Skip to content

cheng81/node2browser

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Node2Browser

Just a simple tool to port node modules to the browser.

What it does

You code your project using the normal Node module structure, then feed the tool with the main file (typically <project path>/lib/index.js or similar). Node2browser figures out the dependency tree, by parseing calls to the require function, and then concatenates the modules in a single fat javascript file, in a way that all the dependencies of a module have been already loaded when the module is required. Modules are loaded using self-invoking anonimous functions.

Features

node2browser simply figures out the dependency tree of your modules, then squash them into a single file. The modules will then be available in the browser using the window.node2browser.require function, which works pretty the same way as Node's require.

Default modules:

  • process: has only one method, nextTick (simulated using window.setTimeout), and an empty array argv
  • util:
    • inspect, just call console.log
    • inherits, stripped from nodejs

Install and Usage

git clone git://github.com/cheng81/node2browser.git
cd node2browser
bin/node2browser <initfile> <outfile> <nodemodulesdir> <requiredAs>

initfile is tipically the index.js in your project. From there, the tool will figure out which file are required where.

outfile is where the "compiled" javascript will be. Default is dist/out.js.

nodemodulesdir is where the node modules that you want to include will be.

requiredAs is how you will require the compiled main module in the browser. For example, if you compile your ~/Projects/myLib/lib/index.js, you might want to set this parameter as myLib. Default main.

Sorry, for now it's very stupid and only single-file modules can be used -it's sufficient for me, if you need more drop a request to frza-AT-itu-DOT-dk-.

A simple example is included the in test directory, which also uses the wu library. Try it out with

bin/node2browser test/test1.js test/out.js test/node myfoo

After the execution the test/out.js file is produced, which includes test1.js, test2.js and node/wu.js modules. To require the initial file test1.js, you can either require ./test1 or myfoo. The file test/index.html uses the produced file, plus shows also how to use node2browser in the browser.

In the Browser

To use the loaded modules in the browser, just use the window.node2browser.require function like you would use the Nodejs one, e.g.:

var require = window.node2browser.require

var mymod = require('./mymod')

By default ./mymod will be resolved against the directory of the initfile above.

Limitations

The grammar that extract the require calls is dumb. Specifically, it will:

  • consider valid a commented call
  • not recognize a call if you alias the require function (e.g. var r = require; r('bla') will not work)

Circular dependencies are not handled. The tool will stop with an exception when a circular dependency is found (and I am not sure if even node itself handle circular dependencies).

The parser has been built using the incredibly cool PEGjs peg parser.

About

simple tool to port node code to the browser

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published