Re: condense_targets and basepath functions

On Mon, Feb 12, 2001 at 09:16:29AM -0600, cmpilato@newton.ch.collab.net wrote:
> Kevin Pilch-Bisson <kevin@pilch-bisson.net> writes:
> > Alright, that sounds reasonable, but how are we handling portability
> > stuff. realpath(3) doesn't exist on win32, but _fullpath does. Should
> > I suggest this to the APR list, since portability stuff is their domain,
> > or should I hack up the file with #ifdef WIN32. That way APR could in
> > theory use the system version if it is not broken, or use their own if
> > it is broken.

Write the code however it is needed, including an #ifdef.

> Nonono. Please don't #ifdef WIN32 if at all possible. I've spent too
> much time already finding and fixing inconsistencies between the
> various Win32 platforms in APR...we don't want to fight the same
> battles on this side of the border. I think APR is indeed the place
> for such a thing.

We move it to APR after we have a working version. We already have several
functions that would best be within APR, and are pending a move. This would
simply be another.

We can't just say "that's APR's job" and sit back. The people that care
about something are the people to write the code. That means we write the
code and move it to the right place later.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/