPEN);
+	if (from !== -1) {
+		const to = html.indexOf(HREFLANG_CLOSE, from);
+		if (to !== -1) {
+			const end = to + 24;
+			if (html.slice(from, end) !== block) ms.overwrite(from, end, block);
+			return;
+		}
+	}
+	const head = /<\/head\s*>/i.exec(html);
+	if (head) ms.appendLeft(head.index, block);
+}
 async function renderSite(cfg, options = {}) {
-	const site = join(cfg.root, options.site ?? "dist");
+	const site = join(cfg.root, options.site ?? cfg.render.site ?? "dist");
 	const locales = options.locales ?? cfg.locales;
+	const attribute = options.attribute ?? cfg.render.attribute;
+	const baseUrl = (options.baseUrl ?? cfg.render.baseUrl)?.replace(/\/+$/, "");
+	const wantHreflang = (options.hreflang ?? cfg.render.hreflang ?? true) && baseUrl !== void 0;
+	const wantSitemap = (options.sitemap ?? cfg.render.sitemap ?? false) && baseUrl !== void 0;
+	const clean = options.clean ?? cfg.render.clean ?? false;
 	const catalogs = loadCatalogs(cfg);
+	if (clean) {
+		for (const locale of locales) if (locale !== cfg.sourceLocale) rmSync(join(site, locale), {
+			recursive: true,
+			force: true
+		});
+	}
 	const files = await glob("**/*.html", {
 		cwd: site,
 		absolute: true,
 		ignore: locales.map((locale) => `${locale}/**`)
 	});
 	const missing = {};
+	const urls = [];
 	for (const file of files) {
 		const html = readFileSync(file, "utf8");
-		const rel = relative(site, file);
+		const rel = relative(site, file).replace(/\\/g, "/");
+		const alternates = wantHreflang ? pageAlternates(baseUrl, rel, locales, cfg.sourceLocale) : [];
+		if (wantHreflang) urls.push({
+			rel,
+			alternates
+		});
 		for (const locale of locales) {
 			const result = renderHtml(html, {
 				locale,
 				catalogs,
 				sourceLocale: cfg.sourceLocale,
-				attribute: options.attribute,
+				attribute,
 				richTags: options.richTags,
-				richLinks: options.richLinks ?? cfg.render.links
+				richLinks: options.richLinks ?? cfg.render.links,
+				alternates
 			});
 			for (const key of result.missing) {
 				const list = missing[locale] ??= [];
@@ -1004,12 +1039,38 @@ async function renderSite(cfg, options = {}) {
 			writeFileSync(out, result.html);
 		}
 	}
+	if (wantSitemap && urls.length) writeFileSync(join(site, typeof wantSitemap === "string" ? wantSitemap : "sitemap-i18n.xml"), buildSitemap(urls));
 	return {
 		files: files.length,
 		locales: [...locales],
 		missing
 	};
 }
+function pageUrl(baseUrl, prefix, rel) {
+	const path = rel.replace(/(^|\/)index\.html$/, "$1");
+	return `${baseUrl}${prefix}${path ? `/${path}` : "/"}`;
+}
+function pageAlternates(baseUrl, rel, locales, sourceLocale) {
+	const alternates = [];
+	for (const locale of locales) {
+		const prefix = locale === sourceLocale ? "" : `/${locale}`;
+		alternates.push({
+			hreflang: locale,
+			href: pageUrl(baseUrl, prefix, rel)
+		});
+	}
+	alternates.push({
+		hreflang: "x-default",
+		href: pageUrl(baseUrl, "", rel)
+	});
+	return alternates;
+}
+function buildSitemap(urls) {
+	return `<?xml version="1.0" encoding="UTF-8"?>\n<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:xhtml="http://www.w3.org/1999/xhtml">${urls.flatMap(({ alternates }) => {
+		const alts = alternates.map((a) => `<xhtml:link rel="alternate" hreflang="${a.hreflang}" href="${escapeAttr(a.href)}"/>`).join("");
+		return alternates.filter((a) => a.hreflang !== "x-default").map((self) => `<url><loc>${escapeAttr(self.href)}</loc>${alts}</url>`);
+	}).join("")}</urlset>\n`;
+}
 function parseAttrs(chunk) {
 	const attrs = /* @__PURE__ */ new Map();
 	ATTR.lastIndex = 0;
@@ -1166,6 +1227,10 @@ Options:
   --dry-run          list what would be translated, write nothing (translate)
   --locale <id>      pseudo-locale id (pseudo, default: en-XA)
   --site <path>      built site directory (render, default: dist)
+  --attribute <name> base data attribute (render, default: data-verbaly)
+  --base-url <url>   site origin — enables hreflang alternates (render)
+  --sitemap          emit sitemap-i18n.xml with per-locale alternates (render)
+  --clean            remove existing locale dirs before mirroring (render)
 
 Config file: verbaly.config.{js,mjs,ts,mts,json} at root (flags win).
 The claude provider needs @anthropic-ai/sdk installed and ANTHROPIC_API_KEY set.
@@ -1182,6 +1247,10 @@ async function main() {
 			model: { type: "string" },
 			locale: { type: "string" },
 			site: { type: "string" },
+			attribute: { type: "string" },
+			"base-url": { type: "string" },
+			sitemap: { type: "boolean" },
+			clean: { type: "boolean" },
 			"dry-run": { type: "boolean" },
 			help: {
 				type: "boolean",
@@ -1288,7 +1357,11 @@ async function main() {
 	if (command === "render") {
 		const result = await renderSite(cfg, {
 			site: values.site,
-			locales: values.locales?.split(",")
+			locales: values.locales?.split(","),
+			attribute: values.attribute,
+			baseUrl: values["base-url"],
+			sitemap: values.sitemap,
+			clean: values.clean
 		});
 		console.log(`[verbaly] ${result.files} pages × ${result.locales.length} locales (${result.locales.join(", ")})`);
 		for (const [locale, keys] of Object.entries(result.missing)) console.warn(`  ${locale}: ${keys.length} keys not pre-filled — ${keys.join(", ")}`);