From: Daniel Berger <djberg96@...>
Date: 2006-11-16T04:10:07+09:00
Subject: Re: Group names in regular expressions


donn@cmscms.com wrote:
> Given the following regex to parse telephone numbers:
>
> /^(\(?(?<area>[0-9]{3})\)?)?(\-| )?(?<exch>[0-9]{3})(\-|
> )?(?<party>[0-9]{4})/
>
> Is there a regex.method to access the parse data by group name?  I was
> hoping something like the following behavior:
>
> rx = /^(\(?(?<area>[0-9]{3})\)?)?(\-| )?(?<exch>[0-9]{3})(\-|
> )?(?<party>[0-9]{4})/
> md = rx.match('800 325-3535')
> puts md['area']+md['exch']+md['party']
>
> I'm not that skilled with regex (yet), so I don't know if I could
> extend the regex class with my own method.  I'm not sure what all data
> is available to me after parsing.
>
> dvn

With Ruby's builtin regex engine? No. With Oniguruma (or the Ruby 1.9
branch)? Yes.

http://www.geocities.jp/kosako3/oniguruma/

Regards,

Dan