From bffbd43672ca9c3366b0e94c20fadb26d86dfdb9 Mon Sep 17 00:00:00 2001
From: erik-nilcoast <138068205+erik-nilcoast@users.noreply.github.com>
Date: Sun, 16 Mar 2025 16:58:18 -0500
Subject: feat(workers): allows videoWorker to use ytdlp command line arguments
 specified in the config. Fixes #775 #792 (#1117)

---
 apps/workers/videoWorker.ts   | 3 ++-
 docs/docs/03-configuration.md | 1 +
 packages/shared/config.ts     | 5 +++++
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/apps/workers/videoWorker.ts b/apps/workers/videoWorker.ts
index 32f16f97..23ead8f8 100644
--- a/apps/workers/videoWorker.ts
+++ b/apps/workers/videoWorker.ts
@@ -63,7 +63,6 @@ export class VideoWorker {
 }
 
 function prepareYtDlpArguments(url: string, assetPath: string) {
-  // TODO allow custom commandline arguments?
   const ytDlpArguments = [url];
   if (serverConfig.crawler.maxVideoDownloadSize > 0) {
     ytDlpArguments.push(
@@ -71,6 +70,8 @@ function prepareYtDlpArguments(url: string, assetPath: string) {
       `best[filesize<${serverConfig.crawler.maxVideoDownloadSize}M]`,
     );
   }
+
+  ytDlpArguments.push(...serverConfig.crawler.ytDlpArguments);
   ytDlpArguments.push("-o", assetPath);
   ytDlpArguments.push("--no-playlist");
   return ytDlpArguments;
diff --git a/docs/docs/03-configuration.md b/docs/docs/03-configuration.md
index 27f9f14e..36de8295 100644
--- a/docs/docs/03-configuration.md
+++ b/docs/docs/03-configuration.md
@@ -87,6 +87,7 @@ Either `OPENAI_API_KEY` or `OLLAMA_BASE_URL` need to be set for automatic taggin
 | CRAWLER_VIDEO_DOWNLOAD_MAX_SIZE    | No       | 50      | The maximum file size for the downloaded video. The quality will be chosen accordingly. Use -1 to disable the limit.                                                                                                                                                                                                                                                          |
 | CRAWLER_VIDEO_DOWNLOAD_TIMEOUT_SEC | No       | 600     | How long to wait for the video download to finish                                                                                                                                                                                                                                                                                                                             |
 | CRAWLER_ENABLE_ADBLOCKER           | No       | true    | Whether to enable an adblocker in the crawler or not. If you're facing troubles downloading the adblocking lists on worker startup, you can disable this.                                                                                                                                                                                                                     |
+| CRAWLER_YTDLP_ARGS           | No       | []    | Include additional yt-dlp arguments to be passed at crawl time separated by %%: https://github.com/yt-dlp/yt-dlp?tab=readme-ov-file#general-options
 
 ## OCR Configs
 
diff --git a/packages/shared/config.ts b/packages/shared/config.ts
index 1295fdbf..f487478f 100644
--- a/packages/shared/config.ts
+++ b/packages/shared/config.ts
@@ -49,6 +49,10 @@ const allEnv = z.object({
   CRAWLER_VIDEO_DOWNLOAD_MAX_SIZE: z.coerce.number().default(50),
   CRAWLER_VIDEO_DOWNLOAD_TIMEOUT_SEC: z.coerce.number().default(10 * 60),
   CRAWLER_ENABLE_ADBLOCKER: stringBool("true"),
+  CRAWLER_YTDLP_ARGS: z
+    .string()
+    .default("")
+    .transform((t) => t.split("%%").filter((a) => a)),
   MEILI_ADDR: z.string().optional(),
   MEILI_MASTER_KEY: z.string().default(""),
   LOG_LEVEL: z.string().default("debug"),
@@ -116,6 +120,7 @@ const serverConfigSchema = allEnv.transform((val) => {
       maxVideoDownloadSize: val.CRAWLER_VIDEO_DOWNLOAD_MAX_SIZE,
       downloadVideoTimeout: val.CRAWLER_VIDEO_DOWNLOAD_TIMEOUT_SEC,
       enableAdblocker: val.CRAWLER_ENABLE_ADBLOCKER,
+      ytDlpArguments: val.CRAWLER_YTDLP_ARGS,
     },
     ocr: {
       langs: val.OCR_LANGS,
-- 
cgit v1.2.3-70-g09d2