[GRASSLIST:7520] Compilation errors of Fedors Core 4

GRASS GRASS user
,
1

Hi all,

I recently upgraded to Fedora Core 4. When tried to restore my system I was
surprised to find out that GRASS 5 and 6 do not compile without significant
errors as they did before. Below is the error.log of grass 5.5 compilation,
while GRASS 6.1 had so much warnings that I did not include them. It is evident
that something is cardinally wrong. Could anybody help me to clear this
problem.

Thanks,
Miha Staut

GRASS GIS compilation log
-------------------------
Start of compilation: e,Bhe(Bet jul 7 15:46:02 CEST 2005
Errors:
Compilation error in module: src/libes/gmath (ignored)
Compilation error in module: src/imagery/i.cca (ignored)
Compilation error in module: src/imagery/i.class (ignored)
Compilation error in module: src/imagery/i.cluster (ignored)
Compilation error in module: src/imagery/i.colors (ignored)
Compilation error in module: src/imagery/i.composite (ignored)
Compilation error in module: src/imagery/i.gensig (ignored)
Compilation error in module: src/imagery/i.gensigset (ignored)
Compilation error in module: src/imagery/i.group (ignored)
Compilation error in module: src/imagery/i.maxlik (ignored)
Compilation error in module: src/imagery/i.ortho.photo (ignored)
Compilation error in module: src/imagery/i.pca (ignored)
Compilation error in module: src/imagery/i.points (ignored)
Compilation error in module: src/imagery/i.quantize (ignored)
Compilation error in module: src/imagery/i.rectify2 (ignored)
Compilation error in module: src/imagery/i.smap (ignored)
Compilation error in module: src/imagery/i.tape.mss (ignored)
Compilation error in module: src/imagery/i.tape.other (ignored)
Compilation error in module: src/imagery/i.tape.spot (ignored)
Compilation error in module: src/imagery/i.tape.tm (ignored)
Compilation error in module: src/imagery/i.tape.tm.fast (ignored)
Compilation error in module: src/imagery/i.target (ignored)
Compilation error in module: src/imagery/i.vpoints (ignored)
Compilation error in module: src/mapdev/v.surf.rst (ignored)
Compilation error in module: src/misc/m.svfit (ignored)
Compilation error in module: src/ps.map (ignored)
Compilation error in module: src/raster/r.grow2 (ignored)
Compilation error in module: src/raster/r.in.gdal (ignored)
Compilation error in module: src/raster/r.param.scale (ignored)
Compilation error in module: src/raster/r.resamp.rst (ignored)
Compilation error in module: src/raster/r.surf.gauss (ignored)
Compilation error in module: src/raster/r.surf.random (ignored)
Compilation error in module: src/sites/s.in.ascii (ignored)
Compilation error in module: src/sites/s.surf.rst (ignored)
Compilation error in module: src.contrib/CERL/raster/r.hydro.CASC2D (ignored)
Compilation error in module: src.contrib/SDTS (ignored)
Compilation error in module: src.garden/grass.postgresql (ignored)
Compilation error in module: src/imagery/i.fft (ignored)
Compilation error in module: src/imagery/i.zc (ignored)
Compilation error in module: src/raster/r.surf.fractal (ignored)
Compilation error in module: src.contrib/DUKE/r.terraflow (ignored)
Compilation error in module: src.contrib/CERL/imagery/i.shape (ignored)
End of compilation: e,Bhe(Bet jul 7 15:54:57 CEST 2005
DONE generating GRASS GIS binary code

___________________________________________________________
How much free photo storage do you get? Store your holiday
snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com

2

Miha Staut wrote:

I recently upgraded to Fedora Core 4. When tried to restore my system I was
surprised to find out that GRASS 5 and 6 do not compile without significant
errors as they did before. Below is the error.log of grass 5.5 compilation,
while GRASS 6.1 had so much warnings that I did not include them. It is evident
that something is cardinally wrong. Could anybody help me to clear this
problem.

We would need to see some of the actual error messages.

For 6.1, you can disable the excessive warnings by clearing
CFLAGS_WARNING when running make, i.e.:

  make CFLAGS_WARNING=

--
Glynn Clements <glynn@gclements.plus.com>

3

Hi Miha

Try the follwing:

1) emacs /grass-6.0.0/lib/gmath/numerical.h
2) comment lines 43 to line 50 and close emacs
3) recompile

Hopefully this will solve all your make errors BUT the
r.terraflow for which you have to apply a patch.

HTH

Jaime

--- Glynn Clements <glynn@gclements.plus.com>
escribió:

Miha Staut wrote:

> I recently upgraded to Fedora Core 4. When tried
to restore my system I was
> surprised to find out that GRASS 5 and 6 do not
compile without significant
> errors as they did before. Below is the error.log
of grass 5.5 compilation,
> while GRASS 6.1 had so much warnings that I did
not include them. It is evident
> that something is cardinally wrong. Could anybody
help me to clear this
> problem.

We would need to see some of the actual error
messages.

For 6.1, you can disable the excessive warnings by
clearing
CFLAGS_WARNING when running make, i.e.:

  make CFLAGS_WARNING=

--
Glynn Clements <glynn@gclements.plus.com>

__________________________________________________
Correo Yahoo!
Espacio para todos tus mensajes, antivirus y antispam ¡gratis!
Regístrate ya - http://correo.espanol.yahoo.com/

4

Jaime,

Thanks for the info, but I am first trying to compile GRASS 5.x.

Miha

--- Jaime Carrera <jaicarrerahdez@yahoo.com> wrote:

Hi Miha

Try the follwing:

1) emacs /grass-6.0.0/lib/gmath/numerical.h
2) comment lines 43 to line 50 and close emacs
3) recompile

Hopefully this will solve all your make errors BUT the
r.terraflow for which you have to apply a patch.

HTH

Jaime

--- Glynn Clements <glynn@gclements.plus.com>
escribió:

>
> Miha Staut wrote:
>
> > I recently upgraded to Fedora Core 4. When tried
> to restore my system I was
> > surprised to find out that GRASS 5 and 6 do not
> compile without significant
> > errors as they did before. Below is the error.log
> of grass 5.5 compilation,
> > while GRASS 6.1 had so much warnings that I did
> not include them. It is evident
> > that something is cardinally wrong. Could anybody
> help me to clear this
> > problem.
>
> We would need to see some of the actual error
> messages.
>
> For 6.1, you can disable the excessive warnings by
> clearing
> CFLAGS_WARNING when running make, i.e.:
>
> make CFLAGS_WARNING=
>
> --
> Glynn Clements <glynn@gclements.plus.com>
>
>

__________________________________________________
Correo Yahoo!
Espacio para todos tus mensajes, antivirus y antispam ¡gratis!
Regístrate ya - http://correo.espanol.yahoo.com/

___________________________________________________________
Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com

5

Thanks Glynn,

I tried to do as you said. So

make CFLAGS_WARNING= 2> std.error.txt

It seemed to me that no less warnings appeared (130 KB of text in all). I am
therefore sending the std. error output as an attachment. I am reminding again
that predominantly imagery commands had problems (according to error.log) while
for the others v.surf.rst, r.grow2, r.in.gdal (normal), r.param.scale,
r.resamp.rst, r.surf.gauss, r.surf.random, s.in.ascii, s.surf.rst,
r.hydro.CASC2D, r.surf.fractal, r.terraflow (normal).

I found it interesting that the CVS snapshot of GRASS 6.1 compiled appearently
without problems while the version from tarball of the CVS source tree did not.
The same was not true for GRASS 5.x.

Miha Staut

--- Glynn Clements <glynn@gclements.plus.com> wrote:

Miha Staut wrote:

> I recently upgraded to Fedora Core 4. When tried to restore my system I was
> surprised to find out that GRASS 5 and 6 do not compile without significant
> errors as they did before. Below is the error.log of grass 5.5 compilation,
> while GRASS 6.1 had so much warnings that I did not include them. It is
evident
> that something is cardinally wrong. Could anybody help me to clear this
> problem.

We would need to see some of the actual error messages.

For 6.1, you can disable the excessive warnings by clearing
CFLAGS_WARNING when running make, i.e.:

  make CFLAGS_WARNING=

--
Glynn Clements <glynn@gclements.plus.com>

___________________________________________________________
Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com

6

Miha,

On Mon, Jul 18, 2005 at 09:08:44AM +0100, Miha Staut wrote:
...

I found it interesting that the CVS snapshot of GRASS 6.1 compiled appearently
without problems while the version from tarball of the CVS source tree did not.
The same was not true for GRASS 5.x.

Brad Douglas and me spent some time to fix compiler warnings/errors:

(Brad worked on image library and others).
http://www.grass.itc.it/pipermail/grass5/2005-July/018955.html
http://www.grass.itc.it/pipermail/grass5/2005-July/018969.html

IMHO that's why 6.1-CVS source tree compiles much better now. Also valid
for the 6.1-CVS snapshot from last Saturday created in the European
morning time. Maybe you got your tarball before that.

To backport this to GRASS 6.0/5.x is too compilated (for me).
Let me recommend to use GRASS 6 instead of GRASS 5.

Markus

7

Miha Staut wrote:

I tried to do as you said. So

make CFLAGS_WARNING= 2> std.error.txt

It seemed to me that no less warnings appeared (130 KB of text in all). I am
therefore sending the std. error output as an attachment. I am reminding again
that predominantly imagery commands had problems (according to error.log) while
for the others v.surf.rst, r.grow2, r.in.gdal (normal), r.param.scale,
r.resamp.rst, r.surf.gauss, r.surf.random, s.in.ascii, s.surf.rst,
r.hydro.CASC2D, r.surf.fractal, r.terraflow (normal).

The first actual error is:

numerical.h:47: error: array type has incomplete element type

This has been fixed recently in 6.x.

That error causes lib/gmath to fail to compile, which in turn causes
the "/usr/bin/ld: cannot find -lgmath" errors (30 in total) as well as
the "No rule to make target .../libgmath.a" errors (5 in total).

Therea are also two other "incomplete element type" errors:

  In file included from matrix.c:3:
  globals.h:12: error: array type has incomplete element type

  In file included from v.out.sdts.c:29:
  externs.h:29: error: array type has incomplete element type

I can't locate the first, as the stderr output doesn't have the
"entering directory ..." messages.

The second is clearly from v.out.sdts; that can be fixed by moving the
definitions of the SDTS_module and SDTS_ddr_elems structures up a few
lines, to before the declarations which use them.

The relevant file is src.contrib/SDTS/mapdev/v.out.sdts/externs.h.

--
Glynn Clements <glynn@gclements.plus.com>

8

On Wed, Jul 20, 2005 at 07:47:42AM +0100, Glynn Clements wrote:

Miha Staut wrote:

> I tried to do as you said. So
>
> make CFLAGS_WARNING= 2> std.error.txt
>
> It seemed to me that no less warnings appeared (130 KB of text in all). I am
> therefore sending the std. error output as an attachment. I am reminding again
> that predominantly imagery commands had problems (according to error.log) while
> for the others v.surf.rst, r.grow2, r.in.gdal (normal), r.param.scale,
> r.resamp.rst, r.surf.gauss, r.surf.random, s.in.ascii, s.surf.rst,
> r.hydro.CASC2D, r.surf.fractal, r.terraflow (normal).

The first actual error is:

numerical.h:47: error: array type has incomplete element type

This has been fixed recently in 6.x.

I have fixed that in 5.-CVS-HEAD.

Paul, can you please update the 5.4.1-release branch (also r.flow)?

Markus