commit 90fd8452e28d057895650bf68e67df69a5f9f3f1
parent 54305634cbeb1a3b3bb50f8833df13213b299f62
Author: I Mean I'm Just Saying <imeanimjustsaying@kiwifarms.st>
Date: Sat, 30 May 2026 00:03:35 -0400
Fix single pipeline-skip transcribe false-detecting audio files with prefix
Diffstat:
6 files changed, 116 insertions(+), 36 deletions(-)
diff --git a/common/controller/transcribeOne.ts b/common/controller/transcribeOne.ts
@@ -10,6 +10,7 @@ import {
getSettings,
} from "../lib/settings";
import { normalizeTranscript } from "./normalizeTranscript";
+import { isRealAudioFile } from "../lib/videoStatus";
const { pathExists, readdir, rename } = fs;
@@ -23,9 +24,7 @@ async function resolveAudioFile(
if (await pathExists(path.join(videoDir, requested))) return requested;
if (strict) return null;
const entries = await readdir(videoDir);
- const candidates = entries.filter(
- (e) => e.startsWith("audio.") && !e.endsWith(".part"),
- );
+ const candidates = entries.filter(isRealAudioFile);
if (candidates.length === 0) return null;
for (const preferred of AUDIO_PREFERENCE) {
if (candidates.includes(preferred)) return preferred;
diff --git a/common/lib/videoStatus.ts b/common/lib/videoStatus.ts
@@ -25,6 +25,35 @@ export const LIVE_CHAT_FILENAME = "transcript.live_chat.json";
export const LIVE_CHAT_CUES_FILENAME = "live_chat.cues.json";
export const META_FILENAME = "metadata.info.json";
+// A finalized media output under data/<id>/audio.<ext>. Excludes yt-dlp partials
+// (.part) and audio-check snapshots (.part.good/.part.testing), the metadata
+// sidecar (*.info.json), temp work files (.tmp-...), and the live-chat sidecar
+// (audio.live_chat.json[.part]) — live_chat ignores the `subtitle:` output prefix
+// and lands under the default audio.%(ext)s template, which is the source of the
+// "no audio file found" whisper crash.
+export function isRealAudioFile(name: string): boolean {
+ if (!name.startsWith("audio.")) return false;
+ if (name.includes(".tmp-")) return false;
+ if (name.endsWith(".info.json")) return false;
+ if (name.endsWith(".live_chat.json")) return false;
+ if (name.endsWith(".part")) return false; // covers .live_chat.json.part too
+ if (name.endsWith(".part.good")) return false;
+ if (name.endsWith(".part.testing")) return false;
+ return true;
+}
+
+// A genuine resumable partial: an interrupted audio download, NOT a live-chat
+// sidecar that merely ends in .part.
+export function isPartAudioFile(name: string): boolean {
+ if (!name.startsWith("audio.")) return false;
+ if (name.includes(".tmp-")) return false;
+ if (name.endsWith(".live_chat.json.part")) return false;
+ if (!name.endsWith(".part")) return false;
+ if (name.endsWith(".part.good")) return false;
+ if (name.endsWith(".part.testing")) return false;
+ return true;
+}
+
export type SubTrack = {
// "live_chat" | language code like "es", "en-orig", etc.
track: string;
@@ -58,26 +87,8 @@ export async function readVideoFiles(
const hasYtVtt = entries.includes(VTT_FILENAME);
const hasWhisper = entries.includes(WHISPER_FILENAME);
const hasCuesJson = entries.includes(CUES_JSON_FILENAME);
- const audioFiles = entries.filter(
- (e) =>
- e.startsWith("audio.") &&
- !e.includes(".tmp-") &&
- !e.endsWith(".info.json") &&
- // yt-dlp writes audio.<ext>.part while downloading; treat those as
- // incomplete so prefilters re-invoke yt-dlp (which resumes via -c).
- !e.endsWith(".part") &&
- // audio-check internals — not user-visible audio outputs.
- !e.endsWith(".part.good") &&
- !e.endsWith(".part.testing"),
- );
- const partAudioFiles = entries.filter(
- (e) =>
- e.startsWith("audio.") &&
- !e.includes(".tmp-") &&
- e.endsWith(".part") &&
- !e.endsWith(".part.good") &&
- !e.endsWith(".part.testing"),
- );
+ const audioFiles = entries.filter(isRealAudioFile);
+ const partAudioFiles = entries.filter(isPartAudioFile);
let isUntranscribable = false;
if (hasWhisper && opts.checkUntranscribable) {
const whisperPath = path.join(videoDir, WHISPER_FILENAME);
diff --git a/editor/CHANGELOG.md b/editor/CHANGELOG.md
@@ -1,6 +1,7 @@
# Changelog
## [Unreleased]
+- **"Audio + Whisper (skip pipeline)" no longer mistakes a live-chat sidecar for audio.** yt-dlp writes the live-chat track as `audio.live_chat.json` (it ignores the subtitle output path), so an interrupted download could leave an `audio.live_chat.json.part` behind. One-click whisper saw the `audio.` prefix, decided audio was already on disk, skipped the download, and then failed with "no audio file found". Live-chat sidecars (`.part` or completed) are now excluded everywhere audio is detected, so whisper downloads the real audio and transcribes as expected.
- **Cleanup tasks on `/actionable`.** The global Actionable view now surfaces two housekeeping sections alongside the download/transcribe backlog: **channels with cleanable transcribed audio** (videos that have a whisper transcript but still keep their audio on disk) and **channels with extra audio formats** (videos with leftover audio files beside the channel's configured format), each with a per-channel count and an inline button. Because these delete files, the buttons pop a confirm dialog before queuing. The dashboard "Needs attention" card is unchanged — it still tracks only download/transcribe work.
- **Per-video "do not clean" archive toggle.** Each video detail page has a new **Archive media** section to mark a video "do not clean". Marked videos are skipped by both cleanup jobs (their audio is preserved for archiving) and excluded from the cleanup counts on `/actionable`. The marker is reversible from the same toggle, and an "archived" badge shows on the video page while it's set.
- **Twitch.tv support.** Twitch is now a first-class platform: Twitch channel/VOD URLs are auto-detected, "Twitch" is selectable in the channel form's platform dropdown and the charts platform filter, videos play via an in-browser Twitch embed, and downloads are queued on a `platform:twitch` queue like the other platforms.
diff --git a/editor/app/channels/[slug]/bulkVideoActions.ts b/editor/app/channels/[slug]/bulkVideoActions.ts
@@ -10,6 +10,7 @@ import {
queueKeyForUrl,
} from "yt-dlp-transcript-common/lib/platform";
import { readChannelConfig } from "yt-dlp-transcript-common/controller/channels";
+import { isRealAudioFile } from "yt-dlp-transcript-common/lib/videoStatus";
import {
downloadVideoPipelineAction,
markVideoUntranscribableAction,
@@ -53,17 +54,7 @@ async function firstAudioFile(slug: string, videoId: string): Promise<string | n
} catch {
return null;
}
- const audio = entries
- .filter(
- (e) =>
- e.startsWith("audio.") &&
- !e.includes(".tmp-") &&
- !e.endsWith(".part") &&
- !e.endsWith(".part.good") &&
- !e.endsWith(".part.testing") &&
- !e.endsWith(".info.json"),
- )
- .sort();
+ const audio = entries.filter(isRealAudioFile).sort();
return audio[0] ?? null;
}
diff --git a/editor/app/channels/[slug]/videos/[id]/videoActions.ts b/editor/app/channels/[slug]/videos/[id]/videoActions.ts
@@ -10,6 +10,7 @@ import type {
} from "yt-dlp-transcript-common/lib/channelConfig";
import { AUDIO_FORMAT_VALUES } from "yt-dlp-transcript-common/lib/channelConfig";
import { getPaths } from "yt-dlp-transcript-common/lib/paths";
+import { isRealAudioFile } from "yt-dlp-transcript-common/lib/videoStatus";
import {
platformQueueKey,
queueKeyForUrl,
@@ -194,7 +195,7 @@ export async function whisperVideoAction(
fn: async (onLog, signal, _setProgress, ctx) => {
const audioFormat = r.config.audioFormat ?? "mp3";
const entries = await readdir(videoDir).catch(() => [] as string[]);
- const hasAudio = entries.some((e: string) => e.startsWith("audio."));
+ const hasAudio = entries.some(isRealAudioFile);
if (!hasAudio) {
const url = await findVideoSourceUrl(paths, slug, videoId);
if (!url) {
diff --git a/editor/e2e/whisper-video.spec.ts b/editor/e2e/whisper-video.spec.ts
@@ -1,4 +1,4 @@
-import { readFile, rm } from "node:fs/promises";
+import { readFile, rm, writeFile } from "node:fs/promises";
import { test, expect } from "@playwright/test";
import { pathExists, resetData, resolvePath } from "./helpers";
@@ -63,3 +63,80 @@ test("one-click whisper skips download when audio is already on disk", async ({
const invocations = await readFile(invocationsPath, "utf8").catch(() => "");
expect(invocations).not.toContain("download-one:");
});
+
+// yt-dlp writes the live-chat track under the default audio.%(ext)s output
+// template (live_chat ignores the `subtitle:` prefix), so an interrupted
+// download leaves an `audio.live_chat.json.part` sidecar. A naive
+// startsWith("audio.") prefilter mistook it for real audio, skipped the
+// download, then crashed in whisper with "no audio file found".
+test("one-click whisper ignores a leftover live-chat .part sidecar", async ({
+ page,
+}) => {
+ await resetData("youtube-with-playlist");
+ await rm(
+ resolvePath(
+ "test-transcripts/channels/test-youtube/data/fake00000001/transcript.en.vtt",
+ ),
+ );
+ await writeFile(
+ resolvePath(
+ "test-transcripts/channels/test-youtube/data/fake00000001/audio.live_chat.json.part",
+ ),
+ "[]",
+ );
+
+ await page.goto("/channels/test-youtube/videos/fake00000001");
+ await page.getByLabel("download mode").selectOption("whisper");
+ await page.getByRole("button", { name: /^Audio \+ Whisper$/ }).click();
+
+ const log = page.getByLabel("Audio + Whisper for fake00000001 output");
+ await expect(log).toContainText("downloading", { timeout: 30_000 });
+ await expect(log).toContainText("Transcribe fake00000001 done", {
+ timeout: 30_000,
+ });
+ await expect(log).not.toContainText("skipping download");
+ await expect(log).not.toContainText("no audio file found");
+
+ expect(
+ await pathExists(
+ "test-transcripts/channels/test-youtube/data/fake00000001/audio.mp3",
+ ),
+ ).toBe(true);
+ expect(
+ await pathExists(
+ "test-transcripts/channels/test-youtube/data/fake00000001/transcript.json",
+ ),
+ ).toBe(true);
+});
+
+// The same blind spot for a completed live-chat sidecar: a finalized
+// audio.live_chat.json (no .part) must not be treated as audio either, or it
+// would be handed to whisper as if it were the audio file.
+test("one-click whisper ignores a completed live-chat sidecar", async ({
+ page,
+}) => {
+ await resetData("youtube-with-playlist");
+ await rm(
+ resolvePath(
+ "test-transcripts/channels/test-youtube/data/fake00000001/transcript.en.vtt",
+ ),
+ );
+ await writeFile(
+ resolvePath(
+ "test-transcripts/channels/test-youtube/data/fake00000001/audio.live_chat.json",
+ ),
+ "[]",
+ );
+
+ await page.goto("/channels/test-youtube/videos/fake00000001");
+ await page.getByLabel("download mode").selectOption("whisper");
+ await page.getByRole("button", { name: /^Audio \+ Whisper$/ }).click();
+
+ const log = page.getByLabel("Audio + Whisper for fake00000001 output");
+ await expect(log).toContainText("downloading", { timeout: 30_000 });
+ await expect(log).toContainText("Transcribe fake00000001 done", {
+ timeout: 30_000,
+ });
+ await expect(log).not.toContainText("skipping download");
+ await expect(log).not.toContainText("no audio file found");
+});
