fix(anki): snapshot mining media clip timing (#197)

src/anki-integration/note-update-workflow.test.ts

@@ -472,6 +472,71 @@ test('NoteUpdateWorkflow uses subtitle sidebar context for sentence media timing
   assert.equal(miscInfoStartTime, 10);
 });
 
+test('NoteUpdateWorkflow snapshots one media range for audio and image without a mining context', async () => {
+  const harness = createWorkflowHarness();
+  const capturedContext: SubtitleMiningContext = {
+    source: 'overlay',
+    text: 'subtitle-text',
+    startTime: 31.5,
+    endTime: 34.25,
+  };
+  let captureCalls = 0;
+  let audioContext: SubtitleMiningContext | null = null;
+  let imageContext: SubtitleMiningContext | null = null;
+  let miscInfoStartTime: number | undefined;
+
+  harness.deps.client.notesInfo = async () =>
+    [
+      {
+        noteId: 42,
+        fields: {
+          Expression: { value: 'taberu' },
+          Sentence: { value: '' },
+          SentenceAudio: { value: '' },
+          Picture: { value: '' },
+          MiscInfo: { value: '' },
+        },
+      },
+    ] satisfies NoteUpdateWorkflowNoteInfo[];
+  harness.deps.getConfig = () => ({
+    fields: {
+      sentence: 'Sentence',
+      image: 'Picture',
+      miscInfo: 'MiscInfo',
+    },
+    media: {
+      generateAudio: true,
+      generateImage: true,
+      imageType: 'avif',
+    },
+    behavior: {},
+  });
+  harness.deps.getResolvedSentenceAudioFieldName = () => 'SentenceAudio';
+  harness.deps.captureSubtitleMediaContext = () => {
+    captureCalls += 1;
+    return capturedContext;
+  };
+  harness.deps.generateAudio = async (context?: SubtitleMiningContext) => {
+    audioContext = context ?? null;
+    return Buffer.from('audio');
+  };
+  harness.deps.generateImage = async (_leadInSeconds?: number, context?: SubtitleMiningContext) => {
+    imageContext = context ?? null;
+    return Buffer.from('image');
+  };
+  harness.deps.formatMiscInfoPattern = (_fallbackFilename, startTimeSeconds) => {
+    miscInfoStartTime = startTimeSeconds;
+    return `start:${startTimeSeconds}`;
+  };
+
+  await harness.workflow.execute(42);
+
+  assert.equal(captureCalls, 1);
+  assert.deepEqual(audioContext, capturedContext);
+  assert.deepEqual(imageContext, capturedContext);
+  assert.equal(miscInfoStartTime, 31.5);
+});
+
 test('NoteUpdateWorkflow queues media updates when YouTube cache is pending', async () => {
   const harness = createWorkflowHarness();
   const queuedUpdates: Array<{

src/anki-integration/note-update-workflow.ts

@@ -87,6 +87,7 @@ export interface NoteUpdateWorkflowDeps {
   ) => Promise<Buffer | null>;
   formatMiscInfoPattern: (fallbackFilename: string, startTimeSeconds?: number) => string;
   consumeSubtitleMiningContext?: () => SubtitleMiningContext | null;
+  captureSubtitleMediaContext?: () => SubtitleMiningContext | null;
   queuePendingYoutubeMediaUpdate?: (job: {
     noteId: number;
     noteInfo: NoteUpdateWorkflowNoteInfo;
@@ -203,6 +204,11 @@ export class NoteUpdateWorkflow {
         sentenceField,
         config.fields?.sentence,
       );
+      // Audio and image generation run sequentially and audio extraction can take tens of
+      // seconds, so resolve the clip range exactly once up front; reading live mpv sub
+      // timings per generator clips whichever line is on screen when each one starts.
+      const mediaTimingContext =
+        subtitleMiningContext ?? this.deps.captureSubtitleMediaContext?.() ?? null;
       const noteLabel = hasExpressionText ? expressionText : noteId;
 
       const currentSubtitleText = subtitleMiningContext?.text ?? this.deps.getCurrentSubtitleText();
@@ -240,7 +246,7 @@ export class NoteUpdateWorkflow {
           ? await this.deps.queuePendingYoutubeMediaUpdate({
               noteId,
               noteInfo,
-              context: subtitleMiningContext ?? undefined,
+              context: mediaTimingContext ?? undefined,
               label: noteLabel,
             })
           : false;
@@ -248,7 +254,7 @@ export class NoteUpdateWorkflow {
       if (!mediaCacheQueued && generateAudio) {
         try {
           const audioFilename = this.deps.generateAudioFilename();
-          const audioBuffer = await this.deps.generateAudio(subtitleMiningContext ?? undefined);
+          const audioBuffer = await this.deps.generateAudio(mediaTimingContext ?? undefined);
 
           if (audioBuffer) {
             await this.deps.client.storeMediaFile(audioFilename, audioBuffer);
@@ -276,7 +282,7 @@ export class NoteUpdateWorkflow {
           const imageFilename = this.deps.generateImageFilename();
           const imageBuffer = await this.deps.generateImage(
             animatedLeadInSeconds,
-            subtitleMiningContext ?? undefined,
+            mediaTimingContext ?? undefined,
           );
 
           if (imageBuffer) {
@@ -308,7 +314,7 @@ export class NoteUpdateWorkflow {
       if (!mediaCacheQueued && config.fields?.miscInfo) {
         const miscInfo = this.deps.formatMiscInfoPattern(
           miscInfoFilename || '',
-          subtitleMiningContext?.startTime ?? this.deps.getCurrentSubtitleStart(),
+          mediaTimingContext?.startTime ?? this.deps.getCurrentSubtitleStart(),
         );
         const miscInfoField = this.deps.resolveConfiguredFieldName(
           noteInfo,