Repository files navigation

I started working on UglifyJS’s successor, version 2. It’s almost a full rewrite (except for the parser which is heavily modified, everything else starts from scratch). I’ve detailed my reasons in the README, see the project page.

https://github.com/mishoo/UglifyJS2

Version 1 will continue to be maintained for fixing show-stopper bugs, but no new features should be expected.

This package implements a general-purpose JavaScript parser/compressor/beautifier toolkit. It is developed onNodeJS, but it should work on any JavaScript platform supporting the CommonJS module system (and if your platform of choice doesn’t support CommonJS, you can easily implement it, or discard theexports.* lines from UglifyJS sources).

The tokenizer/parser generates an abstract syntax tree from JS code. You can then traverse the AST to learn more about the code, or do various manipulations on it. This part is implemented inparse-js.js and it’s a port to JavaScript of the excellentparse-js Common Lisp library fromMarijn Haverbeke.

( Seecl-uglify-js if you’re looking for the Common Lisp version of UglifyJS. )

The second part of this package, implemented inprocess.js, inspects and manipulates the AST generated by the parser to provide the following:

• ability to re-generate JavaScript code from the AST. Optionally indented—you can use this if you want to “beautify” a program that has been compressed, so that you can inspect the source. But you can also run our code generator to print out an AST without any whitespace, so you achieve compression as well.
• shorten variable names (usually to single characters). Our mangler will analyze the code and generate proper variable names, depending on scope and usage, and is smart enough to deal with globals defined elsewhere, or witheval() calls orwith{} statements. In short, ifeval() orwith{} are used in some scope, then all variables in that scope and any variables in the parent scopes will remain unmangled, and any references to such variables remain unmangled as well.
• various small optimizations that may lead to faster code but certainly lead to smaller code. Where possible, we do the following:
  • foo[“bar”] ==> foo.bar
  • remove block brackets{}
  • join consecutive var declarations: var a = 10; var b = 20; ==> var a=10,b=20;
  • resolve simple constant expressions: 1 +2 * 3 ==> 7. We only do the replacement if the result occupies less bytes; for example 1/3 would translate to 0.333333333333, so in this case we don’t replace it.
  • consecutive statements in blocks are merged into a sequence; in many cases, this leaves blocks with a single statement, so then we can remove the block brackets.
  • various optimizations for IF statements:
    • if (foo) bar(); else baz(); ==> foo?bar():baz();
    • if (!foo) bar(); else baz(); ==> foo?baz():bar();
    • if (foo) bar(); ==> foo&&bar();
    • if (!foo) bar(); ==> foo||bar();
    • if (foo) return bar(); else return baz(); ==> return foo?bar():baz();
    • if (foo) return bar(); else something(); ==> {if(foo)return bar();something()}
  • remove some unreachable code and warn about it (code that follows areturn,throw,break orcontinue statement, except function/variable declarations).
  • act a limited version of a pre-processor (c.f. the pre-processor of C/C++) to allow you to safely replace selected global symbols with specified values. When combined with the optimisations above this can make UglifyJS operate slightly more like a compilation process, in that when certain symbols are replaced by constant values, entire code blocks may be optimised away as unreachable.

The following transformations can in theory break code, although they’re probably safe in most practical cases. To enable them you need to pass the--unsafe flag.

The following transformations occur:

newArray(1,2,3,4)=>[1,2,3,4]Array(a,b,c)=>[a,b,c]newArray(5)=>Array(5)newArray(a)=>Array(a)

These are all safe if the Array name isn’t redefined. JavaScript does allow one to globally redefine Array (and pretty much everything, in fact) but I personally don’t see why would anyone do that.

UglifyJS does handle the case where Array is redefined locally, or even globally but with afunction orvar declaration. Therefore, in the following cases UglifyJSdoesn’t touch calls or instantiations of Array:

// case 1.  globally declared variablevarArray;newArray(1,2,3);Array(a,b);// or (can be declared later)newArray(1,2,3);varArray;// or (can be a function)newArray(1,2,3);functionArray(){ ...}// case 2.  declared in a function(function(){a=newArray(1,2,3);b=Array(5,6);varArray;})();// or(function(Array){returnArray(5,6,7);})();// or(function(){returnnewArray(1,2,3,4);functionArray(){ ...}})();// etc.

UglifyJS is now available through NPM —npm install uglify-js@1 should do the job.

NOTE: The NPM package has been upgraded to UglifyJS2. If you need to install version 1.x you need to add `@1` to the command, as I did above. I strongly suggest you to try to upgrade, though this might not be simple (v2 has a completely different AST structure and API).

## clone the repositorymkdir -p /where/you/wanna/put/itcd /where/you/wanna/put/itgit clone git://github.com/mishoo/UglifyJS.git## make the module available to Nodemkdir -p~/.node_libraries/cd~/.node_libraries/ln -s /where/you/wanna/put/it/UglifyJS/uglify-js.js## and if you want the CLI script too:mkdir -p~/bincd~/binln -s /where/you/wanna/put/it/UglifyJS/bin/uglifyjs# (then add ~/bin to your $PATH if it's not there already)

There is a command-line tool that exposes the functionality of this library for your shell-scripting needs:

uglifyjs [ options... ] [ filename ]

filename should be the last argument and should name the file from which to read the JavaScript code. If you don’t specify it, it will read code from STDIN.

Supported options:

• -b or--beautify — output indented code; when passed, additional options control the beautifier:
  • -i N or--indent N — indentation level (number of spaces)
  • -q or--quote-keys — quote keys in literal objects (by default, only keys that cannot be identifier names will be quotes).
• -c or----consolidate-primitive-values — consolidates null, Boolean, and String values. Known as aliasing in the Closure Compiler. Worsens the data compression ratio of gzip.
• --ascii — pass this argument to encode non-ASCII characters as\uXXXX sequences. By default UglifyJS won’t bother to do it and will output Unicode characters instead. (the output is always encoded in UTF8, but if you pass this option you’ll only get ASCII).
• -nm or--no-mangle — don’t mangle names.
• -nmf or--no-mangle-functions – in case you want to mangle variable names, but not touch function names.
• -ns or--no-squeeze — don’t callast_squeeze() (which does various optimizations that result in smaller, less readable code).
• -mt or--mangle-toplevel — mangle names in the toplevel scope too (by default we don’t do this).
• --no-seqs — whenast_squeeze() is called (thus, unless you pass--no-squeeze) it will reduce consecutive statements in blocks into a sequence. For example, “a = 10; b = 20; foo();” will be written as “a=10,b=20,foo();”. In various occasions, this allows us to discard the block brackets (since the block becomes a single statement). This is ON by default because it seems safe and saves a few hundred bytes on some libs that I tested it on, but pass--no-seqs to disable it.
• --no-dead-code — by default, UglifyJS will remove code that is obviously unreachable (code that follows areturn,throw,break orcontinue statement and is not a function/variable declaration). Pass this option to disable this optimization.
• -nc or--no-copyright — by default,uglifyjs will keep the initial comment tokens in the generated code (assumed to be copyright information etc.). If you pass this it will discard it.
• -o filename or--output filename — put the result infilename. If this isn’t given, the result goes to standard output (or see next one).
• --overwrite — if the code is read from a file (not from STDIN) and you pass--overwrite then the output will be written in the same file.
• --ast — pass this if you want to get the Abstract Syntax Tree instead of JavaScript as output. Useful for debugging or learning more about the internals.
• -v or--verbose — output some notes on STDERR (for now just how long each operation takes).
• -d SYMBOL[=VALUE] or--define SYMBOL[=VALUE] — will replace all instances of the specified symbol where used as an identifier (except where symbol has properly declared by a var declaration or use as function parameter or similar) with the specified value. This argument may be specified multiple times to define multiple symbols - if no value is specified the symbol will be replaced with the valuetrue, or you can specify a numeric value (such as1024), a quoted string value (such as"object" or'https://github.com'), or the name of another symbol or keyword (such asnull ordocument). This allows you, for example, to assign meaningful names to key constant values but discard the symbolic names in the uglified version for brevity/efficiency, or when used wth care, allows UglifyJS to operate as a form ofconditional compilation whereby defining appropriate values may, by dint of the constant folding and dead code removal features above, remove entire superfluous code blocks (e.g. completely remove instrumentation or trace code for production use). Where string values are being defined, the handling of quotes are likely to be subject to the specifics of your command shell environment, so you may need to experiment with quoting styles depending on your platform, or you may find the option--define-from-module more suitable for use.
• -define-from-module SOMEMODULE — will load the named module (as per the NodeJSrequire() function) and iterate all the exported properties of the module defining them as symbol names to be defined (as if by the--define option) per the name of each property (i.e. without the module name prefix) and given the value of the property. This is a much easier way to handle and document groups of symbols to be defined rather than a large number of--define options.
• --unsafe — enable other additional optimizations that are known to be unsafe in some contrived situations, but could still be generally useful. For now only these:
  • foo.toString() ==> foo+””
  • new Array(x,…) ==> [x,…]
  • new Array(x) ==> Array(x)
• --max-line-len (default 32K characters) — add a newline after around 32K characters. I’ve seen both FF and Chrome croak when all the code was on a single line of around 670K. Pass –max-line-len 0 to disable this safety feature.
• --reserved-names — some libraries rely on certain names to be used, as pointed out in issue #92 and #81, so this option allow you to exclude such names from the mangler. For example, to keep namesrequire and$super intact you’d specify –reserved-names “require,$super”.
• --inline-script – when you want to include the output literally in an HTML<script> tag you can use this option to prevent</script from showing up in the output.
• --lift-vars – when you pass this, UglifyJS will apply the following transformations (see the notes in API,ast_lift_variables):
  • put allvar declarations at the start of the scope
  • make sure a variable is declared only once
  • discard unused function arguments
  • discard unused inner (named) functions
  • finally, try to merge assignments into that onevar declaration, if possible.

To use the library from JavaScript, you’d do the following (example for NodeJS):

varjsp=require("uglify-js").parser;varpro=require("uglify-js").uglify;varorig_code="... JS code here";varast=jsp.parse(orig_code);// parse code and get the initial ASTast=pro.ast_mangle(ast);// get a new AST with mangled namesast=pro.ast_squeeze(ast);// get an AST with compression optimizationsvarfinal_code=pro.gen_code(ast);// compressed code here

The above performs the full compression that is possible right now. As you can see, there are a sequence of steps which you can apply. For example if you want compressed output but for some reason you don’t want to mangle variable names, you would simply skip the line that callspro.ast_mangle(ast).

Some of these functions take optional arguments. Here’s a description:

• jsp.parse(code, strict_semicolons) – parses JS code and returns an AST.strict_semicolons is optional and defaults tofalse. If you passtrue then the parser will throw an error when it expects a semicolon and it doesn’t find it. For most JS code you don’t want that, but it’s useful if you want to strictly sanitize your code.
• pro.ast_lift_variables(ast) – merge and movevar declarations to the scop of the scope; discard unused function arguments or variables; discard unused (named) inner functions. It also tries to merge assignments following thevar declaration into it.

  If your code is very hand-optimized concerningvar declarations, this lifting variable declarations might actually increase size. For me it helps out. On jQuery it adds 865 bytes (243 after gzip). YMMV. Also note that (since it’s not enabled by default) this operation isn’t yet heavily tested (please report if you find issues!).

  Note that although it might increase the image size (on jQuery it gains 865 bytes, 243 after gzip) it’s technically more correct: in certain situations, dead code removal might drop variable declarations, which would not happen if the variables are lifted in advance.

  Here’s an example of what it does:

functionf(a,b,c,d,e){varq;varw;w=10;q=20;for(vari=1;i<10;++i){varboo=foo(a);}for(vari=0;i<1;++i){varboo=bar(c);}functionfoo(){ ...}functionbar(){ ...}functionbaz(){ ...}}// transforms into ==>functionf(a,b,c){vari,boo,w=10,q=20;for(i=1;i<10;++i){boo=foo(a);}for(i=0;i<1;++i){boo=bar(c);}functionfoo(){ ...}functionbar(){ ...}}

• pro.ast_mangle(ast, options) – generates a new AST containing mangled (compressed) variable and function names. It supports the following options:
  • toplevel – mangle toplevel names (by default we don’t touch them).
  • except – an array of names to exclude from compression.
  • defines – an object with properties named after symbols to replace (see the--define option for the script) and the values representing the AST replacement value. For example,{ defines: { DEBUG: ['name', 'false'], VERSION: ['string', '1.0'] } }
• pro.ast_squeeze(ast, options) – employs further optimizations designed to reduce the size of the code thatgen_code would generate from the AST. Returns a new AST.options can be a hash; the supported options are:
  • make_seqs (default true) which will cause consecutive statements in a block to be merged using the “sequence” (comma) operator
  • dead_code (default true) which will remove unreachable code.
• pro.gen_code(ast, options) – generates JS code from the AST. By default it’s minified, but using theoptions argument you can get nicely formatted output.options is, well, optional :-) and if you pass it it must be an object and supports the following properties (below you can see the default values):
  • beautify: false – passtrue if you want indented output
  • indent_start: 0 (only applies whenbeautify istrue) – initial indentation in spaces
  • indent_level: 4 (only applies whenbeautify istrue) – indentation level, in spaces (pass an even number)
  • quote_keys: false – if you passtrue it will quote all keys in literal objects
  • space_colon: false (only applies whenbeautify istrue) – wether to put a space before the colon in object literals
  • ascii_only: false – passtrue if you want to encode non-ASCII characters as\uXXXX.
  • inline_script: false – passtrue to escape occurrences of</script in strings

The beautifier can be used as a general purpose indentation tool. It’s useful when you want to make a minified file readable. One limitation, though, is that it discards all comments, so you don’t really want to use it to reformat your code, unless you don’t have, or don’t care about, comments.

In fact it’s not the beautifier who discards comments — they are dumped at the parsing stage, when we build the initial AST. Comments don’t really make sense in the AST, and while we could add nodes for them, it would be inconvenient because we’d have to add special rules to ignore them at all the processing stages.

The--define option can be used, particularly when combined with the constant folding logic, as a form of pre-processor to enable or remove particular constructions, such as might be used for instrumenting development code, or to produce variations aimed at a specific platform.

The code below illustrates the way this can be done, and how the symbol replacement is performed.

CLAUSE1:if(typeofDEVMODE==='undefined'){DEVMODE=true;}CLAUSE2:functioninit(){if(DEVMODE){console.log("init() called");}    ....DEVMODE&&console.log("init() complete");}CLAUSE3:functionreportDeviceStatus(device){varDEVMODE=device.mode,DEVNAME=device.name;if(DEVMODE==='open'){        ....}}

When the above code is normally executed, the undeclared global variableDEVMODE will be assigned the valuetrue (seeCLAUSE1) and so theinit() function (CLAUSE2) will write messages to the console log when executed, but inCLAUSE3 a locally declared variable will mask access to theDEVMODE global symbol.

If the above code is processed by UglifyJS with an argument of--define DEVMODE=false then UglifyJS will replaceDEVMODE with the boolean constant valuefalse withinCLAUSE1 andCLAUSE2, but it will leaveCLAUSE3 as it stands because thereDEVMODE resolves to a validly declared variable.

And more so, the constant-folding features of UglifyJS will recognise that theif condition ofCLAUSE1 is thus always false, and so will remove the test and body ofCLAUSE1 altogether (including the otherwise slightly problematical statementfalse = true; which it will have formed by replacingDEVMODE in the body). Similarly, withinCLAUSE2 both calls toconsole.log() will be removed altogether.

In this way you can mimic, to a limited degree, the functionality of the C/C++ pre-processor to enable or completely remove blocks depending on how certain symbols are defined - perhaps using UglifyJS to generate different versions of source aimed at different environments

It is recommmended (but not made mandatory) that symbols designed for this purpose are given names consisting ofUPPER_CASE_LETTERS to distinguish them from other (normal) symbols and avoid the sort of clash thatCLAUSE3 above illustrates.

Here are updated statistics. (I also updated my Google Closure and YUI installations).

We’re still a lot better than YUI in terms of compression, though slightly slower. We’re still a lot faster than Closure, and compression after gzip is comparable.

Unfortunately, for the time being there is no automated test suite. But I ran the compressor manually on non-trivial code, and then I tested that the generated code works as expected. A few hundred times.

DynarchLIB was started in times when there was no good JS minifier. Therefore I was quite religious about trying to write short code manually, and as such DL contains a lot of syntactic hacks[1] such as “foo == bar ? a = 10 : b = 20”, though the more readable version would clearly be to use “if/else”.

Since the parser/compressor runs fine on DL and jQuery, I’m quite confident that it’s solid enough for production use. If you can identify any bugs, I’d love to hear about them (use the Google Group or email me directly).

[1] I even reported a few bugs and suggested some fixes in the originalparse-js library, and Marijn pushed fixes literally in minutes.

• Twitter:@UglifyJS
• Project at GitHub:http://github.com/mishoo/UglifyJS
• Google Group:http://groups.google.com/group/uglifyjs
• Common Lisp JS parser:http://marijn.haverbeke.nl/parse-js/
• JS-to-Lisp compiler:http://github.com/marijnh/js
• Common Lisp JS uglifier:http://github.com/mishoo/cl-uglify-js

UglifyJS is released under the BSD license:

Copyright 2010 (c) Mihai Bazon <mihai.bazon@gmail.com>Based on parse-js (http://marijn.haverbeke.nl/parse-js/).Redistribution and use in source and binary forms, with or withoutmodification, are permitted provided that the following conditionsare met:    * Redistributions of source code must retain the above      copyright notice, this list of conditions and the following      disclaimer.    * Redistributions in binary form must reproduce the above      copyright notice, this list of conditions and the following      disclaimer in the documentation and/or other materials      provided with the distribution.THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER “AS IS” AND ANYEXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THEIMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULARPURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER BELIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY,OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, ORPROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANYTHEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, ORTORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OFTHE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OFSUCH DAMAGE.