mrclay / minify Goto Github PK

Currently, I use a mod_rewrite rule like this:

RewriteRule ^(.*?)\.v[0-9]+\.js$ /js/$1.js [L,E=VERSIONED_FILE:1]
Header add "Expires" "Mon, 28 Jul 2014 23:30:00 GMT" env=VERSIONED_FILE
Header add "Cache-Control" "max-age=315360000" env=VERSIONED_FILE

So I can request my JavaScript file at /js/myjavascript.js as /js/myjavascript.v15.js which forces clients to re-download a new version if I make any important changes. This is great, but I would love to have an easy way to do this with Minified-on-the-fly requests (something like /min/f=/js/myjavascript.js;v15).

I could add a variable to the query string, but some browsers and proxies do not cache resources that contain a query string.

Hi,
is there some sprite implementation (http://www.w3schools.com/css/css_image_sprites.asp)?

Firefox 19.0.2, Windows 7 64bits, PHP 5.4.11, Apache/2.2.23 (Win32)

When using ?debug=1, the file compiles with no error. When I try to use the groupConfig without the ?debug=1, I get a connection reset. The GET request isn't even showing in Apache access.log. There are nothing on error.log as well. I enabled FirePHP, but it doesn't give me any hint.

Works fine for JS though.

i realized that it might be a good idea to actually have a tag that includes the composer.json.

think you can bump the version anytime soon?

Right now when java cannot be executed correctly no exception is raised.

In Minify/YUICompressor.php

I think the code :

private static function _prepare()
{
    if (! is_file(self::$jarFile) 
        || ! is_dir(self::$tempDir)
        || ! is_writable(self::$tempDir)
    ) {
        throw new Exception('Minify_YUICompressor : $jarFile and $tempDir must be set.');
    }
}

Should be replaced by something like :

private static function _prepare()
{
    if (! is_file(self::$jarFile) 
        || ! is_dir(self::$tempDir)
        || ! is_writable(self::$tempDir)
        || ! is_exectuable(self::$jarFile) 
    ) {
        throw new Exception('Minify_YUICompressor : $jarFile and $tempDir must be set.');
    }
}

And

private static function _minify($type, $content, $options)
{
    self::_prepare();
    if (! ($tmpFile = tempnam(self::$tempDir, 'yuic_'))) {
        throw new Exception('Minify_YUICompressor : could not create temp file.');
    }
    file_put_contents($tmpFile, $content);
    //echo self::_getCmd($options, $type, $tmpFile);
    exec(self::_getCmd($options, $type, $tmpFile), $output);
    unlink($tmpFile);
    return implode("\n", $output);
}

By :

private static function _minify($type, $content, $options)
{
    self::_prepare();
    if (! ($tmpFile = tempnam(self::$tempDir, 'yuic_'))) {
        throw new Exception('Minify_YUICompressor : could not create temp file.');
    }
    file_put_contents($tmpFile, $content);
    //echo self::_getCmd($options, $type, $tmpFile);
    exec(self::_getCmd($options, $type, $tmpFile), $output, $return_var);
    unlink($tmpFile);
    if($return_var != 0){
            throw new Exception('Minify_YUICompressor : yuicompressor execution failed.');
    }
    return implode("\n", $output);
}

It seems that my Apache is not correctly logging 500 errors generated by minify, for example when a JS file has a syntax error and the compiler can't minify it.

I'm using ClosureCompiler:

$min_serveOptions['minifiers']['application/x-javascript'] = array('Minify_JS_ClosureCompiler', 'minify');

All my other 500 errors seem to be correctly reported in /var/log/apache2/error.log.
I tried debugging a little but I'm not sure where to look nor I'm sure it's a minify issue.

I use the library for a TYPO3 frontend compressor. Unfortunately the removal of HTML comments is hard coded respectively can't be configured from the caller by using options.

My problem here is, TYPO3 indexed search uses HTML comments to markup the content of a page (<!--TYPO3SEARCH_begin--> and <!--TYPO3SEARCH_end-->). So the only possibility here for me is either to change your code or to derive from Minify_HTML which is not optimal when updates come in.

Better would be a boolean option to enable/disable the removal or even better a regexp (e.g. TYPO3SEARCH_(begin|end)). This would also be nice, since IE 10 no longer supports conditional comments.

Given an array of stylesheet or javascript includes, how can I update cache and retrieve ID so that they can be included like below?

(when gzip is supported)
<link href="/min/cache/904c0ffe0aa5d4cac486e49803aa6eb5.gz" type="text/css" rel="stylesheet" />
or
<link href="/min/cache/904c0ffe0aa5d4cac486e49803aa6eb5" type="text/css" rel="stylesheet" />

There are several motivations for this approach:

• obfuscate URIs
• shorter URIs

Would the performance penalties of this approach be too significant?

Many thanks

Is the issued described at https://groups.google.com/forum/#!msg/minify/cpN-ncKPFZE/kwYVpLMkfDwJ fully addressed in version 2.1.7, or are the revisions to index.php still required with version 2.1.7?

The css .class1 .class2{} is minified to .class1.class2{} which has a different meaning.

I'm using Minify/2.1.7 and CKEditor 4.0 (revision 769d96134b).

There is a line in ckeditor.js (230):

...);return a.replace(/(['"]).*?\1/g,function(a){return ...

JSMin throws an exception:

Unterminated String at byte 107177: '"]).*?\\1/g,function(a){return ...

So, it points to a apostrophe inside RegExp literal. But when I test that line only with your unit test, it passed successfully. It's very strange. Also exception said offset 107117, but that apostrophe is placed in 107002.

I will rename ckeditor.js to ckeditor.min.js, but the bug still exists.

I tried the lastest version of Minify and Raphaël (http://raphaeljs.com/) and I get an error (in the minified source) if I try to use Raphaël on the page.

I was reluctant to post this as an issue, but it may just be one. I've been using Minify for a few years, and this is the first problem I've ever come across. It's possible (and maybe even likely) that the issue is on my end, but judging by the symptoms it may very well be something with Minify itself.

Using the latest version, I have reduced the issue down to a small block of CSS from Eric Meyer's reset:

html, body, div, span, applet, object, iframe,
h1, h2, h3, h4, h5, h6, p, blockquote, pre,
a, abbr, acronym, address, big, cite, code,
del, dfn, em, img, ins, kbd, q, s, samp,
small, strike, strong, sub, sup, tt, var,
b, u, i, center,
dl, dt, dd, ol, ul, li,
fieldset, form, label, legend,
table, caption, tbody, tfoot, thead, tr, th, td,
article, aside, canvas, details, embed, 
figure, figcaption, footer, header, hgroup, 
menu, nav, output, ruby, section, summary,
time, mark, audio, video {
    margin: 0;
    padding: 0;
    border: 0;
    font-size: 100%;
    font: inherit;
    vertical-align: baseline;
}

When this block is added, the stylesheet fails to load through Minify. When viewing the direct URL, the following error appears in Firefox:

The connection was reset
The connection to the server was reset while the page was loading.
* The site could be temporarily unavailable or too busy. Try again in a few moments.
* If you are unable to load any pages, check your computer's network connection.
* If your computer or network is protected by a firewall or proxy, make sure that Firefox is permitted to access the Web.

Of course, it works once you remove the above CSS block. After realizing this, I tried to pin-point the problem, but it seems that there is no single point of failure. It seems that reducing the number of selectors will cause it to work again, but there is no particular selector (or quantity of selectors) that's causing it to break.

Please watch this video so you can see what I'm seeing: http://csnl.net/temp/minify.mov

It seems that the simple act of adding or removing CSS selectors shouldn't cause anything to break, so I'm not sure if this is something to do with Minify or something else on my end.

What do you make of this?

Hi, a question.
what is your main minify repository, the hosted on code.google or github?

Best regards.

currently if using groupsConfig, then cache id becames simply max(mtime) of all files.

so, the cache id will remain the same, if i add more files to the group having identical timestamp.

i guess there should be added * NUMBER_OF_FILES to avoid this problem

so i had in my groupsConfig.php:

  'l' => array(
      '/path/to/lazyload.js'
  ),

then url built with was:

  g=l&1338376276

later i added one more file:

  'l' => array(
      '/path/to/uaft.js',
      '/path/to/lazyload.js'
  ),

and the url remained the same:

  g=l&1338376276

as the files have identical timestamp:

$ ls -l --full uaft.js lazyload.js 
-rw-r--r-- 1 root root 1.3K 2012-05-30 14:11:16.000000000 +0300 lazyload.js
-rw-r--r-- 1 root root 1.7K 2012-05-30 14:11:16.000000000 +0300 uaft.js

expanding that timestamp:

$ LC_ALL=C php -r 'echo strftime("%c %z\n", 1338376276);'
Wed May 30 14:11:16 2012 +0300

Sourcemaps are the new hotness for debugging minified javascript. It would be cool to see that on this library, although it requires Base64 with VLQ. There are implementations of sourcemaps in the closure compiler

Hello,

Given there's a changelog and a semver compatible versioning (https://github.com/mrclay/minify/blob/master/HISTORY.txt), you should tag versions as regular as they are published.

There is currently no tagged version, leading to no stable package available through packagist (see https://packagist.org/packages/mrclay/minify, only dev-master available)

Hello, For whatever reason there is a problem with minify, if bootstrap.css uses (bootstrap 3) from the Less compilation.

No notification but stylesheet not loaded

Trying to minify this:

/* http://example */
alert('success');

Returns:

/* 1 */ /* http://example.com */
/* 2 *| alert('success');

Commenting out the regexp added to Lines.php in d67feea fixes it to return the expected output:

/* 1 */ /* http://example.com */
/* 2 */ alert('success');

I have no trouble getting min to work on my local environment, and localhost:8888/min/f=min/quick-test.css works fine.

I am trying to get this working on the 'dev' environment of my server... so I set up a subdomain at dev.example.com, and created a new web root at /dev_html/ instead of /public_html/. This all seems to be working though, because when I run echo $_SERVER['DOCUMENT_ROOT'], it works as expected. So I don't think the subdomain has anything to do with why this isn't working.

I am testing with an untouched version of min on my server, directly in my subdomain's document root, but when I test dev.mydomain.com/min/f=min/quick-test.css, I receive an Internal Server Error on the request.

Thanks for any help, Nick

Minify currently assumes new file content will arrive with a fresh mtime. This makes it fast, but misses file changes in some situations.

Essentially we want to monitor md5(content) of each source. This is relatively slower than checking timestamps, so we'll make up for it by giving the cache a max-age and only revalidating after that period. Nearly every request will be faster because we only have to check the cache's timestamp.

I think each cacheId (mapping to a URL) will have four files:

• PT and GZ: plaintext and gzipped versions of the output, ready for readfile()
• MD1: a file containing MD5s of each source.
• MD2: a file containing the last time the MD5s were checked, and the ETag and Last-Modified headers to be sent if this file is fresh.

On each request, load MD2. If fresh, send the ETag/Last-Modified and the content (or 304). If MD2 is not fresh, check all sources against MD1, potentially regenerating all caches, or just up the timestamp in MD2.

This is fine if getting an MD5 is cheap, but some Minify configurations use non-files as sources, and we don't want to be fetching source every few seconds. Not sure how to deal with that.

I can see in the comments of Minify_HTML that when a jsMinifier callback is specified, the type attribute for SCRIPT tags is ignored. Why not err on the side of caution, and ONLY run the callback for when type="text/javascript"?

I'm trying to hook up Minify's HTML minification to an application that sometimes uses a javascript templating engine (like mustache), so it has SCRIPT tags that contain HTML content - which causes JSMin to throw an uncaught exception.

Since all sorts of content might appear within a SCRIPT tag, at least if type="text/javascript" one could be reasonably assured that in fact it does only contain Javascript.

Cheers,
John

PHP support prepare content on before/after

Need:

• Set hook get content auto_prepend_file
• Detect mime type text/html and output minify content on auto_append_file

start.php

# ....
ob_start();

end.php

function compress_html($buf)
{
   include_once "minify/min/lib/Minify/HTML.php";
   return Minify_HTML::minify($buf);
}

foreach (headers_list() as $header) {
  if (preg_match('#Content-Type:\s+([\w/+]+)(;\s+charset=(\S+))?#is', $header, $match_mime))
  {
    switch (strtolower($match_mime[1]))
    {
      case "text/html":
         #
      break;
      case "text/css":
         #
      break;
      case "text/javascript":
         #
      break;
      default:
         #
      break;
    }
  }
}

if start hook ob_start(), not work Content-Type

md5 was deprecated nearly a decade ago. However, it is still used as the only digest cryptographic tool.

Hi there,

Since a few days, Composer is returning an error when running 'update' because the location of the GIT repository seems to have changed. Is it possible you could fix that with Packagist?

  [RuntimeException]                                                                                
  Failed to clone http://github.com/mrclay/minify.git via git, https and http protocols, aborting.  

  - git://github.com/mrclay/minify.git                                                              
    fatal: Not a git repository (or any of the parent directories): .git     

it's possible that Minify_Cache_File gets racy when two concurrent thread try to put something to cache. and it reports code error

so, one thread that gets error. first it can't open file (which it assumes exists), gets false from fopen, and passes that false to flock:

PHP Warning:  fopen(/tmp/concurrency_test_500): failed to open stream: No such file or directory in vendor/mrclay/minify/min/lib/Minify/Cache/File.php on line 101
PHP Stack trace:
PHP   1. {main}() htdocs/tests/session.php:0
PHP   2. Minify_Cache_File->store() htdocs/tests/session.php:15
PHP   3. Minify_Cache_File->fetch() vendor/mrclay/minify/min/lib/Minify/Cache/File.php:37
PHP   4. fopen() vendor/mrclay/minify/min/lib/Minify/Cache/File.php:101
FP OPEN FAILED!
PHP Warning:  flock() expects parameter 1 to be resource, boolean given in vendor/mrclay/minify/min/lib/Minify/Cache/File.php on line 106
PHP Stack trace:
PHP   1. {main}() htdocs/tests/session.php:0
PHP   2. Minify_Cache_File->store() htdocs/tests/session.php:15
PHP   3. Minify_Cache_File->fetch() vendor/mrclay/minify/min/lib/Minify/Cache/File.php:37
PHP   4. flock() vendor/mrclay/minify/min/lib/Minify/Cache/File.php:106

the problem appears when one thread updates cachefile, second thread updates cachefile, refetches cachefile to see if there's what it wrote, sees it's not, and unlinks the cachefile. so when third thread does the same, it opens file which is no longer there, gets the false and produces the second warning on flock() call

  /**
    * Fetch the cached content
    *
    * @param string $id cache id (e.g. a filename)
    * 
    * @return string
    */
   public function fetch($id)
   {
       if ($this->_locking) {
           $fp = fopen($this->_path . '/' . $id, 'rb');
           flock($fp, LOCK_SH);
           $ret = stream_get_contents($fp);
           flock($fp, LOCK_UN);
           fclose($fp);
           return $ret;
       } else {
           return file_get_contents($this->_path . '/' . $id);
       }
   }

so perhaps just return false, when $fp = fopen() fails, this would match the behavior file_get_contents() does when it can't read a file.

and in more general fix, the locking should be used on a path that always exists (not unlink()'ed middle of action), i.e the dir containing the cachefile, or separate file for locking.

putting lock on cachedir, may affect two separate threads with different cachefile, which may create too many unwanted locks, so perhaps separate lockfile is better? or make it configurable?

my testcase involved shell script putting code run in background. and for sake of creating race at better rate, i made script to write different contents for each thread. ideally (in production where it happened) the cache value did not vary that much indeed.

<?php
require_once dirname(__FILE__) . '/../../vendor/autoload.php';

define('CACHE_ID', 'concurrency_test_' . posix_getuid());
define('CACHE_DIR', sys_get_temp_dir());
$cache = new Minify_Cache_File(CACHE_DIR, true);
if ($cache->isValid(CACHE_ID, 0)) {
    $session_id = trim($cache->fetch(CACHE_ID));
}
$value = 'value' . rand(1, 10000);
$cache->store(CACHE_ID, $value);

and the shell script to reproduce:

$ rm -f /tmp/concurrency_test_500 && for a in $(seq 1 50); do php session.php $a &  done

Is there a way to make the HTML minifier focus on collapsing whitespace only? And of course not do it for any element that is whitespace sensitive like pre and textarea.

This is because some JS frameworks like AngularJS can have significant comments that indicate a directive.

Also is this minifier tolerant of malformed html?

Hi, I've got a problem with brand new iOS 7 Safari browser. It doesn't want to load minified files like CSS and maybe JS too through your Minify module :( It's tricky to debug it on phone. But in Chrome on iOS 7 is everything OK. Thanks for respond. I'm gonna minify it by hand :D

Can we safely use Minify_HTML::minify($xml, array('xhtml' => true)); in order to minify any xml content ?

This issue might be cosmetic only, but maybe it might have some bigger consequences on other browsers

Please check
http://www.skimbl.com/css/cache/skimbl.1308290908.css (ignoring the first font declarations)
the result of Minify::combine()

ul, ul
li{list-style:none}

//later

.content
h2{margin:6px
0}

// more
img.thumb,.missing-image{float:left;display:block;margin:0
10px 0 0;border:1px
solid #E5E5E5;-moz-box-shadow:1px 1px 0px #F0F0F0;-webkit-box-shadow:1px 1px 0px #F0F0F0;box-shadow:1px 1px 0px #F0F0F0}

It don't expect the whole output to be one line only, but the new lines seem quite random, although it probably isn't

Been struggling with getting this to pass the test-all. I wonder if my server is not allowing..

Warning: file_get_contents(http://closure-compiler.appspot.com/compile): failed to open stream: Connection timed out in /var/www/html/site/min/lib/Minify/JS/ClosureCompiler.php on line 87

Fatal error: Uncaught exception 'Minify_JS_ClosureCompiler_Exception' with message 'No HTTP response from server' in /var/www/html/site/min/lib/Minify/JS/ClosureCompiler.php:104
Stack trace:
#0 /var/www/html/site/min/lib/Minify/JS/ClosureCompiler.php(56): Minify_JS_ClosureCompiler->_getResponse('js_code=%0A%28f...')
#1 /var/www/html/site/min/lib/Minify/JS/ClosureCompiler.php(29): Minify_JS_ClosureCompiler->min('?(function (win...')
#2 /var/www/html/site/min_unit_tests/test_Minify_JS_ClosureCompiler.php(18): Minify_JS_ClosureCompiler::minify('?(function (win...')
#3 /var/www/html/site/min_unit_tests/test_Minify_JS_ClosureCompiler.php(47): test_Minify_JS_ClosureCompiler()
#4 /var/www/html/site/min_unit_tests/test_all.php(12): require('/var/www/html/s...')
#5 {main}

thrown in /var/www/html/site/min/lib/Minify/JS/ClosureCompiler.php on line 104

If possible, it could be useful to enable an option to convert CSS background image to its base64 version to speed up CSS and site loading

Using version 2.1.4. I tried to use the experimental closure compiler by adding this line to my config file:

$min_serveOptions['minifiers']['application/x-javascript'] = array('Minify_JS_ClosureCompiler', 'minify');

But, unfortunately I get the following errors:

Warning: file_get_contents() [function.file-get-contents]: http:// wrapper is disabled in the server configuration by allow_url_fopen=0 in /path/to/min/lib/Minify/JS/ClosureCompiler.php on line 85
Warning: file_get_contents(http://closure-compiler.appspot.com/compile) [function.file-get-contents]: failed to open stream: no suitable wrapper could be found in /path/to/min/lib/Minify/JS/ClosureCompiler.php on line 85
Warning: Cannot modify header information - headers already sent by (output started at /path/to/min/lib/Minify/JS/ClosureCompiler.php:85) in /path/to/min/lib/Minify.php on line 439
Warning: Cannot modify header information - headers already sent by (output started at /path/to/min/lib/Minify/JS/ClosureCompiler.php:85) in /path/to/min/lib/Minify.php on line 440

500 Internal Server Error

Please see http://code.google.com/p/minify/wiki/Debugging.

I realize that it would probably work if we enabled allow_url_fopen, but it would be nice if Minify would gracefully fall back to another method, perhaps cURL.

Event calendar Elgg plugin can't be minified because JSMin throws an 500 error when try to minify fullcalendar script:

https://raw.github.com/lorea/event_calendar/master/views/default/js/event_calendar/fullcalendar.php

I tried to debug it without success.

We've been using minify in production for a while. Recently when a stylesheet is updated under high load many requests race to minify the stylesheet causing 100% CPU usage on all cores.

There's a similar issue at #266

but this seems to be missing in current releases

Hi guys, so i went here and grabed the IconMoon Icon Font. You get a CSS file with all the classes for the icons and the standard font-face font file. Then i tried to minify it using "minify" and it just keept on loading and loading the minifed css file untill after a few seconds it just stops and no css was loaded?

What could be the problem to this?

Hi,

I am using the CLI tool to minify and combine js files.

So I am typing in the cmd:

php tools\minifier\min_extras\cli\minify.php -o tools\myproject-minified.js -t js myproject/js/*.js

as a result I have a file telling me that '*.js' does not exist.

The path is not wrong because if I replace *.js by an existing js file, it is working.

In the CLI help, it seems that we can use the * charcater to select all files in a specifc directory.
Writing that, I have the impression that I misunderstood something from the CLI help and *.js stands for 'type-here-your-js-filename.js'

So the question is, is '*.js' supposed to find all js files from the specific directory ?

I am using minify 2.1.7 on Windows with PHP 5.3.9

Thank you in advance for your answer

If you have a multiline string in javascript such as

var x = "Test content
More content
And content";

The line numbers will be added into this multiline string

/* 100 / var x = "Test content
/ 101 / More content
/ 102 */ And content"

As suggested in issue #46 a quick fix may be to just allow an option disabling line numbers in debug mode.

When acessing directly http://127.0.0.1/app/min?g=js it's giving the content type of text/html. Shouldn't it enforce a text/javascript type instead? The same happens to CSS, returning as text/html

Looks like there is a hack for a more simple case https://github.com/mrclay/minify/blob/master/min/lib/Minify/Lines.php#L45
But that don't help with this problem globally. For example wysibb has lines of code like

    bbClose:"[/*]\n[/list]",

And /* in it is treated as a commentary start.

I honestly have no idea how this can be fixed, except with partially implementing Ecmascript parser in PHP.
Maybe you can make option for disabling line numbers in debug mode altogether? Sometimes you just need your code unfolded to try and add some breakpoints.

this means people can use composer to install this package too.

I'd do it for you, but you should obviously be the owner of the package :)

For sake of remote ClosureCompiler API use, we should bring back behavior of combining before minifying, maybe except for CSS.

See https://github.com/mrclay/minify/blob/master/min/lib/Minify.php#L490

Would be great that the minify module would follow the PSR-0 and be able to be autoloaded in frameworks that support that. the naming convention on the files already exist, just the explicit require/require_once inside the files that make autoloaders choke

Backtrace:

/path/to/min/lib/HTTP/Encoder.php (line 321)
/path/to/min/lib/Minify.php (line 226) HTTP_Encoder::isBuggyIe()
/path/to/min/index.php (line 66) Minify::serve()

$_SERVER:

Array
(
    [REDIRECT_VERSIONED_FILE] => 1
    [REDIRECT_STATUS] => 200
    [HTTP_HOST] => myhost.com
    [HTTP_ACCEPT] => */*
    [HTTP_CONNECTION] => close
    [PATH] => /sbin:/usr/sbin:/bin:/usr/bin
    [SERVER_SIGNATURE] => 
    [SERVER_SOFTWARE] => Apache/2.2.15 (Red Hat)
    [SERVER_NAME] => myhost.com
    [SERVER_ADDR] => ###.###.###.###
    [SERVER_PORT] => 80
    [REMOTE_ADDR] => ###.###.###.###
    [DOCUMENT_ROOT] => /path/to/files/
    [SERVER_ADMIN] => [email protected]
    [SCRIPT_FILENAME] => /path/to/file.php
    [REMOTE_PORT] => 18133
    [REDIRECT_URL] => /css/myfile.v25.css
    [GATEWAY_INTERFACE] => CGI/1.1
    [SERVER_PROTOCOL] => HTTP/1.0
    [REQUEST_METHOD] => GET
    [QUERY_STRING] => 
    [REQUEST_URI] => /css/myfile.v25.css
    [SCRIPT_NAME] => /path/to/file.php
    [PHP_SELF] => /path/to/file.php
    [REQUEST_TIME] => 1318458030
)

It would be great if this project could be added to packagist as a composer package.

It's very easy and requires very little maintenance, you create a composer.json file and it's ready to be added to http://packagist.org. I'd be happy to help out, if you want me to I can fork and send a PR with the necessary changes.

According to docs at http://code.google.com/p/minify/wiki/UriRewriting

$min_serveOptions['rewriteCssUris'] = false;

I should be able to add the above to my configuration file and it will disable uri rewriting. The reason I want to disable it, is because I don't need it. All my css/js files are in one folder and not nested at all. It is rewriting the urls. For example

Example:

minify/min_extras/cli/minify.php -o ../css/all.css ../css/phppos.css ../css/menubar.css ../css/general.css ../css/popupbox.css ../css/register.css ../css/receipt.css ../css/reports.css ../css/tables.css ../css/thickbox.css ../css/datepicker.css ../css/editsale.css ../css/footer.css ../css/css3.css ../css/jquery-ui-1.8.14.custom.css ../css/jquery.loadmask.css ../css/jquery.ui.all.css ../css/paginate.css

Before:

background-image: url("../images/header/header_empty.png");

After:

background-image:url("./header/header_empty.png");

This code on line 244 of builder/_index.js breaks under jQuery 1.9+:

if ($.browser.msie) {
    $('#getBm p:last').append(' Sorry, not supported in MSIE!');
}

$.browser is no longer available; use $.support instead. Giving up on IE entirely is also a bit unsporting; better to give up on the feature that old IE versions are lacking since IE9 and 10 are not so bad, for example using $support.boxModel.

When using a subdirectory like C:\Users\Webdev\www\app, and setting the documentRoot to C:\Users\Webdev\www\app, but the URL is accessed through http://localhost/app/, the rewrites for http://localhost/app/images/ point to http://localhost/images which is wrong. Create a option so the rewrite url can be prepended.

The method serve, line 312, in Minify.php does not properly validate malicious input. As a result, one is able to cause a multitude of injection attacks.

I have a problem minifying Plupload v2.0 (plupload.full.min.js) with JSMin::minify().

The error says:
JSMin: Unterminated String at byte 64434: "===e.typeOf(t)?e.each(t,function(e,t){d.setExif(t,e)' in /Minify/JSMin.php:203

What can I do. Is there a lint problems fixer or something?

Edit: Works fine on a Windows system but the error above still appears on a Debian system.