[vtk-developers] How do I contribute a reader for ParaView / VTK?

Tue Sep 30 01:47:50 EDT 2008

Hi Doug,

First of all, nice to get response from a developer who is concerned
with the OpenFOAM reader. However I have already sent the code and
testcases to Berk yesterday... from now on, to which developer should
I talk to get my modifications checked into ParaView and VTK?

I uploaded what I sent to Berk to the following URLs so that you can
check:

The reader code: http://oshima.eng.niigata-u.ac.jp/OpenFOAM/ParaView/vtkPOpenFOAMReader-VTKcontrib-r48.tar.gz
A parallel decomposed testcase: http://oshima.eng.niigata-u.ac.jp/OpenFOAM/ParaView/OpenFOAMParallelDecomposedCase.tar.gz
A serial testcase: http://oshima.eng.niigata-u.ac.jp/OpenFOAM/ParaView/OpenFOAMSerialCase.tar.gz

A quick installation instruction (INSTALL) is included in the reader
package.

I have been rather interested not only in bugfixes but mainly in
feature enhancements, since the original reader still lacks several
features required to be practically used for, say, 10 million
cells. Here is the list of main (but not all) features I have added to
Terry's original code.

* OpenFOAM 1.5 support. The original reader never works with OpenFOAM
  1.5, the newest vewsion of OpenFOAM, due to the mesh format change
  and extended FormFile directives ($-expansion and
  #include/#inputMode directives, see section 4.2.9 of OpenFOAM user
  guide
  http://www.opencfd.co.uk/openfoam/doc/basicFileFormat.html#x18-1020004.2.9).
  (Testcase: OpenFOAMSerialCase.tar.gz mentioned above)

* Gzip-compressed format support. (Testcase: OpenFOAMSerialCase.tar.gz)

* The reader can read parallel decomposed cases (via a new class
  vtkPOpenFOAMReader). When ParaView is run in server/client mode, it
  can read each decomposed region in parallel. (Testcase:
  OpenFOAMParallelDecomposedCase.tar.gz)

* Multi-region mesh support. (Testcase: icoStructFoam available on the
  OpenFOAM Wiki http://openfoamwiki.net/index.php/Contrib_icoStructFoam)

* Performance enhancements: you can check by loading any two-million
  or larger meshes. The reader should work at least 5 to 10 times
  faster especially for ascii cases.

* Stability enhancements: the reader no longer crashes by typical
  invalid cases (premature end-of-file etc.).

* Better polyhedra handling. (Testcase:
  interDyMFoam/damBreakWithObstacle tutorial available in OF 1.5)

* pointField support. (can be checked by the existence of
  pointMotionUx field after loading icoDyMFoam/movingCone tutorial)

* Lagrangian particle support, especially multiregion lagrangian
  clouds introduced in OF 1.5. (Testcase:
  rhoTurbTwinParcelFoam/simplifiedSiwek tutorial available in OpenFOAM
  1.5)

* Token-based FoamFile parser: as you know OpenFOAM format is a
  free-form format similar to C codes. The original code can't fully
  handle the format since it depends on typical line-by-line structure
  that OpenFOAM writes. This has been a lot problematic for
  hand-written initial condition files. Solved by implementing a
  dedicated token-based parser that semantically parses contents of a
  file.

Again, what should I do to get my modifications to be checked in?

Takuya OSHIMA, Ph.D.
Faculty of Engineering, Niigata University
8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN

From: Doug McCorkle <mccdo at iastate.edu>
Subject: Re: [vtk-developers] How do I contribute a reader for ParaView / VTK?
Date: Mon, 29 Sep 2008 20:54:55 -0500

> Hello Takuya,
> 
> Terry and I still support the OpenFOAM reader. If you run into bugs or need info on what we wrote we would be glad to answer your questions. At this point it still does everything we need it to do so have not submitted any changes.
> 
> Doug
> 
> On Sep 26, 2008, at 8:05 PM, Takuya OSHIMA wrote:
> 
> > Hi Berk and Karthik,
> >
> > Thanks so much for the quick replies. There doesn't seem to be a
> > developer who is interested in vtkOpenFOAMReader as it remains
> > unmaintained for quite a while AFAIK (except coding style fixes), so
> > this time I'll be sending to Berk shortly, upon finishing some
> > remaining cleanups.
> >
> > Takuya OSHIMA, Ph.D.
> > Faculty of Engineering, Niigata University
> > 8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN
> >
> >
> > From: "Karthik Krishnan" <karthik.krishnan at kitware.com>
> > Subject: Re: [vtk-developers] How do I contribute a reader for ParaView / VTK?
> > Date: Fri, 26 Sep 2008 13:27:13 -0400
> >
> >> In addition, it would also be best if you could contribute tests
> >> exercising the reader. Untested code is generally considered broken.
> >>
> >> If you need help getting datasets needed for the tests across to the
> >> developer, that may be too large for emails to handle, you can use the
> >> following link:
> >>
> >>  http://www.kitware.com/cgi-bin/uploadfile.cgi
> >>
> >> Thanks
> >> --
> >> karthik
> >>
> >> On Fri, Sep 26, 2008 at 1:21 PM, Berk Geveci <berk.geveci at kitware.com> wrote:
> >>> Hi Takuya,
> >>>
> >>> Once you have the code is matching the guidelines, you can e-mail it
> >>> to me. There is really no official way to contribute code unless you
> >>> have cvs access. One way is to attach it to the bug tracker but issues
> >>> on the tracker get reviewed somewhat infrequently. The best way is to
> >>> find a developer that is interested in the code and have him or her
> >>> commit it for you. In time, that developer may recommend that you are
> >>> given write access to cvs.
> >>>
> >>> -berk
> >>>
> >>>
> >>> On Fri, Sep 26, 2008 at 11:23 AM, Takuya OSHIMA
> >>> <oshima at eng.niigata-u.ac.jp> wrote:
> >>>> Dear VTK and ParaView developers,
> >>>>
> >>>> I'd like to ask, how do I contribute a reader for ParaView/VTK? It's a
> >>>> modified version of vtkOpenFOAMReader originally written by Terry
> >>>> Jordan and includes performance/stability/compatibility improvements,
> >>>> compressed (gzipped) format support, support for the file format
> >>>> extension in the newest OpenFOAM version etc, etc, with the help of
> >>>> extensive testing by the OpenFOAM community.
> >>>>
> >>>> I have read the "How To Contribute" page on the VTK website
> >>>> (http://www.vtk.org/contribute.php ) and think I can accept all the
> >>>> copyright, commercialization, coding standards discussions, but could
> >>>> not find a tangible procedure.
> >>>>
> >>>> Thanks,
> >>>> Takuya
> >>>>
> >>>> Takuya OSHIMA, Ph.D.
> >>>> Faculty of Engineering, Niigata University
> >>>> 8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN
> >>>> _______________________________________________
> >>>> vtk-developers mailing list
> >>>> vtk-developers at vtk.org
> >>>> http://www.vtk.org/mailman/listinfo/vtk-developers
> >>>>
> >>> _______________________________________________
> >>> vtk-developers mailing list
> >>> vtk-developers at vtk.org
> >>> http://www.vtk.org/mailman/listinfo/vtk-developers
> >>>
> >>
> >>
> >>
> >> --Karthik Krishnan
> >> R&D Engineer,
> >> Kitware Inc.
> >> Ph: 518 371 3971 x119
> >> Fax: 518 371 3971
> > _______________________________________________
> > vtk-developers mailing list
> > vtk-developers at vtk.org
> > http://www.vtk.org/mailman/listinfo/vtk-developers
> 