ytdlp-ejs

YouTube player signature solver implemented in Rust, using SWC for JavaScript parsing and multiple runtime options for execution.

A Rust port of yt-dlp/ejs.

📖 Improving yt-dlp-ejs with Rust: Smaller and Faster

Features

• Parse YouTube player JavaScript code
• Extract and execute signature (sig) decryption functions
• Extract and execute throttle parameter (n) decryption functions
• Multiple JavaScript runtime support: QuickJS, Boa, Node, Deno, Bun
• Cross-platform support (Windows, Linux, macOS)
• Standalone binary under 5MB (with SWC + QuickJS)

Installation

From Source

# Default build (QuickJS + Boa + External runtimes)
cargo build --release

# QuickJS only (smallest binary)
cargo build --release --no-default-features --features qjs

# Boa only
cargo build --release --no-default-features --features boa

Binary output: target/release/ejs

As a Library

[dependencies]
ejs = { git = "https://github.com/ahaoboy/ytdlp-ejs", features = ["qjs"] }

Usage

Command Line

# Basic usage
ejs <player_file> [n:<challenge>] [sig:<challenge>]

# Specify runtime
ejs --runtime qjs player.js n:ZdZIqFPQK-Ty8wId
ejs --runtime boa player.js sig:gN7a-hudCuAuPH6f...
ejs --runtime node player.js n:ZdZIqFPQK-Ty8wId sig:gN7a-hudCuAuPH6f...
ejs --runtime deno player.js n:ZdZIqFPQK-Ty8wId
ejs --runtime bun player.js n:ZdZIqFPQK-Ty8wId

Output (JSON):

{
  "type": "result",
  "responses": [
    { "type": "result", "data": { "ZdZIqFPQK-Ty8wId": "qmtUsIz04xxiNW" } }
  ]
}

As a Library

use ejs::{
    process_input_with_runtime, JsChallengeInput, JsChallengeRequest,
    JsChallengeType, RuntimeType,
};

let input = JsChallengeInput::Player {
    player: player_code.to_string(),
    requests: vec![
        JsChallengeRequest {
            challenge_type: JsChallengeType::N,
            challenges: vec!["ZdZIqFPQK-Ty8wId".to_string()],
        },
    ],
    output_preprocessed: false,
};

let output = process_input_with_runtime(input, RuntimeType::QuickJS);

Runtime Options

Benchmark (Ubuntu)

Latest results: bench.yml