From: Joel VanderWerf <vjoel@...>
Date: 2003-01-12T09:20:42+09:00
Subject: Re: Ruby:  politics & performance [long]

Louis,

Another approach is to avoid the parser, which was probably not intended 
to deal with huge lists of nodes. Ruby has a great feature which allows 
a source file to include arbitrary data. For example:

====================
p DATA.read.split

__END__
1 2 3 4 5
====================

If the __END__ occurs on a line with no other characters, then the 
parser ignores the rest of the file, and the DATA constant refers to an 
IO-like object which contains all the data following the __END__ line.

I'm sure #read and #split will turn out to be more efficient than the 
ruby parser, since they have less to do.