[Paraview-developers] [EXTERNAL] Re: Master is broken - kwProcessXML

Thu Apr 4 14:01:45 EDT 2013

Sorry web should be enable if pvpython is getting build, as web only add a
C++ module and a set of Python modules, nothing more.

To turn it off, you should go to the advanced mode. But as Berk point it
out, it might not be related at all.

You may want to go back to the previous master just before the wrapping
change gets in.

sha 7a15a62

And make sure the VTK submodule is in synch.

Seb

On Thu, Apr 4, 2013 at 1:42 PM, Scott, W Alan <wascott at sandia.gov> wrote:

>  OK, not sure if we are going down the correct rabbit hole, but …****
>
> ** **
>
> Why should enable web be on when python wrapping is on?  If I have a
> pvserver and pvbatch only build, why do I want web?  Um…..****
>
> ** **
>
> Second question, I never enabled ParaViewWeb.  So, specifically, how do I
> turn it off?****
>
> ** **
>
> Third – most plugins now don’t build.  Here are some other examples:****
>
> ** **
>
> [  4%] Built target protobuf****
>
> [  4%] Built target paraview_pyc****
>
> make[2]: *** No rule to make target
> `/projects/viz/paraview/src/ParaView/3.98.2-alan/ParaView3/Plugins/AnalyzeNIfTIReaderWriter/kwProcessXML',
> needed by `Plugins/AnalyzeNIfTIReaderWriter/vtkSMXML_AnalyzeNIfTIIO.h'.
> Stop.****
>
> make[1]: ***
> [Plugins/AnalyzeNIfTIReaderWriter/CMakeFiles/AnalyzeNIfTIIO.dir/all] Error 2
> ****
>
> make[1]: *** Waiting for unfinished jobs....****
>
> make[2]: *** No rule to make target
> `/projects/viz/paraview/src/ParaView/3.98.2-alan/ParaView3/Plugins/ArrowGlyph/kwProcessXML',
> needed by `Plugins/ArrowGlyph/vtkSMXML_ArrowGlyph.h'.  Stop.****
>
> make[1]: *** [Plugins/ArrowGlyph/CMakeFiles/ArrowGlyph.dir/all] Error 2***
> *
>
> make[2]: *** No rule to make target
> `/projects/viz/paraview/src/ParaView/3.98.2-alan/ParaView3/Plugins/GMVReader/kwProcessXML',
> needed by `Plugins/GMVReader/vtkSMXML_GMVReader.h'.  Stop.****
>
> make[1]: *** [Plugins/GMVReader/CMakeFiles/GMVReader.dir/all] Error 2****
>
> [  5%] Built target protoc_compiler****
>
> make: *** [all] Error 2****
>
> ** **
>
> Confused,****
>
> ** **
>
> Alan****
>
> ** **
>
> *From:* Sebastien Jourdain [mailto:sebastien.jourdain at kitware.com]
> *Sent:* Thursday, April 04, 2013 11:16 AM
>
> *To:* Scott, W Alan
> *Cc:* Bauer, Andrew Charles (External Contacts);
> paraview-developers at paraview.org
> *Subject:* Re: [EXTERNAL] Re: [Paraview-developers] Master is broken -
> kwProcessXML****
>
> ** **
>
> No it should be ON when the Python wrapping is ON...****
>
> This will allow cluster users to easily share their viz via ad-hoc web
> apps. ;-)****
>
> ** **
>
> Seb****
>
> ** **
>
> On Thu, Apr 4, 2013 at 1:08 PM, Scott, W Alan <wascott at sandia.gov> wrote:*
> ***
>
> I assume you mean PARAVIEW_ENABLE_WEB.  I just turned it off, and am
> trying.  Shouldn’t this be off by default if a client side gui build is
> disabled (i.e., a cluster build)?****
>
>  ****
>
> Alan****
>
>  ****
>
> *From:* Sebastien Jourdain [mailto:sebastien.jourdain at kitware.com]
> *Sent:* Thursday, April 04, 2013 11:06 AM
> *To:* Scott, W Alan
> *Cc:* Bauer, Andrew Charles (External Contacts);
> paraview-developers at paraview.org
> *Subject:* [EXTERNAL] Re: [Paraview-developers] Master is broken -
> kwProcessXML****
>
>  ****
>
> Hi Alan,****
>
>  ****
>
> can you disable the ParaViewWeb module for now ?****
>
>  ****
>
> Thanks,****
>
>  ****
>
> Seb****
>
>  ****
>
> On Thu, Apr 4, 2013 at 1:02 PM, Scott, W Alan <wascott at sandia.gov> wrote:*
> ***
>
> Next failed compile, after I hack CMake Lists to turn Catalyst off, is as
> follows:****
>
>  ****
>
> [ 83%] Built target vtkParaViewWebCore****
>
> make[2]: *** No rule to make target
> `/projects/viz/paraview/src/ParaView/3.98.2-alan/ParaView3/Plugins/AnalyzeNIfTIReaderWriter/kwProcessXML',
> needed by `Plugins/AnalyzeNIfTIReaderWriter/vtkSMXML_AnalyzeNIfTIIO.h'.
> Stop.****
>
> make[1]: ***
> [Plugins/AnalyzeNIfTIReaderWriter/CMakeFiles/AnalyzeNIfTIIO.dir/all] Error 2
> ****
>
> make[1]: *** Waiting for unfinished jobs....****
>
>  ****
>
> Any ideas?****
>
>  ****
>
> Thanks,****
>
>  ****
>
> Alan****
>
>  ****
>
> *From:* paraview-developers-bounces at paraview.org [mailto:
> paraview-developers-bounces at paraview.org] *On Behalf Of *Scott, W Alan
> *Sent:* Thursday, April 04, 2013 9:50 AM
> *To:* Bauer, Andrew Charles (External Contacts)
> *Cc:* paraview-developers at paraview.org
> *Subject:* Re: [Paraview-developers] [EXTERNAL] Re: Master is broken
> (trying to build and link Fortran)****
>
>  ****
>
> Thanks Andy!  ****
>
>  ****
>
> Mind e-mailing me a patch?  ****
>
>  ****
>
> Alan****
>
>  ****
>
> *From:* Andy Bauer [mailto:andy.bauer at kitware.com]
> *Sent:* Thursday, April 04, 2013 9:01 AM
> *To:* Scott, W Alan
> *Cc:* paraview-developers at paraview.org
> *Subject:* Re: [EXTERNAL] Re: [Paraview-developers] Master is broken
> (trying to build and link Fortran)****
>
>  ****
>
> Hi Alan,****
>
> I have a fix for this in review. It should make it through the gatekeeper
> review tomorrow. Thanks for the information.
>
> Andy****
>
>  ****
>
> On Wed, Apr 3, 2013 at 12:31 PM, Scott, W Alan <wascott at sandia.gov> wrote:
> ****
>
> You are correct with regards to the complaint about MPI fortran libraries.
>  However, it kills my build - thus is more than just an annoyance.
>
> Alan
>
> From: Andy Bauer [mailto:andy.bauer at kitware.com]
> Sent: Tuesday, April 02, 2013 8:32 PM
> To: Scott, W Alan
> Cc: paraview-developers at paraview.org
> Subject: [EXTERNAL] Re: [Paraview-developers] Master is broken (trying to
> build and link Fortran)****
>
>
> Hi Alan,
>
> Thanks for the report. Technically it's complaining about not finding the
> MPI fortran libraries. Either way it's an annoyance that the user shouldn't
> have to worry about. I'll take care of it.
>
> Thanks,
> Andy****
>
> On Tue, Apr 2, 2013 at 6:05 PM, Scott, W Alan <wascott at sandia.gov<mailto:
> wascott at sandia.gov>> wrote:
> Master appears to be broken for an MPI cluster build of ParaView.  When I
> do a fresh pull of Master, we are trying to compile and link against
> Fortran.  I am seeing the following:
>
>
> -- The C compiler identification is GNU 4.1.2
> -- The CXX compiler identification is GNU 4.1.2
> -- Could not use git to determine source version, using version 3.98.1
> -- The Fortran compiler identification is Intel****
>
> -- Checking whether /..../bin/intel64/ifort supports Fortran 90
> -- Checking whether /..../bin/intel64/ifort supports Fortran 90 -- yes
>
> ..........****
>
>
> -- Checking for C type size macros
> -- Checking for C type size macros - compiled
> -- Could NOT find OpenGL (missing:  OPENGL_gl_LIBRARY)
> -- Could NOT find OpenGL (missing:  OPENGL_gl_LIBRARY)
> VTK will be built with OSMesa support!
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   Utilities/VisItBridge/Library/VisItLib/CMake/VisItCommon.cmake:479
> (include)
>   Utilities/VisItBridge/Library/VisItLib/CMakeLists.txt:11 (include)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> -- Could NOT find OpenGL (missing:  OPENGL_gl_LIBRARY)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/ThirdParty/Cosmo/CMakeLists.txt:16 (find_package)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/Parallel/MPI/CMakeLists.txt:1 (find_package)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/IO/MPIImage/CMakeLists.txt:1 (find_package)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/IO/ParallelNetCDF/CMakeLists.txt:1 (find_package)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/ThirdParty/VPIC/CMakeLists.txt:28 (find_package)
>
>
> -- Could NOT find MPI_Fortran (missing:  MPI_Fortran_LIBRARIES)
> CMake Warning at VTK/CMake/FindMPI.cmake:383 (message):
>   Unable to find MPI library imf
> Call Stack (most recent call first):
>   VTK/CMake/FindMPI.cmake:569 (interrogate_mpi_compiler)
>   VTK/IO/VPIC/CMakeLists.txt:2 (find_package)
>
>
>
> However, when I  open ./CMakeLists.txt, line 301, and change the
> definition of PARAVIEW_ENABLE_CATALYST from ON to OFF, all builds well.
>  So, the gross error is the following line:
> option(PARAVIEW_ENABLE_CATALYST "Enable Catalyst CoProcessing modules" ON)
>
> CMakeCache.txt available upon request.
>
> Thanks for looking at this,
>
> Alan
>
>
>
>
> _______________________________________________
> Paraview-developers mailing list****
>
> Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org>
> http://public.kitware.com/mailman/listinfo/paraview-developers****
>
>  ****
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers****
>
>  ****
>
> ** **
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20130404/75b9bc09/attachment-0001.htm>