Skip to content

YuJianrong/node-unrar.js

Repository files navigation

node-unrar-js

Build Status npm version MIT License TypeScript

node-unrar.js is a npm module to extract rar archive in pure JavaScript. It's combined by a Javascript adoption layer and JavaScript unRar library compiled by Emscripten from the C++ unrar library which hosted on http://www.rarlab.com/rar_add.htm .

Installation

You can install the module via npm:

npm install node-unrar-js

Features

  • Fully support for RAR archive, because it comes from the official source code.
  • Unicode support, for both comment and file path/name.
  • API for Extraction to both memory and file system.
  • Both Commonjs module (for nodejs) and ES2015 module (for webpack) are supported.

Unsupported Features

  • Volume archives are not supported.
  • Synchronize File System IO functions are used in File Extraction.

API to create the extractor

  • async function createExtractorFromData(options: ExtractorFromDataOptions): Promise<Extractor<Uint8Array>> - Create the in Memory Extractor

    Options ExtractorFromDataOptions:

    • data: ArrayBuffer : ArrayBuffer object of the RAR archive file
    • password?: string : Optional password
    • wasmBinary? ArrayBuffer; : Optional Use in browser/webpack, the wasm binary must be loaded in the code and send to this function to load the wasm code
  • async function createExtractorFromFile(options: ExtractorFromFileOptions): Promise<Extractor> - Get the File Extractor

    Options ExtractorFromFileOptions:

    • filepath: string : File path of the RAR archive file
    • targetPath?: string : Optional target folder
    • password?: string : Optional password
    • filenameTransform?: (filename: string) => string: Optional transform the file name before it's created on file system
    • wasmBinary? ArrayBuffer; : Optional Use in nodejs/webpack, the wasm binary must be loaded in the code and send to this function to load the wasm code in webpack based nodejs project (please read Used in Webpack-bundled NodeJS Project for more details).

    Node: This function is not available in EM2015 Module since the EM2015 Module is used for webpack in Browser.

API of the extractor

  • getFileList(): ArcList : Get the file header and file list of the archive.

    Members in ArcList:

    • arcHeader: ArcHeader : The header of the archive
    • fileHeaders: Generator<FileHeader> : The iterator of the FileHeader objects
{
  arcHeader: {
    comment: "",
    flags: {
      authInfo: false,
      headerEncrypted: false,
      lock: false,
      recoveryRecord: false,
      solid: false,
      volume: false,
    },
  }, fileHeaders: (Iterator)
    {
      crc: 0,
      flags: {
        directory: false,
        encrypted: false,
        solid: false,
      },
      method: "Storing",
      name: "FileName",
      packSize: 0,
      time: "2017-04-03T10:41:42.000",
      unpSize: 0,
      unpVer: "2.9",
      comment: "",
    },
}
  • extract(options: ExtractOptions): Extract the files.
    • Options ExtractOptions:
      • files?: string[] | ((fileHeader: FileHeader) => boolean) Optional Extract all the files if files is empty
        • string[]: Extract the specific files only
        • (fileHeader: FileHeader) => boolean: Extract only the filtered file (eg. extract only the files without password).
      • password?: string: Optional password for the extracted files only (Different password can be applied to any single file in RAR archive).
    • The return values are different for createExtractorFromData and createExtractorFromFile:
      • ArcFiles<Uint8Array> for createExtractorFromData
      • ArcFiles for createExtractorFromFile
    • Members in ArcFiles:
      • arcHeader: ArcHeader : The header of the archive
      • files: Generator<ArcFile> : The iterator of the ArcFile objects
      • Members in ArcFile:
        • fileHeader: FileHeader : The header of the extracted file
        • extraction?: Uint8Array : The extracted content of the file (createExtractorFromData only). If the ArcFile is a folder (ArcFile.fileHeader.flags.directory is true), the extraction will be undefined, otherwise it will be the content of the file (in `Uint8Array).
{
  arcHeader: {...} // Same as the arcHeader above
  files: (Iterator)
    {
      fileHeader: {...} // Same as the fileHeader above
      extraction?: Uint8Array // createExtractorFromData only
  ]
}

Note: The returned iterators from the two apis are lazy, it means the file header/content will not be parsed/extracted if any file is not iterated yet.

Exception

The customized Error Object UnrarError will be thrown for any exception in extracting. The definition of the Object is:

class UnrarError extends Error {
    reason: FailReason;
    file?: string | undefined; // Will be filled for any exception in extraction of a file
}

The following code is used in the FailReason:

FailReason Message
ERAR_NO_MEMORY Not enough memory
ERAR_BAD_DATA Archive header or data are damaged
ERAR_BAD_ARCHIVE File is not RAR archive
ERAR_UNKNOWN_FORMAT Unknown archive format
ERAR_EOPEN File open error
ERAR_ECREATE File create error
ERAR_ECLOSE File close error
ERAR_EREAD File read error
ERAR_EWRITE File write error
ERAR_SMALL_BUF Buffer for archive comment is too small, comment truncated
ERAR_UNKNOWN Unknown error
ERAR_MISSING_PASSWORD Password for encrypted file or header is not specified
ERAR_EREFERENCE Cannot open file source for reference record
ERAR_BAD_PASSWORD Wrong password is specified

Memory Leak

Note: although the return value fileHeaders and files are iterators, they must be traversed to the end! Otherwise the C++ object for archive extraction will not be destructed and cause memory leak.

Example

const fs = require("fs");
const unrar = require("node-unrar-js");

async function main() {
  // Read the archive file into a typedArray
  const buf = Uint8Array.from(fs.readFileSync("a.rar")).buffer;
  const extractor = await unrar.createExtractorFromData({ data: buf });

  const list = extractor.getFileList();
  const listArcHeader = list.arcHeader; // archive header
  const fileHeaders = [...list.fileHeaders]; // load the file headers

  const extracted = extractor.extract({ files: ["1.txt"] });
  // extracted.arcHeader  : archive header
  const files = [...extracted.files]; //load the files
  files[0].fileHeader; // file header
  files[0].extraction; // Uint8Array content, createExtractorFromData only
}
main();

Demo in Webpack

This package can also be used in browser by Webpack, please visit the demo project to see how to use it in webpack.

Use in Webpack-bundled NodeJS Project

In most cases the exported ESModule is used in browser by Webpack, but in case if the NodeJs project (or an Electron project) is bundled by Webpack, the wasmBinary data must be loaded manually just like the browser by Webpack, it can be loaded like this:

import fs from 'fs';
import { createExtractorFromFile } from 'node-unrar-js/esm';

const wasmBinary = fs.readFileSync(require.resolve('node-unrar-js/esm/js/unrar.wasm'));

const extractor = await createExtractorFromFile({ wasmBinary, filepath: './WithComment.rar' });

// const list = extractor.getFileList();

Note: the package must be loaded from 'node-unrar-js/esm' instead of 'node-unrar-js' to enable the function createExtractorFromFile in ES Module.

TypeScript

This module is written in TypeScript, you can import it directly in TypeScript and get the benefit of static type checking and auto-complete of IDE.

Development and Contribution

If you want to compile the module by yourself, please follow the steps below:

  • Install the docker

  • Download the c++ source of unRar library by:

npm run downloadUnrarSrc

  • Build for debug:

npm run build:debug

  • Build for release

npm run build:release

  • Run Test

npm run test

License

This module is licensed under MIT.