[GRASS5] configure.in Wants testers...

GRASS GRASS developer
1

Okay,

I think I've implemented all of the checks in the hacked configure
script in configure.in. So, now I need others to test it and give me
problem reports. I haven't added anything, like searching for libjpeg,
yet. I want to get the input script at least to where the hacked
configure script is before I start adding things. Probably there will
be search paths for things that I didn't think of. Anyway, make a
backup of the configure script and delete your config.cache before
running autoconf. Also, note that there doesn't seem to be a good way
to have the --bindir default to /usr/local/bin. The macro
AC_DEFAULT_BINDIR() was not recognized by my autoconf and the variable
$bindir is always set to something (defaulting to ${exec_prefix}/bin,
where $exec_prefix defaults to ${prefix} which is set to
/usr/local/grass5 by default. Confused?). So, if you want
/usr/local/bin for $bindir, you need to specify it. Also, see
./configure --help for some of the options (they need testing too).

--
Eric G. Miller <egm2@jps.net>

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

2

Hi Eric

"Eric G . Miller" wrote:

So, now I need others to test it and give me
problem reports.

I just installed m4, autoconf, and automake on our SGI. However, I have
no idea how to use them. If you can give me some pointers in getting
started I'll give configure.in a try here.

Thanks.

--
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

Hi Eric,

thank for your ongoing improvement of configure.in!

Here my report (just updated)

On Thu, Nov 23, 2000 at 09:32:28PM -0800, Eric G . Miller wrote:

Okay,

I think I've implemented all of the checks in the hacked configure
script in configure.in. So, now I need others to test it and give me
problem reports. I haven't added anything, like searching for libjpeg,
yet. I want to get the input script at least to where the hacked
configure script is before I start adding things. Probably there will
be search paths for things that I didn't think of. Anyway, make a
backup of the configure script and delete your config.cache before
running autoconf.

o.k.

Also, note that there doesn't seem to be a good way

to have the --bindir default to /usr/local/bin. The macro
AC_DEFAULT_BINDIR() was not recognized by my autoconf and the variable
$bindir is always set to something (defaulting to ${exec_prefix}/bin,
where $exec_prefix defaults to ${prefix} which is set to
/usr/local/grass5 by default. Confused?). So, if you want
/usr/local/bin for $bindir, you need to specify it. Also, see
./configure --help for some of the options (they need testing too).

Mhhh, here we have the problem that the start script is neither
put into /usr/local/bin nor /usr/bin.
I think we should decide to write it to /usr/bin on default.
Means: The UNIX_BIN would have to be set differently than prefix.

DLLIB is o.k.
ZLIB as well,
XDRLIB, too

PostgreSQL, o.k.
openGL o.k.

not yet o.k.:
ODBCINC not yet (not yet implemented, I think)
TCLINCDIR =
-> found from configure, but here it's empty (it's in /usr/include)
TKINCDIR =
-> found from configure, but here it's empty (it's in /usr/include)
TCLTKLIBS =
-> *not* found from configure (in /usr/lib)

Now I will compile...

  Markus

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

4

On Fri, Nov 24, 2000 at 06:05:25PM +0000, Markus Neteler wrote:

Also, note that there doesn't seem to be a good way
> to have the --bindir default to /usr/local/bin. The macro
> AC_DEFAULT_BINDIR() was not recognized by my autoconf and the variable
> $bindir is always set to something (defaulting to ${exec_prefix}/bin,
> where $exec_prefix defaults to ${prefix} which is set to
> /usr/local/grass5 by default. Confused?). So, if you want
> /usr/local/bin for $bindir, you need to specify it. Also, see
> ./configure --help for some of the options (they need testing too).
Mhhh, here we have the problem that the start script is neither
put into /usr/local/bin nor /usr/bin.
I think we should decide to write it to /usr/bin on default.
Means: The UNIX_BIN would have to be set differently than prefix.

Well, then we're back to using --with-bindir. Under that, it is
possible to know if the user specified something different than the
defaults set by configure.

not yet o.k.:
ODBCINC not yet (not yet implemented, I think)

Ahh, yes, missed that. Not sure what I need to look for.

TCLINCDIR =
-> found from configure, but here it's empty (it's in /usr/include)
TKINCDIR =
-> found from configure, but here it's empty (it's in /usr/include)
TCLTKLIBS =

/usr/include is in the default search path, so nothing extra should need
to be specified, correct???

-> *not* found from configure (in /usr/lib)

I'm still struggling with why this works on my system and fails on yours
when the libraries are in the exact same location (and even the version
is the same).

--
Eric G. Miller <egm2@jps.net>

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

5

> not yet o.k.:
> ODBCINC not yet (not yet implemented, I think)
Ahh, yes, missed that. Not sure what I need to look for.

sql.h, sqlext.h, sqltypes.h

Radim

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

6

Do we think configure.in is close enough at this point to replace the
configure script currently in CVS with a new one generated from
configure.in? There's some breakage right now because some variables
have been defined elsewhere that aren't handled by the current configure
script.

Any objections?

--
Eric G. Miller <egm2@jps.net>

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

7

On Sun, Nov 26, 2000 at 02:10:35PM -0800, Eric G . Miller wrote:

Do we think configure.in is close enough at this point to replace the
configure script currently in CVS with a new one generated from
configure.in? There's some breakage right now because some variables
have been defined elsewhere that aren't handled by the current configure
script.

Any objections?

Hi Eric,

the only objection is that
TCLTKLIBS
is not yet set. If you would manage to fix this, it can be replaced.

Regards

Markus

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

8

Hi Eric, and Markus

Markus Neteler wrote:

On Sun, Nov 26, 2000 at 02:10:35PM -0800, Eric G . Miller wrote:
> Do we think configure.in is close enough at this point to replace
> the configure script currently in CVS with a new one generated from
> configure.in? There's some breakage right now because some
> variables have been defined elsewhere that aren't handled by the
> current configure script.
>
> Any objections?
Hi Eric,

the only objection is that
TCLTKLIBS
is not yet set. If you would manage to fix this, it can be replaced.

Unfortunately I found another problem. I updated from CVS and then ran
make distclean. Then I ran autoconf and make install and then found
these messages:

UNIX_BIN=/usr/people/guest/grassCVS/grass/bin SRC= MAKE=make
                                              ^^^^
NAME_VER=`cat /src/CMD/VERSION | head -1 | sed 's/\..*//'`
              ^^^^^
GISBASE=/usr/people/guest/grassCVS/grass MACHINENAME=`uname -n` /bin/sh
src/CMD/generic/CREATE_GMAKE.sh mips-sgi-irix6.5
/usr/people/guest/grassCVS/grass/bin
Cannot open /src/CMD/VERSION: No such file or directory
        /bin/sh /src/CMD/generic/is64bit.sh mips-sgi-irix6.5
/bin/sh: /src/CMD/generic/is64bit.sh: not found
*** Error code 127 (bu21)

You will notice that SRC variable is empty, thus causing paths for
things in the grass source tree to bigin with /. Checking the Makefile
indicates that SRCDIR=$(pwd) which in this case is empty. I don't know
if this is only happening on SGI's or not. Markus, did you try a make
install on Linux?

Anyway, I hope this helps.

--
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'

9

Justin Hickey wrote:

Unfortunately I found another problem. I updated from CVS and then ran
make distclean. Then I ran autoconf and make install and then found
these messages:

Oops forgot to say I ran configure as well. Sorry.

--
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'

10

Hi Justin,

On Mon, Nov 27, 2000 at 05:26:14PM +0700, Justin Hickey wrote:

Hi Eric, and Markus

Markus Neteler wrote:
> On Sun, Nov 26, 2000 at 02:10:35PM -0800, Eric G . Miller wrote:
> > Do we think configure.in is close enough at this point to replace
> > the configure script currently in CVS with a new one generated from
> > configure.in? There's some breakage right now because some
> > variables have been defined elsewhere that aren't handled by the
> > current configure script.
> >
> > Any objections?
> Hi Eric,
>
> the only objection is that
> TCLTKLIBS
> is not yet set. If you would manage to fix this, it can be replaced.

Unfortunately I found another problem. I updated from CVS and then ran
make distclean. Then I ran autoconf and make install and then found
these messages:

UNIX_BIN=/usr/people/guest/grassCVS/grass/bin SRC= MAKE=make
                                              ^^^^
NAME_VER=`cat /src/CMD/VERSION | head -1 | sed 's/\..*//'`
              ^^^^^

[...]

You will notice that SRC variable is empty, thus causing paths for
things in the grass source tree to bigin with /. Checking the Makefile
indicates that SRCDIR=$(pwd) which in this case is empty. I don't know
if this is only happening on SGI's or not. Markus, did you try a make
install on Linux?

On Linkx it runs well with new configure (except TCLTK):

Makefile
[...]
SRCDIR= /home/neteler/ggg

It is set properly.

In configure.in it is aroung line 417:
SRCDIR=$(pwd)

Yours

Markus

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

11

On Thu, Nov 23, 2000 at 09:32:28PM -0800, Eric G . Miller wrote:

I think I've implemented all of the checks in the hacked configure
script in configure.in. So, now I need others to test it and give me
problem reports.

I am using current CVS version of configure.in rev 1.16.
My problem ist that there is no way to disable opengl.
There also should be an option to disable the SGI mesa widgets and
the corrersponding commands, because not everybody has the mesa build by
themself.

  Bernhard

--
Professional Service around Free Software (intevation.net)
The FreeGIS Project (freegis.org)
Association for a Free Informational Infrastructure (ffii.org)
FSF Europe (www.fsfeurope.org)

12

On Mon, Nov 27, 2000 at 05:31:42PM +0100, Bernhard Reiter wrote:

On Thu, Nov 23, 2000 at 09:32:28PM -0800, Eric G . Miller wrote:
> I think I've implemented all of the checks in the hacked configure
> script in configure.in. So, now I need others to test it and give me
> problem reports.

I am using current CVS version of configure.in rev 1.16.
My problem ist that there is no way to disable opengl.

Generally I suggest to have as most as possible *active*, but
disabling should be possible not to crash the configure.

There also should be an option to disable the SGI mesa widgets and
the corrersponding commands, because not everybody has the mesa build by
themself.

Well, at least SuSe7.0-Mesa contains the SGI mesa widgets :slight_smile:
I can't remember that I had to compile myself. Should a be
*disable* option as well.

Just my 0.02 Euro

Markus

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

13

On Mon, Nov 27, 2000 at 11:43:05AM +0000, Markus Neteler wrote:

> You will notice that SRC variable is empty, thus causing paths for
> things in the grass source tree to bigin with /. Checking the Makefile
> indicates that SRCDIR=$(pwd) which in this case is empty. I don't know
> if this is only happening on SGI's or not. Markus, did you try a make
> install on Linux?

On Linkx it runs well with new configure (except TCLTK):

Makefile
[...]
SRCDIR= /home/neteler/ggg

It is set properly.

In configure.in it is aroung line 417:
SRCDIR=$(pwd)

Yes, I added that. I thought that was supposed to be the "portable" way
to run 'pwd', but apparently it isn't. `pwd` is supposed to be bad, but
I don't know why. I'll take a look again.

--
Eric G. Miller <egm2@jps.net>

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

14

On Mon, Nov 27, 2000 at 05:27:58PM +0000, Markus Neteler wrote:

On Mon, Nov 27, 2000 at 05:31:42PM +0100, Bernhard Reiter wrote:
> I am using current CVS version of configure.in rev 1.16.
> My problem ist that there is no way to disable opengl.

Generally I suggest to have as most as possible *active*, but
disabling should be possible not to crash the configure.

> There also should be an option to disable the SGI mesa widgets and
> the corrersponding commands, because not everybody has the mesa build by
> themself.

Well, at least SuSe7.0-Mesa contains the SGI mesa widgets :slight_smile:
I can't remember that I had to compile myself. Should a be
*disable* option as well.

Well there's two only partially related things here. First, I can add a
--disable-opengl flag. However, the build process will have to know how
to handle this (with say a USE_OPENGL environment variable). I think
only the top-level Makefile can do anything about this. This will have
to prevent NVIZ2.2 and the showdpsf programs from compiling.

The Mesa widgets are only needed by the showdspf stuff I think, so
there'd have to be a variable for that as well. However, the showdspf
stuff is not currently compiled by default, so this should be no big
issue. It will simply fail to link if the user tries to compile it.

--
Eric G. Miller <egm2@jps.net>

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

15

Hi Eric

"Eric G . Miller" wrote:

On Mon, Nov 27, 2000 at 11:43:05AM +0000, Markus Neteler wrote:
> > You will notice that SRC variable is empty, thus causing paths for
> > things in the grass source tree to bigin with /. Checking the
> > Makefile indicates that SRCDIR=$(pwd) which in this case is empty.
> > I don't know if this is only happening on SGI's or not. Markus,
> > did you try a make
> > install on Linux?
>
> On Linkx it runs well with new configure (except TCLTK):
>
> Makefile
> [...]
> SRCDIR= /home/neteler/ggg
>
> It is set properly.
>
> In configure.in it is aroung line 417:
> SRCDIR=$(pwd)

Yes, I added that. I thought that was supposed to be the "portable"
way to run 'pwd', but apparently it isn't. `pwd` is supposed to be
bad, but I don't know why. I'll take a look again.

Since my system seems to be causing the problem I did some poking around
and I may be totally off-base here since my knowledge of autoconf is
very minimal.

What I found was that the @srcdir@ variable (note the lower case) is
preset by autoconf. I commented out the lines concerning the definition
of SRCDIR in configure.in and changed the substitution string in
Makefile.in to lower case. I then ran make distclean, ran autoconf, then
ran configure and the substitution in Makefile turned out to be "." (a
single dot - current directory).

Does this value make sense for our purposes? On my machine, when I ran
make install, it solved the earlier problems, however, in
src/CMD/generic/GISGEN.sh, there is a "cd /tmp" command at the top which
screws up the $GMAKE variable, since it is set to ./src/CMD/gmake5.0
(which does not exist in /tmp).

So, it appears that the . definition would work if all our scripts were
written to take this into account. Does anybody know what kind of effort
this would be (perhaps this is a minimal change)? Does the @srcdir@ get
substituted with a dot on other systems? Or am I way off base and this
is just a looney idea which somebody else has already looked into?

Any comments would be greatly appreciated. Thanks.

--
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'

16

On Tue, Nov 28, 2000 at 03:34:39PM +0700, Justin Hickey wrote:

> Yes, I added that. I thought that was supposed to be the "portable"
> way to run 'pwd', but apparently it isn't. `pwd` is supposed to be
> bad, but I don't know why. I'll take a look again.

Since my system seems to be causing the problem I did some poking around
and I may be totally off-base here since my knowledge of autoconf is
very minimal.

What I found was that the @srcdir@ variable (note the lower case) is
preset by autoconf. I commented out the lines concerning the definition
of SRCDIR in configure.in and changed the substitution string in
Makefile.in to lower case. I then ran make distclean, ran autoconf, then
ran configure and the substitution in Makefile turned out to be "." (a
single dot - current directory).

Does this value make sense for our purposes? On my machine, when I ran
make install, it solved the earlier problems, however, in
src/CMD/generic/GISGEN.sh, there is a "cd /tmp" command at the top which
screws up the $GMAKE variable, since it is set to ./src/CMD/gmake5.0
(which does not exist in /tmp).

So, it appears that the . definition would work if all our scripts were
written to take this into account. Does anybody know what kind of effort
this would be (perhaps this is a minimal change)? Does the @srcdir@ get
substituted with a dot on other systems? Or am I way off base and this
is just a looney idea which somebody else has already looked into?

Any comments would be greatly appreciated. Thanks.

What I gleaned from the AutoConf docs I have, is that $top_srcdir should
be used to reference the TOP src directory. Apparently it's not being
expanded, so I'll look into a way to make it be forcably expanded.

I guess I could just make the configure.in always use $(pwd). That
should always work (right???).

--
Eric G. Miller <egm2@jps.net>

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

17

Hi Eric

"Eric G . Miller" wrote:

What I gleaned from the AutoConf docs I have, is that $top_srcdir
should be used to reference the TOP src directory. Apparently it's
not being expanded, so I'll look into a way to make it be forcably
expanded.

I guess I could just make the configure.in always use $(pwd). That
should always work (right???).

I would have thought so but that is the current problem on my system.
The @SRCDIR@ variable is substituted with $(pwd) without executing the
pwd command. So it is now a make variable called pwd which is empty.
However, I did notice in that section of configure.in, that there are
brace brackets {} used for $top_srcdir but parenthesis () used for $pwd.
Could that be the problem? Just a wild guess.

--
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'

18

On Mon, Nov 27, 2000 at 07:05:19PM -0800, Eric G . Miller wrote:

On Mon, Nov 27, 2000 at 05:27:58PM +0000, Markus Neteler wrote:
> On Mon, Nov 27, 2000 at 05:31:42PM +0100, Bernhard Reiter wrote:
> > I am using current CVS version of configure.in rev 1.16.
> > My problem ist that there is no way to disable opengl.

> > There also should be an option to disable the SGI mesa widgets and
> > the corrersponding commands, because not everybody has the mesa build by
> > themself.
>
> Well, at least SuSe7.0-Mesa contains the SGI mesa widgets :slight_smile:
> I can't remember that I had to compile myself.

  Hu, Suse is always scary.

Well there's two only partially related things here. First, I can add a
--disable-opengl flag. However, the build process will have to know how
to handle this (with say a USE_OPENGL environment variable). I think
only the top-level Makefile can do anything about this. This will have
to prevent NVIZ2.2 and the showdpsf programs from compiling.

Sounds reasonable.

The Mesa widgets are only needed by the showdspf stuff I think, so
there'd have to be a variable for that as well. However, the showdspf
stuff is not currently compiled by default, so this should be no big
issue. It will simply fail to link if the user tries to compile it.

Well, but ./configure shall not stop in that circumstances and report an
error, but just note that there are no SGI mesa widgets around.
And then it should not try to comple itself, because this would be
incorrectly indicated as an error.

We also need another option for the configure to stop when the
compilation fails. This should be the _default_.
The current behaviour to just continue and log the error might
be convenient for some, but is it very surprising for other people and
non-standard. If a build fails somewhere, it must stop,
unless a special flags is given to override this behavious.

  Bernhard

--
Professional Service around Free Software (intevation.net)
The FreeGIS Project (freegis.org)
Association for a Free Informational Infrastructure (ffii.org)
FSF Europe (www.fsfeurope.org)

19

On Tue, Nov 28, 2000 at 01:08:00AM -0800, Eric G . Miller wrote:

On Tue, Nov 28, 2000 at 03:34:39PM +0700, Justin Hickey wrote:

What I gleaned from the AutoConf docs I have, is that $top_srcdir should
be used to reference the TOP src directory. Apparently it's not being
expanded, so I'll look into a way to make it be forcably expanded.

As I am just happen to fill a bug report for autconf, you might be
interested to read it, though it might not be directly related to your
problem.

It is bug #160 and you can get to the bug database from
  http://sources.redhat.com/autoconf/

The mailinglists there might be another good point for specific autoconf
questions.

Here is the text:
         There are two kind of preset output variables for Makefiles
        in autoconf. The first kind delivers(@prefix@)
        and the second kind only works if you are actually
        using it in a makefile and have certain other
        variables already defined.

        MY_PATH=@exec_prefix@

        just alone will not work in a Makefile.in, because it
        might get replaced with ${prefix}, so you need at least

        prefix=@prefix@
        MY_PATH=@exec_prefix@

        this is confusing and inconsistant. I would expect these
        predefined variables for paths just to give me the path
        having done all relevant substitutions already. This way I
        can also use them in other file not just real Makefiles.

        If you think this is the correct behavious it still is a
        documentation bug. There should be a warning that you need to
        have certain variables ready for shell substituion
        if you want to use @exec_prefix@ or @libdir@.

Bernhard

--
Professional Service around Free Software (intevation.net)
The FreeGIS Project (freegis.org)
Association for a Free Informational Infrastructure (ffii.org)
FSF Europe (www.fsfeurope.org)

20

Hi Eric

"Eric G . Miller" wrote:

What I gleaned from the AutoConf docs I have, is that $top_srcdir
should be used to reference the TOP src directory. Apparently it's
not being expanded, so I'll look into a way to make it be forcably
expanded.

I played around a bit with SRCDIR and $top_srcdir and it appears that
$top_srcdir for some reason is not being preset to anything since the
test for it fails. However, the good news is that I found a solution for
setting SRCDIR for my machine. Instead of using

SRCDIR=$(pwd)

I used the following:

SRCDIR=`pwd`

This will properly define the SRCDIR variable (an expanded path) in my
Makefile. However, the question is whether this works on other
platforms. I think it should since it is basic shell syntax. What do you
think?

--
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'