Allow scoping the proxy to lookups only

YouTube's bot checks started once traffic went through a proxy exit IP,
while direct downloads had worked. YTDLP_PROXY_SCOPE=search keeps the
proxy on search and metadata — where it is needed to get past filtered
results — and lets the audio stream go out directly. Default stays "all",
so nothing changes unless it is set.

Searches also now run against a throwaway copy of the cookie file: they
run in parallel and yt-dlp rewrites that file on exit, so two of them
could clobber the jar the downloads depend on.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

.env.example

@@ -52,6 +52,12 @@ YTDLP_PATH=yt-dlp
 # стороне прокси — обычно нужен именно он). Можно с логином и паролем:
 # socks5h://user:pass@host:1080
 # YTDLP_PROXY=socks5h://127.0.0.1:1080
+
+# Где применять прокси: all — везде (по умолчанию), search — только поиск и
+# метаданные, а сам аудиопоток качать напрямую. Второй вариант выручает, когда
+# выходной IP прокси ловит от YouTube «Sign in to confirm you're not a bot»,
+# а без прокси скачивание работает.
+# YTDLP_PROXY_SCOPE=search
 #
 # Внимание: 127.0.0.1 внутри контейнера — это сам контейнер. Если прокси поднят
 # на хосте, используйте socks5h://host.docker.internal:1080 и добавьте в

README.md

@@ -292,8 +292,16 @@ YTDLP_PROXY=socks5h://127.0.0.1:1080
    ```
 
 3. **Подозрительный IP.** Выходные узлы Psiphon и прочих публичных прокси YouTube знает и
-   проверяет чаще. Здесь помогают именно cookies; экспортировать их лучше из браузера,
-   работающего через тот же прокси, чтобы сессия не выглядела прыгающей между странами.
+   проверяет чаще. Если без прокси скачивание работало, а с ним пошли бот-проверки — оставьте
+   прокси только поиску:
+
+   ```dotenv
+   YTDLP_PROXY_SCOPE=search
+   ```
+
+   Тогда через прокси идут поиск и метаданные (ради обхода фильтрации выдачи), а аудиопоток
+   качается напрямую. Второй путь — cookies, экспортированные из браузера, работающего через
+   тот же прокси, чтобы сессия не выглядела прыгающей между странами.
 
 ## Учётка YouTube (cookies)
 

src/config.ts

@@ -43,6 +43,11 @@ const schema = z.object({
 
   YTDLP_PATH: z.string().default("yt-dlp"),
   YTDLP_COOKIES: z.string().optional(),
+  /**
+   * Where the proxy applies: "all" for every request, "search" to keep the audio
+   * stream direct — useful when a proxy exit IP triggers YouTube's bot checks.
+   */
+  YTDLP_PROXY_SCOPE: z.enum(["all", "search"]).default("all"),
   /** Proxy for every yt-dlp request: http://, https://, socks5:// or socks5h://. */
   YTDLP_PROXY: z
     .string()

src/sources/ytdlp.ts

@@ -2,7 +2,9 @@ import { spawn } from "node:child_process";
 import { connect } from "node:net";
 import { randomUUID } from "node:crypto";
 import { constants } from "node:fs";
-import { access } from "node:fs/promises";
+import { access, copyFile, rm } from "node:fs/promises";
+import { tmpdir } from "node:os";
+import path from "node:path";
 import type { Readable } from "node:stream";
 import { config } from "../config.js";
 import { logger } from "../logger.js";
@@ -51,10 +53,21 @@ async function detectJsRuntime(): Promise<string | null> {
   }
 }
 
-function baseArgs(): string[] {
+interface ArgOptions {
+  /** Streaming audio; the proxy may be scoped away from this path. */
+  download?: boolean;
+  /** Cookie file to use instead of the configured one. */
+  cookies?: string;
+}
+
+function baseArgs(options: ArgOptions = {}): string[] {
   const args = ["--no-warnings", "--no-playlist-reverse", "--ignore-config", "--no-color", ...jsRuntimeArgs];
-  if (config.YTDLP_COOKIES) args.push("--cookies", config.YTDLP_COOKIES);
-  if (config.YTDLP_PROXY) args.push("--proxy", config.YTDLP_PROXY);
+  const cookies = options.cookies ?? config.YTDLP_COOKIES;
+  if (cookies) args.push("--cookies", cookies);
+  // A proxy exit IP is often flagged by YouTube, so it can be limited to the
+  // lookups that actually need it while playback goes out directly.
+  const proxied = config.YTDLP_PROXY_SCOPE === "all" || !options.download;
+  if (config.YTDLP_PROXY && proxied) args.push("--proxy", config.YTDLP_PROXY);
   for (const value of config.YTDLP_EXTRACTOR_ARGS?.split(";") ?? []) {
     const trimmed = value.trim();
     if (trimmed) args.push("--extractor-args", trimmed);
@@ -261,12 +274,21 @@ export async function search(
   requestedBy: Requester,
 ): Promise<Track[]> {
   const prefix = kind === "soundcloud" ? "scsearch" : "ytsearch";
-  const { stdout, stderr } = await runYtDlp([
-    ...baseArgs(),
-    "--flat-playlist",
-    "--dump-json",
-    `${prefix}${limit}:${query}`,
-  ]);
+  // Searches run in parallel and yt-dlp rewrites the cookie file when it exits,
+  // so give each one a throwaway copy: only downloads update the real file.
+  const cookies = await copyCookies();
+  let stdout: string;
+  let stderr: string;
+  try {
+    ({ stdout, stderr } = await runYtDlp([
+      ...baseArgs(cookies ? { cookies } : {}),
+      "--flat-playlist",
+      "--dump-json",
+      `${prefix}${limit}:${query}`,
+    ]));
+  } finally {
+    if (cookies) await rm(cookies, { force: true }).catch(() => {});
+  }
   const entries = parseNdjson(stdout);
   // yt-dlp can exit 0 with nothing to show (bot checks, region blocks). Without
   // this the panel would just render an empty list and say nothing at all.
@@ -321,12 +343,32 @@ export async function resolveUrl(
 
 /** Resolves a direct, time-limited media URL for a page URL (used when seeking). */
 export async function resolveStreamUrl(pageUrl: string): Promise<string> {
-  const { stdout } = await runYtDlp([...baseArgs(), "-f", "bestaudio/best", "--no-playlist", "-g", pageUrl]);
+  const { stdout } = await runYtDlp([
+    ...baseArgs({ download: true }),
+    "-f",
+    "bestaudio/best",
+    "--no-playlist",
+    "-g",
+    pageUrl,
+  ]);
   const url = stdout.split(/\r?\n/).map((l) => l.trim()).find(Boolean);
   if (!url) throw new UserFacingError("Не удалось получить ссылку на аудиопоток");
   return url;
 }
 
+/** Private, disposable copy of the cookie jar for concurrent reads. */
+async function copyCookies(): Promise<string | null> {
+  if (!config.YTDLP_COOKIES) return null;
+  const target = path.join(tmpdir(), `mbot-cookies-${randomUUID()}.txt`);
+  try {
+    await copyFile(config.YTDLP_COOKIES, target);
+    return target;
+  } catch (err) {
+    log.warn({ err }, "could not copy cookie file, using it directly");
+    return null;
+  }
+}
+
 export interface AudioProcess {
   stream: Readable;
   kill(): void;
@@ -338,7 +380,7 @@ export interface AudioProcess {
 export function openAudioStream(pageUrl: string): AudioProcess {
   const child = spawn(
     config.YTDLP_PATH,
-    [...baseArgs(), "-f", "bestaudio/best", "--no-playlist", "--quiet", "-o", "-", pageUrl],
+    [...baseArgs({ download: true }), "-f", "bestaudio/best", "--no-playlist", "--quiet", "-o", "-", pageUrl],
     { stdio: ["ignore", "pipe", "pipe"], windowsHide: true },
   );