From: Charles Mills <cmills@...>
Date: 2005-06-21T01:45:35+09:00
Subject: Re: Accessing MatchData results in StringScanner

Phrogz wrote:
> Is it possible using #scan or any other method of StringScanner to get
> access to captured sub-expressions in the supplied regexp?
>
> For example, something like:
>   ss = StringScanner.new( "Hello __World__" )
>
>   while ...
>     if md = ss.scan( /__([^_]+)__/ )
>       # I would like md to be a MatchData, not a String
>       # because I want access to the first capture;
>       # $1 (which I dislike on principle anyway) is not
>       # updated by the call to #scan
>     end
>   end
>
>
> If the answer is 'no'...might this be a good addition? Perhaps as a new
> method (in case returning a MatchData was much slower than a String)?

Read the docs :)
http://www.ruby-doc.org/stdlib/libdoc/strscan/rdoc/classes/StringScanner.html
http://www.ruby-doc.org/stdlib/libdoc/strscan/rdoc/classes/StringScanner.html#M000039

use #[]
ss[1] is like $1

-Charlie