From: Hee-Sob Park <phasis@...>
Date: 2001-04-21T12:41:04+09:00
Subject: [ruby-talk:13954] Re: regexp


"James Ponder" <james@squish.net> wrote in message
news:20010421024536.C14119@snowy.squish.net...
> Hi, I'm new to ruby and am coming from a perl background - therefore I
> was hoping to use the perlisms in ruby when it came to regexps.
>
> Quite often you have an input line, and you want to pull data out from it
> using a regexp.  In perl I would do the following, naming my fields
without
> ever using $1, $2, etc.:
>
>   if (($key, $val) = $var =~ /^(.+)=(.+)$/) {
>     # ...
>   }
>
> In ruby, the =~ String instance method returns the position of the match
> (which perl does when in scalar context) rather than an array of matches,
> so I'm finding it quite difficult to find a way to do the above perl
> code, which in my mind is very readable and understandable.
>
> I thought about doing
>
>   key, val = var =~ /^(.+)=(.+)$/ && $~[1..-1]
>   if key
>     ...
>
> but nobody would be able to read my code (besides I started feeling
sick)...
>
> so at the moment I just do:
>
>   if var =~ /^(.+)=(.+)$/
>     key, val = $1, $2
>
> but that's no where near as nice as perl's one-liner (and I really hate
using
> $1, $2 etc.).  Any ideas?
>
>
How about

key, val = /^(.+)=(.+)$/.match(var)[1..-1]
if key
...

Or

if (key, val = /^(.+)=(.+)$/.match(var)[1..-1])[0]
...

> Best wishes, James
> --
> James Ponder; www.squish.net

Park Hee-Sob