From: Robert Klemme <shortcutter@...>
Date: 2008-04-05T17:55:03+09:00
Subject: Re: [regexp] How to extract...

On 05.04.2008 10:41, Todd Benson wrote:
> On Fri, Apr 4, 2008 at 8:00 PM, 7stud -- <bbxx789_05ss@yahoo.com> wrote:
>> Jan Kowalski wrote:
>>  > From:
>>  >
>>  > #header h1 a {
>>  >   display: block;
>>  >   background: url(images/wordpress-logo.png) center left no-repeat;
>>  >   width: 301px;
>>  >   height: 88px;
>>  >   text-indent: -9999px;
>>  >   float: left;
>>  > }
>>  >
>>  > #header ul li#download a {
>>  >   background: #d54e21 url(images/download-tab-bg.png) bottom left
>>  > repeat-x;
>>  >   color: #fff;
>>  >   -moz-border-radius-topleft: 3px;
>>  >   -khtml-border-top-left-radius: 3px;
>>  >   -webkit-border-top-left-radius: 3px;
>>  >   border-top-left-radius: 3px;
>>  >   -moz-border-radius-topright: 3px;
>>  >   -khtml-border-top-right-radius: 3px;
>>  >   -webkit-border-top-right-radius: 3px;
>>  >   border-top-right-radius: 3px;
>>  >   text-shadow: #b5421c 1px 1px 1px;
>>  > }
>>  >
>>  > i need each occurence of text enclosed between url(************), for
>>  > this example:
>>  >
>>  > images/wordpress-logo.png
>>  > images/download-tab-bg.png
>>
>>  str = <<CSS
>>
>> #header h1 a {
>>   display: block;
>>   background: url(images/wordpress-logo.png) center left no-repeat;
>>   width: 301px;
>>   height: 88px;
>>   text-indent: -9999px;
>>   float: left;
>>  }
>>
>>  #header ul li#download a {
>>   background: #d54e21 url(images/download-tab-bg.png) bottom left
>>  repeat-x;
>>   color: #fff;
>>   -moz-border-radius-topleft: 3px;
>>   -khtml-border-top-left-radius: 3px;
>>   -webkit-border-top-left-radius: 3px;
>>   border-top-left-radius: 3px;
>>   -moz-border-radius-topright: 3px;
>>   -khtml-border-top-right-radius: 3px;
>>   -webkit-border-top-right-radius: 3px;
>>   border-top-right-radius: 3px;
>>   text-shadow: #b5421c 1px 1px 1px;
>>  }
>>  CSS
>>
>>  pattern = /url\((.*)\)/ #to match a parenthesis escape it with a '\'
>>
>>  str.each do |line|
>>   match = line[pattern, 1]  #1 is the parenthesized sub pattern
>>   if match
>>     puts match
>>   end
>>  end
>>
>>  --output:--
>>
>>
>> images/wordpress-logo.png
>>  images/download-tab-bg.png
> 
> You can use a non-greedy multi-line #scan pattern as well...
> 
> str.scan /url\((.*?)\)/m
> => [["images/wordpress-logo.png"], [images/download-tab-bg.png]]
> 
> ..which you can #flatten if you want.

I'd rather use a more specific match as I believe this could be a tad 
more efficient

str.scan /url\(([^)]*)\)/m

Cheers

	robert