fix(media): tolerate slow MKV audio extraction (#195)

changes/audio-generation-network-mounts.md

@@ -0,0 +1,4 @@
+type: fixed
+area: Anki media
+
+- Fixed sentence-audio generation timing out on slow network-mounted MKV files with many subtitle and font-attachment streams. Selected audio tracks now use bounded FFmpeg probing and a two-minute extraction budget, and missing output reports a clear FFmpeg error instead of raw `ENOENT`.

docs-site/troubleshooting.md

@@ -171,7 +171,9 @@ Without FFmpeg, card creation still works but audio and image fields will be emp
 
 **Audio or screenshot generation hangs**
 
-Media generation has a 30-second timeout (60 seconds for animated AVIF). If your video file is on a slow network mount or the codec requires software decoding, generation may time out. Try:
+Audio extraction has a 2-minute timeout. SubMiner also limits FFmpeg probing when mpv provides the selected audio stream, which avoids scanning unrelated subtitle and font-attachment streams in large MKV files. Screenshots retain a 30-second timeout, and animated AVIF uses 60 seconds.
+
+If your video file is on a slow or unresponsive network mount, generation may still time out. Try:
 
 - Using a local copy of the video file.
 - Reducing `ankiConnect.media.imageQuality` or switching from `avif` to `static` image type.

src/media-generator.test.ts

@@ -4,13 +4,18 @@ import * as os from 'node:os';
 import * as path from 'node:path';
 import test from 'node:test';
 
-import { buildAnimatedImageVideoFilter, MediaGenerator } from './media-generator';
+import {
+  AUDIO_GENERATION_TIMEOUT_MS,
+  buildAnimatedImageVideoFilter,
+  MediaGenerator,
+  type MediaGeneratorOptions,
+} from './media-generator';
 
 async function withStubbedFfmpeg(
   run: (generator: MediaGenerator, argsPath: string) => Promise<void>,
-  options: {
+  options: MediaGeneratorOptions = {},
-    logDebug?: (message: string) => void;
+  stubOptions: {
-    now?: () => number;
+    skipOutput?: boolean;
   } = {},
 ): Promise<void> {
   const root = fs.mkdtempSync(path.join(os.tmpdir(), 'subminer-media-generator-test-'));
@@ -31,7 +36,9 @@ async function withStubbedFfmpeg(
       '}',
       "fs.writeFileSync(process.env.SUBMINER_TEST_FFMPEG_ARGS, JSON.stringify(args), 'utf8');",
       'const outputPath = args.at(-1);',
-      "fs.writeFileSync(outputPath, 'avif', 'utf8');",
+      "if (process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT !== '1') {",
+      "  fs.writeFileSync(outputPath, 'avif', 'utf8');",
+      '}',
     ].join('\n'),
     'utf8',
   );
@@ -46,8 +53,14 @@ async function withStubbedFfmpeg(
 
   const originalPath = process.env.PATH;
   const originalArgsPath = process.env.SUBMINER_TEST_FFMPEG_ARGS;
+  const originalSkipOutput = process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT;
   process.env.PATH = `${binDir}${path.delimiter}${originalPath ?? ''}`;
   process.env.SUBMINER_TEST_FFMPEG_ARGS = argsPath;
+  if (stubOptions.skipOutput) {
+    process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT = '1';
+  } else {
+    delete process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT;
+  }
   const generator = new MediaGenerator(tempDir, options);
 
   try {
@@ -60,6 +73,11 @@ async function withStubbedFfmpeg(
     } else {
       process.env.SUBMINER_TEST_FFMPEG_ARGS = originalArgsPath;
     }
+    if (originalSkipOutput === undefined) {
+      delete process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT;
+    } else {
+      process.env.SUBMINER_TEST_FFMPEG_SKIP_OUTPUT = originalSkipOutput;
+    }
     fs.rmSync(root, { recursive: true, force: true });
   }
 }
@@ -316,6 +334,65 @@ test('generateAudio keeps explicit audio stream maps for normal media paths', as
   });
 });
 
+test('generateAudio bounds probing when the selected local audio stream is known', async () => {
+  await withStubbedFfmpeg(async (generator, argsPath) => {
+    await generator.generateAudio('/video.mkv', 10, 12, 0, 2);
+
+    const args = readFfmpegArgs(argsPath);
+    const inputIndex = args.indexOf('-i');
+    assert.ok(args.indexOf('-probesize') > -1);
+    assert.ok(args.indexOf('-probesize') < inputIndex);
+    assert.equal(args[args.indexOf('-probesize') + 1], '32768');
+    assert.ok(args.indexOf('-analyzeduration') < inputIndex);
+    assert.equal(args[args.indexOf('-analyzeduration') + 1], '0');
+  });
+});
+
+test('generateAudio retains normal probing for non-Matroska local media', async () => {
+  await withStubbedFfmpeg(async (generator, argsPath) => {
+    await generator.generateAudio('/video.mp4', 10, 12, 0, 2);
+
+    const args = readFfmpegArgs(argsPath);
+    assert.equal(args.includes('-probesize'), false);
+    assert.equal(args.includes('-analyzeduration'), false);
+  });
+});
+
+test('generateAudio retains a two-minute extraction timeout', async () => {
+  let observedTimeout: number | undefined;
+
+  await withStubbedFfmpeg(
+    async (generator) => {
+      await generator.generateAudio('/video.mp4', 10, 12);
+    },
+    {
+      execFile: (_file, args, options, callback) => {
+        observedTimeout = options.timeout;
+        const outputPath = args.at(-1);
+        assert.ok(outputPath);
+        fs.writeFileSync(outputPath, 'mp3', 'utf8');
+        queueMicrotask(() => callback(null));
+      },
+    },
+  );
+
+  assert.equal(AUDIO_GENERATION_TIMEOUT_MS, 120_000);
+  assert.equal(observedTimeout, AUDIO_GENERATION_TIMEOUT_MS);
+});
+
+test('generateAudio reports when ffmpeg exits without creating output', async () => {
+  await withStubbedFfmpeg(
+    async (generator) => {
+      await assert.rejects(
+        generator.generateAudio('/video.mp4', 10, 12),
+        /FFmpeg audio generation failed: FFmpeg exited without creating an output file/,
+      );
+    },
+    {},
+    { skipOutput: true },
+  );
+});
+
 test('generateAudio debug-logs cached input and completion timing', async () => {
   const logs: string[] = [];
   const times = [1000, 1052];

src/media-generator.ts

@@ -26,9 +26,17 @@ import { normalizeMediaInput, type MediaInput } from './media-input';
 const log = createLogger('media');
 const AUDIO_NORMALIZATION_FILTER = 'loudnorm=I=-23:TP=-2:LRA=11';
 const AUDIO_AMPLIFICATION_LIMITER_FILTER = 'alimiter=limit=0.891251:level=false';
+export const AUDIO_GENERATION_TIMEOUT_MS = 120_000;
 
 export type { MediaInput, MediaInputOptions } from './media-input';
 
+type MediaGeneratorExecFile = (
+  file: string,
+  args: readonly string[],
+  options: { timeout: number },
+  callback: (error: ExecFileException | null) => void,
+) => void;
+
 function normalizeAnimatedImageFps(fps: number | undefined): number {
   const fallbackFps = 10;
   const safeFps = typeof fps === 'number' && Number.isFinite(fps) ? fps : fallbackFps;
@@ -77,6 +85,7 @@ export function buildAnimatedImageVideoFilter(options: {
 export interface MediaGeneratorOptions {
   logDebug?: (message: string) => void;
   now?: () => number;
+  execFile?: MediaGeneratorExecFile;
 }
 
 function sanitizeDebugToken(value: string, fallback: string): string {
@@ -274,6 +283,14 @@ export class MediaGenerator {
     const duration = endTime - start + safePadding;
     const mediaInput = normalizeMediaInput(videoPath);
     const inputDescription = describeMediaInputForDebugLog(videoPath);
+    const hasSelectedAudioStream =
+      !mediaInput.singleResolvedStream &&
+      typeof audioStreamIndex === 'number' &&
+      Number.isInteger(audioStreamIndex) &&
+      audioStreamIndex >= 0;
+    const isLocalMatroskaMedia =
+      !/^[A-Za-z][A-Za-z\d+.-]*:\/\//.test(mediaInput.path) &&
+      /\.(?:mkv|mka|mks|webm)$/i.test(mediaInput.path);
 
     return new Promise((resolve, reject) => {
       const outputPath = this.createTempOutputPath('audio', 'mp3');
@@ -284,16 +301,14 @@ export class MediaGenerator {
         '-t',
         duration.toString(),
         ...mediaInput.inputArgs,
+        ...(hasSelectedAudioStream && isLocalMatroskaMedia
+          ? ['-probesize', '32768', '-analyzeduration', '0']
+          : []),
         '-i',
         mediaInput.path,
       ];
 
-      if (
+      if (hasSelectedAudioStream) {
-        !mediaInput.singleResolvedStream &&
-        typeof audioStreamIndex === 'number' &&
-        Number.isInteger(audioStreamIndex) &&
-        audioStreamIndex >= 0
-      ) {
         args.push('-map', `0:${audioStreamIndex}`);
       }
 
@@ -321,7 +336,8 @@ export class MediaGenerator {
       this.logMediaDebug(
         `audio start ${inputDescription} start=${start} duration=${duration} padding=${safePadding}`,
       );
-      execFile('ffmpeg', args, { timeout: 30000 }, (error) => {
+      const runExecFile: MediaGeneratorExecFile = this.options.execFile ?? execFile;
+      runExecFile('ffmpeg', args, { timeout: AUDIO_GENERATION_TIMEOUT_MS }, (error) => {
         if (error) {
           this.logMediaDebug(
             `audio failed ${inputDescription} elapsedMs=${this.elapsedMs(startedAt)} ${describeFfmpegFailureForDebugLog(error)}`,
@@ -338,7 +354,15 @@ export class MediaGenerator {
           );
           resolve(data);
         } catch (err) {
-          reject(err);
+          if ((err as NodeJS.ErrnoException).code === 'ENOENT') {
+            reject(
+              new Error(
+                'FFmpeg audio generation failed: FFmpeg exited without creating an output file.',
+              ),
+            );
+          } else {
+            reject(err);
+          }
         }
       });
     });