add Cloudflare Pages setup instructions and OG/Twitter tags generation in Cloudflare Worker

cloudflare-worker.js

@@ -0,0 +1,127 @@
+async function generateOGTags(url, SITE_URL, DOCS_URL) {
+    const params = new URLSearchParams(url.search);
+    let slug = '';
+    if (params.has('')) {
+        slug = params.get('');
+    } else {
+        for (const [key] of params) {
+            slug = key;
+            break;
+        }
+    }
+
+    const indexResponse = await fetch(`${DOCS_URL}/index.json`);
+    const indexData = await indexResponse.json();
+    const defaultPage = indexData.defaultPage || 'welcome';
+    slug = slug || defaultPage;
+
+    function findDocumentBySlug(documents, targetSlug) {
+        for (const doc of documents) {
+            if (doc.slug === targetSlug) return doc;
+            if (doc.type === 'folder' && doc.items) {
+                const found = findDocumentBySlug(doc.items, targetSlug);
+                if (found) return found;
+            }
+        }
+        return null;
+    }
+
+    const currentDoc = findDocumentBySlug(indexData.documents, slug);
+    let description = indexData.metadata?.description || 'Documentation reader';
+    let thumbnail = indexData.metadata?.thumbnail || 'img/og-image.png';
+    if (currentDoc) {
+        try {
+            const docResponse = await fetch(`${DOCS_URL}/${currentDoc.path}`);
+            const content = await docResponse.text();
+
+            const metadataMatch = content.match(/---\n([\s\S]*?)\n---/);
+            if (metadataMatch) {
+                const metadataStr = metadataMatch[1];
+                console.log(metadataMatch);
+                const thumbnailMatch = metadataStr.match(/thumbnail:\s*(.*)/);
+                if (thumbnailMatch) {
+                    thumbnail = thumbnailMatch[1].trim();
+                }
+                const descriptionMatch = metadataStr.match(/description:\s*(.*)/);
+                if (descriptionMatch) {
+                    description = descriptionMatch[1].trim();
+                }
+                if (!description) {
+                    // Get content after metadata
+                    const afterMetadata = content.substring(
+                        metadataMatch.index + metadataMatch[0].length
+                    );
+                    const paragraphMatch = afterMetadata
+                        .split('\n\n')
+                        .find(p => p.trim() && !p.startsWith('#'));
+                    if (paragraphMatch) {
+                        description = paragraphMatch.trim().replace(/\[|\]|\(|\)/g, '').substring(0, 160);
+                    }
+                }
+            }
+        } catch (e) {
+            console.error('Error fetching document:', e);
+        }
+    }
+    const site = new URL(SITE_URL);
+    site.pathname += (url.pathname === '/' ? '' : url.pathname);
+    site.search = url.search;
+    site.hash = url.hash;
+
+    const title = currentDoc?.title || 'Documentation';
+    const ogImage = new URL(thumbnail, SITE_URL).href;
+    const siteName = indexData.metadata?.site_name || 'Documentation';
+
+    return `
+    <meta property="og:title" content="${title} - ${siteName}">
+    <meta property="og:description" content="${description}">
+    <meta property="og:type" content="article">
+    <meta property="og:url" content="${site.href}">
+    <meta property="og:image" content="${ogImage}">
+    <meta property="og:site_name" content="${siteName}">
+    <meta name="twitter:card" content="summary_large_image">
+    <meta name="twitter:title" content="${title} - ${siteName}">
+    <meta name="twitter:description" content="${description}">
+    <meta name="twitter:image" content="${ogImage}">`;
+}
+
+async function handleRequest(request, SITE_URL, DOCS_URL) {
+    const url = new URL(request.url);
+    const siteUrlObj = new URL(SITE_URL);
+    const pathPrefix = siteUrlObj.pathname;
+    
+    const strippedPath = url.pathname.startsWith(pathPrefix) 
+        ? url.pathname.slice(pathPrefix.length - 1)
+        : url.pathname;
+
+    const docUrl = new URL(DOCS_URL);
+    docUrl.pathname = strippedPath;
+    docUrl.search = url.search;
+    docUrl.hash = url.hash;
+    
+    const response = await fetch(docUrl.toString(), request);
+    
+    if (url.pathname === pathPrefix || url.pathname === '/') {
+        const contentType = response.headers.get('content-type') || '';
+        if (contentType.includes('text/html')) {
+            let html = await response.text();
+            const ogTags = await generateOGTags(url, SITE_URL, DOCS_URL);
+            html = html.replace('<!--ogmetadata-->', ogTags);
+            return new Response(html, {
+                headers: {
+                    'content-type': 'text/html;charset=UTF-8',
+                    ...Object.fromEntries(response.headers)
+                }
+            });
+        }
+    }
+    return new Response(response.body, response);
+}
+
+export default {
+    async fetch(request, env, ctx) {
+        const SITE_URL = env.SITE_URL;
+        const DOCS_URL = env.DOCS_URL;
+        return handleRequest(request, SITE_URL, DOCS_URL);
+    }
+};