Installation

$ [sudo] pip install slimit

Or the bleeding edge version from the git master branch:

$ [sudo] pip install git+https://github.com/rspivak/slimit.git#egg=slimit

Let’s minify some code

From the command line:

$ slimit -h
Usage: slimit [options] [input file]

If no input file is provided STDIN is used by default.
Minified JavaScript code is printed to STDOUT.

Options:
  -h, --help            show this help message and exit
  -m, --mangle          mangle names
  -t, --mangle-toplevel
                        mangle top level scope (defaults to False)

$ cat test.js
var foo = function( obj ) {
        for ( var name in obj ) {
                return false;
        }
        return true;
};
$
$ slimit --mangle < test.js
var foo=function(a){for(var b in a)return false;return true;};

Or using library API:

>>> from slimit import minify
>>> text = """
... var foo = function( obj ) {
...         for ( var name in obj ) {
...                 return false;
...         }
...         return true;
... };
... """
>>> print minify(text, mangle=True, mangle_toplevel=True)
var a=function(a){for(var b in a)return false;return true;};

Iterate over, modify a JavaScript AST and pretty print it

>>> from slimit.parser import Parser
>>> from slimit.visitors import nodevisitor
>>> from slimit import ast
>>>
>>> parser = Parser()
>>> tree = parser.parse('for(var i=0; i<10; i++) {var x=5+i;}')
>>> for node in nodevisitor.visit(tree):
...     if isinstance(node, ast.Identifier) and node.value == 'i':
...         node.value = 'hello'
...
>>> print tree.to_ecma() # print awesome javascript :)
for (var hello = 0; hello < 10; hello++) {
  var x = 5 + hello;
}
>>>

Writing custom node visitor

>>> from slimit.parser import Parser
>>> from slimit.visitors.nodevisitor import ASTVisitor
>>>
>>> text = """
... var x = {
...     "key1": "value1",
...     "key2": "value2"
... };
... """
>>>
>>> class MyVisitor(ASTVisitor):
...     def visit_Object(self, node):
...         """Visit object literal."""
...         for prop in node:
...             left, right = prop.left, prop.right
...             print 'Property key=%s, value=%s' % (left.value, right.value)
...             # visit all children in turn
...             self.visit(prop)
...
>>>
>>> parser = Parser()
>>> tree = parser.parse(text)
>>> visitor = MyVisitor()
>>> visitor.visit(tree)
Property key="key1", value="value1"
Property key="key2", value="value2"

Using lexer in your project

>>> from slimit.lexer import Lexer
>>> lexer = Lexer()
>>> lexer.input('a = 1;')
>>> for token in lexer:
...     print token
...
LexToken(ID,'a',1,0)
LexToken(EQ,'=',1,2)
LexToken(NUMBER,'1',1,4)
LexToken(SEMI,';',1,5)

You can get one token at a time using token method:

>>> lexer.input('a = 1;')
>>> while True:
...     token = lexer.token()
...     if not token:
...         break
...     print token
...
LexToken(ID,'a',1,0)
LexToken(EQ,'=',1,2)
LexToken(NUMBER,'1',1,4)
LexToken(SEMI,';',1,5)

LexToken instance has different attributes:

>>> lexer.input('a = 1;')
>>> token = lexer.token()
>>> token.type, token.value, token.lineno, token.lexpos
('ID', 'a', 1, 0)

Benchmarks

SAM - JQuery size after minification in bytes (the smaller number the better)

Roadmap

• when doing name mangling handle cases with ‘eval’ and ‘with’

• foo[“bar”] ==> foo.bar

• consecutive declarations: var a = 10; var b = 20; ==> var a=10,b=20;

• reduce simple constant expressions if the result takes less space: 1 +2 * 3 ==> 7

• IF statement optimizations

  1. if (foo) bar(); else baz(); ==> foo?bar():baz();

  2. if (!foo) bar(); else baz(); ==> foo?baz():bar();

  3. if (foo) bar(); ==> foo&&bar();

  4. if (!foo) bar(); ==> foo||bar();

  5. if (foo) return bar(); else return baz(); ==> return foo?bar():baz();

  6. if (foo) return bar(); else something(); ==> {if(foo)return bar();something()}

• remove unreachable code that follows a return, throw, break or continue statement, except function/variable declarations

• parsing speed improvements

Acknowledgments

• The lexer and parser are built with PLY

• Several test cases and regexes from jslex

• Some visitor ideas - pycparser

• Many grammar rules are taken from rkelly

• Name mangling and different optimization ideas - UglifyJS

• ASI implementation was inspired by pyjsparser

License

The MIT License (MIT)

0.6.1 (2012-03-15)

• Added command-line option -t/–mangle-toplevel to turn on global scope name mangling. As of this version it’s off by default: https://github.com/rspivak/slimit/issues/27

• Removed dependency on a ‘distribute’ package

• Bug fix: https://github.com/rspivak/slimit/issues/26

• Bug fix: https://github.com/rspivak/slimit/issues/25

0.6 (2012-02-04)

• Added optimization: foo[“bar”] ==> foo.bar

• Added base class for custom AST node visitors

• Documentation updates

• Bug fix: https://github.com/rspivak/slimit/issues/22

• Bug fix: https://github.com/rspivak/slimit/issues/21

0.5.5 (2011-10-05)

• Bugfix: https://github.com/rspivak/slimit/issues/7

0.5.4 (2011-10-01)

• Bugfix: https://github.com/rspivak/slimit/issues/6 Division with “this” fails

0.5.3 (2011-06-29)

• Bugfix: https://github.com/rspivak/slimit/issues/5

0.5.2 (2011-06-14)

• Bugfix: https://github.com/rspivak/slimit/issues/4

• Bugfix: https://github.com/rspivak/slimit/issues/3

0.5.1 (2011-06-06)

• Bugfix: https://github.com/rspivak/slimit/issues/2

0.5 (2011-06-06)

• Added name mangling

0.4 (2011-05-12)

• Minify more by removing block braces { }

• More tests

0.3.2 (2011-05-09)

• More hacks to use pre-generated lex and yacc tables when called from the command line

0.3.1 (2011-05-09)

• Use pre-generated lex and yacc tables when called from the command line

0.3 (2011-05-09)

• Added minifier

0.2 (2011-05-07)

• Added a JavaScript parser

• Added pretty printer

• Added node visitor

0.1 (2011-05-02)

• Initial public version. It contains only a JavaScript lexer