complexion-js

Complexion - JavaScript

Extends the Complexion tokenizing library so it is able to parse JavaScript into tokens. Useful if you want to create a JavaScript pretty printer, parser or anything that would need to understand the syntax of JavaScript.

Usage

Install this package and Complexion with npm or bower or maybe just clone the two repositories. Then, use the library in your JavaScript. There is UMD markup (via FidUmd) that should make this library available under any module system that you prefer to use.

// NPM
var Complexion = require('complexion');
var instance = new Complexion();
require('complexion-js')(instance);
var tokenList = instance.tokenize('... JavaScript goes here ...');

// Browser
var instance = new Complexion();
complexionJs(instance);
var tokenList = instance.tokenize('... JavaScript goes here ...');

Options may be passed as well.

complexionJs(instance, {
    bom: false,
    shebang: false
});

Options

• bom (boolean, enabled by default) - Allow searching for the byte order mark in files.
• shebang (boolean, enabled by default) - Allow searching for a shebang at the top of a file.

Token Types

The following are all of the available tokens that the tokenizer produces.

• BOM - UTF8 byte order mark at the very beginning of the file.
• BOOLEAN_LITERAL - Either true or false.
• IDENTIFIER_NAME - The names of functions, variables, properties. This does not include keywords.
• IMPLICIT_SEMICOLON - A special token that has no content but indicates that there should have been a semicolon here. It's necessary for determining context and if something is division or a regular expression.
• KEYWORD - Special words that the language reserves and are not allowed as variable names. Does not include null nor booleans.
• LINE_TERMINATOR - Newlines (\r\n, \n and \r).
• MULTI_LINE_COMMENT - A comment that uses /* */.
• NULL_LITERAL - Only matches null.
• NUMERIC_LITERAL - A number in decimal, octal or hex. Decimals may have exponents.
• PUNCTUATOR - Most symbols including math, array literals, object literals, function calls.
• REGULAR_EXPRESSION_LITERAL - Regular expressions can be tricky because they sometimes look like division and have lots of allowed escape sequences.
• SHEBANG - Only matches the first line when it starts with #! and can be turned off.
• SINGLE_LINE_COMMENT - A comment that starts with //.
• STRING_LITERAL - A single or double quoted string.
• UNKNOWN - An unknown character. If you hit this, you have poorly formed JavaScript. This is not a full lint test, just the beginning of one.
• WHITESPACE - Spaces and tabs. Not newlines.

Token Objects

The tokens returned from tokenization will all be instances of a ComplexionJsToken class. They retain the same properties that the original token object has and get additional methods.

ComplexionJsToken.prototype.isAnyType(tokenTypes)

Returns true if the token is any of the types passed in.

if (token.isAnyType([ 'WHITESPACE', 'LINE_TERMINATOR' ])) {
    // This token does not appear on screens
}

ComplexionJsToken.prototype.isAnyType(tokenTypes)

Returns true if the token is exactly the type passed in.

if (token.isType('BOOLEAN_LITERAL')) {
    // token.content === "true" || token.content === "false"
}

ComplexionJsToken.prototype.isAnyType(tokenTypes)

Returns true when the token could be eliminated entirely from the output.

// Simple minifier
tokenList.forEach(function (token) {
    if (!token.isUnimportant()) {
        if (token.isType('IMPLICIT_SEMICOLON')) {
            console.log(';');
        } else {
            console.log(token.content);
        }
    }
});

Development

If you want to work on this library, you need to check out the repository and run npm install to get the dependencies.

Tests are always included. Make sure tests cover your changes. To run the current tests, just use npm test or grunt test (they will run the same test suite).

Make sure the matchers are fast. This library is tweaked to work on unminified source, so take that into consideration before shuffling the matchers too much.

License

This software is licensed under an MIT license with an additional non-advertising clause.