[Paraview] Re: CVS Compile Error

Fabian Braennstroem f.braennstroem at gmx.de
Sun Oct 7 15:31:09 EDT 2007


Hi Clinton,

clinton at elemtech.com schrieb am 10/04/2007 09:07 PM:
> Its just been fixed in CVS.

Just tried it, now it stopped at around 60% when it started
to use and look for the 'Python.h' file (sorry, I forgot to
copy the error output). I am pretty sure, that I defined the
exact path for pythons executable, lib and include...!?
I'll try it on Monday with a new CVS... do you know, if a
3.3 binary is released the next week :-)
Fabian




> On Thursday 04 October 2007 4:34:13 pm Fabian Braennstroem wrote:
>> On the xeon dual core 64bit I get this error message:
>>
>> [ 86%] Built target QtTesting
>> [ 88%] Built target pqWidgets
>> [ 88%] Building CXX object
>> Qt/Widgets/Testing/CMakeFiles/pqWidgetsTest.dir/QTestApp.o
>> /home/gcae504/SCR1/Software/ParaView/ParaView3/Qt/Widgets/Testing/QTestApp.
>> cxx:
>>
>> In static member function `static void
>> QTestApp::messageHandler(QtMsgType,
>> const char*)':
>> /home/gcae504/SCR1/Software/ParaView/ParaView3/Qt/Widgets/Testing/QTestApp.
>> cxx:50:
>>
>> error: `stderr' undeclared (first use this function)
>> /home/gcae504/SCR1/Software/ParaView/ParaView3/Qt/Widgets/Testing/QTestApp.
>> cxx:50:
>>
>> error: (Each undeclared identifier is reported only once for
>> each function
>> it appears in.)
>> /home/gcae504/SCR1/Software/ParaView/ParaView3/Qt/Widgets/Testing/QTestApp.
>> cxx:50:
>>
>> error: `fprintf' undeclared (first use this function)
>> make[2]: ***
>> [Qt/Widgets/Testing/CMakeFiles/pqWidgetsTest.dir/QTestApp.o]
>> Error 1
>> make[1]: ***
>> [Qt/Widgets/Testing/CMakeFiles/pqWidgetsTest.dir/all] Error 2
>> make: *** [all] Error 2
>>
>>
>> I would say, that I used the same settings (for paraview and
>> qt 4.2.3) like I did for the 32bit machine... strange!?
>> Maybe, someone did have the same problem before and can help!?
>> Fabian
>>
>> Fabian Braennstroem schrieb am 10/03/2007 08:28 PM:
>>> It works now; thanks for your help!
>>> ... now, I'll see, if I get it compiled on a 64Bit...
>>>
>>> Fabian
>>>
>>> Fabian Braennstroem schrieb am 10/01/2007 08:58 PM:
>>>> clinton at elemtech.com schrieb am 10/01/2007 06:42 PM:
>>>>> Try configuring Qt with -prefix, not --prefix.
>>>>> Do a "make confclean" then configure, make, make install.
>>>> I will takes a while though ... thanks for your help!
>>>> Fabian
>>>>
>>>>> On Monday 01 October 2007 2:39:35 pm Fabian Braennstroem wrote:
>>>>>> clinton at elemtech.com schrieb am 10/01/2007 06:22 PM:
>>>>>>> In
>>>>>>> /home/fab/HOME/Dissertation/simulation_calc/qt4_2/mkspecs
>>>>>>> do you have a default symlink -> to linux-g++ or something?
>>>>>> It is a normal directory:
>>>>>> localhost~/HOME/Dissertation/simulation_calc/qt4_2/mkspecs$
>>>>>> cd linux-g++
>>>>>> localhost~/HOME/Dissertation/simulation_calc/qt4_2/mkspecs/linux-g++$
>>>>>> ls -rlt
>>>>>> total 20
>>>>>> -rw-------  1 fab users 4395 Sep 29 17:55 qplatformdefs.h
>>>>>> -rw-------  1 fab users  260 Sep 29 17:55 qmake.conf
>>>>>>
>>>>>>> There's something wrong with your Qt setup if qmake keeps complaining
>>>>>>> about this.  Did you simply do
>>>>>>> "./configure -prefix
>>>>>>> /home/fab/HOME/Dissertation/simulation_calc/qt4_2 && make && make
>>>>>>> install" ?
>>>>>> Yes, kind of
>>>>>> /configure
>>>>>> --prefix=/home/fab/HOME/Dissertation/simulation_calc/qt4_2
>>>>>> afterwards:
>>>>>> gmake
>>>>>> and
>>>>>> gmake install
>>>>>>
>>>>>> Thanks!
>>>>>> Fabian
>>>>>>
>>>>>>> On Monday 01 October 2007 2:15:27 pm Fabian Braennstroem wrote:
>>>>>>>> Hi Clinton,
>>>>>>>>
>>>>>>>> clinton at elemtech.com schrieb am 10/01/2007 03:25 PM:
>>>>>>>>> And if you run
>>>>>>>>> qmake -query QMAKE_MKSPECS
>>>>>>>>> what directory(ies) does it give you?  And that directory should
>>>>>>>>> contain a subdirectory "linux-g++"
>>>>>>>> Thanks for your help!
>>>>>>>> I get:
>>>>>>>>
>>>>>>>> localhost~/HOME/Dissertation/simulation_calc$
>>>>>>>> ./qt4_2/bin/qmake -query QMAKE_MKSPECS
>>>>>>>> /home/fab/HOME/Dissertation/simulation_calc/qt4_2/mkspecs
>>>>>>>>
>>>>>>>> localhost~/HOME/Dissertation/simulation_calc/qt4_2/mkspecs$ ls
>>>>>>>> total 428
>>>>>>>> ...
>>>>>>>> linux-g++
>>>>>>>> ...
>>>>>>>>
>>>>>>>> It seems to be all right!?
>>>>>>>> Fabian
>>>>>>>>
>>>>>>>>> Clint
>>>>>>>>>
>>>>>>>>> On Sunday 30 September 2007 8:48:04 am Fabian Braennstroem wrote:
>>>>>>>>>> Hi to both,
>>>>>>>>>>
>>>>>>>>>> Clinton Stimpson schrieb am 09/29/2007 10:13 PM:
>>>>>>>>>>> Is this your own build of Qt?  If so, did you "make
>>>>>>>>>> install" it?
>>>>>>>>>>
>>>>>>>>>>> To double check, if you do "echo $QMAKESPEC" that gives
>>>>>>>>>> you what you
>>>>>>>>>>
>>>>>>>>>>> set, right?
>>>>>>>>>>> What version of CMake?
>>>>>>>>>> I build qt 4.2.3 by my own, could not find any package for
>>>>>>>>>> scientific linux (redhat clone) 4.4 and I used an binary
>>>>>>>>>> cmake version 2.4.7.  Echoing QMAKESPEC gives:
>>>>>>>>>> node1~$ echo $QMAKESPEC
>>>>>>>>>> linux-g++
>>>>>>>>>>
>>>>>>>>>> Thanks for your help!
>>>>>>>>>> Fabian
>>>>>>>>>>
>>>>>>>>>>> Clint
>>>>>>>>>>>
>>>>>>>>>>> Fabian Braennstroem wrote:
>>>>>>>>>>>> Hi,
>>>>>>>>>>>> I just tried to compile paraview cvs on a scientific-linux
>>>>>>>>>>>> (redhat enterprise clone) using qt 4.2.
>>>>>>>>>>>> Unfortunately, it stopped with this error:
>>>>>>>>>>>>
>>>>>>>>>>>> [ 22%] Generating moc_QVTKWidget.cxx
>>>>>>>>>>>> /bin/sh: QMAKESPEC has not been set, so configuration cannot
>>>>>>>>>>>> be deduced.: command not found
>>>>>>>>>>>> make[2]: *** [VTK/GUISupport/Qt/moc_QVTKWidget.cxx] Error 127
>>>>>>>>>>>> make[1]: *** [VTK/GUISupport/Qt/CMakeFiles/QVTK.dir/all] Error 2
>>>>>>>>>>>> make: *** [all] Error 2
>>>>>>>>>>>> 1419.376 user, 160.518 system, 28:05.85 real
>>>>>>>>>>>>
>>>>>>>>>>>> So I set it with in my .tcshrc (setenv QMAKESPEC
>>>>>>>>>>>> 'linux-g++') and tried it in my .bashrc... but I get always
>>>>>>>>>>>> the same error message!?
>>>>>>>>>>>> Does anyone know, what kind of problem I got?
>>>>>>>>>>>> Greetings!
>>>>>>>>>>>> Fabian


More information about the ParaView mailing list