Re: make vs. gmake in 6-CURRENT...

Hi, Ruslan, Harti, (and all):

Ruslan Ermilov wrote:
> On Wed, Dec 15, 2004 at 11:18:33PM -0500, Chuck Swiger wrote:
[ ... ]
>>However, it seems like that space-stuffing when generating paths would 
>>break lots of other things.  Anyway, I wanted to bring up the issue here 
>>and ask whether this change to make's behavior is expected or whether it 
>>should be considered a bug...?
> 
> FWIW, I cannot reproduce this problem:

Did you comment out the "USE_GMAKE=   YES" line?  As I mentioned, a fix to the 
port has already been applied, but you can see the problem reproduced on the 
ports build cluster at this URL:

http://pointyhat.freebsd.org/errorlogs/i386-errorlogs/e.6.2004121303/dvd+rw-tools-5.21.4.10.8.log

	-----

Harti Brandt wrote:
[ ... ]
> I cannot reproduce this (I have not looked at the port's makefile). Can 
> you make a small makefile that shows this behaviour?

Sure, I'll give it a shot.  Try a "make install" against this:

# Makefile attempt to reproduce bug

BIN_FILES=      foo bar baz
WRKSRC=         /tmp

install:
.for i in ${BIN_FILES}
         cp ${WRKSRC}/${i} /var/tmp/
.endfor


...and see whether 6's make space-stuffs "/tmp/foo" into "/tmp/ foo".  If it 
doesn't, the port's Makefile isn't much bigger than this example.

Thanks for the interest...

-- 
-Chuck