feat(studio): server-side waveform generation with caching

packages/core/src/studio-api/helpers/mime.ts

@@ -18,6 +18,9 @@ export const MIME_TYPES: Record<string, string> = {
   ".wav": "audio/wav",
   ".ogg": "audio/ogg",
   ".m4a": "audio/mp4",
+  ".aac": "audio/aac",
+  ".flac": "audio/flac",
+  ".opus": "audio/ogg",
   ".woff": "font/woff",
   ".woff2": "font/woff2",
   ".ttf": "font/ttf",
@@ -30,3 +33,7 @@ export function getMimeType(path: string): string {
   const ext = path.slice(path.lastIndexOf(".")).toLowerCase();
   return MIME_TYPES[ext] || "application/octet-stream";
 }
+
+export function isAudioFile(name: string): boolean {
+  return (getMimeType(name) ?? "").startsWith("audio/");
+}

packages/core/src/studio-api/helpers/waveform.ts

@@ -0,0 +1,82 @@
+import { spawn } from "node:child_process";
+import { existsSync, writeFileSync, mkdirSync } from "node:fs";
+import { join } from "node:path";
+
+const SAMPLE_RATE = 4000;
+const PEAK_COUNT = 4000;
+export const WAVEFORM_CACHE_VERSION = "v2";
+
+export function buildWaveformCacheKey(assetPath: string): string {
+  return `${WAVEFORM_CACHE_VERSION}_${assetPath.replace(/[/\\]/g, "_")}.json`;
+}
+
+function computePeaks(floats: Float32Array, count: number): number[] {
+  const step = floats.length / count;
+  const peaks: number[] = [];
+  for (let i = 0; i < count; i++) {
+    const start = Math.floor(i * step);
+    const end = Math.min(Math.floor((i + 1) * step), floats.length);
+    let max = 0;
+    for (let j = start; j < end; j++) {
+      const abs = Math.abs(floats[j] ?? 0);
+      if (abs > max) max = abs;
+    }
+    peaks.push(max);
+  }
+  const maxPeak = Math.max(...peaks, 0.001);
+  return peaks.map((p) => p / maxPeak);
+}
+
+export function decodeAudioPeaks(audioPath: string): Promise<number[]> {
+  return new Promise((resolve, reject) => {
+    const proc = spawn(
+      "ffmpeg",
+      [
+        "-i",
+        audioPath,
+        "-af",
+        "atrim=start_sample=1152",
+        "-f",
+        "f32le",
+        "-ac",
+        "1",
+        "-ar",
+        String(SAMPLE_RATE),
+        "-vn",
+        "pipe:1",
+      ],
+      { stdio: ["ignore", "pipe", "ignore"] },
+    );
+
+    const chunks: Buffer[] = [];
+    proc.stdout?.on("data", (chunk: Buffer) => chunks.push(chunk));
+    proc.on("close", (code) => {
+      if (code !== 0 && chunks.length === 0) {
+        reject(new Error(`ffmpeg exited with code ${code}`));
+        return;
+      }
+      const buf = Buffer.concat(chunks);
+      const numSamples = Math.floor(buf.length / 4);
+      if (numSamples === 0) {
+        reject(new Error("ffmpeg produced no audio samples"));
+        return;
+      }
+      const ab = buf.buffer.slice(buf.byteOffset, buf.byteOffset + numSamples * 4);
+      resolve(computePeaks(new Float32Array(ab), PEAK_COUNT));
+    });
+    proc.on("error", reject);
+  });
+}
+
+export async function generateWaveformCache(projectDir: string, assetPath: string): Promise<void> {
+  const audioPath = join(projectDir, assetPath);
+  if (!existsSync(audioPath)) return;
+
+  const cacheDir = join(projectDir, ".waveform-cache");
+  const cachePath = join(cacheDir, buildWaveformCacheKey(assetPath));
+  if (existsSync(cachePath)) return;
+
+  const peaks = await decodeAudioPeaks(audioPath);
+  mkdirSync(cacheDir, { recursive: true });
+  writeFileSync(cachePath, JSON.stringify(peaks));
+}