Skip to content

Latest commit

Β 

History

History
432 lines (330 loc) Β· 14.6 KB

single-executable-applications.md

File metadata and controls

432 lines (330 loc) Β· 14.6 KB

Single executable applications

Stability: 1.1 - Active development

This feature allows the distribution of a Node.js application conveniently to a system that does not have Node.js installed.

Node.js supports the creation of single executable applications by allowing the injection of a blob prepared by Node.js, which can contain a bundled script, into the node binary. During start up, the program checks if anything has been injected. If the blob is found, it executes the script in the blob. Otherwise Node.js operates as it normally does.

The single executable application feature currently only supports running a single embedded script using the CommonJS module system.

Users can create a single executable application from their bundled script with the node binary itself and any tool which can inject resources into the binary.

Here are the steps for creating a single executable application using one such tool, postject:

  1. Create a JavaScript file:

    echo 'console.log(`Hello, ${process.argv[2]}!`);' > hello.js
  2. Create a configuration file building a blob that can be injected into the single executable application (see Generating single executable preparation blobs for details):

    echo '{ "main": "hello.js", "output": "sea-prep.blob" }' > sea-config.json
  3. Generate the blob to be injected:

    node --experimental-sea-config sea-config.json
  4. Create a copy of the node executable and name it according to your needs:

    • On systems other than Windows:
    cp $(command -v node) hello
    • On Windows:
    node -e "require('fs').copyFileSync(process.execPath, 'hello.exe')"
    

    The .exe extension is necessary.

  5. Remove the signature of the binary (macOS and Windows only):

    • On macOS:
    codesign --remove-signature hello
    • On Windows (optional):

    signtool can be used from the installed Windows SDK. If this step is skipped, ignore any signature-related warning from postject.

    signtool remove /s hello.exe
  6. Inject the blob into the copied binary by running postject with the following options:

    • hello / hello.exe - The name of the copy of the node executable created in step 4.
    • NODE_SEA_BLOB - The name of the resource / note / section in the binary where the contents of the blob will be stored.
    • sea-prep.blob - The name of the blob created in step 1.
    • --sentinel-fuse NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2 - The fuse used by the Node.js project to detect if a file has been injected.
    • --macho-segment-name NODE_SEA (only needed on macOS) - The name of the segment in the binary where the contents of the blob will be stored.

    To summarize, here is the required command for each platform:

    • On Linux:

      npx postject hello NODE_SEA_BLOB sea-prep.blob \
          --sentinel-fuse NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2
    • On Windows - PowerShell:

      npx postject hello.exe NODE_SEA_BLOB sea-prep.blob `
          --sentinel-fuse NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2
    • On Windows - Command Prompt:

      npx postject hello.exe NODE_SEA_BLOB sea-prep.blob ^
          --sentinel-fuse NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2
      
    • On macOS:

      npx postject hello NODE_SEA_BLOB sea-prep.blob \
          --sentinel-fuse NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2 \
          --macho-segment-name NODE_SEA
  7. Sign the binary (macOS and Windows only):

    • On macOS:
    codesign --sign - hello
    • On Windows (optional):

    A certificate needs to be present for this to work. However, the unsigned binary would still be runnable.

    signtool sign /fd SHA256 hello.exe
  8. Run the binary:

    • On systems other than Windows
    $ ./hello world
    Hello, world!
    • On Windows
    $ .\hello.exe world
    Hello, world!

Generating single executable preparation blobs

Single executable preparation blobs that are injected into the application can be generated using the --experimental-sea-config flag of the Node.js binary that will be used to build the single executable. It takes a path to a configuration file in JSON format. If the path passed to it isn't absolute, Node.js will use the path relative to the current working directory.

The configuration currently reads the following top-level fields:

{
  "main": "/path/to/bundled/script.js",
  "output": "/path/to/write/the/generated/blob.blob",
  "disableExperimentalSEAWarning": true, // Default: false
  "useSnapshot": false,  // Default: false
  "useCodeCache": true, // Default: false
  "assets": {  // Optional
    "a.dat": "/path/to/a.dat",
    "b.txt": "/path/to/b.txt"
  }
}

If the paths are not absolute, Node.js will use the path relative to the current working directory. The version of the Node.js binary used to produce the blob must be the same as the one to which the blob will be injected.

Assets

Users can include assets by adding a key-path dictionary to the configuration as the assets field. At build time, Node.js would read the assets from the specified paths and bundle them into the preparation blob. In the generated executable, users can retrieve the assets using the sea.getAsset() and sea.getAssetAsBlob() APIs.

{
  "main": "/path/to/bundled/script.js",
  "output": "/path/to/write/the/generated/blob.blob",
  "assets": {
    "a.jpg": "/path/to/a.jpg",
    "b.txt": "/path/to/b.txt"
  }
}

The single-executable application can access the assets as follows:

const { getAsset } = require('node:sea');
// Returns a copy of the data in an ArrayBuffer.
const image = getAsset('a.jpg');
// Returns a string decoded from the asset as UTF8.
const text = getAsset('b.txt', 'utf8');
// Returns a Blob containing the asset.
const blob = getAssetAsBlob('a.jpg');
// Returns an ArrayBuffer containing the raw asset without copying.
const raw = getRawAsset('a.jpg');

See documentation of the sea.getAsset() and sea.getAssetAsBlob() APIs for more information.

Startup snapshot support

The useSnapshot field can be used to enable startup snapshot support. In this case the main script would not be when the final executable is launched. Instead, it would be run when the single executable application preparation blob is generated on the building machine. The generated preparation blob would then include a snapshot capturing the states initialized by the main script. The final executable with the preparation blob injected would deserialize the snapshot at run time.

When useSnapshot is true, the main script must invoke the v8.startupSnapshot.setDeserializeMainFunction() API to configure code that needs to be run when the final executable is launched by the users.

The typical pattern for an application to use snapshot in a single executable application is:

  1. At build time, on the building machine, the main script is run to initialize the heap to a state that's ready to take user input. The script should also configure a main function with v8.startupSnapshot.setDeserializeMainFunction(). This function will be compiled and serialized into the snapshot, but not invoked at build time.
  2. At run time, the main function will be run on top of the deserialized heap on the user machine to process user input and generate output.

The general constraints of the startup snapshot scripts also apply to the main script when it's used to build snapshot for the single executable application, and the main script can use the v8.startupSnapshot API to adapt to these constraints. See documentation about startup snapshot support in Node.js.

V8 code cache support

When useCodeCache is set to true in the configuration, during the generation of the single executable preparation blob, Node.js will compile the main script to generate the V8 code cache. The generated code cache would be part of the preparation blob and get injected into the final executable. When the single executable application is launched, instead of compiling the main script from scratch, Node.js would use the code cache to speed up the compilation, then execute the script, which would improve the startup performance.

Note: import() does not work when useCodeCache is true.

In the injected main script

Single-executable application API

The node:sea builtin allows interaction with the single-executable application from the JavaScript main script embedded into the executable.

sea.isSea()

  • Returns: {boolean} Whether this script is running inside a single-executable application.

sea.getAsset(key[, encoding])

This method can be used to retrieve the assets configured to be bundled into the single-executable application at build time. An error is thrown when no matching asset can be found.

  • key {string} the key for the asset in the dictionary specified by the assets field in the single-executable application configuration.
  • encoding {string} If specified, the asset will be decoded as a string. Any encoding supported by the TextDecoder is accepted. If unspecified, an ArrayBuffer containing a copy of the asset would be returned instead.
  • Returns: {string|ArrayBuffer}

sea.getAssetAsBlob(key[, options])

Similar to sea.getAsset(), but returns the result in a Blob. An error is thrown when no matching asset can be found.

  • key {string} the key for the asset in the dictionary specified by the assets field in the single-executable application configuration.
  • options {Object}
    • type {string} An optional mime type for the blob.
  • Returns: {Blob}

sea.getRawAsset(key)

This method can be used to retrieve the assets configured to be bundled into the single-executable application at build time. An error is thrown when no matching asset can be found.

Unlike sea.getRawAsset() or sea.getAssetAsBlob(), this method does not return a copy. Instead, it returns the raw asset bundled inside the executable.

For now, users should avoid writing to the returned array buffer. If the injected section is not marked as writable or not aligned properly, writes to the returned array buffer is likely to result in a crash.

  • key {string} the key for the asset in the dictionary specified by the assets field in the single-executable application configuration.
  • Returns: {string|ArrayBuffer}

require(id) in the injected main script is not file based

require() in the injected main script is not the same as the require() available to modules that are not injected. It also does not have any of the properties that non-injected require() has except require.main. It can only be used to load built-in modules. Attempting to load a module that can only be found in the file system will throw an error.

Instead of relying on a file based require(), users can bundle their application into a standalone JavaScript file to inject into the executable. This also ensures a more deterministic dependency graph.

However, if a file based require() is still needed, that can also be achieved:

const { createRequire } = require('node:module');
require = createRequire(__filename);

__filename and module.filename in the injected main script

The values of __filename and module.filename in the injected main script are equal to process.execPath.

__dirname in the injected main script

The value of __dirname in the injected main script is equal to the directory name of process.execPath.

Notes

Single executable application creation process

A tool aiming to create a single executable Node.js application must inject the contents of the blob prepared with --experimental-sea-config" into:

  • a resource named NODE_SEA_BLOB if the node binary is a PE file
  • a section named NODE_SEA_BLOB in the NODE_SEA segment if the node binary is a Mach-O file
  • a note named NODE_SEA_BLOB if the node binary is an ELF file

Search the binary for the NODE_SEA_FUSE_fce680ab2cc467b6e072b8b5df1996b2:0 fuse string and flip the last character to 1 to indicate that a resource has been injected.

Platform support

Single-executable support is tested regularly on CI only on the following platforms:

This is due to a lack of better tools to generate single-executables that can be used to test this feature on other platforms.

Suggestions for other resource injection tools/workflows are welcomed. Please start a discussion at https://github.com/nodejs/single-executable/discussions to help us document them.