[GRASS5] make vs make install

GRASS GRASS developer
1

The convention is that "make" does compilation work, and that "make
install" does the copying of files into the place they are meant to go.

Packaging is normally done without running the compiler as root.

Would it be possible to fix the Grass Makefile so that this convention
is kept?

I may have mentioned this before, I don't know.

--
James Cameron mailto:quozl@us.netrek.org http://quozl.netrek.org/

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

2

Hello James

On Jul 27, 2:44pm, James Cameron wrote:

The convention is that "make" does compilation work, and that "make
install" does the copying of files into the place they are meant to go.

Yes this is known and is mentioned in the INSTALL file, but unfortunately, it
is a large task to change GRASS to do this. All the makefiles for all the
modules need to be changed.

I'm not sure if someone is actually working on this or not, but it is something
that we want to do.

--
Sincerely,

Jazzman (a.k.a. Justin Hickey) e-mail: jhickey@hpcc.nectec.or.th
High Performance Computing Center
National Electronics and Computer Technology Center (NECTEC)
Bangkok, Thailand

People who think they know everything are very irritating to those
of us who do. ---Anonymous

Jazz and Trek Rule!!!

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

3

On Thu, Jul 27, 2000 at 12:25:59PM +0000, Justin Hickey wrote:

Yes this is known and is mentioned in the INSTALL file, but unfortunately, it
is a large task to change GRASS to do this. All the makefiles for all the
modules need to be changed.

Could automake and friends be applied here?
Is there anything unusual about the per-module Makefiles?

--
James Cameron mailto:quozl@us.netrek.org http://quozl.netrek.org/

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

4

Hello James

On Jul 27, 4:14pm, James Cameron wrote:

Could automake and friends be applied here?

I don't know. I'm not that familiar with automake. However, we are using
autoconf to configure part of the system. Is autoconf part of automake?

Is there anything unusual about the per-module Makefiles?

The problem is outlined below, which comes from the INSTALL file

===============================

THINGS TO DO

A major task remaining is to replace the GISGEN system with a
proper set of Makefiles, all of which include a common one containing
the system dependencies.

================================

So, the problem is ensuring that the system dependencies are available to all
Makefiles. Currently, the GISGEN.sh shell script and the gmake.sh shell script
handle this task. Therefore, these scripts have to be replaced with a proper
Makefile system.

Note that GRASS was developed over 20 years ago when automake and friends did
not exist. With over 300 modules in the system, it is a daunting task to change
the way GRASS compiles now.

But we are improving. When I first compiled GRASS over 4 years ago, it was a
nightmare! At least we have a simple 3 step process now (assuming there are no
errors).

If you have any insights on how to solve this problem, they would be most
welcome.

--
Sincerely,

Jazzman (a.k.a. Justin Hickey) e-mail: jhickey@hpcc.nectec.or.th
High Performance Computing Center
National Electronics and Computer Technology Center (NECTEC)
Bangkok, Thailand

People who think they know everything are very irritating to those
of us who do. ---Anonymous

Jazz and Trek Rule!!!

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

5

On Thu, Jul 27, 2000 at 01:31:30PM +0000, Justin Hickey wrote:

I don't know. I'm not that familiar with automake. However, we are using
autoconf to configure part of the system. Is autoconf part of automake?

Separate package. Normally applied before autoconf. It is used to
generate Makefile.in that autoconf's generated configure script uses.
The input to automake is Makefile.am.

A major task remaining is to replace the GISGEN system with a
proper set of Makefiles, all of which include a common one containing
the system dependencies.

Common one containing the system dependencies? What does that mean?
If it is dependencies based on header files, there are automatic tools
to handle that now. makedepend is one. There are better ones.

But we are improving. When I first compiled GRASS over 4 years ago, it was a
nightmare!

I remember doing that! On a 486/33.

If you have any insights on how to solve this problem, they would be most
welcome.

I will have a bit of a think.

--
James Cameron mailto:quozl@us.netrek.org http://quozl.netrek.org/

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

6

Justin Hickey wrote:

Hello James

On Jul 27, 4:14pm, James Cameron wrote:
> Could automake and friends be applied here?

They are applied, and even in a fairly reasonable manner.

I have some notes on what should be done to ease this
process, but work has gotten busier and I'm (still) trying
to buy a house, so I haven't polished them yet. I'll try
to post a detailed description of what (I think) should be
done tomorrow night, to get comments, find help, etc.

Ideally, a New Make System for grass should allow distribution
of new programs independent of the grass module. For example,
the NVIZ program should be able to be distributed outside of
the grass "core" set of programs and libraries. We have a
decent extensible Make system here at work, I just need a night
or two free to hack some initial support into grass.

Markus, I promise to post something tomorrow, even if it is
just my disjointed thoughts. =)

-- ebm
+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+
| Eric B. Mitchell mailto:emitchell@altaira.com |
| tel: (301) 809 - 3534 Altair Aerospace Corporation |
| tel: (800) 7 - ALTAIR 4201 Northview Dr. Suite 410 |
| fax: (301) 805 - 8122 Bowie, MD 20716 |
+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+
              ,___
          /"\ / o=\ /"""---===/
         / \_/ \__/ ---===/
         | //\ || /""TT""/ //\ || ||""\
         | // \ || || // \ || ||__/
         | //--==\ |L--/ || //--==\ || || "=,
          \ ---===/
           \____---===/

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

7

On Thu, 27 Jul 2000, Justin Hickey wrote:

I don't know. I'm not that familiar with automake. However, we are using
autoconf to configure part of the system. Is autoconf part of automake?

  Autoconf and automake are both products of the FSF under the GNU label.
Each does a different set of related tasks. Both are well documented at the
Free Software Foundation's web site. I usually get there via
http://www.gnu.org.

HTH,

Rich

Dr. Richard B. Shepard, President

                       Applied Ecosystem Services, Inc. (TM)
              Making environmentally-responsible mining happen. (SM)
                       --------------------------------
            2404 SW 22nd Street | Troutdale, OR 97060-1247 | U.S.A.
+ 1 503-667-4517 (voice) | + 1 503-667-8863 (fax) | rshepard@appl-ecosys.com

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'