feat: add URL classifier for embed type detection

2026-03-20 23:30:13 +01:00
parent de1c0b4ed8
commit 29cfd88a15
1 changed files with 108 additions and 0 deletions
@@ -0,0 +1,108 @@
+import type { EmbedType, EmbedProvider } from '@backspace/shared';
+
+export interface EmbedClassification {
+  embedType: EmbedType;
+  provider: EmbedProvider | null;
+  embedUrl: string | null;
+  needsMetadataFetch: boolean;
+}
+
+function extractYouTubeId(url: URL): string | null {
+  const host = url.hostname.replace(/^www\./, '');
+
+  if (host === 'youtu.be') {
+    const id = url.pathname.slice(1).split('/')[0];
+    return id || null;
+  }
+
+  if (host === 'youtube.com' || host === 'm.youtube.com') {
+    const path = url.pathname;
+
+    // /watch?v=ID
+    if (path === '/watch') {
+      return url.searchParams.get('v');
+    }
+
+    // /shorts/ID
+    const shortsMatch = path.match(/^\/shorts\/([A-Za-z0-9_-]+)/);
+    if (shortsMatch) return shortsMatch[1];
+
+    // /embed/ID
+    const embedMatch = path.match(/^\/embed\/([A-Za-z0-9_-]+)/);
+    if (embedMatch) return embedMatch[1];
+
+    // /v/ID (legacy)
+    const vMatch = path.match(/^\/v\/([A-Za-z0-9_-]+)/);
+    if (vMatch) return vMatch[1];
+  }
+
+  return null;
+}
+
+export function classifyUrl(rawUrl: string): EmbedClassification {
+  // Direct media extension checks
+  if (/\.(jpg|jpeg|png|gif|webp|avif)(\?.*)?$/i.test(rawUrl)) {
+    return { embedType: 'image', provider: null, embedUrl: null, needsMetadataFetch: false };
+  }
+
+  if (/\.(mp3|ogg|wav|flac|opus)(\?.*)?$/i.test(rawUrl)) {
+    return { embedType: 'audio', provider: null, embedUrl: null, needsMetadataFetch: false };
+  }
+
+  if (/\.(mp4|webm|mov)(\?.*)?$/i.test(rawUrl)) {
+    return { embedType: 'video', provider: null, embedUrl: null, needsMetadataFetch: false };
+  }
+
+  // Provider matching requires a valid URL
+  let url: URL;
+  try {
+    url = new URL(rawUrl);
+  } catch {
+    return { embedType: 'generic', provider: null, embedUrl: null, needsMetadataFetch: true };
+  }
+
+  const host = url.hostname.replace(/^www\./, '');
+
+  // YouTube
+  if (host === 'youtube.com' || host === 'm.youtube.com' || host === 'youtu.be') {
+    const videoId = extractYouTubeId(url);
+    if (videoId) {
+      return {
+        embedType: 'video',
+        provider: 'youtube',
+        embedUrl: `https://www.youtube-nocookie.com/embed/${videoId}`,
+        needsMetadataFetch: true,
+      };
+    }
+  }
+
+  // Vimeo
+  if (host === 'vimeo.com') {
+    const match = url.pathname.match(/^\/(\d+)/);
+    if (match) {
+      return {
+        embedType: 'video',
+        provider: 'vimeo',
+        embedUrl: `https://player.vimeo.com/video/${match[1]}`,
+        needsMetadataFetch: true,
+      };
+    }
+  }
+
+  // Spotify
+  if (host === 'open.spotify.com') {
+    const match = url.pathname.match(/^\/(track|album|playlist)\/([A-Za-z0-9]+)/);
+    if (match) {
+      const [, type, id] = match;
+      return {
+        embedType: 'rich',
+        provider: 'spotify',
+        embedUrl: `https://open.spotify.com/embed/${type}/${id}`,
+        needsMetadataFetch: true,
+      };
+    }
+  }
+
+  // Everything else
+  return { embedType: 'generic', provider: null, embedUrl: null, needsMetadataFetch: true };
+}