Store embed titles and search them

media-scrape/src/store.ts

@@ -136,6 +136,9 @@ export interface MediaFilterOptions {
     limit?: number;
     offset?: number;
     sort?: MediaSort;
+    /** Title search: multi-word prefix tokens by default, a phrase when `titleExact`. */
+    titleQuery?: string;
+    titleExact?: boolean;
 }
 
 export interface MediaRow {
@@ -166,9 +169,22 @@ export interface MediaRow {
     download_error: string | null;
     preview_error: string | null;
     preview_attempts: number;
+    /** Embed title from Discord (YouTube and friends). */
+    title: string | null;
+    /** Lowercased, punctuation-collapsed title for token-prefix search. */
+    title_key: string | null;
     created_at: string;
 }
 
+/** Normalize a title for prefix-token search: lowercase, punctuation to spaces, collapsed. */
+export function normalizeTitleKey(title: string): string {
+    return title
+        .toLowerCase()
+        .replace(/[^a-z0-9]+/g, " ")
+        .trim()
+        .replace(/\s+/g, " ");
+}
+
 const SCHEMA = `
 CREATE TABLE IF NOT EXISTS messages (
     id TEXT PRIMARY KEY,
@@ -206,6 +222,8 @@ CREATE TABLE IF NOT EXISTS media (
     download_error TEXT,
     preview_error TEXT,
     preview_attempts INTEGER NOT NULL DEFAULT 0,
+    title TEXT,
+    title_key TEXT,
     created_at TEXT NOT NULL
 );
 
@@ -251,6 +269,10 @@ export class ScrapeStore {
         if (!names.has("preview_attempts")) {
             this.db.exec("ALTER TABLE media ADD COLUMN preview_attempts INTEGER NOT NULL DEFAULT 0");
         }
+        if (!names.has("title")) {
+            this.db.exec("ALTER TABLE media ADD COLUMN title TEXT");
+            this.db.exec("ALTER TABLE media ADD COLUMN title_key TEXT");
+        }
     }
 
     close(): void {
@@ -264,7 +286,7 @@ export class ScrapeStore {
             "INSERT OR REPLACE INTO messages (id, author_id, author_name, timestamp, content) VALUES (?, ?, ?, ?, ?)"
         );
         const insertMedia = this.db.prepare(
-            "INSERT OR IGNORE INTO media (kind, source, author_id, author_name, posted_at, url, play_url, provider, content_type, size, width, height, aspect_ratio, duration_secs, downloaded, preview_url, created_at) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"
+            "INSERT OR IGNORE INTO media (kind, source, author_id, author_name, posted_at, url, play_url, provider, content_type, size, width, height, aspect_ratio, duration_secs, downloaded, preview_url, title, title_key, created_at) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"
         );
         const findMedia = this.db.prepare("SELECT id FROM media WHERE url = ?");
         const linkSource = this.db.prepare(
@@ -301,6 +323,8 @@ export class ScrapeStore {
                         item.durationSecs,
                         item.downloaded ? 1 : 0,
                         item.previewUrl,
+                        item.title,
+                        item.title ? normalizeTitleKey(item.title) : null,
                         createdAt
                     );
                     if (result.changes > 0) added[item.kind]++;
@@ -640,6 +664,18 @@ export class ScrapeStore {
             conditions.push("duration_secs > ?");
             values.push(options.lengthGreaterThan);
         }
+        if (options.titleQuery?.trim()) {
+            const query = normalizeTitleKey(options.titleQuery);
+            if (options.titleExact) {
+                conditions.push("title_key LIKE ?");
+                values.push(`%${query}%`);
+            } else {
+                for (const token of query.split(" ").filter(Boolean)) {
+                    conditions.push("(' ' || COALESCE(title_key, '')) LIKE ?");
+                    values.push(`% ${token}%`);
+                }
+            }
+        }
         if (options.after) {
             conditions.push("posted_at >= ?");
             values.push(options.after);