npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

tiny-module-compiler

v1.3.1

Published

Compile, archive, unpack, and load compiled modules leveraging v8 cached data.

Downloads

16

Readme

Compile, archive, unpack, and load compiled modules leveraging v8 cached data.

Table of Contents

Status

Stable/Documentation

Actions Status

Installation

$ npm install tiny-module-compiler

Abstract

The tiny-module-compiler module is small toolkit for compiling JavaScript CommonJs modules into standalone binaries that leverage that v8 cache data format exposed by the vm API. Compiled module objects can be archived into a single file based on the TinyBox file format and then unpacked later to the file system.

This toolkit allows for the compilation of an entire project into a single compiled binary object file. Multiple binary object files and various assets (*.node, *.so, etc) can be packaged into an archive and unpacked to file system for later use making it suitable as a delivery mechanism.

Compiled modules objects can be loaded and executed but must be in a runtime that uses the same version of v8.

Basic Usage

const tmc = require('tiny-module-compiler')

const targets = '*.js'
const archiveName = 'modules.a'

// compile module targets into compiled module objects
tmc.compile(targets, (err, objects) => {
  // archive objects into file specified by `archiveName`
  tmc.archive(archiveName, objects, (err) => {
    tmc.load(archiveName, (err, archive) => {
      // `archive` contains a mapping of compiled module object
      // filenames to loaded exports
    })
  })
})

API

The tiny-module-compiler module exports a public API described in this section.

compile(target[, options], callback)

Compiles a file specified at target. The default behavior is to write the output to a file of the same name as target with .out appended to the end. This behaviour can be configured by specifying an options.output option or options.storage as a custom random-access-storage factory function.

The value of options is optional and can be:

{
  // current working directory for compilation
  cwd: process.cwd(),

  // output filename (single file compilation) or directory (multiple files)
  output: target + '.out',

  // custom storage factory function to return
  // 'random-access-storage' compliant object
  storage(filename) {
    return require('random-access-file')(filename)
  }
}

Examples

Simple Compilation

A simple compilation example that compiles a target input file to an output file.

const { compile } = require('tiny-module-compiler')

// compile this file
const target = __filename
const output = __filename + '.out'
compile(target, { output }, (err) => {
  // `target` compiled to `output`
})
Simple Compilation to Memory

A simple compilation example that compiles a target input file to an in memory random-access-memory storage.

const { compile } = require('tiny-module-compiler')
const ram = require('random-access-memory')

// compile this file
const target = __filename
const storage = ram()
compile(target, { storage: () => storage }, (err) => {
  // `target` compiled and written to `storage`
})

archive(target, inputs[, options], callback)

Archives the entries found in a given inputs Map into target. The default behaviour is to enumerate the entries in inputs and write them to a TinyBox instance specified at target where the keys and values of the entries are "put" into the TinyBox storage that lives on the file system. This behaviour can be configured by specifying options.storage as a custom random-access-storage instance.

The value of options is optional and can be:

{
  // if `false`, will not truncate archive storage
  truncate: true,

  // custom 'random-access-storage' compliant object
  storage: require('random-access-file')(target)
}

Examples

Simple Archive

A simple example that archives the compiled objects of a compiled file.

const { compile, archive } = require('tiny-module-compiler')

// compile this file
const target = __filename

compile(target, (err, objects) => {
  archive(__filename + '.a', objects, (err) => {
    // `target` is compiled and then archived
  })
})
Simple Archive in Memory

A simple example that archives the compiled objects of a compiled file into an in memory random-access-memory storage.

const { compile, archive } = require('tiny-module-compiler')
const ram = require('random-access-memory')

// compile this file
const target = __filename
compile(target, { storage: ram }, (err) => {
  const storage = ram()
  archive(__filename + '.a', objects, { storage}, (err) => {
    // `target` is compiled and then archived
    // `storage` contains archived objects
  })
})

load(target[, options], callback)

Loads the exports from a module or the entries in archive specified at target. The default behaviour is to load the contents of the file specified at target as a compiled module and call callback(err, exports) upon success or error with the module exports. If target is a path to an archive represented by a TinyBox, then the entries are loaded and callback(err, archive) is called upon success or error. The storage of the compiled module or archive can be explicitly set by specifying options.storage as a custom random-access-storage instance.

The value of options is optional and can be:

{
  // custom 'random-access-storage' compliant object
  storage: require('random-access-file')(target),

  // custom cache `Map` store for loaded modules
  cache: new Map(),

  // current working directory to load module in
  cwd: process.cwd()
}

Example

Simple Compile and Load

A simple compilation and load example that compiles a target input file to an output file and then loads the exports.

file.js:

module.exports = {
  hello() {
    return 'world'
  }
}

compile-and-load.js:

const { compile, load } = require('tiny-module-compiler')

// compile this file
const target = 'file.js'
const output = target + '.out'
compile(target, { output }, (err) => {
  // `target` compiled to `output`
  load(output, (err, exports) => {
    // `exports` points to `module.exports` in `output`
    console.log(exports.hello()) // 'world'
  })
})
Simple Compile and Load in Memory

A simple compilation and load example that compiles a target input file to an in memory storage and then loads the exports from it.

file.js:

module.exports = {
  hello() {
    return 'world'
  }
}

compile-and-load-in-memory.js:

const { compile, load } = require('tiny-module-compiler')
const ram = require('random-access-memory')

// compile this file
const target = 'file.js'
const storage = ram()
compile(target, { storage: () => storage }, (err) => {
  // `target` compiled to `storage`
  load('loaded from memory', { storage }, (err, exports) => {
    // `exports` points to `module.exports` in `output`
    console.log(exports.hello()) // 'world'
  })
})

unpack(target[, options], callback)

Unpacks target archive entries. The default behaviour is to enumerate the entries in the archive specified at target and copy them to the file system. This behaviour can be configured by specifying a options.storage random-access-storage factory function to provide custom storage for the archive entries. If target is a random-access-storage instance, it will be used instead of reading from the file system.

The value of options is optional and can be:

{
  // output path for default storage
  output: process.cwd(),

  // custom storage factory function to return
  // 'random-access-storage' compliant object
  storage(filename) {
    return require('random-access-file')(filename)
  }
}

Example

Simple Unpack

A simple example to unpack an archive's entries to file system.

const { unpack } = require('tiny-module-compiler')

const archive = '/path/to/archive'
unpack(archive, (err, entries) => {
  // `archive` entries unpacked to file system
  console.log(entries) // array of unpacked file names
})
Simple Unpack in Memory

A simple example to unpack an archive's entries to an in memory file store.

const { unpack } = require('tiny-module-compiler')
const ram = require('random-access-memory'

const archive = '/path/to/archive'
const files = new Map()

unpack(archive, { storage: createStorage }, (err, entries) => {
  // `archive` entries unpacked to file system
  console.log(entries) // array of unpacked file names
  for (const entry of entries) {
    console.log(files[entry.filename])
  }
})

function createStorage(filename) {
  return files.set(filename, ram()).get(filename)
}

Class: Compiler

The Compiler class represents a container of compile targets that can be compiled into a single binary file containing v8 cache data and header information about the compiled output.

Constructor: new Compiler([options])

Creates a new Compiler instance where options can be:

{
  // current working directory for compilation
  cwd: process.cwd()
}
Example
const compiler = new Compiler()

Accessor: compiler.targets

  • Array<String>

All opened targets in the compiler.

Example
for (const target of compiler.targets) {
  // handle opened `target`
}

compiler.ready(callback)

Waits for compiler to be ready and calls callback() upon success.

Example
compiler.ready(() => {
  // `compiler` is opened and ready
})

compiler.target(filename[, options], callback)

Creates and returns a new compile target that is added to compiler pool calling callback(err, target) when the target resource is opened or an error occurs. The target will be compiled when compiler.compiler() is called.

The value of options can be:

{
  // the default output for a compilation target
  output: target + '.out',

  // custom storage factory function to return
  // 'random-access-storage' compliant object
  storage(filename) {
    return require('random-access-file')(filename)
  }
}
Example
compiler.target('/path/to/file.js', (err, target) => {
  // `target` is an opened `nanoresource`
})

compiler.compile([options], callback)

Compiles all pending compile targets calling callback(err, objects, assets) upon success or error. Callback will be given a Map of compiled objects and a Map of extracted assets that should live with the compiled objects on the file system.

The value of options can be:

{
  // if `true`, will produce a source map stored in the `assets` map
  map: false,

  // if `false`, `ncc` will produce verbose output
  quiet: true,

  // if `true`, will use `ncc` cache for faster builds
  cache: false,

  // if `true`, will produce compiled javascript source debug output
  debug: false,

  // if `true`, will minify compiled javascript source before creating binary
  optimize: false,

  // an array of external modules that should _not_ be compiled
  externals: [],
}
Example
compiler.compile({ externals: ['sodium-native'] }, (err, objects, assets) => {
  // `objects` is a `Map` mapping file names to compiled module objects
  // `assets` is a `Map` mapping file names to assets that should be copied
})

Class: Archiver

The Archiver class represents an abstraction for storing compiled objects into a TinyBox

Constructor: new Archiver([options])

Creates a new Archiver instance where options can be:

{
  // default custom storage factory function to return
  // 'random-access-storage' compliant object used in
  // `archiver.archive()` calls. This storage can be overloaded
  // by supplying a storage factory function to `archiver.archive()`
  storage(filename) {
    return require('random-access-file')(filename)
  }
}
Example
const ram = require('random-access-memory')

// in memory archiver
const archiver = new Archiver({ storage: ram })

archiver.ready(callback)

Waits for archiver to be ready and calling callback() when it is.

Example
archiver.ready(() => {
  // `archiver` is opened and ready
})

archiver.archive(filename, inputs[, options], callback)]

Archives the entries found in a given inputs Map into target. The underlying storage for filename can be given by options.storage or a new one is created by the storage factory function given to the Archiver class constructor. callback(err) is called upon success or error.

The value of options can be:

{
  // if `false`, will not truncate archive storage
  truncate: true,

  // custom 'random-access-storage' compliant object
  // where `inputs` are archived to
  storage: require('random-access-file')(filename)
}
Example
const extend = require('map-extend')

compiler.compile((err, objects, assets) => {
  // merge `objects` and `assets` and archive
  archiver.archive('/path/to/archive', extend(objects, assets), (err) => {
    // inputs should be archived
  })

Class: Loader

The Loader class represents an abstraction for loading compiled module objects and JavaScript sources as node modules.

Constructor: new Loader([options])

Creates a new Loader instance where options can be:

{
  // custom cache `Map` store for loaded modules
  cache: new Map()
}

loader.ready(callback)

Waits for loader to be ready and calling callback() when it is.

Example
loader.ready(() => {
  // `loader` is opened and ready
})

loader.load(filename[, options], callback)

Loads a compiled module object or JavaScript source module specified at filename calling callback(err, exports) upon success or error. Success loads will cache resulting module for subsequent requests to load the module.

The value of options can be:

{
  // custom 'random-access-storage' compliant object
  // where module is loaded from
  storage: require('random-access-file')(filename)
}
Example
loader.load('/path/to/compiled/module.js', (err, exports) => {
  // `exports` points to `module.exports` of loaded module
})

Class: Target

The Target class represents a nanoresource to a target file backed by a random-access-storage instance.

Constructor: new Target(filename[, options])

Creates a new Target instance where filename is the name of the target file and options can be:

{
  // custom 'random-access-storage' compliant object
  // where file is loaded from
  storage: require('random-access-file')(filename)
}

Note: This class is intended for internal and advanced use. You will most likely not use this directly.

Example
const target = new Target('/path/to/file.js')

Accessor: target.fd

  • ?(Number)

The active file descriptor for the target resource. Will be null if not opened.

target.stat(callback)

Queries for stats from the underlying target storage calling callback(err, stats) upon success or error.

Example
target.stat((err, stats) => {
  console.log(stats.size)
})

target.read(offset, size, callback)

Reads data from the underlying target storage at a specified offset and size calling callback(err, buffer) upon success or error.

Example
taret.read(32, 64, (err, buffer) => {
  console.log(buffer)
})

target.ready(callback)

Waits for loader to be ready and calling callback() when it is.

Example
target.ready(() => {
  // `target` is opened and ready
})

Command Line Interface

The tiny-module-compiler exposes a command line interface through the tmc command that is suitable for compiling, archiving, and unpacking compiled modules and their assets.

This section describes the command line interface and a few workflows for making the best use of the tmc command.

The tmc(1) Command

The tmc command has a command line signature of the following:

tmc [-hV] [-acu] [-vCDMO] [options] ...input

Where options can be:

  -a, --archive             If present, will archive input into "tinybox" format
  -c, --compile             If present, will compile input into header prefixed v8 cached data
    , --concurrency <jobs>  An alias for '--jobs'
  -C, --copy-assets         If present, will copy assets to directory of output
  -D, --debug               If present, will enable debug output (DEBUG=tiny-module-compiler)
  -e, --external <module>   Specifies an external dependency that will be linked at runtime
  -h, --help                If present, will print this message
  -l, --load                If present, will load inputs
  -j, --jobs <jobs>         Specifies the number of concurrent jobs for batch tasks (--load, --archive, --copy-assets)
  -M, --source-map          If present, a source map will be generated
  -o, --output <path>       If present, will change the output path. Assumes directory if multiple inputs given
  -O, --optimize            If present, will optimize output by minifying JavaScript source prior to compilation
  -u, --unpack              If present, will treat input as an archive and will unpack files to path specified by '--output'
  -v, --verbose             If present, will emit verbose output to stdout/stderr
  -V, --version             If present, will print the version number
  -x                        An alias for '--external'

Compiling Modules

The simplest use of the tmc command is to compile a single file. By default, the -c (or --compile) is assumed if -a (or --archive) and -u (or --unpack) flags are not present.

$ tmc file.js ## will write ./file.js.out

The output of the compiled file can be configured using the -o (or --output) flag to set the output name.

$ tmc file.js -o file.compiled.js

If multiple inputs were given, then the output value will be a directory and the original file names are preserved.

$ tmc *.js -o build/

Archiving Compiled Modules

After compiling modules it may be useful to archive them into a single file. This is often beneficial if the compiled module contained static assets that need to live on the file system alongside the compiled module file during runtime.

The tmc command makes it easy to specify a number of files that should be added to an archive by making use of the -a (or --archive) flag to indicate that inputs should be archived.

$ tmc -a modules.archive *.compiled.js ## archives all `*.compiled.js` files into `modules.archive`

Any asset can be added to an archive as well.

$ tmc -a modules.archive *.node ## archives all `*.node` files into `modules.archive`

The modules.archive file contains an index of entries added to it by file name. The archive can be unpacked using the tmc command as well.

Unpacking Archives

Compiled modules and their assets living in an archive can be easily unpacked to the file system. The tmc command makes it easy to unpack entries in an archive.

$ tmc -u modules.archive

The output of the entries can be specified with the -o (or --output) flag.

$ tmc -u modules.archive -o build/

Loading Modules & Archives

Compiled modules and packed archives can be loaded at runtime in an application. The tmc command provides a simple way to do this from the command line from array of inputs by making use of the -l (or --load) flag to indicate the inputs should be loaded.

$ tmc -l module.compiled.js

Archives can be loaded as well.

$ tmc -l modules.archive

Regular JavaScript files an be given as input.

$ tmc -l file.js

Multiple inputs can be loaded.

$ tmc -l modules.archive file.js module.compiled

The load concurrency can be set by specifying the -j (or --jobs or --concurrency) flag to indicate the number of concurrent loads that can occur when loading the inputs.

$ tmc -l -j1 modules.archive file.js module.compiled ## one at a time

Constraints

The tiny-module-compile module ships with a few constrains due to the nature of how v8 cache data works. They are described in this section.

v8 Version

The v8 version (process.versions.v8) in the runtime of node used to compile modules to v8 cache data must match the one used to load. The module will throw an error if the versions do not match as unexpected behaviour may occur.

Function.prototype.toString()

For user-defined functions, Function.prototype.toString() may not work as expected or not at all because the string source code representation of the function is lost after compilation.

See Also

Prior Art

  • https://github.com/OsamaAbbas/bytenode
  • https://github.com/zertosh/v8-compile-cache
  • https://github.com/v8/v8/blob/master/src/snapshot/code-serializer.h
  • https://hackernoon.com/how-to-compile-node-js-code-using-bytenode-11dcba856fa9

License

MIT