[GRASS5] Re: [GRASSLIST:1998] Re: Installing grass57_exp_2003_10_25 on MacX

Hello Scott

On Wed, 10 Dec 2003, Scott W Mitchell wrote:

On 9-Dec-03, at 05:36, Markus Neteler wrote:
> On Mon, Dec 08, 2003 at 09:39:22PM +0000, Paul Kelly wrote:
>> I also think this means all the EXTRA_LIBS lines can be removed from
>> the
>> library Makefiles again now but since I haven't got an OSX system to
>> test it I don't want to risk doing this. It's nice to see the
>> Makefiles
>> tidy and as short as possible though.
>
> Generally agreed - someone with MacOSX system may try and post the
> result. I am too far from a Mac to try myself.
>
>
> Markus
>

What constitutes a test ? On an os 10.3 system, I went in to lib/ogsf,
then moved to lib/gis figuring that's a better test, and removed the
EXTRA_LIBS file from the 2 makefiles, then recompiled. Went smoothly,
and I can digitize a line, display rasters and run r.mapcalc.

Just as long as it compiles (you are using shared libraries of course?) I
think should be OK. I will remove just one or two from CVS then (maybe gis
and proj) and see if there are any bug reports.

The main thing I worried about the EXTRA_LIBS was it might make the libraries
bigger if you were compiling statically, but I didn't test this and could be
completely wrong.

Paul

Oops, sorry, I THOUGHT I was using shared libraries, but your email made me check my config.status, and sure enough, that time around I hadn't used the shared flag.

Just did a rebuild with it turned back on, and it still seems fine. Oh, I also deleted the EXTRA_LIBS line from lib/proj/Makefile first. All looks good.

Cheers,
Scott

On 10-Dec-03, at 10:24, Paul Kelly wrote:

Hello Scott

On Wed, 10 Dec 2003, Scott W Mitchell wrote:

On 9-Dec-03, at 05:36, Markus Neteler wrote:

On Mon, Dec 08, 2003 at 09:39:22PM +0000, Paul Kelly wrote:

I also think this means all the EXTRA_LIBS lines can be removed from
the library Makefiles again now but since I haven't got an OSX system to
test it I don't want to risk doing this. It's nice to see the
Makefiles tidy and as short as possible though.

Generally agreed - someone with MacOSX system may try and post the
result. I am too far from a Mac to try myself.

What constitutes a test ? On an os 10.3 system, I went in to lib/ogsf,
then moved to lib/gis figuring that's a better test, and removed the
EXTRA_LIBS file from the 2 makefiles, then recompiled. Went smoothly,
and I can digitize a line, display rasters and run r.mapcalc.

Just as long as it compiles (you are using shared libraries of course?) I
think should be OK. Paul

------
Scott W. Mitchell Scott_Mitchell@carleton.ca
Department of Geography and Environmental Studies
Carleton University, B349 Loeb Building (Office A209)
1125 Colonel By Drive, Ottawa, ON Canada K1S 5B6
+1-613-520-2600 ext 2695 Fax: 1-613-520-4301