From dave.demarle at kitware.com Tue Mar 1 10:18:21 2016 From: dave.demarle at kitware.com (David E DeMarle) Date: Tue, 1 Mar 2016 10:18:21 -0500 Subject: [Paraview-developers] google summer of code 2016 - project ideas and mentor volunteers In-Reply-To: References: Message-ID: Hello, Unfortunately VTK did not make the cut for GSoC 2016. We will apply again next year, so if you come up with a great project ideas please do add it to the idea list wiki page. thanks David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Tue, Feb 9, 2016 at 11:02 AM, David E DeMarle wrote: > Hey gang, > > It is GSoC application time again and we need project ideas and mentors > for those projects. See http://www.vtk.org/Wiki/VTK/GSoC_2016 for the > idea list and do feel free to edit it directly if you like. > > The project ideas list is a key part of our application to have VTK taken > up and for students to be funded under GSoC. Since most students base their > proposals directly on ideas in it, it is also a big part of what we can > expect to get out of the summer. > > So, if you've got ideas for things that you would like to see > college/graduate students do with or to VTK over the summer, please add > them to the list. Ideally the ideas should have some sort of wow factor > (like Andy Bauer's proposal last year for motion amplification in VTK) in > order to entice students to apply to VTK over all the other organizations. > > Each project needs at least one mentor. Advisors should know VTK well and > be able to commit to spending something like two to six hours per week > advising a remote student. Keep in mind that the coaching and talent > development aspect is perhaps the most important goal of GSoC, so mentors > should be able to engage their students throughout and attempt to groom > them to become regular contributors. > > thanks > David E DeMarle > Kitware, Inc. > R&D Engineer > 21 Corporate Drive > Clifton Park, NY 12065-8662 > Phone: 518-881-4909 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cory.quammen at kitware.com Tue Mar 1 11:57:40 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Tue, 1 Mar 2016 10:57:40 -0600 Subject: [Paraview-developers] Customs Filters In-Reply-To: References: Message-ID: Orlando, It doesn't appear that PV_PLUGIN_PATH can be used to load custom filters. In principle, ParaView could be made to do this, but I do not know of any impending work that will make this possible. You would have to add the XML description of the custom filter to the ParaView?.?.?.ini file. This is a lot more complicated than navigating to the location of the custom filter and loading it, though. - Cory On Fri, Feb 19, 2016 at 2:51 AM, RIVERA ROLDAN, Jorge Orlando < Orlando.RIVERA at mtu.de> wrote: > Hi > > > > I have, hopefully a not so stupid question. > > > > > > A colleague of mine has saved 2 Custom filters, and packed them in one > single xml (cpd file), and I would like to make them available for all > users > > When I use the ?Manage custom plugins? windows it shows the CWD . > > > > 1) There is a way to load custom filters without User interaction , > like using $PV_PLUGIN_PATH ? > > and, or > > 2) ?Manage custom plugin? window will point to a specific path > > > > Thank you so much > > > > Regards > > > > Orlando Rivera > > > -- > *MTU Aero Engines AG* > Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer > Martens, Michael Schreyoegg > Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: Klaus > Eberhardt > Sitz der Gesellschaft/Registered Office: Muenchen > Handelsregister/Commercial Register: Muenchen HRB 157206 > > Diese E-Mail sowie ihre Anhaenge enthalten MTU-eigene vertrauliche oder > rechtlich geschuetzte Informationen. > Wenn Sie nicht der beabsichtigte Empfaenger sind, informieren Sie bitte > den Absender und loeschen Sie diese > E-Mail sowie die Anhaenge. Das unbefugte Speichern, Kopieren oder > Weiterleiten ist nicht gestattet. > > This e-mail and any attached documents are proprietary to MTU, > confidential or protected by law. > If you are not the intended recipient, please advise the sender and delete > this message and its attachments. > Any unauthorised storing, copying or distribution is prohibited. > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > > -- Cory Quammen R&D Engineer Kitware, Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.g.hennessey2.ctr at mail.mil Tue Mar 1 12:20:34 2016 From: joseph.g.hennessey2.ctr at mail.mil (Hennessey, Joseph G CTR USARMY RDECOM ARL (US)) Date: Tue, 1 Mar 2016 17:20:34 +0000 Subject: [Paraview-developers] Status of new Xdmf3 plugin for ParaView (UNCLASSIFIED) Message-ID: <10A03274360DCF47A6EE78C9952A31CA416F183A@umechpao4.easf.csd.disa.mil> CLASSIFICATION: UNCLASSIFIED Hello, I have been working on upgrading the support for Xdmf3 in ParaView. Towards that end I have integrated the latest version of Xdmf3 developed here at ARL into ParaView. Xdmf3 included a parallel reader and a non-parallel writer, I have upgraded this writer to make it parallel also. I have tested and built my new version of Xdmf3 and found that it works correctly when running in client server mode on a localhost. But, when I use the superbuild script to build it, something is not working and the parallel writer only works in non-parallel mode. I can understand why this might happen as until now the Xdmf3 writer was non-parallel capable and therefore there was no way to test this. What steps would you recommend that I take next, I could fix the superbuild process, but I want to ensure that my fixes would be integrated into ParaView as I would rather not maintain my own branch of ParaView and/or its superbuild scripts. Please let me know what you think. Thanks, Joe Hennessey ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Joseph G. Hennessey Ph.D., Lockheed Martin Army Research Lab CLASSIFICATION: UNCLASSIFIED -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5615 bytes Desc: not available URL: From utkarsh.ayachit at kitware.com Sat Mar 5 15:31:16 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Sat, 5 Mar 2016 15:31:16 -0500 Subject: [Paraview-developers] Customs Filters In-Reply-To: References: Message-ID: > 1) There is a way to load custom filters without User interaction , > like using $PV_PLUGIN_PATH ? You can "convert" a cpd file to plugin xml by minor modifications to the XML. Attached is an example. My example shows a source; for a filter, you'd put it under the "filters" group, instead of "sources". This plugin XML can then be loaded automatically using PV_PLUGIN_PATH. -------------- next part -------------- A non-text attachment was scrubbed... Name: mywavelet.xml Type: text/xml Size: 2848 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: mywavelet.cpd Type: application/octet-stream Size: 2887 bytes Desc: not available URL: From david.lonie at kitware.com Mon Mar 7 13:03:42 2016 From: david.lonie at kitware.com (David Lonie) Date: Mon, 7 Mar 2016 13:03:42 -0500 Subject: [Paraview-developers] New vtkDataArray implementations, tools Message-ID: Hi folks, As part of the ongoing effort to support in-situ toolkits such as Catalyst in VTK, we have made some changes to how vtkDataArrays are implemented and used in VTK. In a nutshell, the array-of-structs (AOS) memory layout should no longer be assumed when using vtkDataArrays, and there are new SOA (struct-of-arrays) vtkDataArray implementations included in VTK. This means that methods like GetVoidPointer, WritePointer, etc should no longer be used, and may enter a deprecation phase at some point in the future. This also means that vtkTemplateMacro is no longer the preferred way of interacting with array data efficiently, since it indirectly relies on GetVoidPointer. Existing code using vtkTemplateMacro will continue to work, though it will be very inefficient when used with some of the new array subclasses. As a replacement for vtkTemplateMacro, there are some new tools that are used to provide efficient, compile-time optimized access to array data. These tools and their usage are detailed here: http://www.vtk.org/Wiki/VTK/Tutorials/DataArrays (wiki version) https://gitlab.kitware.com/vtk/vtk/blob/master/Documentation/Doxygen/ArrayDispatch-VTK-7-1.md (raw doxygen markdown -- will replace wiki page eventually) The new vtkArrayDispatch tool allows optimized algorithms to be written that work on both vtkAOSDataArrayTemplate (the AOS replacement for vtkDataArrayTemplate) and vtkSOADataArrayTemplate (the new SOA data array backend). The vtkDataArrayAccessor wrapper will then allow the same implementation to fallback to using the generic vtkDataArray API in cases where template-explosion may be an issue. Also note that there is an API change involved with this patch. The typed vtkDataArrayTemplate API methods: void GetTupleValue(vtkIdType tupleIdx, ValueType *tuple) void SetTupleValue(vtkIdType tupleIdx, const ValueType *tuple) void InsertTupleValue(vtkIdType tupleIdx, const ValueType *tuple) vtkIdType InsertNextTupleValue(const ValueType *tuple) have been renamed to more clearly indicate their purpose. They are now "GetTypedTuple", etc, and there is a new set of corresponding "GetTypedComponent"-style methods. The old signatures are still available on vtkAOSDataArrayTemplate subclasses and have been marked deprecated. VTK and ParaView have both been updated to use the new APIs. The last iteration of in-situ data array interfaces, namely vtkTypedDataArray and vtkMappedDataArray, will be deprecated at some point due to performance concerns, but continue to be available for the time being. If you encounter any issues with the new implementations or have questions about them, please do not hesitate to email the vtk-developers list and let us know. Best, Dave -------------- next part -------------- An HTML attachment was scrubbed... URL: From m.schlottke-lakemper at aia.rwth-aachen.de Tue Mar 8 06:22:27 2016 From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael) Date: Tue, 8 Mar 2016 11:22:27 +0000 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters Message-ID: Hi everyone, We have a custom parallel reader plugin that works well with our files even on large-ish (100-1000) number of cores for the pvserver. Recently we?ve added the capability to also create ghost cells with this reader such that MPI domain boundaries become transparent to filters that need neighbor data, e.g., for gradient computations. However, this ghost cell generation takes up quite some time, so we would like to keep the number of MPI ranks at a minimum and use hybrid MPI-OpenMP parallelism for our plugin. My question is thus if it is possible to use OpenMP inside a custom plugin? And if yes, is there a canonical way to do it (maybe an example plugin) or is it going to be a hack? When searching the interwebs for ParaView and OpenMP, I only came up with an old VTK mailing list thread from 2014 (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), in which David (DeMarle) refers to some initial documentation of the SMP framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). However, the linked PDF does not contain references to OpenMP. Thanks a lot in advance! Michael -- Michael Schlottke-Lakemper Chair of Fluid Mechanics and Institute of Aerodynamics RWTH Aachen University W?llnerstra?e 5a 52062 Aachen Germany Phone: +49 (241) 80 95188 Fax: +49 (241) 80 92257 Mail: m.schlottke-lakemper at aia.rwth-aachen.de Web: http://www.aia.rwth-aachen.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From utkarsh.ayachit at kitware.com Tue Mar 8 08:39:26 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Tue, 8 Mar 2016 08:39:26 -0500 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters In-Reply-To: References: Message-ID: Michael, The VTK-SMP approach is indeed the way to go for this. To make the backend use OpenMP, you simply compile VTK/ParaView with VTK_SMP_IMPLEMENTATION_TYPE set to OpenMP. The programming model described in the SMP Guide PDF has been implemented for all supported backends including OpenMP and TBB. Utkarsh On Tue, Mar 8, 2016 at 6:22 AM, Schlottke-Lakemper, Michael wrote: > Hi everyone, > > We have a custom parallel reader plugin that works well with our files even > on large-ish (100-1000) number of cores for the pvserver. Recently we?ve > added the capability to also create ghost cells with this reader such that > MPI domain boundaries become transparent to filters that need neighbor data, > e.g., for gradient computations. However, this ghost cell generation takes > up quite some time, so we would like to keep the number of MPI ranks at a > minimum and use hybrid MPI-OpenMP parallelism for our plugin. > > My question is thus if it is possible to use OpenMP inside a custom plugin? > And if yes, is there a canonical way to do it (maybe an example plugin) or > is it going to be a hack? > > When searching the interwebs for ParaView and OpenMP, I only came up with an > old VTK mailing list thread from 2014 > (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), in > which David (DeMarle) refers to some initial documentation of the SMP > framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). However, > the linked PDF does not contain references to OpenMP. > > Thanks a lot in advance! > > Michael > > -- > Michael Schlottke-Lakemper > > Chair of Fluid Mechanics and Institute of Aerodynamics > RWTH Aachen University > W?llnerstra?e 5a > 52062 Aachen > Germany > > Phone: +49 (241) 80 95188 > Fax: +49 (241) 80 92257 > Mail: m.schlottke-lakemper at aia.rwth-aachen.de > Web: http://www.aia.rwth-aachen.de > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > From m.schlottke-lakemper at aia.rwth-aachen.de Tue Mar 8 09:08:16 2016 From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael) Date: Tue, 8 Mar 2016 14:08:16 +0000 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters In-Reply-To: References: Message-ID: <224B9381-4C0D-4A8C-8D01-CEEE091B1131@aia.rwth-aachen.de> Hi Utkarsh, Thanks for the fast answer. I have a a few follow-up questions, if I may: - What is the default SMP backend if I do not configure anything during ParaView compilation (I assume ?Sequential?)? - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) explicitly with nThreads determined by myself, or will it honor e.g. OMP_NUM_THREADS automatically? - If I have to call vtkSMPTools::Initialize(nThreads) explicitly, where should it be done in case of a plugin? In the plugin?s ctor, in the RequestData method, or someplace else? - Am I correct in the assumption that the backend type should not change the results, i.e. that on hosts where I do not know which SMP backend is configured, I can assume my algorithms ?just work?? Thanks a lot in advance Michael > On 08 Mar 2016, at 14:39 , Utkarsh Ayachit wrote: > > Michael, > > The VTK-SMP approach is indeed the way to go for this. To make the > backend use OpenMP, you simply compile VTK/ParaView with > VTK_SMP_IMPLEMENTATION_TYPE set to OpenMP. The programming model > described in the SMP Guide PDF has been implemented for all supported > backends including OpenMP and TBB. > > Utkarsh > > On Tue, Mar 8, 2016 at 6:22 AM, Schlottke-Lakemper, Michael > wrote: >> Hi everyone, >> >> We have a custom parallel reader plugin that works well with our files even >> on large-ish (100-1000) number of cores for the pvserver. Recently we?ve >> added the capability to also create ghost cells with this reader such that >> MPI domain boundaries become transparent to filters that need neighbor data, >> e.g., for gradient computations. However, this ghost cell generation takes >> up quite some time, so we would like to keep the number of MPI ranks at a >> minimum and use hybrid MPI-OpenMP parallelism for our plugin. >> >> My question is thus if it is possible to use OpenMP inside a custom plugin? >> And if yes, is there a canonical way to do it (maybe an example plugin) or >> is it going to be a hack? >> >> When searching the interwebs for ParaView and OpenMP, I only came up with an >> old VTK mailing list thread from 2014 >> (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), in >> which David (DeMarle) refers to some initial documentation of the SMP >> framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). However, >> the linked PDF does not contain references to OpenMP. >> >> Thanks a lot in advance! >> >> Michael >> >> -- >> Michael Schlottke-Lakemper >> >> Chair of Fluid Mechanics and Institute of Aerodynamics >> RWTH Aachen University >> W?llnerstra?e 5a >> 52062 Aachen >> Germany >> >> Phone: +49 (241) 80 95188 >> Fax: +49 (241) 80 92257 >> Mail: m.schlottke-lakemper at aia.rwth-aachen.de >> Web: http://www.aia.rwth-aachen.de >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> From sujin.philip at kitware.com Tue Mar 8 09:22:52 2016 From: sujin.philip at kitware.com (Sujin Philip) Date: Tue, 8 Mar 2016 09:22:52 -0500 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters In-Reply-To: <224B9381-4C0D-4A8C-8D01-CEEE091B1131@aia.rwth-aachen.de> References: <224B9381-4C0D-4A8C-8D01-CEEE091B1131@aia.rwth-aachen.de> Message-ID: Hi Michael, - What is the default SMP backend if I do not configure anything during ParaView compilation (I assume ?Sequential?)? Sequential is the default backend. - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) explicitly with nThreads determined by myself, or will it honor e.g. OMP_NUM_THREADS automatically? It is not necessary to call vtkSMPTools::Initialize. If initialize is not called and OMP_NUM_THREADS is set, it will be used. - Am I correct in the assumption that the backend type should not change the results, i.e. that on hosts where I do not know which SMP backend is configured, I can assume my algorithms ?just work?? Yes, the different backends should produce the same results. The only noticeable difference should be in the performance. If the results are not the same then its a bug. Thanks Sujin On Tue, Mar 8, 2016 at 9:08 AM, Schlottke-Lakemper, Michael < m.schlottke-lakemper at aia.rwth-aachen.de> wrote: > Hi Utkarsh, > > Thanks for the fast answer. I have a a few follow-up questions, if I may: > > - What is the default SMP backend if I do not configure anything during > ParaView compilation (I assume ?Sequential?)? > - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) > explicitly with nThreads determined by myself, or will it honor e.g. > OMP_NUM_THREADS automatically? > - If I have to call vtkSMPTools::Initialize(nThreads) explicitly, where > should it be done in case of a plugin? In the plugin?s ctor, in the > RequestData method, or someplace else? > - Am I correct in the assumption that the backend type should not change > the results, i.e. that on hosts where I do not know which SMP backend is > configured, I can assume my algorithms ?just work?? > > Thanks a lot in advance > > Michael > > > On 08 Mar 2016, at 14:39 , Utkarsh Ayachit > wrote: > > > > Michael, > > > > The VTK-SMP approach is indeed the way to go for this. To make the > > backend use OpenMP, you simply compile VTK/ParaView with > > VTK_SMP_IMPLEMENTATION_TYPE set to OpenMP. The programming model > > described in the SMP Guide PDF has been implemented for all supported > > backends including OpenMP and TBB. > > > > Utkarsh > > > > On Tue, Mar 8, 2016 at 6:22 AM, Schlottke-Lakemper, Michael > > wrote: > >> Hi everyone, > >> > >> We have a custom parallel reader plugin that works well with our files > even > >> on large-ish (100-1000) number of cores for the pvserver. Recently we?ve > >> added the capability to also create ghost cells with this reader such > that > >> MPI domain boundaries become transparent to filters that need neighbor > data, > >> e.g., for gradient computations. However, this ghost cell generation > takes > >> up quite some time, so we would like to keep the number of MPI ranks at > a > >> minimum and use hybrid MPI-OpenMP parallelism for our plugin. > >> > >> My question is thus if it is possible to use OpenMP inside a custom > plugin? > >> And if yes, is there a canonical way to do it (maybe an example plugin) > or > >> is it going to be a hack? > >> > >> When searching the interwebs for ParaView and OpenMP, I only came up > with an > >> old VTK mailing list thread from 2014 > >> (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), > in > >> which David (DeMarle) refers to some initial documentation of the SMP > >> framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). > However, > >> the linked PDF does not contain references to OpenMP. > >> > >> Thanks a lot in advance! > >> > >> Michael > >> > >> -- > >> Michael Schlottke-Lakemper > >> > >> Chair of Fluid Mechanics and Institute of Aerodynamics > >> RWTH Aachen University > >> W?llnerstra?e 5a > >> 52062 Aachen > >> Germany > >> > >> Phone: +49 (241) 80 95188 > >> Fax: +49 (241) 80 92257 > >> Mail: m.schlottke-lakemper at aia.rwth-aachen.de > >> Web: http://www.aia.rwth-aachen.de > >> > >> > >> _______________________________________________ > >> Powered by www.kitware.com > >> > >> Visit other Kitware open-source projects at > >> http://www.kitware.com/opensource/opensource.html > >> > >> Search the list archives at: > >> http://markmail.org/search/?q=Paraview-developers > >> > >> Follow this link to subscribe/unsubscribe: > >> http://public.kitware.com/mailman/listinfo/paraview-developers > >> > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > -------------- next part -------------- An HTML attachment was scrubbed... URL: From m.schlottke-lakemper at aia.rwth-aachen.de Tue Mar 8 09:36:51 2016 From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael) Date: Tue, 8 Mar 2016 14:36:51 +0000 Subject: [Paraview-developers] "Unable to build k-d tree structure" error with D3 filter and PV 4.4.0 on large datasets Message-ID: <5178E5D9-D3BB-471B-AAA1-5D22B458906F@aia.rwth-aachen.de> Dear all, We have recently come across the following error when using the D3 filter on large and distributed vtkUnstructuredGrids. We have our own reader that locally creates a vtkUnstructuredGrid by reading the according subdomain from a file in netcdf format and I would like to use the D3 filter to have halo cells available on all domains (e.g. for gradient computations). For small scale meshes containing about 260 x 10^6 cells there is no problem. However, when I want to execute the filter on 1.8 x 10^9 cells I get the following error messages: Warning: In /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/Parallel/vtkPKdTree.cxx, line 207 vtkPKdTree (0x2943d0c0): (process 490) memory allocation on my node (BreadthFirstDivide) ERROR: In /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/ParallelMPI/vtkDistributedDataFilter.cxx, line 796 vtkDistributedDataFilter (0x29433cd0): Unable to build k-d tree structure the same is happening with other remote nodes, i.e., Warning: In /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/Parallel/vtkPKdTree.cxx, line 207 vtkPKdTree (0x2a9fdf20): (process 137) memory allocation on a remote node (BreadthFirstDivide) I run a server on 576 cores using ParaView version 4.4.0. The program does not crash after this error, however, it is not creating any output. I have tracked down the error to the following (potential) location. I noticed that this error seems to be generated in the function BreadthFirstDivision, which somehow returns the error code 1, which is then in AllCheckForFailure summarized to a value > 0 on all MPI ranks leading to the error message. I also noticed that this cannot be a VTKERROR as it is not printed to the terminal. I believe that the error might reside in the check for midpt < -1, which would also return an error code of 1 in BreatdhFirstDivision. I hope you can help me with this problem. Thanks a lot in advance! Michael -- Michael Schlottke-Lakemper Chair of Fluid Mechanics and Institute of Aerodynamics RWTH Aachen University W?llnerstra?e 5a 52062 Aachen Germany Phone: +49 (241) 80 95188 Fax: +49 (241) 80 92257 Mail: m.schlottke-lakemper at aia.rwth-aachen.de Web: http://www.aia.rwth-aachen.de -------------- next part -------------- An HTML attachment was scrubbed... URL: From m.schlottke-lakemper at aia.rwth-aachen.de Tue Mar 8 09:39:38 2016 From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael) Date: Tue, 8 Mar 2016 14:39:38 +0000 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters In-Reply-To: References: <224B9381-4C0D-4A8C-8D01-CEEE091B1131@aia.rwth-aachen.de> Message-ID: Hi Sujin, Thanks again for the fast and helpful answer! One last question: if I do not call vtkSMPTools::Initialize() and OMP_NUM_THREADS is not set, which number of threads will be used? Will it be the same as if running a normal OpenMP program with OMP_NUM_THREADS not set or will a different value be used? Michael On 08 Mar 2016, at 15:22 , Sujin Philip > wrote: Hi Michael, - What is the default SMP backend if I do not configure anything during ParaView compilation (I assume ?Sequential?)? Sequential is the default backend. - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) explicitly with nThreads determined by myself, or will it honor e.g. OMP_NUM_THREADS automatically? It is not necessary to call vtkSMPTools::Initialize. If initialize is not called and OMP_NUM_THREADS is set, it will be used. - Am I correct in the assumption that the backend type should not change the results, i.e. that on hosts where I do not know which SMP backend is configured, I can assume my algorithms ?just work?? Yes, the different backends should produce the same results. The only noticeable difference should be in the performance. If the results are not the same then its a bug. Thanks Sujin On Tue, Mar 8, 2016 at 9:08 AM, Schlottke-Lakemper, Michael > wrote: Hi Utkarsh, Thanks for the fast answer. I have a a few follow-up questions, if I may: - What is the default SMP backend if I do not configure anything during ParaView compilation (I assume ?Sequential?)? - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) explicitly with nThreads determined by myself, or will it honor e.g. OMP_NUM_THREADS automatically? - If I have to call vtkSMPTools::Initialize(nThreads) explicitly, where should it be done in case of a plugin? In the plugin?s ctor, in the RequestData method, or someplace else? - Am I correct in the assumption that the backend type should not change the results, i.e. that on hosts where I do not know which SMP backend is configured, I can assume my algorithms ?just work?? Thanks a lot in advance Michael > On 08 Mar 2016, at 14:39 , Utkarsh Ayachit > wrote: > > Michael, > > The VTK-SMP approach is indeed the way to go for this. To make the > backend use OpenMP, you simply compile VTK/ParaView with > VTK_SMP_IMPLEMENTATION_TYPE set to OpenMP. The programming model > described in the SMP Guide PDF has been implemented for all supported > backends including OpenMP and TBB. > > Utkarsh > > On Tue, Mar 8, 2016 at 6:22 AM, Schlottke-Lakemper, Michael > > wrote: >> Hi everyone, >> >> We have a custom parallel reader plugin that works well with our files even >> on large-ish (100-1000) number of cores for the pvserver. Recently we?ve >> added the capability to also create ghost cells with this reader such that >> MPI domain boundaries become transparent to filters that need neighbor data, >> e.g., for gradient computations. However, this ghost cell generation takes >> up quite some time, so we would like to keep the number of MPI ranks at a >> minimum and use hybrid MPI-OpenMP parallelism for our plugin. >> >> My question is thus if it is possible to use OpenMP inside a custom plugin? >> And if yes, is there a canonical way to do it (maybe an example plugin) or >> is it going to be a hack? >> >> When searching the interwebs for ParaView and OpenMP, I only came up with an >> old VTK mailing list thread from 2014 >> (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), in >> which David (DeMarle) refers to some initial documentation of the SMP >> framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). However, >> the linked PDF does not contain references to OpenMP. >> >> Thanks a lot in advance! >> >> Michael >> >> -- >> Michael Schlottke-Lakemper >> >> Chair of Fluid Mechanics and Institute of Aerodynamics >> RWTH Aachen University >> W?llnerstra?e 5a >> 52062 Aachen >> Germany >> >> Phone: +49 (241) 80 95188 >> Fax: +49 (241) 80 92257 >> Mail: m.schlottke-lakemper at aia.rwth-aachen.de >> Web: http://www.aia.rwth-aachen.de >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> _______________________________________________ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Search the list archives at: http://markmail.org/search/?q=Paraview-developers Follow this link to subscribe/unsubscribe: http://public.kitware.com/mailman/listinfo/paraview-developers -------------- next part -------------- An HTML attachment was scrubbed... URL: From sujin.philip at kitware.com Tue Mar 8 09:43:26 2016 From: sujin.philip at kitware.com (Sujin Philip) Date: Tue, 8 Mar 2016 09:43:26 -0500 Subject: [Paraview-developers] Using OpenMP for custom plugins/filters In-Reply-To: References: <224B9381-4C0D-4A8C-8D01-CEEE091B1131@aia.rwth-aachen.de> Message-ID: Hi Michael It will be the same as a regular OpenMP program without the OMP_NUM_THREADS environment variable. -Sujin On Tue, Mar 8, 2016 at 9:39 AM, Schlottke-Lakemper, Michael < m.schlottke-lakemper at aia.rwth-aachen.de> wrote: > Hi Sujin, > > Thanks again for the fast and helpful answer! One last question: if I do > not call vtkSMPTools::Initialize() and OMP_NUM_THREADS is not set, which > number of threads will be used? Will it be the same as if running a normal > OpenMP program with OMP_NUM_THREADS not set or will a different value be > used? > > Michael > > On 08 Mar 2016, at 15:22 , Sujin Philip wrote: > > Hi Michael, > > - What is the default SMP backend if I do not configure anything during > ParaView compilation (I assume ?Sequential?)? > Sequential is the default backend. > > - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) > explicitly with nThreads determined by myself, or will it honor e.g. > OMP_NUM_THREADS automatically? > It is not necessary to call vtkSMPTools::Initialize. If initialize is not > called and OMP_NUM_THREADS is set, it will be used. > > - Am I correct in the assumption that the backend type should not change > the results, i.e. that on hosts where I do not know which SMP backend is > configured, I can assume my algorithms ?just work?? > Yes, the different backends should produce the same results. The only > noticeable difference should be in the performance. If the results are not > the same then its a bug. > > Thanks > Sujin > > > On Tue, Mar 8, 2016 at 9:08 AM, Schlottke-Lakemper, Michael < > m.schlottke-lakemper at aia.rwth-aachen.de> wrote: > >> Hi Utkarsh, >> >> Thanks for the fast answer. I have a a few follow-up questions, if I may: >> >> - What is the default SMP backend if I do not configure anything during >> ParaView compilation (I assume ?Sequential?)? >> - With OpenMP, will I have to call vtkSMPTools::Initialize(nThreads) >> explicitly with nThreads determined by myself, or will it honor e.g. >> OMP_NUM_THREADS automatically? >> - If I have to call vtkSMPTools::Initialize(nThreads) explicitly, where >> should it be done in case of a plugin? In the plugin?s ctor, in the >> RequestData method, or someplace else? >> - Am I correct in the assumption that the backend type should not change >> the results, i.e. that on hosts where I do not know which SMP backend is >> configured, I can assume my algorithms ?just work?? >> >> Thanks a lot in advance >> >> Michael >> >> > On 08 Mar 2016, at 14:39 , Utkarsh Ayachit >> wrote: >> > >> > Michael, >> > >> > The VTK-SMP approach is indeed the way to go for this. To make the >> > backend use OpenMP, you simply compile VTK/ParaView with >> > VTK_SMP_IMPLEMENTATION_TYPE set to OpenMP. The programming model >> > described in the SMP Guide PDF has been implemented for all supported >> > backends including OpenMP and TBB. >> > >> > Utkarsh >> > >> > On Tue, Mar 8, 2016 at 6:22 AM, Schlottke-Lakemper, Michael >> > wrote: >> >> Hi everyone, >> >> >> >> We have a custom parallel reader plugin that works well with our files >> even >> >> on large-ish (100-1000) number of cores for the pvserver. Recently >> we?ve >> >> added the capability to also create ghost cells with this reader such >> that >> >> MPI domain boundaries become transparent to filters that need neighbor >> data, >> >> e.g., for gradient computations. However, this ghost cell generation >> takes >> >> up quite some time, so we would like to keep the number of MPI ranks >> at a >> >> minimum and use hybrid MPI-OpenMP parallelism for our plugin. >> >> >> >> My question is thus if it is possible to use OpenMP inside a custom >> plugin? >> >> And if yes, is there a canonical way to do it (maybe an example >> plugin) or >> >> is it going to be a hack? >> >> >> >> When searching the interwebs for ParaView and OpenMP, I only came up >> with an >> >> old VTK mailing list thread from 2014 >> >> (http://public.kitware.com/pipermail/vtkusers/2014-March/083399.html), >> in >> >> which David (DeMarle) refers to some initial documentation of the SMP >> >> framework (http://www.vtk.org/Wiki/images/3/3b/VTK_SMP_Guide.pdf). >> However, >> >> the linked PDF does not contain references to OpenMP. >> >> >> >> Thanks a lot in advance! >> >> >> >> Michael >> >> >> >> -- >> >> Michael Schlottke-Lakemper >> >> >> >> Chair of Fluid Mechanics and Institute of Aerodynamics >> >> RWTH Aachen University >> >> W?llnerstra?e 5a >> >> 52062 Aachen >> >> Germany >> >> >> >> Phone: +49 (241) 80 95188 >> >> Fax: +49 (241) 80 92257 >> >> Mail: m.schlottke-lakemper at aia.rwth-aachen.de >> >> Web: http://www.aia.rwth-aachen.de >> >> >> >> >> >> _______________________________________________ >> >> Powered by www.kitware.com >> >> >> >> Visit other Kitware open-source projects at >> >> http://www.kitware.com/opensource/opensource.html >> >> >> >> Search the list archives at: >> >> http://markmail.org/search/?q=Paraview-developers >> >> >> >> Follow this link to subscribe/unsubscribe: >> >> http://public.kitware.com/mailman/listinfo/paraview-developers >> >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From maik.pohl87 at gmx.de Thu Mar 10 05:32:23 2016 From: maik.pohl87 at gmx.de (mp) Date: Thu, 10 Mar 2016 02:32:23 -0800 (PST) Subject: [Paraview-developers] Carrying calculated array through to the last time step in programmable filter Message-ID: <1457605943043-4213.post@n6.nabble.com> Hey, I have to write a programmable filter, that takes the values of an array from a source in the pipeline and calculates the difference of it's values at different times. Here is what i have so far: from paraview.simple import * filter = ProgrammableFilter() RenameSource("Diference between time steps" , filter) filter.Script = """ import os #Number of Array to be subtracted arr = 2 # tStart is the time that contains the minuend tStart = 0 # tEnd is the time that contains the subtrahend tEnd = 900 tempFilePath = "" tCurrent = inputs[0].GetInformation().Get(vtk.vtkDataObject.DATA_TIME_STEP()) if tCurrent == tStart: f = open(tempFilePath + "test.txt" , 'w') for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): f.write(str(self.GetInput().GetCellData().GetArray(arr).GetValue(i)) + "\\n") elif tCurrent==tEnd: subtrahend = list() for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): subtrahend.append(float(self.GetInput().GetCellData().GetArray(arr).GetValue(i))) f = open(tempFilePath + "test.txt" , 'r') minuend = [line.rstrip("\\n") for line in f] for j in range(len(minuend)): minuend[j] = float(minuend[j]) difference = vtk.vtkDoubleArray() difference.SetNumberOfComponents(0) difference.SetName("difference") d = map(lambda x,y: x-y,minuend,subtrahend) for k in range(len(d)): difference.InsertNextValue(d[k]) self.GetOutput().GetCellData().AddArray(difference) f.close() os.remove(tempFilePath + "test.txt") """ filter_repr = Show() In this example, the difference between the 3rd array at 0 and 900 is calculated. Now the user has to go back to the time step, where "difference" was calculated. Is there a way to display the array, that was calculated in another time step, after the last time step? Thank you, mp -- View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Carrying-calculated-array-through-to-the-last-time-step-in-programmable-filter-tp4213.html Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com. From cory.quammen at kitware.com Fri Mar 11 08:34:56 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Fri, 11 Mar 2016 08:34:56 -0500 Subject: [Paraview-developers] Carrying calculated array through to the last time step in programmable filter In-Reply-To: <1457605943043-4213.post@n6.nabble.com> References: <1457605943043-4213.post@n6.nabble.com> Message-ID: While this isn't a direct answer to your question, this blog post may have some answers for you. https://blog.kitware.com/the-multiple-uses-of-the-new-forcetime-filter/ On Thu, Mar 10, 2016 at 5:32 AM, mp wrote: > Hey, > > I have to write a programmable filter, that takes the values of an array > from a source in the pipeline and calculates the difference of it's values > at different times. > > Here is what i have so far: > > from paraview.simple import * > filter = ProgrammableFilter() > RenameSource("Diference between time steps" , filter) > filter.Script = """ > import os > #Number of Array to be subtracted > arr = 2 > # tStart is the time that contains the minuend > tStart = 0 > # tEnd is the time that contains the subtrahend > tEnd = 900 > > tempFilePath = "" > tCurrent = > inputs[0].GetInformation().Get(vtk.vtkDataObject.DATA_TIME_STEP()) > if tCurrent == tStart: > f = open(tempFilePath + "test.txt" , 'w') > for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): > f.write(str(self.GetInput().GetCellData().GetArray(arr).GetValue(i)) > + "\\n") > elif tCurrent==tEnd: > subtrahend = list() > for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): > > subtrahend.append(float(self.GetInput().GetCellData().GetArray(arr).GetValue(i))) > f = open(tempFilePath + "test.txt" , 'r') > minuend = [line.rstrip("\\n") for line in f] > for j in range(len(minuend)): > minuend[j] = float(minuend[j]) > difference = vtk.vtkDoubleArray() > difference.SetNumberOfComponents(0) > difference.SetName("difference") > d = map(lambda x,y: x-y,minuend,subtrahend) > for k in range(len(d)): > difference.InsertNextValue(d[k]) > self.GetOutput().GetCellData().AddArray(difference) > f.close() > os.remove(tempFilePath + "test.txt") > """ > filter_repr = Show() > > > > > In this example, the difference between the 3rd array at 0 and 900 is > calculated. Now the user has to go back to the time step, where "difference" > was calculated. > > Is there a way to display the array, that was calculated in another time > step, after the last time step? > > Thank you, mp > > > > -- > View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Carrying-calculated-array-through-to-the-last-time-step-in-programmable-filter-tp4213.html > Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com. > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers -- Cory Quammen R&D Engineer Kitware, Inc. From maik.pohl87 at gmx.de Wed Mar 16 06:57:55 2016 From: maik.pohl87 at gmx.de (mp) Date: Wed, 16 Mar 2016 03:57:55 -0700 (PDT) Subject: [Paraview-developers] Carrying calculated array through to the last time step in programmable filter In-Reply-To: References: <1457605943043-4213.post@n6.nabble.com> Message-ID: <1458125875917-4215.post@n6.nabble.com> OK, first: Thank you for your answer. I should probably have mentioned, that i am a total beginner in python and paraview. Though i kind of understood the theory of the article, i have no idea how to implement it, and when i searched for the "vtkForceTimeFilter" i got exactly one result, the article itself. -- View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Carrying-calculated-array-through-to-the-last-time-step-in-programmable-filter-tp4213p4215.html Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com. From maik.pohl87 at gmx.de Wed Mar 16 08:03:06 2016 From: maik.pohl87 at gmx.de (mp) Date: Wed, 16 Mar 2016 05:03:06 -0700 (PDT) Subject: [Paraview-developers] Carrying calculated array through to the last time step in programmable filter In-Reply-To: <1458125875917-4215.post@n6.nabble.com> References: <1457605943043-4213.post@n6.nabble.com> <1458125875917-4215.post@n6.nabble.com> Message-ID: <1458129786059-4216.post@n6.nabble.com> For now, i used a different (very inelegant and inefficient) way, but it's the only one i can come up with for now. Instead of writing the difference into an array and trying to get the array to the end, i write the results of my calculation to a file and when the simulation reaches the last time step (for now, the user has to specify this too), the file is read and written into an array which is then committed to the output. The source code now looks like this: from paraview.simple import * filter = ProgrammableFilter() RenameSource("Diference between time steps" , filter) filter.Script = """ import os import os.path #Number of Array to be subtracted arr = 2 # t1 is the time that contains the minuend t1 = 100 # t2 is the time that contains the subtrahend t2 = 800 # tEnd is the last timestep of the simulation tEnd = 1000 tempFilePath = "" tCurrent = inputs[0].GetInformation().Get(vtk.vtkDataObject.DATA_TIME_STEP()) if tCurrent == t1: f = open(tempFilePath + "test.txt" , 'w') for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): f.write(str(self.GetInput().GetCellData().GetArray(arr).GetValue(i)) + "\\n") elif tCurrent==t2: subtrahend = list() for i in range(self.GetInput().GetCellData().GetArray(arr).GetSize()): subtrahend.append(float(self.GetInput().GetCellData().GetArray(arr).GetValue(i))) f = open(tempFilePath + "test.txt" , 'r') minuend = [line.rstrip("\\n") for line in f] for j in range(len(minuend)): minuend[j] = float(minuend[j]) d = map(lambda x,y: x-y,minuend,subtrahend) f.close() f = open(tempFilePath + "difference.txt" ,'w') for k in range(len(d)): f.write(str(d[k]) + "\\n") os.remove(tempFilePath + "test.txt") elif tCurrent==tEnd: f = open(tempFilePath + "difference.txt" , 'r') dif = [line.rstrip("\\n") for line in f] for j in range(len(dif)): dif[j] = float(dif[j]) f.close() difference = vtk.vtkDoubleArray() difference.SetNumberOfComponents(0) difference.SetName("difference") for i in range(len(dif)): difference.InsertNextValue(dif[i]) self.GetOutput().GetCellData().AddArray(difference) os.remove(tempFilePath + "difference.txt") """ filter_repr = Show() Is there a way to get the last time step without having to rely on the user? Thanks, mp -- View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Carrying-calculated-array-through-to-the-last-time-step-in-programmable-filter-tp4213p4216.html Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com. From utkarsh.ayachit at kitware.com Wed Mar 16 16:45:20 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Wed, 16 Mar 2016 16:45:20 -0400 Subject: [Paraview-developers] ANN: ParaView 5.0.1-RC2 Tagged Message-ID: Folks, ParaView 5.0.1-RC2 has been tagged! Git tag name: v5.0.1-RC2 Source tar/zip files are available for download: http://www.paraview.org/download/ Binaries will be available for testing within the next few days. - ParaView Team From joseph.g.hennessey2.ctr at mail.mil Thu Mar 17 16:04:08 2016 From: joseph.g.hennessey2.ctr at mail.mil (Hennessey, Joseph G CTR USARMY RDECOM ARL (US)) Date: Thu, 17 Mar 2016 20:04:08 +0000 Subject: [Paraview-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) Message-ID: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> CLASSIFICATION: UNCLASSIFIED Hello, I have modified Rendering/Core/vtkCamera.cxx:: ComputeOffAxisProjectionFrustum To fix a problem with eyeseparation and near and far planes for use with immersive paraview The merge request is called fix ComputeOffAxisProjectionFrustum eyeseparation and near and far planes. https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740ec988d060c5c296 d36afb Commit 4fe4fabb420bba77740ec988d060c5c296d36afb Authored by Joseph G. Hennessey about an hour ago 1 parent d52198e1 Please add this to VTK and indirectly to ParaView Thanks, Joe ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Joseph G. Hennessey Ph.D., Lockheed Martin Army Research Lab DOD Supercomputing Resource Center CLASSIFICATION: UNCLASSIFIED -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5615 bytes Desc: not available URL: From aashish.chaudhary at kitware.com Thu Mar 17 16:17:02 2016 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Thu, 17 Mar 2016 16:17:02 -0400 Subject: [Paraview-developers] [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) In-Reply-To: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> References: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> Message-ID: Hi Joe, Thank for you for the PR. Could you please describe the problem in bit more detail? Thanks, On Thu, Mar 17, 2016 at 4:04 PM, Hennessey, Joseph G CTR USARMY RDECOM ARL (US) wrote: > CLASSIFICATION: UNCLASSIFIED > > Hello, > > I have modified > > Rendering/Core/vtkCamera.cxx:: ComputeOffAxisProjectionFrustum > > To fix a problem with eyeseparation and near and far planes for use with > immersive paraview > > The merge request is called > > fix ComputeOffAxisProjectionFrustum eyeseparation and near and far planes. > > > https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740ec988d060c5c296 > d36afb > > Commit 4fe4fabb420bba77740ec988d060c5c296d36afb > > Authored by Joseph G. Hennessey about an hour ago > 1 parent d52198e1 > > Please add this to VTK and indirectly to ParaView > > Thanks, > > Joe > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Joseph G. Hennessey Ph.D., Lockheed Martin > Army Research Lab > DOD Supercomputing Resource Center > > > CLASSIFICATION: UNCLASSIFIED > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Thu Mar 17 17:15:09 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Thu, 17 Mar 2016 15:15:09 -0600 Subject: [Paraview-developers] [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) In-Reply-To: References: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> Message-ID: Considering that vtkCamera doesn't know the physical size of the screen, I certainly hope that EyeSeparation is not in meters or any other absolute units! It should be in units relative to the screen size. And, since only the application itself knows the physical screen size, it is up to the application to set it appropriately. The eye separation is much smaller relative to a huge wall screen, than to a workstation display. Just agreeing with the other David here... - David On Thu, Mar 17, 2016 at 2:44 PM, David Cole via vtk-developers < vtk-developers at vtk.org> wrote: > OffAxis projection in VTK is woefully underdocumented, and accepting > this pull request will only add to the mystery. > > I would suggest documenting that the coordinates used to specify it > are in vtkCamera coordinates (not world), and clarifying with comments > in the code what some of the constants used are for. Also, > EyeSeparation appears to be specified in meters, although it's unclear > how it relates to any of the VTK coordinate systems or to the size of > the actual display being used. > > > 2 cents, > David C. > > > On Thu, Mar 17, 2016 at 4:17 PM, Aashish Chaudhary > wrote: > > Hi Joe, > > > > Thank for you for the PR. Could you please describe the problem in bit > more > > detail? > > > > Thanks, > > > > > > On Thu, Mar 17, 2016 at 4:04 PM, Hennessey, Joseph G CTR USARMY RDECOM > ARL > > (US) wrote: > >> > >> CLASSIFICATION: UNCLASSIFIED > >> > >> Hello, > >> > >> I have modified > >> > >> Rendering/Core/vtkCamera.cxx:: ComputeOffAxisProjectionFrustum > >> > >> To fix a problem with eyeseparation and near and far planes for use with > >> immersive paraview > >> > >> The merge request is called > >> > >> fix ComputeOffAxisProjectionFrustum eyeseparation and near and far > planes. > >> > >> > >> > https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740ec988d060c5c296 > >> d36afb > >> > >> Commit 4fe4fabb420bba77740ec988d060c5c296d36afb > >> > >> Authored by Joseph G. Hennessey about an hour ago > >> 1 parent d52198e1 > >> > >> Please add this to VTK and indirectly to ParaView > >> > >> Thanks, > >> > >> Joe > >> > >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > >> Joseph G. Hennessey Ph.D., Lockheed Martin > >> Army Research Lab > >> DOD Supercomputing Resource Center > >> > >> > >> CLASSIFICATION: UNCLASSIFIED > >> > >> _______________________________________________ > >> Powered by www.kitware.com > >> > >> Visit other Kitware open-source projects at > >> http://www.kitware.com/opensource/opensource.html > >> > >> Search the list archives at: > http://markmail.org/search/?q=vtk-developers > >> > >> Follow this link to subscribe/unsubscribe: > >> http://public.kitware.com/mailman/listinfo/vtk-developers > >> > >> > > > > > > > > -- > > | Aashish Chaudhary > > | Technical Leader > > | Kitware Inc. > > | http://www.kitware.com/company/team/chaudhary.html > > > > _______________________________________________ > > Powered by www.kitware.com > > > > Visit other Kitware open-source projects at > > http://www.kitware.com/opensource/opensource.html > > > > Search the list archives at: > http://markmail.org/search/?q=vtk-developers > > > > Follow this link to subscribe/unsubscribe: > > http://public.kitware.com/mailman/listinfo/vtk-developers > > > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseph.g.hennessey2.ctr at mail.mil Fri Mar 18 08:47:11 2016 From: joseph.g.hennessey2.ctr at mail.mil (Hennessey, Joseph G CTR USARMY RDECOM ARL (US)) Date: Fri, 18 Mar 2016 12:47:11 +0000 Subject: [Paraview-developers] [Non-DoD Source] Re: [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) In-Reply-To: References: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> Message-ID: <10A03274360DCF47A6EE78C9952A31CA41726AC3@UMECHPA7C.easf.csd.disa.mil> CLASSIFICATION: UNCLASSIFIED The modifications that I have made are designed to bring the server side rendering in line with the client rendering in ParaView. They correct an issue where the server side render used in immersive ParaView Display does not match what is generated and Displayed in the ParaView client RenderWindow. The OffAxis projection code is used on the server side, but is not used to generate Images for the client display. My goal with these changes is to bring them closer, but it does not bring them to pixel by pixel perfection, which is perhaps not even possible given that there are using different routines to calculate what to display. I do not know why constants were originally chosen for the near and far planes, for example, but I do know that they do not work well on different scaled data clipping the data out of visibility. My changes allow data across a large range of scales to be visualized and closely approximate what would be displayed in the ParaView client RenderWindow. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Joseph G. Hennessey Ph.D., Lockheed Martin Army Research Lab DOD Supercomputing Resource Center -----Original Message----- From: David Cole [mailto:DLRdave at aol.com] Sent: Thursday, March 17, 2016 4:44 PM To: Aashish Chaudhary Cc: Hennessey, Joseph G CTR USARMY RDECOM ARL (US) ; paraview-developers at paraview.org; vtk-developers at vtk.org Subject: [Non-DoD Source] Re: [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) All active links contained in this email were disabled. Please verify the identity of the sender, and confirm the authenticity of all links contained within the message prior to copying and pasting the address to a Web browser. ---- OffAxis projection in VTK is woefully underdocumented, and accepting this pull request will only add to the mystery. I would suggest documenting that the coordinates used to specify it are in vtkCamera coordinates (not world), and clarifying with comments in the code what some of the constants used are for. Also, EyeSeparation appears to be specified in meters, although it's unclear how it relates to any of the VTK coordinate systems or to the size of the actual display being used. 2 cents, David C. On Thu, Mar 17, 2016 at 4:17 PM, Aashish Chaudhary wrote: > Hi Joe, > > Thank for you for the PR. Could you please describe the problem in bit > more detail? > > Thanks, > > > On Thu, Mar 17, 2016 at 4:04 PM, Hennessey, Joseph G CTR USARMY RDECOM > ARL > (US) wrote: >> >> CLASSIFICATION: UNCLASSIFIED >> >> Hello, >> >> I have modified >> >> Rendering/Core/vtkCamera.cxx:: ComputeOffAxisProjectionFrustum >> >> To fix a problem with eyeseparation and near and far planes for use >> with immersive paraview >> >> The merge request is called >> >> fix ComputeOffAxisProjectionFrustum eyeseparation and near and far planes. >> >> >> Caution-https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740 >> ec988d060c5c296 >> d36afb >> >> Commit 4fe4fabb420bba77740ec988d060c5c296d36afb >> >> Authored by Joseph G. Hennessey about an hour ago >> 1 parent d52198e1 >> >> Please add this to VTK and indirectly to ParaView >> >> Thanks, >> >> Joe >> >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> Joseph G. Hennessey Ph.D., Lockheed Martin Army Research Lab DOD >> Supercomputing Resource Center >> >> >> CLASSIFICATION: UNCLASSIFIED >> >> _______________________________________________ >> Powered by Caution-www.kitware.com >> >> Visit other Kitware open-source projects at >> Caution-http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> Caution-http://markmail.org/search/?q=vtk-developers >> >> Follow this link to subscribe/unsubscribe: >> Caution-http://public.kitware.com/mailman/listinfo/vtk-developers >> >> > > > > -- > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > | Caution-http://www.kitware.com/company/team/chaudhary.html > > _______________________________________________ > Powered by Caution-www.kitware.com > > Visit other Kitware open-source projects at > Caution-http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > Caution-http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > Caution-http://public.kitware.com/mailman/listinfo/vtk-developers > > CLASSIFICATION: UNCLASSIFIED -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5615 bytes Desc: not available URL: From joseph.g.hennessey2.ctr at mail.mil Fri Mar 18 08:42:43 2016 From: joseph.g.hennessey2.ctr at mail.mil (Hennessey, Joseph G CTR USARMY RDECOM ARL (US)) Date: Fri, 18 Mar 2016 12:42:43 +0000 Subject: [Paraview-developers] [Non-DoD Source] Re: [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) In-Reply-To: References: <10A03274360DCF47A6EE78C9952A31CA417254ED@UMECHPA7C.easf.csd.disa.mil> Message-ID: <10A03274360DCF47A6EE78C9952A31CA41726A99@UMECHPA7C.easf.csd.disa.mil> CLASSIFICATION: UNCLASSIFIED The application would still set the eyeseparation with my changes. What I have done is correct a mismatch between what is shown on the server side screen with what is generated on the same size screen when visualized in the ParaView client RenderWindow. The correction I have added for eyeseparation scaling with distance to the object is a close approximation of what is already being displayed when the ParaView client RenderWindow displays stereo data. There were 2 may problems a mismatch of separation across all scales, which I have corrected as well as an additional mismatch for very small data (of sizes less than 1) which I have also corrected. The eyeseparation now produces similar results to what is generated for the ParaView client RenderWIndow. The near and far planes changes were necessary as they were hard coded constants and did not scale with the size of the data. This was particularly noticeable for data scaled at 0.001 or smaller, where the data would be clipped out of visibility on the server side display but would show up just fine in the ParaView client RenderWindow. By changing the near and far plane to be relative to the data, I am now more closely approximating what can be seen in the ParaView client RenderWindow and can visually data across a wide range of scales. I have tested this routing across scales from 1,000,000 X to 1/1,000,000 X and at every intermediate scale of 10 and the data now clips in a way that closely approximates the way it clips in the ParaView client RenderWindow. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Joseph G. Hennessey Ph.D., Lockheed Martin Army Research Lab DOD Supercomputing Resource Center -----Original Message----- From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of David Gobbi Sent: Thursday, March 17, 2016 5:15 PM To: David Cole Cc: paraview-developers at paraview.org; vtk-developers at vtk.org Subject: [Non-DoD Source] Re: [vtk-developers] merge request for vtkCamera.cxx eyeseparation and near far planes correction (UNCLASSIFIED) All active links contained in this email were disabled. Please verify the identity of the sender, and confirm the authenticity of all links contained within the message prior to copying and pasting the address to a Web browser. ________________________________ Considering that vtkCamera doesn't know the physical size of the screen, I certainly hope that EyeSeparation is not in meters or any other absolute units! It should be in units relative to the screen size. And, since only the application itself knows the physical screen size, it is up to the application to set it appropriately. The eye separation is much smaller relative to a huge wall screen, than to a workstation display. Just agreeing with the other David here... - David On Thu, Mar 17, 2016 at 2:44 PM, David Cole via vtk-developers > wrote: OffAxis projection in VTK is woefully underdocumented, and accepting this pull request will only add to the mystery. I would suggest documenting that the coordinates used to specify it are in vtkCamera coordinates (not world), and clarifying with comments in the code what some of the constants used are for. Also, EyeSeparation appears to be specified in meters, although it's unclear how it relates to any of the VTK coordinate systems or to the size of the actual display being used. 2 cents, David C. On Thu, Mar 17, 2016 at 4:17 PM, Aashish Chaudhary > wrote: > Hi Joe, > > Thank for you for the PR. Could you please describe the problem in bit more > detail? > > Thanks, > > > On Thu, Mar 17, 2016 at 4:04 PM, Hennessey, Joseph G CTR USARMY RDECOM ARL > (US) > wrote: >> >> CLASSIFICATION: UNCLASSIFIED >> >> Hello, >> >> I have modified >> >> Rendering/Core/vtkCamera.cxx:: ComputeOffAxisProjectionFrustum >> >> To fix a problem with eyeseparation and near and far planes for use with >> immersive paraview >> >> The merge request is called >> >> fix ComputeOffAxisProjectionFrustum eyeseparation and near and far planes. >> >> >> Caution-https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740ec988d060c5c296 < Caution-https://gitlab.kitware.com/vtk/vtk/commit/4fe4fabb420bba77740ec988d060c5c296 > >> d36afb >> >> Commit 4fe4fabb420bba77740ec988d060c5c296d36afb >> >> Authored by Joseph G. Hennessey about an hour ago >> 1 parent d52198e1 >> >> Please add this to VTK and indirectly to ParaView >> >> Thanks, >> >> Joe >> >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> Joseph G. Hennessey Ph.D., Lockheed Martin >> Army Research Lab >> DOD Supercomputing Resource Center >> >> >> CLASSIFICATION: UNCLASSIFIED >> >> _______________________________________________ >> Powered by Caution-www.kitware.com < Caution-http://www.kitware.com > >> >> Visit other Kitware open-source projects at >> Caution-http://www.kitware.com/opensource/opensource.html < Caution-http://www.kitware.com/opensource/opensource.html > >> >> Search the list archives at: Caution-http://markmail.org/search/?q=vtk-developers < Caution-http://markmail.org/search/?q=vtk-developers > >> >> Follow this link to subscribe/unsubscribe: >> Caution-http://public.kitware.com/mailman/listinfo/vtk-developers < Caution-http://public.kitware.com/mailman/listinfo/vtk-developers > >> >> > > > > -- > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > | Caution-http://www.kitware.com/company/team/chaudhary.html < Caution-http://www.kitware.com/company/team/chaudhary.html > > > _______________________________________________ > Powered by Caution-www.kitware.com < Caution-http://www.kitware.com > > > Visit other Kitware open-source projects at > Caution-http://www.kitware.com/opensource/opensource.html < Caution-http://www.kitware.com/opensource/opensource.html > > > Search the list archives at: Caution-http://markmail.org/search/?q=vtk-developers < Caution-http://markmail.org/search/?q=vtk-developers > > > Follow this link to subscribe/unsubscribe: > Caution-http://public.kitware.com/mailman/listinfo/vtk-developers < Caution-http://public.kitware.com/mailman/listinfo/vtk-developers > > > _______________________________________________ Powered by Caution-www.kitware.com < Caution-http://www.kitware.com > Visit other Kitware open-source projects at Caution-http://www.kitware.com/opensource/opensource.html < Caution-http://www.kitware.com/opensource/opensource.html > Search the list archives at: Caution-http://markmail.org/search/?q=vtk-developers < Caution-http://markmail.org/search/?q=vtk-developers > Follow this link to subscribe/unsubscribe: Caution-http://public.kitware.com/mailman/listinfo/vtk-developers < Caution-http://public.kitware.com/mailman/listinfo/vtk-developers > CLASSIFICATION: UNCLASSIFIED -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 5615 bytes Desc: not available URL: From mathieu.westphal at kitware.com Mon Mar 21 07:08:54 2016 From: mathieu.westphal at kitware.com (Mathieu Westphal) Date: Mon, 21 Mar 2016 12:08:54 +0100 Subject: [Paraview-developers] Paraview Qt5 Plugin Exit bug Message-ID: Hello I'm experiencing a bug trying to port a plugin to paraview Qt5. The plugin works correctly, but when paraview exits, i have this : Inconsistency detected by ld.so: dl-close.c: 765: _dl_close: Assertion `map->l_init_called' failed! I have reduced the bug to its most pure expression. A faulty plugins needs : - A dummy xml file to parse - A QProperty containing a Qt Container, and associated dummy method. - Paraview Qt 5.X (tested 5.6) Attached is a archive containing a faulty plugin as an example. This bug can be found here as well : http://www.paraview.org/Bug/view.php?id=14643 Any help from qt experts would be much appreciated ! Mathieu Westphal -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: pluginBugQt5.tar.gz Type: application/x-gzip Size: 803 bytes Desc: not available URL: From utkarsh.ayachit at kitware.com Mon Mar 21 09:05:18 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Mon, 21 Mar 2016 09:05:18 -0400 Subject: [Paraview-developers] Paraview Qt5 Plugin Exit bug In-Reply-To: References: Message-ID: Thanks for tracking it down as much as you did, Mathieu. I've been seeing this on the dashboard but didn't get a chance to track it down. Now that I know what causes the segfault I can dig in. I'll get back to you. Utkarsh On Mon, Mar 21, 2016 at 7:08 AM, Mathieu Westphal wrote: > Hello > > I'm experiencing a bug trying to port a plugin to paraview Qt5. > The plugin works correctly, but when paraview exits, i have this : > Inconsistency detected by ld.so: dl-close.c: 765: _dl_close: Assertion > `map->l_init_called' failed! > > I have reduced the bug to its most pure expression. > A faulty plugins needs : > - A dummy xml file to parse > - A QProperty containing a Qt Container, and associated dummy method. > - Paraview Qt 5.X (tested 5.6) > > Attached is a archive containing a faulty plugin as an example. > This bug can be found here as well : > http://www.paraview.org/Bug/view.php?id=14643 > > Any help from qt experts would be much appreciated ! > > Mathieu Westphal From christophe.bourcier.pv at gmail.com Mon Mar 21 10:23:42 2016 From: christophe.bourcier.pv at gmail.com (Christophe Bourcier) Date: Mon, 21 Mar 2016 15:23:42 +0100 Subject: [Paraview-developers] SEGFAULT when applying a slice on a dataset with 2.2e9 cells Message-ID: Hi, I have an ImageData with 2.2e9 cells that loads well in Paraview compiled with VTK_USE_64BIT_IDS=ON. But when I apply a slice on it, I get a segmentation fault. I have modified VTK/Filters/Core/vtkSynchronizedTemplatesCutter3D.cxx to use vtkIdType instead of int where needed. See the patch in attachment. Now, I can make the slice on my dataset. You can download my sample file here (2,1Gb): http://bourcier.eu/VOL0_fft.vti I wonder if other filters will need some patch too with this number of cells. Christophe -------------- next part -------------- --- ParaView-v5.0.0-source.orig/VTK/Filters/Core/vtkSynchronizedTemplatesCutter3D.cxx 2016-01-08 21:22:21.000000000 +0100 +++ ParaView-v5.0.0-source/VTK/Filters/Core/vtkSynchronizedTemplatesCutter3D.cxx 2016-03-21 13:59:36.476965953 +0100 @@ -132,14 +132,14 @@ int numContours = self->GetNumberOfContours(); T *inPtrX, *inPtrY, *inPtrZ; T *s0, *s1, *s2, *s3; - int xMin, xMax, yMin, yMax, zMin, zMax; - int xInc, yInc, zInc; + vtkIdType xMin, xMax, yMin, yMax, zMin, zMax; + vtkIdType xInc, yInc, zInc; int xIncFunc, yIncFunc, zIncFunc, scalarZIncFunc; double *origin = data->GetOrigin(); double *spacing = data->GetSpacing(); int *isect1Ptr, *isect2Ptr; double y, z, t; - int i, j, k; + vtkIdType i, j, k; int zstep, yisectstep; int offsets[12]; int *tablePtr; @@ -149,7 +149,7 @@ vtkIdType ptIds[3]; double value; // We need to know the edgePointId's for interpolating attributes. - int edgePtId, inCellId, outCellId; + vtkIdType edgePtId, inCellId, outCellId; vtkPointData *inPD = data->GetPointData(); vtkCellData *inCD = data->GetCellData(); vtkPointData *outPD = output->GetPointData(); @@ -309,15 +309,15 @@ for (j = yMin; j <= yMax; j++) { // Should not impact performance here/ - edgePtId = (xMin-inExt[0])*xInc + (j-inExt[2])*yInc + (k-inExt[4])*zInc; + edgePtId = (xMin-(vtkIdType)inExt[0])*xInc + (j-(vtkIdType)inExt[2])*yInc + (k-(vtkIdType)inExt[4])*zInc; // Increments are different for cells. Since the cells are not // contoured until the second row of templates, subtract 1 from // i,j,and k. Note: first cube is formed when i=0, j=1, and k=1. inCellId = - (xMin-inExt[0]) + (inExt[1]-inExt[0])* - ( (j-inExt[2]-1) + (k-inExt[4]-1)*(inExt[3]-inExt[2]) ); - + (xMin-(vtkIdType)inExt[0]) + ((vtkIdType)inExt[1]-(vtkIdType)inExt[0])* + ( (j-(vtkIdType)inExt[2]-1) + (k-(vtkIdType)inExt[4]-1)*((vtkIdType)inExt[3]-(vtkIdType)inExt[2]) ); + y = origin[1] + j*spacing[1]; xz[1] = y; From utkarsh.ayachit at kitware.com Mon Mar 21 13:43:06 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Mon, 21 Mar 2016 13:43:06 -0400 Subject: [Paraview-developers] Paraview Qt5 Plugin Exit bug In-Reply-To: References: Message-ID: A WIP fix here: https://gitlab.kitware.com/paraview/paraview/merge_requests/695 On Mon, Mar 21, 2016 at 9:05 AM, Utkarsh Ayachit wrote: > Thanks for tracking it down as much as you did, Mathieu. I've been > seeing this on the dashboard but didn't get a chance to track it down. > Now that I know what causes the segfault I can dig in. I'll get back > to you. > > Utkarsh > > On Mon, Mar 21, 2016 at 7:08 AM, Mathieu Westphal > wrote: >> Hello >> >> I'm experiencing a bug trying to port a plugin to paraview Qt5. >> The plugin works correctly, but when paraview exits, i have this : >> Inconsistency detected by ld.so: dl-close.c: 765: _dl_close: Assertion >> `map->l_init_called' failed! >> >> I have reduced the bug to its most pure expression. >> A faulty plugins needs : >> - A dummy xml file to parse >> - A QProperty containing a Qt Container, and associated dummy method. >> - Paraview Qt 5.X (tested 5.6) >> >> Attached is a archive containing a faulty plugin as an example. >> This bug can be found here as well : >> http://www.paraview.org/Bug/view.php?id=14643 >> >> Any help from qt experts would be much appreciated ! >> >> Mathieu Westphal From christophe.bourcier.pv at gmail.com Wed Mar 23 11:41:12 2016 From: christophe.bourcier.pv at gmail.com (Christophe Bourcier) Date: Wed, 23 Mar 2016 16:41:12 +0100 Subject: [Paraview-developers] Wrong image slice representation of cells fields since ParaView 4.4.0 Message-ID: Hi, On an ImageData, the result of an image slice representation is wrong for cells fields. It is reproduced on ParaView 4.4.0 to 5.0.1-rc2, it worked fine in 4.3.1. And it works fine with nodes fields. See pictures in links [1] [2]. In debug mode, I have an error in the assertion in VTK/Common/DataModel/vtkExtractStructuredGridHelper.cxx :525 : void vtkExtractStructuredGridHelper::CopyCellData(int*, int*, vtkCellData*, vtkCellData*): // Sanity checks assert( "pre: srcStart out of bounds" && (srcStart >= 0) && (srcStart < inSize) ); Any clue on how to fix this? Steps to reproduce: - load the vtk file in link [3] - in Display, change Representation from Outline to Slice Thanks, Christophe [1] http://bourcier.eu/slice_431_ok.png [2] http://bourcier.eu/slice_501rc2_pb.png [3] http://bourcier.eu/arlequin_N2_r16.vtk From utkarsh.ayachit at kitware.com Wed Mar 23 13:17:18 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Wed, 23 Mar 2016 13:17:18 -0400 Subject: [Paraview-developers] Wrong image slice representation of cells fields since ParaView 4.4.0 In-Reply-To: References: Message-ID: I certainly can reproduce this issue. Looking into it. I'll report back once I know more. Utkarsh On Wed, Mar 23, 2016 at 11:41 AM, Christophe Bourcier wrote: > Hi, > > On an ImageData, the result of an image slice representation is wrong > for cells fields. It is reproduced on ParaView 4.4.0 to 5.0.1-rc2, it > worked fine in 4.3.1. And it works fine with nodes fields. > > See pictures in links [1] [2]. > > In debug mode, I have an error in the assertion in > VTK/Common/DataModel/vtkExtractStructuredGridHelper.cxx :525 : void > vtkExtractStructuredGridHelper::CopyCellData(int*, int*, vtkCellData*, > vtkCellData*): > > > // Sanity checks > assert( "pre: srcStart out of bounds" && (srcStart >= 0) && > (srcStart < inSize) ); > > > Any clue on how to fix this? > > Steps to reproduce: > - load the vtk file in link [3] > - in Display, change Representation from Outline to Slice > > Thanks, > > Christophe > > [1] http://bourcier.eu/slice_431_ok.png > [2] http://bourcier.eu/slice_501rc2_pb.png > [3] http://bourcier.eu/arlequin_N2_r16.vtk > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers From nazarian2 at gmail.com Thu Mar 24 00:09:54 2016 From: nazarian2 at gmail.com (nazarian2) Date: Wed, 23 Mar 2016 21:09:54 -0700 (PDT) Subject: [Paraview-developers] Disabled buttons on ParaView 5.0.0 on Mac Os X 10.11.3 Message-ID: <1458792594769-4230.post@n6.nabble.com> Hi all, I'm currently using ParaView 5.0.0 on Mac Os X 10.11.3 and I'm having difficulties with the attached example: all the filters appear disabled, as well as the third row buttons (Calculator, contour, clip, slice, etc..). What am I doing wrong? Regards. -- View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Disabled-buttons-on-ParaView-5-0-0-on-Mac-Os-X-10-11-3-tp4230.html Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com. From Oliver.Fernandes at visus.uni-stuttgart.de Thu Mar 24 04:56:54 2016 From: Oliver.Fernandes at visus.uni-stuttgart.de (Oliver Fernandes) Date: Thu, 24 Mar 2016 08:56:54 +0000 Subject: [Paraview-developers] Passing values between filters Message-ID: <1458809799131.52423@visus.uni-stuttgart.de> Hello all, So I wrote a couple of ParaView filters in C++, namely a reader A for a custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter B that takes two of these outputs as input and does some calculation on them. Everything is working fine, but now I noticed that I need to pass a single string value (and 3-4 numbers) from each reader to the actual filter B doing stuff. A use case would be, for example, passing the input data filename from a reader to a filter on to a writer plugin, which can use it to come up with an appropriate output filename. Now the work-around I'm using at the moment to pass these data to the filter B is abusing CellData or PointData arrays of the vtkUnstructuredGrid generated in A. Needles to say this is not very elegant and creates other problems too (memory issues), and I'm fairly sure there is a better way to do it. I think that somehow I need to create an appropriate vtkInformationKey storing the data. However, everything I tried seems to be wrong, the key seems to disappear in the pipeline between the filters. Here is some pseudocode snippet to show what I tried to add a value to an vtkInformation. Here the reader class A: class A : vtkUnstructuredGridAlgorithm { static vtkInformationStringKey* MYSTRING(); ... } vtkInformationKeyMacro(A, MYSTRING, String); A::RequestInformation(...) { vtkSmartPointer outInfo = outputVector->GetInformationObject(0); outInfo->Set(A::MYSTRING(),"test"); } And the filter class B: #include "classA.h" class B : vtkUnstructuredGridAlgorithm { ... } B::RequestInformation(...) { vtkSmartPointer inInfo = inputVector[0]->GetInformationObject(0); if (!inInfo->Has(A::MYSTRING())) std::cout << "fail" std::endl; } I keep getting fails printed to my console :( So here is the question(s): A colleague of mine mentioned to check out the python interface, but I assume this can be done within vtk? Is this the correct approach using vtkInformation? -If yes, what am I doing wrong? I derived my code from an example which I can't seem to find anymore... Can someone give me a detailed explanation of the steps necessary, or a well commented example? -If no,is there a better, or more paraviewesque way to do it (not the python way), and can I get hint at what I need to look at please? Any other comments on how to pass user defined data along the pipeline? Thanks a lot for the help, Oli P.S.: I'm using ParaView 5.0.0 -------------- next part -------------- An HTML attachment was scrubbed... URL: From cory.quammen at kitware.com Thu Mar 24 09:03:37 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Thu, 24 Mar 2016 08:03:37 -0500 Subject: [Paraview-developers] Passing values between filters In-Reply-To: <1458809799131.52423@visus.uni-stuttgart.de> References: <1458809799131.52423@visus.uni-stuttgart.de> Message-ID: Oliver, I often pass these kinds of values through the vtkUnstructuredGrid's field data. Field data consists of data arrays, just like point or cell data, but the number of elements in the arrays is not expected to correspond to the number of points or cells, as is the case with point and cell data arrays. You can get the field data with vtkUnstructuredGrid::GetFieldData() Docs for vtkFieldData are here: http://www.vtk.org/doc/nightly/html/classvtkFieldData.html The nice thing with this approach is that the field data gets passed automatically through most filters, so you'll be able to pick up the values set by your reader in your writer, for example. HTH, Cory On Thu, Mar 24, 2016 at 3:56 AM, Oliver Fernandes wrote: > Hello all, > > So I wrote a couple of ParaView filters in C++, namely a reader A for a > custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter > B that takes two of these outputs as input and does some calculation on > them. > > Everything is working fine, but now I noticed that I need to pass a single > string value (and 3-4 numbers) from each reader to the actual filter B doing > stuff. > A use case would be, for example, passing the input data filename from a > reader to a filter on to a writer plugin, which can use it to come up with > an appropriate output filename. > > Now the work-around I'm using at the moment to pass these data to the filter > B is abusing CellData or PointData arrays of the vtkUnstructuredGrid > generated in A. Needles to say this is not very elegant and creates other > problems too (memory issues), and I'm fairly sure there is a better way to > do it. > > I think that somehow I need to create an appropriate vtkInformationKey > storing the data. > However, everything I tried seems to be wrong, the key seems to disappear in > the pipeline between the filters. > > Here is some pseudocode snippet to show what I tried to add a value to an > vtkInformation. > Here the reader class A: > > class A : vtkUnstructuredGridAlgorithm { > static vtkInformationStringKey* MYSTRING(); > ... > } > > vtkInformationKeyMacro(A, MYSTRING, String); > > A::RequestInformation(...) { > vtkSmartPointer outInfo = > outputVector->GetInformationObject(0); > outInfo->Set(A::MYSTRING(),"test"); > } > > And the filter class B: > > #include "classA.h" > class B : vtkUnstructuredGridAlgorithm { > ... > } > > B::RequestInformation(...) { > vtkSmartPointer inInfo = > inputVector[0]->GetInformationObject(0); > if (!inInfo->Has(A::MYSTRING())) > std::cout << "fail" std::endl; > > } > > I keep getting fails printed to my console :( > > So here is the question(s): > A colleague of mine mentioned to check out the python interface, but I > assume this can be done within vtk? > Is this the correct approach using vtkInformation? > > -If yes, what am I doing wrong? I derived my code from an example which I > can't seem to find anymore... > Can someone give me a detailed explanation of the steps necessary, or a well > commented example? > > -If no,is there a better, or more paraviewesque way to do it (not the python > way), and can I get hint at what I need to look at please? > Any other comments on how to pass user defined data along the pipeline? > > Thanks a lot for the help, > Oli > > P.S.: I'm using ParaView 5.0.0 > > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > -- Cory Quammen R&D Engineer Kitware, Inc. From Oliver.Fernandes at visus.uni-stuttgart.de Thu Mar 24 09:19:15 2016 From: Oliver.Fernandes at visus.uni-stuttgart.de (Oliver Fernandes) Date: Thu, 24 Mar 2016 13:19:15 +0000 Subject: [Paraview-developers] Passing values between filters In-Reply-To: References: <1458809799131.52423@visus.uni-stuttgart.de>, Message-ID: <1458825539440.49033@visus.uni-stuttgart.de> Thanks for the help, Cory! I will try out vtkFieldData member. Just for educational reasons, is there a reason to learn about vtkInformation? Or what would be a use case for that? (No details, just want to know what it's intended use is) Thanks again for the quick reply! -Oli ________________________________________ Von: Cory Quammen Gesendet: Donnerstag, 24. M?rz 2016 14:03 An: Oliver Fernandes Cc: paraview-developers at paraview.org Betreff: Re: [Paraview-developers] Passing values between filters Oliver, I often pass these kinds of values through the vtkUnstructuredGrid's field data. Field data consists of data arrays, just like point or cell data, but the number of elements in the arrays is not expected to correspond to the number of points or cells, as is the case with point and cell data arrays. You can get the field data with vtkUnstructuredGrid::GetFieldData() Docs for vtkFieldData are here: http://www.vtk.org/doc/nightly/html/classvtkFieldData.html The nice thing with this approach is that the field data gets passed automatically through most filters, so you'll be able to pick up the values set by your reader in your writer, for example. HTH, Cory On Thu, Mar 24, 2016 at 3:56 AM, Oliver Fernandes wrote: > Hello all, > > So I wrote a couple of ParaView filters in C++, namely a reader A for a > custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter > B that takes two of these outputs as input and does some calculation on > them. > > Everything is working fine, but now I noticed that I need to pass a single > string value (and 3-4 numbers) from each reader to the actual filter B doing > stuff. > A use case would be, for example, passing the input data filename from a > reader to a filter on to a writer plugin, which can use it to come up with > an appropriate output filename. > > Now the work-around I'm using at the moment to pass these data to the filter > B is abusing CellData or PointData arrays of the vtkUnstructuredGrid > generated in A. Needles to say this is not very elegant and creates other > problems too (memory issues), and I'm fairly sure there is a better way to > do it. > > I think that somehow I need to create an appropriate vtkInformationKey > storing the data. > However, everything I tried seems to be wrong, the key seems to disappear in > the pipeline between the filters. > > Here is some pseudocode snippet to show what I tried to add a value to an > vtkInformation. > Here the reader class A: > > class A : vtkUnstructuredGridAlgorithm { > static vtkInformationStringKey* MYSTRING(); > ... > } > > vtkInformationKeyMacro(A, MYSTRING, String); > > A::RequestInformation(...) { > vtkSmartPointer outInfo = > outputVector->GetInformationObject(0); > outInfo->Set(A::MYSTRING(),"test"); > } > > And the filter class B: > > #include "classA.h" > class B : vtkUnstructuredGridAlgorithm { > ... > } > > B::RequestInformation(...) { > vtkSmartPointer inInfo = > inputVector[0]->GetInformationObject(0); > if (!inInfo->Has(A::MYSTRING())) > std::cout << "fail" std::endl; > > } > > I keep getting fails printed to my console :( > > So here is the question(s): > A colleague of mine mentioned to check out the python interface, but I > assume this can be done within vtk? > Is this the correct approach using vtkInformation? > > -If yes, what am I doing wrong? I derived my code from an example which I > can't seem to find anymore... > Can someone give me a detailed explanation of the steps necessary, or a well > commented example? > > -If no,is there a better, or more paraviewesque way to do it (not the python > way), and can I get hint at what I need to look at please? > Any other comments on how to pass user defined data along the pipeline? > > Thanks a lot for the help, > Oli > > P.S.: I'm using ParaView 5.0.0 > > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > -- Cory Quammen R&D Engineer Kitware, Inc. From cory.quammen at kitware.com Thu Mar 24 09:23:52 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Thu, 24 Mar 2016 08:23:52 -0500 Subject: [Paraview-developers] Passing values between filters In-Reply-To: <1458825539440.49033@visus.uni-stuttgart.de> References: <1458809799131.52423@visus.uni-stuttgart.de> <1458825539440.49033@visus.uni-stuttgart.de> Message-ID: vtkInformation is mainly used within the pipeline execution and I would mostly consider it to be part of the internal implementation of VTK/ParaView that is subject to change. I have gotten by in 10+ years of VTK usage/development without really needing to do much with vtkInformation, so you may not benefit from learning about it. Cory On Thu, Mar 24, 2016 at 8:19 AM, Oliver Fernandes wrote: > Thanks for the help, Cory! I will try out vtkFieldData member. > Just for educational reasons, is there a reason to learn about vtkInformation? > Or what would be a use case for that? > (No details, just want to know what it's intended use is) > > Thanks again for the quick reply! > -Oli > > ________________________________________ > Von: Cory Quammen > Gesendet: Donnerstag, 24. M?rz 2016 14:03 > An: Oliver Fernandes > Cc: paraview-developers at paraview.org > Betreff: Re: [Paraview-developers] Passing values between filters > > Oliver, > > I often pass these kinds of values through the vtkUnstructuredGrid's > field data. Field data consists of data arrays, just like point or > cell data, but the number of elements in the arrays is not expected to > correspond to the number of points or cells, as is the case with point > and cell data arrays. > > You can get the field data with > > vtkUnstructuredGrid::GetFieldData() > > Docs for vtkFieldData are here: > > http://www.vtk.org/doc/nightly/html/classvtkFieldData.html > > The nice thing with this approach is that the field data gets passed > automatically through most filters, so you'll be able to pick up the > values set by your reader in your writer, for example. > > HTH, > Cory > > On Thu, Mar 24, 2016 at 3:56 AM, Oliver Fernandes > wrote: >> Hello all, >> >> So I wrote a couple of ParaView filters in C++, namely a reader A for a >> custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter >> B that takes two of these outputs as input and does some calculation on >> them. >> >> Everything is working fine, but now I noticed that I need to pass a single >> string value (and 3-4 numbers) from each reader to the actual filter B doing >> stuff. >> A use case would be, for example, passing the input data filename from a >> reader to a filter on to a writer plugin, which can use it to come up with >> an appropriate output filename. >> >> Now the work-around I'm using at the moment to pass these data to the filter >> B is abusing CellData or PointData arrays of the vtkUnstructuredGrid >> generated in A. Needles to say this is not very elegant and creates other >> problems too (memory issues), and I'm fairly sure there is a better way to >> do it. >> >> I think that somehow I need to create an appropriate vtkInformationKey >> storing the data. >> However, everything I tried seems to be wrong, the key seems to disappear in >> the pipeline between the filters. >> >> Here is some pseudocode snippet to show what I tried to add a value to an >> vtkInformation. >> Here the reader class A: >> >> class A : vtkUnstructuredGridAlgorithm { >> static vtkInformationStringKey* MYSTRING(); >> ... >> } >> >> vtkInformationKeyMacro(A, MYSTRING, String); >> >> A::RequestInformation(...) { >> vtkSmartPointer outInfo = >> outputVector->GetInformationObject(0); >> outInfo->Set(A::MYSTRING(),"test"); >> } >> >> And the filter class B: >> >> #include "classA.h" >> class B : vtkUnstructuredGridAlgorithm { >> ... >> } >> >> B::RequestInformation(...) { >> vtkSmartPointer inInfo = >> inputVector[0]->GetInformationObject(0); >> if (!inInfo->Has(A::MYSTRING())) >> std::cout << "fail" std::endl; >> >> } >> >> I keep getting fails printed to my console :( >> >> So here is the question(s): >> A colleague of mine mentioned to check out the python interface, but I >> assume this can be done within vtk? >> Is this the correct approach using vtkInformation? >> >> -If yes, what am I doing wrong? I derived my code from an example which I >> can't seem to find anymore... >> Can someone give me a detailed explanation of the steps necessary, or a well >> commented example? >> >> -If no,is there a better, or more paraviewesque way to do it (not the python >> way), and can I get hint at what I need to look at please? >> Any other comments on how to pass user defined data along the pipeline? >> >> Thanks a lot for the help, >> Oli >> >> P.S.: I'm using ParaView 5.0.0 >> >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> > > > > -- > Cory Quammen > R&D Engineer > Kitware, Inc. -- Cory Quammen R&D Engineer Kitware, Inc. From andy.bauer at kitware.com Thu Mar 24 10:09:35 2016 From: andy.bauer at kitware.com (Andy Bauer) Date: Thu, 24 Mar 2016 10:09:35 -0400 Subject: [Paraview-developers] Disabled buttons on ParaView 5.0.0 on Mac Os X 10.11.3 In-Reply-To: <1458792594769-4230.post@n6.nabble.com> References: <1458792594769-4230.post@n6.nabble.com> Message-ID: Hi, Have you hit the "Apply" button after you've loaded or created your data set? What kind of data set have you loaded? Certain filters will be disabled if you don't have a data set which they support (e..g streamlines isn't available if there isn't a vector/3 component field to compute over). Cheers, Andy On Thu, Mar 24, 2016 at 12:09 AM, nazarian2 wrote: > Hi all, > > I'm currently using ParaView 5.0.0 on Mac Os X 10.11.3 and I'm having > difficulties with the attached example: all the filters appear disabled, as > well as the third row buttons (Calculator, contour, clip, slice, etc..). > > What am I doing wrong? > > Regards. > > < > http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/file/n4229/brazil_2d_nl.zip > > > > > > > -- > View this message in context: > http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Disabled-buttons-on-ParaView-5-0-0-on-Mac-Os-X-10-11-3-tp4230.html > Sent from the The Unofficial ParaView Developers Forum mailing list > archive at Nabble.com. > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kmorel at sandia.gov Thu Mar 24 10:51:15 2016 From: kmorel at sandia.gov (Moreland, Kenneth) Date: Thu, 24 Mar 2016 14:51:15 +0000 Subject: [Paraview-developers] Passing values between filters Message-ID: Oliver, One way to pass this type of information is through the unassociated field data. vtkDataObject (from which vtkUnstructuredGrid inherits) has a method named GetFieldData() (http://www.vtk.org/doc/nightly/html/classvtkDataObject.html#a72f64ab58126f72e048410751091011b). This returns a vtkFieldData object, which is a superclass of vtkCellData and vtkPointData. You can store your data there in the same way you are storing in CellData or PointData, but it should cause less problems. -Ken From: Paraview-developers > on behalf of Oliver Fernandes > Date: Thursday, March 24, 2016 at 2:56 AM To: "paraview-developers at paraview.org" > Subject: [EXTERNAL] [Paraview-developers] Passing values between filters Hello all, So I wrote a couple of ParaView filters in C++, namely a reader A for a custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter B that takes two of these outputs as input and does some calculation on them. Everything is working fine, but now I noticed that I need to pass a single string value (and 3-4 numbers) from each reader to the actual filter B doing stuff. A use case would be, for example, passing the input data filename from a reader to a filter on to a writer plugin, which can use it to come up with an appropriate output filename. Now the work-around I'm using at the moment to pass these data to the filter B is abusing CellData or PointData arrays of the vtkUnstructuredGrid generated in A. Needles to say this is not very elegant and creates other problems too (memory issues), and I'm fairly sure there is a better way to do it. I think that somehow I need to create an appropriate vtkInformationKey storing the data. However, everything I tried seems to be wrong, the key seems to disappear in the pipeline between the filters. Here is some pseudocode snippet to show what I tried to add a value to an vtkInformation. Here the reader class A: class A : vtkUnstructuredGridAlgorithm { static vtkInformationStringKey* MYSTRING(); ... } vtkInformationKeyMacro(A, MYSTRING, String); A::RequestInformation(...) { vtkSmartPointer outInfo = outputVector->GetInformationObject(0); outInfo->Set(A::MYSTRING(),"test"); } And the filter class B: #include "classA.h" class B : vtkUnstructuredGridAlgorithm { ... } B::RequestInformation(...) { vtkSmartPointer inInfo = inputVector[0]->GetInformationObject(0); if (!inInfo->Has(A::MYSTRING())) std::cout << "fail" std::endl; } I keep getting fails printed to my console :( So here is the question(s): A colleague of mine mentioned to check out the python interface, but I assume this can be done within vtk? Is this the correct approach using vtkInformation? -If yes, what am I doing wrong? I derived my code from an example which I can't seem to find anymore... Can someone give me a detailed explanation of the steps necessary, or a well commented example? -If no,is there a better, or more paraviewesque way to do it (not the python way), and can I get hint at what I need to look at please? Any other comments on how to pass user defined data along the pipeline? Thanks a lot for the help, Oli P.S.: I'm using ParaView 5.0.0 -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.lonie at kitware.com Thu Mar 24 14:20:59 2016 From: david.lonie at kitware.com (David Lonie) Date: Thu, 24 Mar 2016 14:20:59 -0400 Subject: [Paraview-developers] Wrong image slice representation of cells fields since ParaView 4.4.0 In-Reply-To: References: Message-ID: Thanks for the report! This VTK patch fixes the issue: https://gitlab.kitware.com/vtk/vtk/merge_requests/1370 On Wed, Mar 23, 2016 at 1:17 PM, Utkarsh Ayachit < utkarsh.ayachit at kitware.com> wrote: > I certainly can reproduce this issue. Looking into it. I'll report > back once I know more. > > Utkarsh > > On Wed, Mar 23, 2016 at 11:41 AM, Christophe Bourcier > wrote: > > Hi, > > > > On an ImageData, the result of an image slice representation is wrong > > for cells fields. It is reproduced on ParaView 4.4.0 to 5.0.1-rc2, it > > worked fine in 4.3.1. And it works fine with nodes fields. > > > > See pictures in links [1] [2]. > > > > In debug mode, I have an error in the assertion in > > VTK/Common/DataModel/vtkExtractStructuredGridHelper.cxx :525 : void > > vtkExtractStructuredGridHelper::CopyCellData(int*, int*, vtkCellData*, > > vtkCellData*): > > > > > > // Sanity checks > > assert( "pre: srcStart out of bounds" && (srcStart >= 0) && > > (srcStart < inSize) ); > > > > > > Any clue on how to fix this? > > > > Steps to reproduce: > > - load the vtk file in link [3] > > - in Display, change Representation from Outline to Slice > > > > Thanks, > > > > Christophe > > > > [1] http://bourcier.eu/slice_431_ok.png > > [2] http://bourcier.eu/slice_501rc2_pb.png > > [3] http://bourcier.eu/arlequin_N2_r16.vtk > > _______________________________________________ > > Powered by www.kitware.com > > > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > > > Follow this link to subscribe/unsubscribe: > > http://public.kitware.com/mailman/listinfo/paraview-developers > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > -------------- next part -------------- An HTML attachment was scrubbed... URL: From cory.quammen at kitware.com Thu Mar 24 14:59:07 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Thu, 24 Mar 2016 14:59:07 -0400 Subject: [Paraview-developers] "Unable to build k-d tree structure" error with D3 filter and PV 4.4.0 on large datasets In-Reply-To: <5178E5D9-D3BB-471B-AAA1-5D22B458906F@aia.rwth-aachen.de> References: <5178E5D9-D3BB-471B-AAA1-5D22B458906F@aia.rwth-aachen.de> Message-ID: Hi Michael, Sorry to just getting around to responding to your question. It looks like you are running out of memory on at least one compute node, but I'm guessing you already figured that out. You'll either need to reduce the data size somehow, increase the memory per core, or increase the number of cores. HTH, Cory On Tue, Mar 8, 2016 at 9:36 AM, Schlottke-Lakemper, Michael < m.schlottke-lakemper at aia.rwth-aachen.de> wrote: > Dear all, > > We have recently come across the following error when using the D3 filter > on large and distributed vtkUnstructuredGrids. We have our own reader that > locally creates a vtkUnstructuredGrid by reading the according subdomain > from a file in netcdf format and I would like to use the D3 filter to have > halo cells available on all domains (e.g. for gradient computations). For > small scale meshes containing about 260 x 10^6 cells there is no problem. > However, when I want to execute the filter on 1.8 x 10^9 cells I get the > following error messages: > > Warning: In > /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/Parallel/vtkPKdTree.cxx, > line 207 > vtkPKdTree (0x2943d0c0): (process 490) memory allocation on my node > (BreadthFirstDivide) > > ERROR: In > /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/ParallelMPI/vtkDistributedDataFilter.cxx, > line 796 > vtkDistributedDataFilter (0x29433cd0): Unable to build k-d tree structure > > > the same is happening with other remote nodes, i.e., > > Warning: In > /dev/shm/domini/ParaView/4.4.0/intel-para-2015.07-mpi-devel/ParaView-v4.4.0-source/VTK/Filters/Parallel/vtkPKdTree.cxx, > line 207 > vtkPKdTree (0x2a9fdf20): (process 137) memory allocation on a remote node > (BreadthFirstDivide) > > > I run a server on 576 cores using ParaView version 4.4.0. The program does > not crash after this error, however, it is not creating any output. I have > tracked down the error to the following (potential) location. I noticed > that this error seems to be generated in the function BreadthFirstDivision, > which somehow returns the error code 1, which is then in AllCheckForFailure > summarized to a value > 0 on all MPI ranks leading to the error message. I > also noticed that this cannot be a VTKERROR as it is not printed to the > terminal. I believe that the error might reside in the check for midpt < > -1, which would also return an error code of 1 in BreatdhFirstDivision. > > I hope you can help me with this problem. Thanks a lot in advance! > > Michael > > > -- > Michael Schlottke-Lakemper > > Chair of Fluid Mechanics and Institute of Aerodynamics > RWTH Aachen University > W?llnerstra?e 5a > 52062 Aachen > Germany > > Phone: +49 (241) 80 95188 > Fax: +49 (241) 80 92257 > Mail: m.schlottke-lakemper at aia.rwth-aachen.de > Web: http://www.aia.rwth-aachen.de > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: > http://markmail.org/search/?q=Paraview-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/paraview-developers > > -- Cory Quammen R&D Engineer Kitware, Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: From nazarian2 at gmail.com Thu Mar 24 23:39:27 2016 From: nazarian2 at gmail.com (Francisco Nazar) Date: Fri, 25 Mar 2016 00:39:27 -0300 Subject: [Paraview-developers] Disabled buttons on ParaView 5.0.0 on Mac Os X 10.11.3 In-Reply-To: References: <1458792594769-4230.post@n6.nabble.com> Message-ID: Hi, Yes I did. In any case, I just downloaded version 5.0.1 and it worked like a charm. Strange. Regards. 2016-03-24 11:09 GMT-03:00 Andy Bauer : > Hi, > > Have you hit the "Apply" button after you've loaded or created your data > set? > > What kind of data set have you loaded? Certain filters will be disabled if > you don't have a data set which they support (e..g streamlines isn't > available if there isn't a vector/3 component field to compute over). > > Cheers, > Andy > > On Thu, Mar 24, 2016 at 12:09 AM, nazarian2 wrote: > >> Hi all, >> >> I'm currently using ParaView 5.0.0 on Mac Os X 10.11.3 and I'm having >> difficulties with the attached example: all the filters appear disabled, >> as >> well as the third row buttons (Calculator, contour, clip, slice, etc..). >> >> What am I doing wrong? >> >> Regards. >> >> < >> http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/file/n4229/brazil_2d_nl.zip >> > >> >> >> >> >> -- >> View this message in context: >> http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Disabled-buttons-on-ParaView-5-0-0-on-Mac-Os-X-10-11-3-tp4230.html >> Sent from the The Unofficial ParaView Developers Forum mailing list >> archive at Nabble.com. >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joachim.pouderoux at kitware.com Fri Mar 25 00:36:31 2016 From: joachim.pouderoux at kitware.com (Joachim Pouderoux) Date: Thu, 24 Mar 2016 22:36:31 -0600 Subject: [Paraview-developers] "Unable to build k-d tree structure" error with D3 filter and PV 4.4.0 on large datasets In-Reply-To: <5178E5D9-D3BB-471B-AAA1-5D22B458906F@aia.rwth-aachen.de> References: <5178E5D9-D3BB-471B-AAA1-5D22B458906F@aia.rwth-aachen.de> Message-ID: Michael, If you just intend to compute ghost cells on your unstructured grid, I would recommend you to use the new "Generate Ghost Cells" filter available in ParaView 5.0. It requires less memory than D3 and will perform faster if you don't require data redistribution. Having it as a plugin on PV 4.4 should be straightforward as this is a single class (vtkPUnstructuredGridGhostCellsGenerator) you will have to pick in VTK 7.0. Best, Joachim -------------- next part -------------- An HTML attachment was scrubbed... URL: From Oliver.Fernandes at visus.uni-stuttgart.de Fri Mar 25 08:26:43 2016 From: Oliver.Fernandes at visus.uni-stuttgart.de (Oliver Fernandes) Date: Fri, 25 Mar 2016 12:26:43 +0000 Subject: [Paraview-developers] Passing values between filters In-Reply-To: References: <1458809799131.52423@visus.uni-stuttgart.de> <1458825539440.49033@visus.uni-stuttgart.de>, Message-ID: <1458908786753.24970@visus.uni-stuttgart.de> Ok good, thats exactly what I wanted to hear :) Thanks again to Cory and Kenneth! -Oli ________________________________________ Von: Cory Quammen Gesendet: Donnerstag, 24. M?rz 2016 14:23 An: Oliver Fernandes Cc: paraview-developers at paraview.org Betreff: Re: [Paraview-developers] Passing values between filters vtkInformation is mainly used within the pipeline execution and I would mostly consider it to be part of the internal implementation of VTK/ParaView that is subject to change. I have gotten by in 10+ years of VTK usage/development without really needing to do much with vtkInformation, so you may not benefit from learning about it. Cory On Thu, Mar 24, 2016 at 8:19 AM, Oliver Fernandes wrote: > Thanks for the help, Cory! I will try out vtkFieldData member. > Just for educational reasons, is there a reason to learn about vtkInformation? > Or what would be a use case for that? > (No details, just want to know what it's intended use is) > > Thanks again for the quick reply! > -Oli > > ________________________________________ > Von: Cory Quammen > Gesendet: Donnerstag, 24. M?rz 2016 14:03 > An: Oliver Fernandes > Cc: paraview-developers at paraview.org > Betreff: Re: [Paraview-developers] Passing values between filters > > Oliver, > > I often pass these kinds of values through the vtkUnstructuredGrid's > field data. Field data consists of data arrays, just like point or > cell data, but the number of elements in the arrays is not expected to > correspond to the number of points or cells, as is the case with point > and cell data arrays. > > You can get the field data with > > vtkUnstructuredGrid::GetFieldData() > > Docs for vtkFieldData are here: > > http://www.vtk.org/doc/nightly/html/classvtkFieldData.html > > The nice thing with this approach is that the field data gets passed > automatically through most filters, so you'll be able to pick up the > values set by your reader in your writer, for example. > > HTH, > Cory > > On Thu, Mar 24, 2016 at 3:56 AM, Oliver Fernandes > wrote: >> Hello all, >> >> So I wrote a couple of ParaView filters in C++, namely a reader A for a >> custom file format deriving from vtkUnstructuredGridAlgorithm, and a filter >> B that takes two of these outputs as input and does some calculation on >> them. >> >> Everything is working fine, but now I noticed that I need to pass a single >> string value (and 3-4 numbers) from each reader to the actual filter B doing >> stuff. >> A use case would be, for example, passing the input data filename from a >> reader to a filter on to a writer plugin, which can use it to come up with >> an appropriate output filename. >> >> Now the work-around I'm using at the moment to pass these data to the filter >> B is abusing CellData or PointData arrays of the vtkUnstructuredGrid >> generated in A. Needles to say this is not very elegant and creates other >> problems too (memory issues), and I'm fairly sure there is a better way to >> do it. >> >> I think that somehow I need to create an appropriate vtkInformationKey >> storing the data. >> However, everything I tried seems to be wrong, the key seems to disappear in >> the pipeline between the filters. >> >> Here is some pseudocode snippet to show what I tried to add a value to an >> vtkInformation. >> Here the reader class A: >> >> class A : vtkUnstructuredGridAlgorithm { >> static vtkInformationStringKey* MYSTRING(); >> ... >> } >> >> vtkInformationKeyMacro(A, MYSTRING, String); >> >> A::RequestInformation(...) { >> vtkSmartPointer outInfo = >> outputVector->GetInformationObject(0); >> outInfo->Set(A::MYSTRING(),"test"); >> } >> >> And the filter class B: >> >> #include "classA.h" >> class B : vtkUnstructuredGridAlgorithm { >> ... >> } >> >> B::RequestInformation(...) { >> vtkSmartPointer inInfo = >> inputVector[0]->GetInformationObject(0); >> if (!inInfo->Has(A::MYSTRING())) >> std::cout << "fail" std::endl; >> >> } >> >> I keep getting fails printed to my console :( >> >> So here is the question(s): >> A colleague of mine mentioned to check out the python interface, but I >> assume this can be done within vtk? >> Is this the correct approach using vtkInformation? >> >> -If yes, what am I doing wrong? I derived my code from an example which I >> can't seem to find anymore... >> Can someone give me a detailed explanation of the steps necessary, or a well >> commented example? >> >> -If no,is there a better, or more paraviewesque way to do it (not the python >> way), and can I get hint at what I need to look at please? >> Any other comments on how to pass user defined data along the pipeline? >> >> Thanks a lot for the help, >> Oli >> >> P.S.: I'm using ParaView 5.0.0 >> >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: >> http://markmail.org/search/?q=Paraview-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/paraview-developers >> > > > > -- > Cory Quammen > R&D Engineer > Kitware, Inc. -- Cory Quammen R&D Engineer Kitware, Inc. From christophe.bourcier.pv at gmail.com Fri Mar 25 08:27:05 2016 From: christophe.bourcier.pv at gmail.com (Christophe Bourcier) Date: Fri, 25 Mar 2016 13:27:05 +0100 Subject: [Paraview-developers] Wrong image slice representation of cells fields since ParaView 4.4.0 In-Reply-To: References: Message-ID: Thank you for the fix. It works now. Christophe 2016-03-25 12:47 GMT+01:00 Christophe Bourcier : > Thank you for the fix. It works now. > > Christophe > > 2016-03-24 19:20 GMT+01:00 David Lonie : >> Thanks for the report! This VTK patch fixes the issue: >> >> https://gitlab.kitware.com/vtk/vtk/merge_requests/1370 >> >> On Wed, Mar 23, 2016 at 1:17 PM, Utkarsh Ayachit >> wrote: >>> >>> I certainly can reproduce this issue. Looking into it. I'll report >>> back once I know more. >>> >>> Utkarsh >>> >>> On Wed, Mar 23, 2016 at 11:41 AM, Christophe Bourcier >>> wrote: >>> > Hi, >>> > >>> > On an ImageData, the result of an image slice representation is wrong >>> > for cells fields. It is reproduced on ParaView 4.4.0 to 5.0.1-rc2, it >>> > worked fine in 4.3.1. And it works fine with nodes fields. >>> > >>> > See pictures in links [1] [2]. >>> > >>> > In debug mode, I have an error in the assertion in >>> > VTK/Common/DataModel/vtkExtractStructuredGridHelper.cxx :525 : void >>> > vtkExtractStructuredGridHelper::CopyCellData(int*, int*, vtkCellData*, >>> > vtkCellData*): >>> > >>> > >>> > // Sanity checks >>> > assert( "pre: srcStart out of bounds" && (srcStart >= 0) && >>> > (srcStart < inSize) ); >>> > >>> > >>> > Any clue on how to fix this? >>> > >>> > Steps to reproduce: >>> > - load the vtk file in link [3] >>> > - in Display, change Representation from Outline to Slice >>> > >>> > Thanks, >>> > >>> > Christophe >>> > >>> > [1] http://bourcier.eu/slice_431_ok.png >>> > [2] http://bourcier.eu/slice_501rc2_pb.png >>> > [3] http://bourcier.eu/arlequin_N2_r16.vtk >>> > _______________________________________________ >>> > Powered by www.kitware.com >>> > >>> > Visit other Kitware open-source projects at >>> > http://www.kitware.com/opensource/opensource.html >>> > >>> > Search the list archives at: >>> > http://markmail.org/search/?q=Paraview-developers >>> > >>> > Follow this link to subscribe/unsubscribe: >>> > http://public.kitware.com/mailman/listinfo/paraview-developers >>> _______________________________________________ >>> Powered by www.kitware.com >>> >>> Visit other Kitware open-source projects at >>> http://www.kitware.com/opensource/opensource.html >>> >>> Search the list archives at: >>> http://markmail.org/search/?q=Paraview-developers >>> >>> Follow this link to subscribe/unsubscribe: >>> http://public.kitware.com/mailman/listinfo/paraview-developers >> >> From utkarsh.ayachit at kitware.com Mon Mar 28 11:11:47 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Mon, 28 Mar 2016 11:11:47 -0400 Subject: [Paraview-developers] ANN: ParaView 5.0.1 Tagged Message-ID: Folks, ParaView 5.0.1 has been tagged! Git tag name: v5.0.1 Source tar/zip files are available for download: http://www.paraview.org/download/ Binaries will be available for testing within the next few days. Utkarsh From utkarsh.ayachit at kitware.com Wed Mar 30 14:23:01 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Wed, 30 Mar 2016 14:23:01 -0400 Subject: [Paraview-developers] ANN: ParaView 5.0.1 available for download Message-ID: Folks, ParaView 5.0.1 is now available for download [1]. Please refer to the release notes on the Kitware blog [2]. As always, we look forward to your feedback [3]. Also stay tuned to the Kitware Blog [4] for upcoming features and enhancements to ParaView, ParaView Catalyst, ParaViewWeb and much more! - ParaView Team [1] http://www.paraview.org/download/ [2] https://blog.kitware.com/paraview-5-0-1-release-notes/ [3] http://paraview.uservoice.com [4] https://blog.kitware.com/