From: Kevin Smith <kevinbsmith@...>
Date: 2001-09-20T02:05:14+09:00
Subject: [ruby-talk:21401] Re: nested require files need path internally

--- Yukihiro Matsumoto <matz@ruby-lang.org> wrote:
> |But, I think what he is looking for is the ability to
> require a file from the same directory as the file that
> does the require, not from the current directory.  I have
> experienced this problem too...
> 
> Your statement is understandable.  I think it's trade off
> between
> library search path complexity and saving few strokes of
> grep and
> edit.  I prefer behavior simplicity this case.

That's ok, if we can come up with a simple pattern to allow
multiple-file apps to WORK reliably, regardless of where
they are executed from, and what the LIBPATH is. This is
important.

I'll take a stab based on a bit of reading and only a bit
of actual (inconclusive) coding: It would seem that the
"main" module should examine __FILE__ or $0, and extract
the path portion (if any). If there's no path, "." should
be assumed. That path should then be inserted at the front
of $LOAD_PATH.

Now, I don't know if the path information is preserved
correctly in either __FILE__ or $0. If not, then I can't
think of any way to accomplish the goal, and some mechanism
should be added. If my approach does work, perhaps someone
(maybe me, but no promises) could bundle it up into a
reusable code facet.

Kevin


__________________________________________________
Terrorist Attacks on U.S. - How can you help?
Donate cash, emergency relief information
http://dailynews.yahoo.com/fc/US/Emergency_Information/