fix(youtube): parse mpv EDL stream URLs with byte-length guards (#134)

changes/youtube-stream-edl-options.md

@@ -0,0 +1,4 @@
+type: fixed
+area: youtube
+
+- Fixed direct YouTube stream media extraction by parsing mpv EDL stream URLs with their byte-length guards, preventing trailing EDL segment options from corrupting signed googlevideo URLs and causing ffmpeg 403 errors.

release/release-notes.md

@@ -1,28 +0,0 @@
-## Highlights
-
-### Fixed
-
-- **YouTube Background Cache:** Fixed Windows background media cache startup for YouTube URLs opened directly in mpv.
-  - Resolved stream URLs are now tracked even when mpv still exposes the original YouTube playlist entry.
-  - Queued Anki media updates can append audio and images after the cache finishes instead of staying text-only.
-
-- **YouTube Subtitle Picker Notifications:** Manual subtitle picker requests now show immediate status while SubMiner probes tracks and opens the modal.
-  - Subtitle download progress is replaced with a transient success notification after tracks load.
-
-## What's Changed
-
-- feat(youtube): notify on manual picker open and show success after track load by @ksyasuda in #133
-- fix(youtube): recover source URL for background media cache on direct mpv open by @ksyasuda in #132
-
-## Installation
-
-See the README and docs/installation guide for full setup steps.
-
-## Assets
-
-- Linux: `SubMiner.AppImage`
-- macOS: `SubMiner-*.dmg` and `SubMiner-*.zip`
-- Windows: `SubMiner-*.exe` and `SubMiner-*-win.zip`
-- Optional extras: `subminer-assets.tar.gz` and the `subminer` launcher
-
-Note: the `subminer` wrapper script uses Bun (`#!/usr/bin/env bun`), so `bun` must be installed and on `PATH`.

src/anki-integration/card-creation-manual-update.test.ts

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict';
 import test from 'node:test';
 
 import { CardCreationService } from './card-creation';
+import { toMpvEdlValue } from './mpv-edl-test-utils';
 import type { MediaInput } from '../media-generator';
 import type { AnkiConnectConfig } from '../types/anki';
 
@@ -269,9 +270,11 @@ test('manual clipboard subtitle update uses resolved mpv stream URLs for remote
   const imagePaths: string[] = [];
   const recordMediaPath = (mediaInput: MediaInput): string =>
     typeof mediaInput === 'string' ? mediaInput : mediaInput.path;
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const videoUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
   const edlSource = [
-    'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
-    '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
+    `edl://!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(audioUrl)}`,
+    `!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(videoUrl)}`,
     '!global_tags,title=test',
   ].join(';');
 
@@ -354,8 +357,8 @@ test('manual clipboard subtitle update uses resolved mpv stream URLs for remote
 
   await service.updateLastAddedFromClipboard('一行目\n\n二行目');
 
-  assert.deepEqual(audioPaths, ['https://audio.example/videoplayback?mime=audio%2Fwebm']);
-  assert.deepEqual(imagePaths, ['https://video.example/videoplayback?mime=video%2Fmp4']);
+  assert.deepEqual(audioPaths, [audioUrl]);
+  assert.deepEqual(imagePaths, [videoUrl]);
   assert.equal(storedMedia.length, 2);
   assert.equal(updatedFields.length, 1);
   assert.equal(updatedFields[0]?.Sentence, '一行目 二行目');

src/anki-integration/card-creation.test.ts

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict';
 import test from 'node:test';
 
 import { CardCreationService } from './card-creation';
+import { toMpvEdlValue } from './mpv-edl-test-utils';
 import type { MediaInput } from '../media-generator';
 import type { AnkiConnectConfig } from '../types/anki';
 
@@ -290,9 +291,11 @@ test('CardCreationService uses stream-open-filename for remote media generation'
   const imagePaths: string[] = [];
   const recordMediaPath = (mediaInput: MediaInput): string =>
     typeof mediaInput === 'string' ? mediaInput : mediaInput.path;
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const videoUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
   const edlSource = [
-    'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
-    '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
+    `edl://!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(audioUrl)}`,
+    `!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(videoUrl)}`,
     '!global_tags,title=test',
   ].join(';');
 
@@ -397,8 +400,8 @@ test('CardCreationService uses stream-open-filename for remote media generation'
   const created = await service.createSentenceCard('テスト', 0, 1);
 
   assert.equal(created, true);
-  assert.deepEqual(audioPaths, ['https://audio.example/videoplayback?mime=audio%2Fwebm']);
-  assert.deepEqual(imagePaths, ['https://video.example/videoplayback?mime=video%2Fmp4']);
+  assert.deepEqual(audioPaths, [audioUrl]);
+  assert.deepEqual(imagePaths, [videoUrl]);
 });
 
 test('CardCreationService does not use mpv stream indexes for ready cached YouTube media', async () => {

src/anki-integration/media-source.test.ts

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict';
 import test from 'node:test';
 
 import * as mediaSource from './media-source';
+import { toMpvEdlValue } from './mpv-edl-test-utils';
 
 const { resolveMediaGenerationInputPath } = mediaSource;
 
@@ -53,9 +54,11 @@ test('resolveMediaGenerationInputPath prefers stream-open-filename for remote me
 });
 
 test('resolveMediaGenerationInputPath unwraps mpv edl source for audio and video', async () => {
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const videoUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
   const edlSource = [
-    'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
-    '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
+    `edl://!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(audioUrl)}`,
+    `!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(videoUrl)}`,
     '!global_tags,title=test',
   ].join(';');
 
@@ -74,8 +77,52 @@ test('resolveMediaGenerationInputPath unwraps mpv edl source for audio and video
     'video',
   );
 
-  assert.equal(audioResult, 'https://audio.example/videoplayback?mime=audio%2Fwebm');
-  assert.equal(videoResult, 'https://video.example/videoplayback?mime=video%2Fmp4');
+  assert.equal(audioResult, audioUrl);
+  assert.equal(videoResult, videoUrl);
+});
+
+test('resolveMediaGenerationInputPath strips mpv edl segment options from unwrapped streams', async () => {
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const signedVideoUrl =
+    'https://rr1---sn.example.googlevideo.com/videoplayback?mime=video%2Fmp4&mn=sn-a,sn-b&lsig=abc%3D';
+  const edlSource = [
+    `edl://!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(audioUrl)}`,
+    `!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(signedVideoUrl)},title=clip,length=73,timestamps=chapters`,
+    '!global_tags,title=test',
+  ].join(';');
+
+  const result = await resolveMediaGenerationInputPath(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async () => edlSource,
+    },
+    'video',
+  );
+
+  assert.equal(result, signedVideoUrl);
+});
+
+test('resolveMediaGenerationInputPath ignores length-guarded URLs in mpv edl headers', async () => {
+  const initUrl = 'https://init.example/init.mp4';
+  const audioUrl = 'https://audio.example/stream';
+  const videoUrl = 'https://video.example/stream';
+  const edlSource = [
+    `edl://!mp4_dash,init=${toMpvEdlValue(initUrl)}`,
+    '!new_stream',
+    toMpvEdlValue(audioUrl),
+    '!new_stream',
+    toMpvEdlValue(videoUrl),
+  ].join(';');
+
+  const audioResult = await resolveMediaGenerationInputPath(
+    {
+      currentVideoPath: 'https://www.youtube.com/watch?v=abc123',
+      requestProperty: async () => edlSource,
+    },
+    'audio',
+  );
+
+  assert.equal(audioResult, audioUrl);
 });
 
 test('resolveMediaGenerationInputPath falls back to currentVideoPath when stream-open-filename fails', async () => {
@@ -97,9 +144,11 @@ test('resolveMediaGenerationInput returns single-stream metadata for mpv EDL URL
   ).resolveMediaGenerationInput;
   assert.equal(typeof resolver, 'function');
 
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const videoUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
   const edlSource = [
-    'edl://!new_stream;!no_clip;!no_chapters;%70%https://audio.example/videoplayback?mime=audio%2Fwebm',
-    '!new_stream;!no_clip;!no_chapters;%69%https://video.example/videoplayback?mime=video%2Fmp4',
+    `edl://!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(audioUrl)}`,
+    `!new_stream;!no_clip;!no_chapters;${toMpvEdlValue(videoUrl)}`,
   ].join(';');
 
   const result = await resolver!(
@@ -117,7 +166,7 @@ test('resolveMediaGenerationInput returns single-stream metadata for mpv EDL URL
     'audio',
   );
 
-  assert.equal(result?.path, 'https://audio.example/videoplayback?mime=audio%2Fwebm');
+  assert.equal(result?.path, audioUrl);
   assert.equal(result?.singleResolvedStream, true);
   assert.equal(result?.inputOptions?.reconnect, true);
   assert.equal(result?.inputOptions?.userAgent, 'Mozilla/5.0');

src/anki-integration/media-source.ts

@@ -1,6 +1,7 @@
 import { isRemoteMediaPath } from '../jimaku/utils';
 import type { MediaInput, MediaInputOptions } from '../media-input';
 import type { MpvClient } from '../types/runtime';
+import { extractFileUrlsFromMpvEdlSource } from './mpv-edl';
 
 export type MediaGenerationKind = 'audio' | 'video';
 export type MediaGenerationInputSource =
@@ -73,9 +74,8 @@ function normalizeHeaderName(value: string): string | null {
 }
 
 function extractUrlsFromMpvEdlSource(source: string): string[] {
-  const matches = source.matchAll(/%\d+%(https?:\/\/.*?)(?=;!new_stream|;!global_tags|$)/gms);
-  return [...matches]
-    .map((match) => trimToNonEmptyString(match[1]))
+  return extractFileUrlsFromMpvEdlSource(source)
+    .map((value) => trimToNonEmptyString(value))
     .filter((value): value is string => value !== null);
 }
 

src/anki-integration/mpv-edl-test-utils.ts

@@ -0,0 +1,3 @@
+export function toMpvEdlValue(value: string): string {
+  return `%${Buffer.byteLength(value, 'utf8')}%${value}`;
+}

src/anki-integration/mpv-edl.test.ts

@@ -0,0 +1,37 @@
+import assert from 'node:assert/strict';
+import test from 'node:test';
+
+import { extractFileUrlsFromMpvEdlSource } from './mpv-edl';
+import { toMpvEdlValue } from './mpv-edl-test-utils';
+
+test('extractFileUrlsFromMpvEdlSource honors length-guarded file values', () => {
+  const url =
+    'https://rr1---sn.example.googlevideo.com/videoplayback?mime=video%2Fmp4&mn=sn-a,sn-b&lsig=abc%3D';
+  const source = `edl://!new_stream;${toMpvEdlValue(url)},title=clip,length=73`;
+
+  assert.deepEqual(extractFileUrlsFromMpvEdlSource(source), [url]);
+});
+
+test('extractFileUrlsFromMpvEdlSource reads file parameters', () => {
+  const initUrl = 'https://init.example/init.mp4';
+  const fileUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
+  const source = `edl://!mp4_dash,init=${toMpvEdlValue(initUrl)};file=${toMpvEdlValue(
+    fileUrl,
+  )},length=42`;
+
+  assert.deepEqual(extractFileUrlsFromMpvEdlSource(source), [fileUrl]);
+});
+
+test('extractFileUrlsFromMpvEdlSource aggregates file URLs across entries', () => {
+  const audioUrl = 'https://audio.example/videoplayback?mime=audio%2Fwebm';
+  const videoUrl = 'https://video.example/videoplayback?mime=video%2Fmp4';
+  const source = [
+    'edl://!new_stream',
+    toMpvEdlValue(audioUrl),
+    '!new_stream',
+    `file=${toMpvEdlValue(videoUrl)},length=50`,
+    '!global_tags,title=test',
+  ].join(';');
+
+  assert.deepEqual(extractFileUrlsFromMpvEdlSource(source), [audioUrl, videoUrl]);
+});

src/anki-integration/mpv-edl.ts

@@ -0,0 +1,195 @@
+const EDL_URI_PREFIX = 'edl://';
+
+const BYTE_COMMA = ','.charCodeAt(0);
+const BYTE_CR = '\r'.charCodeAt(0);
+const BYTE_EQUALS = '='.charCodeAt(0);
+const BYTE_EXCLAMATION = '!'.charCodeAt(0);
+const BYTE_LF = '\n'.charCodeAt(0);
+const BYTE_PERCENT = '%'.charCodeAt(0);
+const BYTE_SEMICOLON = ';'.charCodeAt(0);
+
+function isDigitByte(value: number | undefined): value is number {
+  return value !== undefined && value >= 48 && value <= 57;
+}
+
+function isEntrySeparator(value: number | undefined): boolean {
+  return value === BYTE_SEMICOLON || value === BYTE_LF || value === BYTE_CR;
+}
+
+function isParamSeparator(value: number | undefined): boolean {
+  return value === BYTE_COMMA || isEntrySeparator(value);
+}
+
+function decodeBytes(buffer: Buffer, start: number, end: number): string {
+  return buffer.subarray(start, end).toString('utf8');
+}
+
+function isHttpUrl(value: string): boolean {
+  return /^https?:\/\//i.test(value);
+}
+
+function toEdlDataBuffer(source: string): Buffer {
+  const data = source.startsWith(EDL_URI_PREFIX) ? source.slice(EDL_URI_PREFIX.length) : source;
+  return Buffer.from(data, 'utf8');
+}
+
+function parseLengthGuardedValue(
+  buffer: Buffer,
+  position: number,
+): { value: string; end: number } | null {
+  if (buffer[position] !== BYTE_PERCENT) {
+    return null;
+  }
+
+  let cursor = position + 1;
+  if (!isDigitByte(buffer[cursor])) {
+    return null;
+  }
+
+  let byteLength = 0;
+  while (true) {
+    const digit = buffer[cursor];
+    if (!isDigitByte(digit)) {
+      break;
+    }
+    byteLength = byteLength * 10 + (digit - 48);
+    cursor += 1;
+  }
+
+  if (buffer[cursor] !== BYTE_PERCENT) {
+    return null;
+  }
+
+  const valueStart = cursor + 1;
+  const valueEnd = valueStart + byteLength;
+  if (valueEnd > buffer.length) {
+    return null;
+  }
+
+  return {
+    value: decodeBytes(buffer, valueStart, valueEnd),
+    end: valueEnd,
+  };
+}
+
+function skipEntrySeparators(buffer: Buffer, position: number): number {
+  let cursor = position;
+  while (cursor < buffer.length && isEntrySeparator(buffer[cursor])) {
+    cursor += 1;
+  }
+  return cursor;
+}
+
+function skipEntry(buffer: Buffer, position: number): number {
+  let cursor = position;
+  while (cursor < buffer.length) {
+    const guardedValue = parseLengthGuardedValue(buffer, cursor);
+    if (guardedValue) {
+      cursor = guardedValue.end;
+      continue;
+    }
+    if (isEntrySeparator(buffer[cursor])) {
+      break;
+    }
+    cursor += 1;
+  }
+  return cursor;
+}
+
+function parseRawValue(buffer: Buffer, position: number): { value: string; end: number } {
+  let cursor = position;
+  while (
+    cursor < buffer.length &&
+    !isParamSeparator(buffer[cursor]) &&
+    buffer[cursor] !== BYTE_EXCLAMATION
+  ) {
+    cursor += 1;
+  }
+  return {
+    value: decodeBytes(buffer, position, cursor),
+    end: cursor,
+  };
+}
+
+function parseParamValue(buffer: Buffer, position: number): { value: string; end: number } {
+  return parseLengthGuardedValue(buffer, position) ?? parseRawValue(buffer, position);
+}
+
+function parseOptionalParamName(
+  buffer: Buffer,
+  position: number,
+): { name: string | null; valueStart: number } {
+  let cursor = position;
+  while (
+    cursor < buffer.length &&
+    !isParamSeparator(buffer[cursor]) &&
+    buffer[cursor] !== BYTE_PERCENT &&
+    buffer[cursor] !== BYTE_EXCLAMATION
+  ) {
+    if (buffer[cursor] === BYTE_EQUALS) {
+      return {
+        name: decodeBytes(buffer, position, cursor),
+        valueStart: cursor + 1,
+      };
+    }
+    cursor += 1;
+  }
+
+  return { name: null, valueStart: position };
+}
+
+function parseSegmentEntry(buffer: Buffer, position: number): { urls: string[]; end: number } {
+  const urls: string[] = [];
+  let cursor = position;
+  let unnamedParamIndex = 0;
+
+  while (cursor < buffer.length && !isEntrySeparator(buffer[cursor])) {
+    const { name, valueStart } = parseOptionalParamName(buffer, cursor);
+    const value = parseParamValue(buffer, valueStart);
+    const lowerName = name?.toLowerCase() ?? null;
+    const isFileParam = lowerName === 'file' || (lowerName === null && unnamedParamIndex === 0);
+
+    if (isFileParam && isHttpUrl(value.value)) {
+      urls.push(value.value);
+    }
+
+    if (lowerName === null) {
+      unnamedParamIndex += 1;
+    }
+
+    cursor = value.end;
+    if (buffer[cursor] === BYTE_COMMA) {
+      cursor += 1;
+      continue;
+    }
+    if (!isEntrySeparator(buffer[cursor])) {
+      cursor = skipEntry(buffer, cursor);
+    }
+  }
+
+  return { urls, end: cursor };
+}
+
+export function extractFileUrlsFromMpvEdlSource(source: string): string[] {
+  const buffer = toEdlDataBuffer(source);
+  const urls: string[] = [];
+  let cursor = 0;
+
+  while (cursor < buffer.length) {
+    cursor = skipEntrySeparators(buffer, cursor);
+    if (cursor >= buffer.length) {
+      break;
+    }
+
+    if (buffer[cursor] === BYTE_EXCLAMATION) {
+      cursor = skipEntry(buffer, cursor);
+      continue;
+    }
+
+    const segment = parseSegmentEntry(buffer, cursor);
+    urls.push(...segment.urls);
+    cursor = segment.end;
+  }
+
+  return urls;
+}