This is plugin for Acorn - a tiny, fast JavaScript parser, written completely in JavaScript.
It was created as an experimental alternative, faster React.js JSX parser.
According to benchmarks, Acorn-JSX is 2x faster than official Esprima-based parser when location tracking is turned on in both (call it “source maps enabled mode”). At the same time, it consumes all the ES6+JSX syntax that can be consumed by Esprima-FB (this is proved by official tests).
UPDATE [14-Apr-2015]: Facebook implementation started deprecation process in favor of Acorn + Acorn-JSX + Babel for parsing and transpiling JSX syntax.
Please note that this tool only parses source code to JSX AST, which is useful for various language tools and services. If you want to transpile your code to regular ES5-compliant JavaScript with source map, check out the babel transpiler which uses acorn-jsx
under the hood.
Requiring this module provides you with an Acorn plugin that you can use like this:
var acorn = require("acorn");
var jsx = require("acorn-jsx");
acorn.Parser.extend(jsx()).parse("my(<jsx/>, 'code');");
Note that official spec doesn’t support mix of XML namespaces and object-style access in tag names (#27) like in <namespace:Object.Property />
, so it was deprecated in acorn-jsx@3.0
. If you still want to opt-in to support of such constructions, you can pass the following option:
acorn.Parser.extend(jsx({ allowNamespacedObjects: true }))
Also, since most apps use pure React transformer, a new option was introduced that allows to prohibit namespaces completely:
acorn.Parser.extend(jsx({ allowNamespaces: false }))
Note that by default allowNamespaces
is enabled for spec compliancy.
This plugin is issued under the MIT license.