[GRASS5] grass cvs snapshot from 26. Feb.

GRASS GIS GRASS GIS developer
1

Hallo Markus,

ich muss dich leider wieder mal bitten, die angehaengte Nachricht
weiterzuleiten. Ich stelle es dir frei, die Nachricht an die Liste oder
an die entspr. Leute weiterzuleiten.
Viele Gruesse,

Andreas
-----

Hi 2 all,

i am currently not able to work on grass bugfixing,
but i try to follow the developers' list.
And i give up on the code merger for ipc and fifo
setup too, i did not work on it for weeks,
so it is not in sync with the CVS tree any more.
Will the sockets setup be default for the stable
version?

I yesterday downloaded the CVS snapshot and did a "mass compile"
on Linux, IRIX, FreeBSD/alpha and a new CYGWIN/Windows 2000 setup.
On Cygwin with some manual tweaking of the header file the
setup compiles without problems (still the yacc/bison problem
and missing "more", the X11 include directory must be
corrected because the tcl/tk setup has some headers that
are found by default, but in the wrong directory).
But it does not work on my machine. I have a Windows 2000
PIII machine with the latest CYGWIN and the latest
xfree86-4.0 installed.

The "make install" with the sockets setup dies with
some error for a missing create_sockets.sh script,
this happens on Linux, IRIX, FreeBSD and Cygwin, but obviously only
if ./configure --with-socket is used!

If i fire up the X11 server i can only log into grass
if i load explicitly bash in the xterm or in the rxvt.
If i just issue /usr/local/bin/grass5 i get the login screen,
but get thrown out of the grass session immediatly.
The last message is see is "cleaning up temporary files" or such.

If i issue "bash", next log into grass5 and do a "d.mon start=x0"
i get the following error message:
Usage: x0 [nlev]
No socket to connect to for monitor <x0>
Trying again
No socket to connect to for monitor <x0>
..

Any idea what to do?

If i recompile using ipc/cygipc setup, i have to edit
the header file again manually to get the XDRIVER to compile,
the install works, but if i start the ipc-daemon.exe, fire
up X11 server, log into grass and issue "d.mon start=x0" i
get the following error message:

using default visual which is TrueColor
Visual is read-only or using a private colormap
ncolors: 32768
allocating memory...
Graphics driver [x0] started
      0 [main] x0 1672 fixup_mmaps_after_fork: base address fails to match
requested address 0x45F0000
C:\cygwin\usr\local\grass5\driver\XDRIVER.exe: ***
recreate_mmaps_after_fork_failed

Please start graphics monitor <x0>.
Error - Graphics monitor <x0> not running!
Problem selecting x0. Will try once more
Please start graphics monitor <x0>.
Error - Graphics monitor <x0> not running!

This must be some internal cygipc error, i sadly am not able to
debug this.

Generally i must admit that the X11 server is still beta
quality and i think that the average windows user will not
be able to install the server.
I myself was not able to configure it to use a german keyboard,
backspace and delete do not work too, so it is a real pain to
work in the xterm.

And as far as i know the cygwish80 shell is linked against the
Windows API, so it does not display on X11, but in the Win32 Window.
I was able to tweak tcltkgrass in a way to start with cygwish80,
but some functionalty will obviously not available on Win32
(process management, cmp. proc kill in gui.tcl, xlsclients in
proc search_xdrivers in gui.tcl, xwininfo in gui.tcl).
It would be much easier to create a tcltkgrass that switches functions
on win32/unix and displays in the native Win32 windows.
This would require to write a new driver (e. g. a driver that
uses OpenGL and the GLUT toolkit). I still believe that an OpenGL
driver would be a nice solution as it would work on X11 and on Win32.
The OpenGL and the X11 concept are not that different conceptually.

The other tests in summary:

Linux, Red Hat 7.0, Kernel 2.2.17-7, i686, gcc 2.96:
fifo: compiles and works. confirmed.
ipc: compiles error free and works, confirmed.
sockets: compiles, d.mon start=x0 does not work.
Message:
"No socket to connect to for monitor <x0>
Problem selecting x0. Will try once more" (twice)

IRIX 6.5, IP22/R44k, gcc 2.95.2:
fifo: compiles and works OK, confirmed.
ipc: compiles OK, works, but is slow.
sockets: does not compile, message below:

./configure --bindir=/usr/people/andreas/bin \
--with-socket \
--enable-postgres=no \
--with-jpeg-includes=/usr/freeware/include \
--with-jpeg-libs=/usr/freeware/lib32 \
--with-tiff-includes=/usr/freeware/include \
--with-tiff-libs=/usr/freeware/lib32 \
--with-tcltk-includes=/usr/freeware/include \
--with-tcltk-libs=/usr/freeware/lib32 \
--with-opengl-includes=/usr/include/GL \
--with-opengl-libs=/usr/lib32

make install dies with error 1:
src/scripts/shells/create_sockets.sh is missing!

ERROR with XDRIVER:
gcc -L/usr/freeware/lib32 -L/usr/lib32 -o
/disk2/opt/grass.src/dist.mips-sgi-irix6.5/driver/XDRIVER OBJ.mips-sgi-irix6.5/SWITCHER.o
OBJ.mips-sgi-irix6.5/Box_abs.o OBJ.mips-sgi-irix6.5/Can_do.o OBJ.mips-sgi-irix6.5/Color.o
OBJ.mips-sgi-irix6.5/Cont_abs.o OBJ.mips-sgi-irix6.5/Clr_table.o
OBJ.mips-sgi-irix6.5/Draw_line.o OBJ.mips-sgi-irix6.5/Get_w_box.o
OBJ.mips-sgi-irix6.5/Get_w_line.o OBJ.mips-sgi-irix6.5/Get_w_pnt.o OBJ.mips-sgi-irix6.5/Graph_Clse.o
OBJ.mips-sgi-irix6.5/Graph_Set.o OBJ.mips-sgi-irix6.5/Panel.o
OBJ.mips-sgi-irix6.5/Plylne_abs.o OBJ.mips-sgi-irix6.5/Polygn_abs.o OBJ.mips-sgi-irix6.5/Returns.o
OBJ.mips-sgi-irix6.5/Serve_Xevent.o OBJ.mips-sgi-irix6.5/alloc.o
OBJ.mips-sgi-irix6.5/command_pend.o OBJ.mips-sgi-irix6.5/Raster.o
./lib/LIB.mips-sgi-irix6.5/driverlib.a /disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libD.a
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a -lSM -lICE
-lX11 -lm
ld32: WARNING 84: /disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libD.a
is not used for resolving any symbol.
ld32: WARNING 84: /usr/lib32/libSM.so is not used for resolving any symbol.
ld32: WARNING 84: /usr/lib32/libm.so is not used for resolving any symbol.
ld32: ERROR 33: Unresolved text symbol "__INTADDR__" -- 1st referenced by
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a(unix_socks.o).
        Use linker option -v to see when and which objects, archives and
dsos are loaded.
ld32: INFO 152: Output file removed because of error.
collect2: ld returned 2 exit status
make[1]: *** [/disk2/opt/grass.src/dist.mips-sgi-irix6.5/driver/XDRIVER]
Error 1
make[1]: Leaving directory
`/disk2/opt/grass.src/src/display/devices/XDRIVER/XDRIVER24'
make: *** [all] Error 2
#################################################################
/disk2/opt/grass.src/src/display/d.3d
  make -f OBJ.mips-sgi-irix6.5/make.rules
gcc -L/usr/freeware/lib32 -L/usr/lib32 -o
/disk2/opt/grass.src/dist.mips-sgi-irix6.5/etc/bin/cmd/d.3d OBJ.mips-sgi-irix6.5/Dmain.o
OBJ.mips-sgi-irix6.5/command.o OBJ.mips-sgi-irix6.5/threed.o OBJ.mips-sgi-irix6.5/transform.o
OBJ.mips-sgi-irix6.5/get_range.o OBJ.mips-sgi-irix6.5/clip.o
OBJ.mips-sgi-irix6.5/check_opt.o OBJ.mips-sgi-irix6.5/sigint.o OBJ.mips-sgi-irix6.5/get_row.o
OBJ.mips-sgi-irix6.5/draw_face.o
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libraster.a /disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a -lm -I/usr/include
-L/usr/lib -lz
ld32: WARNING 84:
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a is not used for resolving any symbol.
ld32: ERROR 33: Unresolved text symbol "__INTADDR__" -- 1st referenced by
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a(unix_socks.o).
        Use linker option -v to see when and which objects, archives and
dsos are loaded.
ld32: INFO 152: Output file removed because of error.
collect2: ld returned 2 exit status
make: *** [/disk2/opt/grass.src/dist.mips-sgi-irix6.5/etc/bin/cmd/d.3d]
Error 1

all d.* modules do not link!
many imagery, mapdev, paint, raster etc. modules do not link too.

With all setups (ipc/fifo/sockets):
Problem starting grass5:
"Sorry, I need to know a more specific terminal type than ""."
grass5 -tcltk works!
grass5 -text does not work.
export TERMINAL=ansi does not help.
r.fill.dir does not compile (needs f77/g77, f77 is always found, i think
that the detection order should be g77 first, f77 second.)

FreeBSD/alpha, 4.2-RELEASE, gcc 2.95.2:
fifo: compiles and works, confirmed.
ipc: compiles error free except for NVIZ, works OK, confirmed. Slow.
Problem with backing store?
sockets: compiles and installs ok, but does not work:
d.mon start=x0
Usage: x0 [nlev]
No socket to connect to for monitor <x0>.
Problem selecting x0. Will try once more
No socket to connect to for monitor <x0>.

Windows 2000 Professional, 5.00.2195, Cygwin 1.1.8, i686, gcc 2.95.2-6:
fifo: compiles ok, working not possible
ipc: compiles ok so far, does not work.
sockets: does not work.

(on each machine i did:
tar xzf grass_src, ./configure; make; make install for default fifo setup
then:
./README.xdriver fifo|ipc|socket ; make ; make install)

You are allowed to forward this to the developers list.
Please answer to the list and cc to the above address.

cu,

Andreas

--
Sent through GMX FreeMail - http://www.gmx.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 Andreas,

hi all bison/yacc friends,

[...]

On Cygwin with some manual tweaking of the header file the
setup compiles without problems (still the yacc/bison problem)

I have added a "bison" test borrowed from PHP4. It searches for
bison and replaces the name in the head file.

However, it doesn't seem to recognise "yacc" properly if bison is
not present (I don't fully understand the bison test).

On Linux no problem, but the CRAY complains (as usual):
checking for flex... lex
checking for yywrap in -ll... yes
checking for bison... no
checking for byacc... no
configure: warning: You will need bison/yacc as required for various GRASS
modules.

Could someone please check the test in configure.in so that yacc is
searched if bison is not there?

[...]

cu,

Andreas

Regards

Markus

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

3

Hi again,

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:
[...]

And as far as i know the cygwish80 shell is linked against the
Windows API, so it does not display on X11, but in the Win32 Window.
I was able to tweak tcltkgrass in a way to start with cygwish80,
but some functionalty will obviously not available on Win32
(process management, cmp. proc kill in gui.tcl, xlsclients in
proc search_xdrivers in gui.tcl, xwininfo in gui.tcl).
It would be much easier to create a tcltkgrass that switches functions
on win32/unix and displays in the native Win32 windows.
This would require to write a new driver (e. g. a driver that
uses OpenGL and the GLUT toolkit). I still believe that an OpenGL
driver would be a nice solution as it would work on X11 and on Win32.
The OpenGL and the X11 concept are not that different conceptually.

This seems to be the only realistic way to get GRASS working properly
on Windows and Unix with the same driver.
Can't we use the NVIZ window as it is openGL based? Of course this couldn't
done in a few minutes!

[...]

cu,

Andreas

Regards

Markus

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

4

Hi Andreas,
The /bin/configure-grass.sh script that I use is as follows:

LIBS="-L/usr/X11R6/lib -lGLw -lGLU -lGL -lXm -lXt -lX11 -lSM -lICE -lXext"
export LIBS
rm -i config.cache
./configure -with-ipc --prefix=/usr/local --exec-prefix=/usr/local
--with-x-libs=/usr/X11R6/lib --with-x-includes=/usr/X11R6/include --with-x

and yes, I edit the head file to make sure that everything is correct.

In particular, the "-g -O2" option does not work with the current gdb
so I remove the "-g" , fix the bison, and other configurations for
my pc.

John Huddleston

----- Original Message -----
From: "Andreas Lange" <A.C.Lange@gmx.net>
To: <neteler@geog.uni-hannover.de>
Sent: Friday, March 02, 2001 6:39 AM
Subject: [GRASS5] grass cvs snapshot from 26. Feb.

Hallo Markus,

ich muss dich leider wieder mal bitten, die angehaengte Nachricht
weiterzuleiten. Ich stelle es dir frei, die Nachricht an die Liste oder
an die entspr. Leute weiterzuleiten.
Viele Gruesse,

Andreas
-----

Hi 2 all,

i am currently not able to work on grass bugfixing,
but i try to follow the developers' list.
And i give up on the code merger for ipc and fifo
setup too, i did not work on it for weeks,
so it is not in sync with the CVS tree any more.
Will the sockets setup be default for the stable
version?

I yesterday downloaded the CVS snapshot and did a "mass compile"
on Linux, IRIX, FreeBSD/alpha and a new CYGWIN/Windows 2000 setup.
On Cygwin with some manual tweaking of the header file the
setup compiles without problems (still the yacc/bison problem
and missing "more", the X11 include directory must be
corrected because the tcl/tk setup has some headers that
are found by default, but in the wrong directory).
But it does not work on my machine. I have a Windows 2000
PIII machine with the latest CYGWIN and the latest
xfree86-4.0 installed.

The "make install" with the sockets setup dies with
some error for a missing create_sockets.sh script,
this happens on Linux, IRIX, FreeBSD and Cygwin, but obviously only
if ./configure --with-socket is used!

If i fire up the X11 server i can only log into grass
if i load explicitly bash in the xterm or in the rxvt.
If i just issue /usr/local/bin/grass5 i get the login screen,
but get thrown out of the grass session immediatly.
The last message is see is "cleaning up temporary files" or such.

If i issue "bash", next log into grass5 and do a "d.mon start=x0"
i get the following error message:
Usage: x0 [nlev]
No socket to connect to for monitor <x0>
Trying again
No socket to connect to for monitor <x0>
..

Any idea what to do?

If i recompile using ipc/cygipc setup, i have to edit
the header file again manually to get the XDRIVER to compile,
the install works, but if i start the ipc-daemon.exe, fire
up X11 server, log into grass and issue "d.mon start=x0" i
get the following error message:

using default visual which is TrueColor
Visual is read-only or using a private colormap
ncolors: 32768
allocating memory...
Graphics driver [x0] started
      0 [main] x0 1672 fixup_mmaps_after_fork: base address fails to match
requested address 0x45F0000
C:\cygwin\usr\local\grass5\driver\XDRIVER.exe: ***
recreate_mmaps_after_fork_failed

Please start graphics monitor <x0>.
Error - Graphics monitor <x0> not running!
Problem selecting x0. Will try once more
Please start graphics monitor <x0>.
Error - Graphics monitor <x0> not running!

This must be some internal cygipc error, i sadly am not able to
debug this.

Generally i must admit that the X11 server is still beta
quality and i think that the average windows user will not
be able to install the server.
I myself was not able to configure it to use a german keyboard,
backspace and delete do not work too, so it is a real pain to
work in the xterm.

And as far as i know the cygwish80 shell is linked against the
Windows API, so it does not display on X11, but in the Win32 Window.
I was able to tweak tcltkgrass in a way to start with cygwish80,
but some functionalty will obviously not available on Win32
(process management, cmp. proc kill in gui.tcl, xlsclients in
proc search_xdrivers in gui.tcl, xwininfo in gui.tcl).
It would be much easier to create a tcltkgrass that switches functions
on win32/unix and displays in the native Win32 windows.
This would require to write a new driver (e. g. a driver that
uses OpenGL and the GLUT toolkit). I still believe that an OpenGL
driver would be a nice solution as it would work on X11 and on Win32.
The OpenGL and the X11 concept are not that different conceptually.

The other tests in summary:

Linux, Red Hat 7.0, Kernel 2.2.17-7, i686, gcc 2.96:
fifo: compiles and works. confirmed.
ipc: compiles error free and works, confirmed.
sockets: compiles, d.mon start=x0 does not work.
Message:
"No socket to connect to for monitor <x0>
Problem selecting x0. Will try once more" (twice)

IRIX 6.5, IP22/R44k, gcc 2.95.2:
fifo: compiles and works OK, confirmed.
ipc: compiles OK, works, but is slow.
sockets: does not compile, message below:

./configure --bindir=/usr/people/andreas/bin \
--with-socket \
--enable-postgres=no \
--with-jpeg-includes=/usr/freeware/include \
--with-jpeg-libs=/usr/freeware/lib32 \
--with-tiff-includes=/usr/freeware/include \
--with-tiff-libs=/usr/freeware/lib32 \
--with-tcltk-includes=/usr/freeware/include \
--with-tcltk-libs=/usr/freeware/lib32 \
--with-opengl-includes=/usr/include/GL \
--with-opengl-libs=/usr/lib32

make install dies with error 1:
src/scripts/shells/create_sockets.sh is missing!

ERROR with XDRIVER:
gcc -L/usr/freeware/lib32 -L/usr/lib32 -o
/disk2/opt/grass.src/dist.mips-sgi-irix6.5/driver/XDRIVER

OBJ.mips-sgi-irix6.5/SWITCHER.o

OBJ.mips-sgi-irix6.5/Box_abs.o OBJ.mips-sgi-irix6.5/Can_do.o

OBJ.mips-sgi-irix6.5/Color.o

OBJ.mips-sgi-irix6.5/Cont_abs.o OBJ.mips-sgi-irix6.5/Clr_table.o
OBJ.mips-sgi-irix6.5/Draw_line.o OBJ.mips-sgi-irix6.5/Get_w_box.o
OBJ.mips-sgi-irix6.5/Get_w_line.o OBJ.mips-sgi-irix6.5/Get_w_pnt.o

OBJ.mips-sgi-irix6.5/Graph_Clse.o

OBJ.mips-sgi-irix6.5/Graph_Set.o OBJ.mips-sgi-irix6.5/Panel.o
OBJ.mips-sgi-irix6.5/Plylne_abs.o OBJ.mips-sgi-irix6.5/Polygn_abs.o

OBJ.mips-sgi-irix6.5/Returns.o

OBJ.mips-sgi-irix6.5/Serve_Xevent.o OBJ.mips-sgi-irix6.5/alloc.o
OBJ.mips-sgi-irix6.5/command_pend.o OBJ.mips-sgi-irix6.5/Raster.o
./lib/LIB.mips-sgi-irix6.5/driverlib.a

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libD.a

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a -lSM -lICE
-lX11 -lm
ld32: WARNING 84:

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libD.a

is not used for resolving any symbol.
ld32: WARNING 84: /usr/lib32/libSM.so is not used for resolving any

symbol.

ld32: WARNING 84: /usr/lib32/libm.so is not used for resolving any symbol.
ld32: ERROR 33: Unresolved text symbol "__INTADDR__" -- 1st referenced by

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a(unix_socks.o).

        Use linker option -v to see when and which objects, archives and
dsos are loaded.
ld32: INFO 152: Output file removed because of error.
collect2: ld returned 2 exit status
make[1]: *** [/disk2/opt/grass.src/dist.mips-sgi-irix6.5/driver/XDRIVER]
Error 1
make[1]: Leaving directory
`/disk2/opt/grass.src/src/display/devices/XDRIVER/XDRIVER24'
make: *** [all] Error 2
#################################################################
/disk2/opt/grass.src/src/display/d.3d
  make -f OBJ.mips-sgi-irix6.5/make.rules
gcc -L/usr/freeware/lib32 -L/usr/lib32 -o
/disk2/opt/grass.src/dist.mips-sgi-irix6.5/etc/bin/cmd/d.3d

OBJ.mips-sgi-irix6.5/Dmain.o

OBJ.mips-sgi-irix6.5/command.o OBJ.mips-sgi-irix6.5/threed.o

OBJ.mips-sgi-irix6.5/transform.o

OBJ.mips-sgi-irix6.5/get_range.o OBJ.mips-sgi-irix6.5/clip.o
OBJ.mips-sgi-irix6.5/check_opt.o OBJ.mips-sgi-irix6.5/sigint.o

OBJ.mips-sgi-irix6.5/get_row.o

OBJ.mips-sgi-irix6.5/draw_face.o
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libraster.a

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a

disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a -lm -I/usr/
include

-L/usr/lib -lz
ld32: WARNING 84:
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libdisplay.a is not

used for resolving any symbol.

ld32: ERROR 33: Unresolved text symbol "__INTADDR__" -- 1st referenced by

/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a(unix_socks.o).

        Use linker option -v to see when and which objects, archives and
dsos are loaded.
ld32: INFO 152: Output file removed because of error.
collect2: ld returned 2 exit status
make: *** [/disk2/opt/grass.src/dist.mips-sgi-irix6.5/etc/bin/cmd/d.3d]
Error 1

all d.* modules do not link!
many imagery, mapdev, paint, raster etc. modules do not link too.

With all setups (ipc/fifo/sockets):
Problem starting grass5:
"Sorry, I need to know a more specific terminal type than ""."
grass5 -tcltk works!
grass5 -text does not work.
export TERMINAL=ansi does not help.
r.fill.dir does not compile (needs f77/g77, f77 is always found, i think
that the detection order should be g77 first, f77 second.)

FreeBSD/alpha, 4.2-RELEASE, gcc 2.95.2:
fifo: compiles and works, confirmed.
ipc: compiles error free except for NVIZ, works OK, confirmed. Slow.
Problem with backing store?
sockets: compiles and installs ok, but does not work:
d.mon start=x0
Usage: x0 [nlev]
No socket to connect to for monitor <x0>.
Problem selecting x0. Will try once more
No socket to connect to for monitor <x0>.

Windows 2000 Professional, 5.00.2195, Cygwin 1.1.8, i686, gcc 2.95.2-6:
fifo: compiles ok, working not possible
ipc: compiles ok so far, does not work.
sockets: does not work.

(on each machine i did:
tar xzf grass_src, ./configure; make; make install for default fifo setup
then:
./README.xdriver fifo|ipc|socket ; make ; make install)

You are allowed to forward this to the developers list.
Please answer to the list and cc to the above address.

cu,

Andreas

--
Sent through GMX FreeMail - http://www.gmx.net

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write

to:

minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

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

5

Hi all,

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:
[...]

r.fill.dir does not compile (needs f77/g77, f77 is always found, i think
that the detection order should be g77 first, f77 second.)

[...]

I have reordered this in configure[.in] (was requested by Justin as
well some time ago).

Hope it's ok,

Markus

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

6

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:

Will the sockets setup be default for the stable
version?

???

The "make install" with the sockets setup dies with
some error for a missing create_sockets.sh script,
this happens on Linux, IRIX, FreeBSD and Cygwin, but obviously only
if ./configure --with-socket is used!

I don't know what this create_sockets.sh is about. The sockets are
created as needed in the current mapset's .tmp directory. Just comment
out the line in Makefile.in that looks like:

test ${XDRIVER} != socket && ${SHELL} -c "${SHELL}
${SRCDIR}/src/scripts/shells/create_${XDRIVER}s.sh ${INST_DIR}
2>/dev/null ; true"

I have no idea what the purpose of that is supposed to be, but the
"test" should be protected in an "if test ..." so it doesn't kill
the make when ${XDRIVER} == socket.

If i issue "bash", next log into grass5 and do a "d.mon start=x0"
i get the following error message:
Usage: x0 [nlev]
No socket to connect to for monitor <x0>
Trying again
No socket to connect to for monitor <x0>
..

This problem can occur if there's a mismatch with the communications
method. At the moment, if you do --with-socket for configure, and later
do, --with-fifo or --with-ipc, you have to make sure at least the d.mon
object files are deleted before continuing with "make". Make can't know
it should recompile some of the d.mon sources as there's a USE_G_SOCKS
define that triggers conditional compilation in d.mon sources when
--with-socket is specified. This conditional compile was necessary, as
the sockets drivers take at most two arguments (argv[0] and argv[1]),
where argv[0] is the name the driver was invoked as (and will be the
name used in creating the socket file) and argv[1] is the optional nlev
argument -- No paths to "fifos" are needed.

The other tests in summary:

Linux, Red Hat 7.0, Kernel 2.2.17-7, i686, gcc 2.96:
fifo: compiles and works. confirmed.
ipc: compiles error free and works, confirmed.
sockets: compiles, d.mon start=x0 does not work.
Message:
"No socket to connect to for monitor <x0>
Problem selecting x0. Will try once more" (twice)

See above about cleaning d.mon and rebuilding it.

ld32: ERROR 33: Unresolved text symbol "__INTADDR__" -- 1st referenced by
/disk2/opt/grass.src/src/libes/LIB.mips-sgi-irix6.5/libgis.a(unix_socks.o).
        Use linker option -v to see when and which objects, archives and
dsos are loaded.

Bob Covill mentioned something about needing a -lsocket linker option on
a Solaris build. Maybe the same issue? Guess we need another configure
check...

FreeBSD/alpha, 4.2-RELEASE, gcc 2.95.2:
fifo: compiles and works, confirmed.
ipc: compiles error free except for NVIZ, works OK, confirmed. Slow.
Problem with backing store?
sockets: compiles and installs ok, but does not work:
d.mon start=x0
Usage: x0 [nlev]
No socket to connect to for monitor <x0>.
Problem selecting x0. Will try once more
No socket to connect to for monitor <x0>.

See above about cleaning d.mon sources first.

Windows 2000 Professional, 5.00.2195, Cygwin 1.1.8, i686, gcc 2.95.2-6:
fifo: compiles ok, working not possible
ipc: compiles ok so far, does not work.
sockets: does not work.

(on each machine i did:
tar xzf grass_src, ./configure; make; make install for default fifo setup
then:
./README.xdriver fifo|ipc|socket ; make ; make install)

Don't do this for sockets. You must do ./configure --with-socket.
Otherwise a symbol needed for d.mon conditional compile will not be
defined. And, as already mentioned, d.mon sources need to be cleaned of
object files.

Yes, it's all pretty messy right now...

--
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

The grad student here has installed the cvs snapshot from Feb 26
on LINUX with Redhat 7 with no problems except for setting-up the
correct links to Mesa for his graphics card to get nviz2.2 running
(if there is an interest, I can ask him to write to the list what he had to do,
it was related to the graphics card, not GRASS.)
We had previously problems with fifos, no problems now.

I tried most of the d.* commands and in general they work well, except for:

1. GRASS:~ > d.dm help
File /home/helena/grassdata/helena/.d.dmrc doesn't exist.
There is no manual page so I could not tell whether this is a bug
or I am missing something

---------------
2. monitor does not close on mouse click: this is not a problem, I just thought
that I have seen in this list that it does. but I was running it through the net
on Sun,
so that may be the problem.
---------------
3. d.text : Please enter text instructions. Enter EOF on last line to quit -
does anybody know what the EOF means here? (I hope that this is not too stupid
question)
---------------
4. manual says: d.leg.thin [-nsm] map=name [color=name] [lines=value]
[thin=value]
when I run help
d.leg.thin map=name [color=name] [lines=value]

it does not recognize ms flags and it does nothing, maybe just manual needs to
be updated
-----------------------
5. not included:
xganim (I miss this one!!!!) - should I get lesstif and compile separately?
d.param.scale (would be really useful!), but I have never seen it included in a
release
------------------------------------------------------

thanks for all the great work - the installation is now fast and simple
compared with what one
had to go through in previous releases and there is so much great new stuff!

Helena

P.S. nviz works fine (no masked out areas, only lights are still a little bit
weird, not a big
problem though)

Eric G. Miller" wrote:

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:
> Will the sockets setup be default for the stable
> version?

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

8

Helena wrote:

1. GRASS:~ > d.dm help
File /home/helena/grassdata/helena/.d.dmrc doesn't exist.

It only says that d.dm is started fist time in current mapset
or last d.dm session was not saved.
So, should be correct if your grassdata/ is location and grassdata/helena/
is current mapset and .d.dmrc really doesn't exist and d.dm window appeared
on screen with set0 in set frame.

help option for d.dm doesn't exist (and I am not going to add it now).

There is no manual page so I could not tell whether this is a bug
or I am missing something

I will write some short manual.

---------------
4. manual says: d.leg.thin [-nsm] map=name [color=name] [lines=value]
[thin=value]
when I run help
d.leg.thin map=name [color=name] [lines=value]

it does not recognize ms flags and it does nothing, maybe just manual needs to
be updated

In d.leg.thin is used main.c (which is code for d.legend) but new code
for d.leg.thin is in main_new.c and box.c. New code works for all
options but only for non FP rasters. I'll try to fix it.

Question: for FP rasters is not number of categories done, so I want
to use some default value for lines= option, which would specify (for FP only)
number of legend rows. Or should be legend for FP always -s (smooth
gradient)?

Thanks for your tests.

Radim

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

9

Hi Markus, Hi all,

thank you very much for the all the useful answers.

I can now compile and use the cvs snapshot on linux, freeBSD and Cygwin.
You have to either do a ./configure --with-socket on a clean source tree
_or_ a
make distclean; ./configure --with-socket; make; make install
and ignore the error with install.
The sockets setup works ok. I can now even fire up X11 on Cygwin, issue bash
and login into grass and start a grass monitor. But this bogs down the
machine extremely and it is very unusual to work in a full-window X11 on Windows
2000.

I will have a look on the IRIX problem another time. Will try the -lsocket
tip first.

But i would ask to fix the install script before making sockets the default
:-). And the socket option to ./README.xdriver should be disabled or echo an
information message that it does not work.

The sockets setup is really great work.

Andreas

Hi all,

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:
[...]
> r.fill.dir does not compile (needs f77/g77, f77 is always found, i think

> that the detection order should be g77 first, f77 second.)
[...]

I have reordered this in configure[.in] (was requested by Justin as
well some time ago).

Hope it's ok,

Markus

--
Sent through GMX FreeMail - http://www.gmx.net

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

10

Hi Markus

Markus Neteler wrote:

Hi Andreas,

hi all bison/yacc friends,

I have added a "bison" test borrowed from PHP4. It searches for
bison and replaces the name in the head file.

However, it doesn't seem to recognise "yacc" properly if bison is
not present (I don't fully understand the bison test).

On Linux no problem, but the CRAY complains (as usual):
checking for flex... lex
checking for yywrap in -ll... yes
checking for bison... no
checking for byacc... no

               ^
Is the "b" correct, or should it be looking for just "yacc"?

Just a wild guess without looking at any code.

--
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'

11

On Mon, Mar 05, 2001 at 11:53:36AM +0700, Justin Hickey wrote:

Hi Markus

Markus Neteler wrote:
>
> Hi Andreas,
>
> hi all bison/yacc friends,
>
> I have added a "bison" test borrowed from PHP4. It searches for
> bison and replaces the name in the head file.
>
> However, it doesn't seem to recognise "yacc" properly if bison is
> not present (I don't fully understand the bison test).
>
> On Linux no problem, but the CRAY complains (as usual):
> checking for flex... lex
> checking for yywrap in -ll... yes
> checking for bison... no
> checking for byacc... no
               ^
Is the "b" correct, or should it be looking for just "yacc"?

Just a wild guess without looking at any code.

Bingo! But I just changed in "configure"...

checking for bison... no
checking for yacc... yacc
configure: warning: You will need bison/yacc as required for various GRASS
modules.

However, how to tell configure.in about it?

It seems to be an autoconf bug (see configure):

  for ac_prog in 'bison -y' byacc

should be

  for ac_prog in 'bison -y' byacc yacc

How can we fix this?

Markus

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

12

On Sat, Mar 03, 2001 at 12:19:21PM -0800, Eric G. Miller wrote:

On Fri, Mar 02, 2001 at 02:39:35PM +0100, Andreas Lange wrote:
> Will the sockets setup be default for the stable
> version?
???

Hopefully today or tomorrow :slight_smile:

> The "make install" with the sockets setup dies with
> some error for a missing create_sockets.sh script,
> this happens on Linux, IRIX, FreeBSD and Cygwin, but obviously only
> if ./configure --with-socket is used!

I don't know what this create_sockets.sh is about. The sockets are
created as needed in the current mapset's .tmp directory.

[...]

Eric, I think this is again confusion: The
create_sockets.sh script should be named create_ipcbuffer.sh
or so.

[...]

Eric

Simply let's remove the fifos and IPC to clean up the situation.
Huidae and others: Your work on IPC was very important (as already
written by Eric) to stimulate the sockets. Please don't worry if
it is moved to the archives now (if all agree).

Markus

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

13

Helena,

On Sat, Mar 03, 2001 at 03:42:33PM -0600, Helena wrote:

The grad student here has installed the cvs snapshot from Feb 26
on LINUX with Redhat 7 with no problems except for setting-up the
correct links to Mesa for his graphics card to get nviz2.2 running
(if there is an interest, I can ask him to write to the list what he had to do,
it was related to the graphics card, not GRASS.)
We had previously problems with fifos, no problems now.

I tried most of the d.* commands and in general they work well, except for:

[...]

---------------
2. monitor does not close on mouse click: this is not a problem, I just thought
that I have seen in this list that it does. but I was running it through the net
on Sun, so that may be the problem.
From above it seems you compiled using the fifos. The monitor close on mouse

click is only supported with sockets:

configure --with-socket
make
make install

---------------
3. d.text : Please enter text instructions. Enter EOF on last line to quit -
does anybody know what the EOF means here? (I hope that this is not too stupid
question)

Try
Ctrl-D

(this is not very user convenient, should be changed!)

-----------------------
5. not included:
xganim (I miss this one!!!!) - should I get lesstif and compile separately?

Yes...

d.param.scale (would be really useful!), but I have never seen it included
in a release

Reason is that r/d.param.scale are using functions from "Numerical receipes
in C" which are copyright protected. If anyone would modify the routines to
use the "new" GRASS -gmath library (src/libes/gmath) which are based on
LAPACK/BLAS, we could put the code back fro distribution.

Both modules are in
src.nonGPL/raster/...
src.nonGPL/display/...

thanks for all the great work - the installation is now fast and simple
compared with what one had to go through in previous releases and there is
so much great new stuff!

Congrats to all! :slight_smile:

Helena

P.S. nviz works fine (no masked out areas, only lights are still a little
bit weird, not a big problem though)

Huh, happy to hear that! Kudos to Bob!

Markus

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

14

Hi Markus

Markus Neteler wrote:

However, how to tell configure.in about it?

It seems to be an autoconf bug (see configure):

  for ac_prog in 'bison -y' byacc

should be

  for ac_prog in 'bison -y' byacc yacc

How can we fix this?

I'm basically brain dead when it comes to using configure.in but
according to the texinfo file for autoconf we have the following:

- Macro: AC_PROG_YACC
     If `bison' is found, set output variable `YACC' to `bison -y'.
     Otherwise, if `byacc' is found, set `YACC' to `byacc'. Otherwise
     set `YACC' to `yacc'.

So I don't think there should be a problem. To eliminate the warning
message maybe we need to add a check for byacc in configure.in as
follows

AC_PROG_YACC
if test "$YACC" != "bison -y"; then
  if test "$YACC" != "byacc" && test "$YACC" != "yacc"; then
    AC_MSG_WARN(You will need bison/yacc as required for various GRASS
modules.)
  fi

The rest of the code in configure.in checks the version of bison. I
don't know, just guessing.

--
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'

15

On Mon, Mar 05, 2001 at 05:00:01PM +0700, Justin Hickey wrote:

Hi Markus

Markus Neteler wrote:
> However, how to tell configure.in about it?
>
> It seems to be an autoconf bug (see configure):
>
> for ac_prog in 'bison -y' byacc
>
> should be
>
> for ac_prog in 'bison -y' byacc yacc
>
> How can we fix this?

I'm basically brain dead when it comes to using configure.in but
according to the texinfo file for autoconf we have the following:

- Macro: AC_PROG_YACC
     If `bison' is found, set output variable `YACC' to `bison -y'.
     Otherwise, if `byacc' is found, set `YACC' to `byacc'. Otherwise
     set `YACC' to `yacc'.

Yes, I have seen that right now.

So I don't think there should be a problem. To eliminate the warning
message maybe we need to add a check for byacc in configure.in as
follows

AC_PROG_YACC
if test "$YACC" != "bison -y"; then
  if test "$YACC" != "byacc" && test "$YACC" != "yacc"; then
    AC_MSG_WARN(You will need bison/yacc as required for various GRASS
modules.)
  fi

The rest of the code in configure.in checks the version of bison. I
don't know, just guessing.

O.k, if you check the current version, it should run fine and print out
the yacc message as well (using simply "echo").

Linux, Solaris and CRAY are satisfied. :slight_smile:

However, do we need to keep the bison version check?
checking bison version... configure: warning: Bison 1.25 or newer needed to
regenerate parsers (found 1.22).
-> "my" SUN

Thanks for your assistance

Markus

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

16

Hi Markus

Markus Neteler wrote:

Linux, Solaris and CRAY are satisfied. :slight_smile:

I haven't tested IRIX but it didn't complain this morning with your
first version.

However, do we need to keep the bison version check?
checking bison version... configure: warning: Bison 1.25 or newer needed to
regenerate parsers (found 1.22).
-> "my" SUN

Sorry, no idea. I'm not sure what the problem was before version 1.25.

--
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'

17

On Mon, Mar 05, 2001 at 06:03:56PM +0700, Justin Hickey wrote:

Hi Markus

Markus Neteler wrote:
> Linux, Solaris and CRAY are satisfied. :slight_smile:

I haven't tested IRIX but it didn't complain this morning with your
first version.

> However, do we need to keep the bison version check?
> checking bison version... configure: warning: Bison 1.25 or newer needed to
> regenerate parsers (found 1.22).
> -> "my" SUN

Sorry, no idea. I'm not sure what the problem was before version 1.25.

This was taken over from PHP4.... nothing to do with GRASS.
I'll remove the check,

Markus

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

18

Hi,

before making sockets default, could someone help me to get the sockets
setup to compile on IRIX with gcc?

Adding -lsocket to XEXTRALIBS does not help to compile all modules, as the
d.* modules do not link the Xlibs. I can compile the XDRIVER24 now, but many
other modules still do not compile.
Adding -lsocket to all modules does not help too, there seems to be some
problem with the library order on the linker command line?

How was the problem solved with Solaris?
Please do not assume that everything is ok if you recieve no feedback within
a day ;-).
I vote for keeping the fifo setup as an option, as it worked ok for a long
time (and i have written a G2DRIVER that still uses fifos).
Are all the other drivers (HTMLMAP, PNGdriver, CELL) working with sockets?
Any hints on how to adapt them to sockets?

Thanks a lot in advance.

cu,

Andreas

--
Sent through GMX FreeMail - http://www.gmx.net

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

19

Hi Andreas

Andreas Lange wrote:

before making sockets default, could someone help me to get the
sockets setup to compile on IRIX with gcc?

Adding -lsocket to XEXTRALIBS does not help to compile all modules,
as the d.* modules do not link the Xlibs. I can compile the XDRIVER24
now, but many other modules still do not compile.

Strange. I don't need -lsocket at all for IRIX with MIPSPro compilers. I
didn't think different compilers would link differently. I thought
linking was more system dependent than compiler dependent. In fact, even
with gcc, don't you need the SGI C libraies installed?

Adding -lsocket to all modules does not help too, there seems to be
some problem with the library order on the linker command line?

Where did you add it? I would try adding it to the MATHLIB definition in
src/CMD/head/head.in. Of course, the "correct" way to do it is to add it
to each module that needs it. I suppose configure.in should set a make
variable and this variable be added to the compile lines of modules that
need socket code. A bit tedious but not that daunting a task. I'd do it
myself, but I don't have time this week.

How was the problem solved with Solaris?
Please do not assume that everything is ok if you recieve no feedback
within a day ;-).

But we had positive feedback for IRIX from me. We just didn't think a
different compiler would cause a problem. :frowning:

Taken from your earlier post

./configure --bindir=/usr/people/andreas/bin \
--with-socket \
--enable-postgres=no \
--with-jpeg-includes=/usr/freeware/include \
--with-jpeg-libs=/usr/freeware/lib32 \
--with-tiff-includes=/usr/freeware/include \
--with-tiff-libs=/usr/freeware/lib32 \
--with-tcltk-includes=/usr/freeware/include \
--with-tcltk-libs=/usr/freeware/lib32 \
--with-opengl-includes=/usr/include/GL \
--with-opengl-libs=/usr/lib32

Did you try running configure without specifying the freeware
directories? I changed configure.in to add the freeware directories if
the machine is an SGI. Simply running configure without options now
works for me.

Problem starting grass5:
"Sorry, I need to know a more specific terminal type than ""."
grass5 -tcltk works!
grass5 -text does not work.
export TERMINAL=ansi does not help.

Shouldn't that be TERM=ansi? At least that is the environment variable I
have on my machine.

I hope some of 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'

20

Hi,

i found the cause of the problem on IRIX.
(__INTADDR__ wrong reference).
See here for a description:
http://gcc.gnu.org/ml/gcc-bugs/1999-04/msg00157.html

I hope that i can fix it tomorrow. Will report success or failure.
So nothing to worry about for the sockets setup.

cu,

Andreas

--
Sent through GMX FreeMail - http://www.gmx.net

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