mini-sass v1.0.1

mini-sass

An attempt at building a scss-like compiler in Javascript. SCSS/SASS is a superset of CSS that provides additional functionality like variables and functions.

Install

For CLI:

npm install mini-sass -g

For programmatic use:

npm install mini-sass

Usage

CLI:

mini-sass [options] <files...>

If no input files are specified, mini-sass will read from STDIN. If no flags are specified, mini-sass will print to STDOUT.

Options

Options:
  -o, --output <file>  Output file (default: STDOUT)
  -s, --separate       Generate separate output file for each input

When the -o flag is present, result of compiling all input files will be concatenated into a single file.

When the -s flag is present, each file compiled will be put into its own file terminated with the .css extension

Examples

mini-sass myfile.scss -o output.css

Will generate a single file output.css

mini-sass myfile1.scss myfile2.scss -s

Will generate two files: myfile1.css and myfile2.css corresponding to each input.

API References

Assuming proper installation, load as follows:

var miniSass = require("mini-sass");

There is a single high-level function compile that takes a SCSS string and returns the compiled CSS string:

var scss = "$var: 2px; div { width: $var; }"
var css = miniSass.compile(scss);
console.log(css) // "div { \n width: 2px; \n }"

Features

• Variables

$var: 20px;

div{
    width: $var;
}

• Nested Styles

div {
    ul{
        background: green;
    }
}

• Mixins

@mixin reset-list {
  margin: 0;
  padding: 0;
  list-style: none;
}

div{
    @include reset-list;
}

IMPORTANT

This compiler is not entirely rigorous and will most likely fail in certain cases.

How it works

First, a lexical analysis is performed, converting source code into a stream of tokens, such as IDs, brackets, etc. This corresponds to src/lexer.js

Then a parser converts the stream of tokens into an abstract syntax tree (AST), a data structure that represents the input as a tree based on a Context-Free Grammer. More accurately, the parser gets the next token in the stream by calling the token() method. The parser and corresponding grammar exist in src/parser.js

Finally, the generator takes the AST and produces an output string based on various rules (i.e. the features listed above). This corresponds to src/generator.js

TO DO

• Interpolation
• Partials and extended @import
• Extend API to allow retrieval of AST and tokens

Contributions welcome

Tests

This project uses mocha and chai with babel for test located in test/index.js. Run tests using the following command

npm run test