Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Lucas, Do you mind if I use the coproc files you sent me as an example for Catalyst? Thanks, Andy On Tue, Mar 26, 2013 at 3:29 PM, Andy Bauer andy.ba...@kitware.com wrote: Hi Lucas, I was able to replicate your issue on the mac. There is a fix for it that is waiting a code review and dashboard feedback. I'll let you know when it's in the main ParaView repo (which should be a couple of days). Andy On Mon, Mar 25, 2013 at 5:56 PM, Andy Bauer andy.ba...@kitware.comwrote: Hi Lucas, I'm getting rid of the history as it was getting really long and making it hard to read. There is a bug with the vtkMPIImageReader -- http://paraview.org/Bug/view.php?id=13968. I think it should be fixed in a couple of days. I'm not getting the issues with the script that was in the tarball on my linux machine but I'll try it on a Mac. I did have to have the python script in the directory I was working in (a relative path to it didn't work when I tried to run ./CoProcessingExample ../coproc.py). Andy === Hi Andy, So I got the repo version of ParaView built on both my Mac laptop (10.7.5) and Mac desktop (10.8.3). The CoProc code from the tar file you sent compiles just fine on both machines. On execution, both machines generate the error: lucass-mac-pro:coprocf90 lucaspettey$ ./CoProcessingExample Traceback (most recent call last): File string, line 2, in module File /Users/lucaspettey/codes/BEC/ coproc.py, line 9, in module import vtkCoProcessorPython ImportError: No module named vtkCoProcessorPython Traceback (most recent call last): File string, line 2, in module So I thought that I would remake the images and generate a new coprocessing script. The second problem is that ParaView crashes when it tries to open .RAW files on either machine. I get the following error when I select the file: ERROR in .../vtkSIProxy.cxx line 307 vtkSISourceProxy: failed to create vtkMPIImageReader XDMF and other file types seem to open ok. These files opened fine under the release version of ParaView. Thanks, Lucas ___ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView Follow this link to subscribe/unsubscribe: http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Andy, I do not mind at all. In fact, I am quite happy to have contributed something given all the help that you have provided to us. Thanks, Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Wednesday, March 27, 2013 3:31 PM To: paraview@paraview.org; Pettey . Lucas Subject: Re: CoProcessing with Fortran and ParaView 3.98 Hi Lucas, Do you mind if I use the coproc files you sent me as an example for Catalyst? Thanks, Andy On Tue, Mar 26, 2013 at 3:29 PM, Andy Bauer andy.ba...@kitware.commailto:andy.ba...@kitware.com wrote: Hi Lucas, I was able to replicate your issue on the mac. There is a fix for it that is waiting a code review and dashboard feedback. I'll let you know when it's in the main ParaView repo (which should be a couple of days). Andy On Mon, Mar 25, 2013 at 5:56 PM, Andy Bauer andy.ba...@kitware.commailto:andy.ba...@kitware.com wrote: Hi Lucas, I'm getting rid of the history as it was getting really long and making it hard to read. There is a bug with the vtkMPIImageReader -- http://paraview.org/Bug/view.php?id=13968. I think it should be fixed in a couple of days. I'm not getting the issues with the script that was in the tarball on my linux machine but I'll try it on a Mac. I did have to have the python script in the directory I was working in (a relative path to it didn't work when I tried to run ./CoProcessingExample ../coproc.py). Andy === Hi Andy, So I got the repo version of ParaView built on both my Mac laptop (10.7.5) and Mac desktop (10.8.3). The CoProc code from the tar file you sent compiles just fine on both machines. On execution, both machines generate the error: lucass-mac-pro:coprocf90 lucaspettey$ ./CoProcessingExample Traceback (most recent call last): File string, line 2, in module File /Users/lucaspettey/codes/BEC/ coproc.py, line 9, in module import vtkCoProcessorPython ImportError: No module named vtkCoProcessorPython Traceback (most recent call last): File string, line 2, in module So I thought that I would remake the images and generate a new coprocessing script. The second problem is that ParaView crashes when it tries to open .RAW files on either machine. I get the following error when I select the file: ERROR in .../vtkSIProxy.cxx line 307 vtkSISourceProxy: failed to create vtkMPIImageReader XDMF and other file types seem to open ok. These files opened fine under the release version of ParaView. Thanks, Lucas This electronic message transmission and any attachments that accompany it contain information from DRC® (Dynamics Research Corporation) or its subsidiaries, or the intended recipient, which is privileged, proprietary, business confidential, or otherwise protected from disclosure and is the exclusive property of DRC and/or the intended recipient. The information in this email is solely intended for the use of the individual or entity that is the intended recipient. If you are not the intended recipient, any use, dissemination, distribution, retention, or copying of this communication, attachments, or substance is prohibited. If you have received this electronic transmission in error, please immediately reply to the author via email that you received the message by mistake and also promptly and permanently delete this message and all copies of this email and any attachments. We thank you for your assistance and apologize for any inconvenience. ___ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView Follow this link to subscribe/unsubscribe: http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Lucas, I was able to replicate your issue on the mac. There is a fix for it that is waiting a code review and dashboard feedback. I'll let you know when it's in the main ParaView repo (which should be a couple of days). Andy On Mon, Mar 25, 2013 at 5:56 PM, Andy Bauer andy.ba...@kitware.com wrote: Hi Lucas, I'm getting rid of the history as it was getting really long and making it hard to read. There is a bug with the vtkMPIImageReader -- http://paraview.org/Bug/view.php?id=13968. I think it should be fixed in a couple of days. I'm not getting the issues with the script that was in the tarball on my linux machine but I'll try it on a Mac. I did have to have the python script in the directory I was working in (a relative path to it didn't work when I tried to run ./CoProcessingExample ../coproc.py). Andy === Hi Andy, So I got the repo version of ParaView built on both my Mac laptop (10.7.5) and Mac desktop (10.8.3). The CoProc code from the tar file you sent compiles just fine on both machines. On execution, both machines generate the error: lucass-mac-pro:coprocf90 lucaspettey$ ./CoProcessingExample Traceback (most recent call last): File string, line 2, in module File /Users/lucaspettey/codes/BEC/ coproc.py, line 9, in module import vtkCoProcessorPython ImportError: No module named vtkCoProcessorPython Traceback (most recent call last): File string, line 2, in module So I thought that I would remake the images and generate a new coprocessing script. The second problem is that ParaView crashes when it tries to open .RAW files on either machine. I get the following error when I select the file: ERROR in .../vtkSIProxy.cxx line 307 vtkSISourceProxy: failed to create vtkMPIImageReader XDMF and other file types seem to open ok. These files opened fine under the release version of ParaView. Thanks, Lucas ___ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView Follow this link to subscribe/unsubscribe: http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Andy, I have attached a tar file with all 3 source code files, the Python script and the CMakeLists.txt file used to build the executable. Thanks so much for helping with this. Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.com] Sent: Friday, March 22, 2013 8:01 AM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Lucas, Could you send me your CMakeLists.txt or your Makefile used to build this? I think I should be able to diagnose the problem fairly quickly in a debugger. Andy On Thu, Mar 21, 2013 at 4:44 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: Also something weird that I have noted. The Initialize Helper Script line prints twice when I run the Fortran code and no print statements anywhere in my Python script execute (not even if they are at the top). When I run the C++ example located here: http://www.itk.org/Wiki/Coprocessing_example, the Initialize Helper Script only appears once and print statements in the Python script execute normally. I'm not sure if this helps or makes things worse. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.com] Sent: Thursday, March 21, 2013 3:12 PM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Nothing printed at all. Does anything special have to be done during the compile/link steps for the python script to work properly? Thanks, Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 3:00 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 How about trying: psi01_.UpdatePipeline() print psi01_.GetDataInformation().DataInformation.GetNumberOfPoints() right about where the writer was created. This will give print out the number of points in the data set that you have. Andy On Thu, Mar 21, 2013 at 3:47 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Sorry, Andy, that didn't seem to work either. Let me know what information would be useful to print. Thanks Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:36 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:45 AM To: Andy Bauer Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 So after some troubleshooting it seems that C++ doesn't like the complex data type the way Fortran is sending it. The code runs when I just send a real value. Images appear, but of course they are meaningless since it isn't the correct data. I may try passing the real and imaginary separate. I may also look into row/column order issues although this grid is 100x100x100 so it shouldn't affect anything. Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:12 AM To: Andy Bauer Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Andy and Berk, Here is the error that I am receiving. I know that it is initializing ok and failing at the first call to the co-processor. I have attached my python script output as well as my test code and adaptors. CoProcessingExample(19247,0x7fff77470180) malloc: *** mmap(size=3739605888335872) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Memory allocation failed Thanks for the help, Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Wednesday, March 20, 2013 9:08 AM To: Pettey . Lucas Cc: Berk Geveci; paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Lucas, Can you share your co-processing script and any run output? There may be warnings in the output or info in the script that will help figure that out. Thanks, Andy On Wed, Mar 20, 2013 at 10:04 AM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.com wrote: Hi Berk, Thanks for the help. I found that this example is written based on an earlier version of VTK so some of the function calls have changed. Based on your advice and some of the examples contained in the ParaView wiki, I have gotten it to compile and run. Now I just have to figure out why it isn't generating any VTK objects Probably something wrong with my python co-processing script. Thanks Lucas From: Berk Geveci [berk.gev...@kitware.commailto:berk.gev...@kitware.com ] Sent: Tuesday, March 19, 2013 6:48 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 You can simply delete this line: Grid-SetNumberOfScalarComponents(2); It is redundant given the following later: field-SetNumberOfComponents(2); You also need to #include vtkSmartPointer.h. Best, -berk On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hello everyone, I am trying to run a simple Fortran coprocessing example with a complex data type. I am following the example here: http://en.wikibooks.org/wiki
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:45 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 So after some troubleshooting it seems that C++ doesn't like the complex data type the way Fortran is sending it. The code runs when I just send a real value. Images appear, but of course they are meaningless since it isn't the correct data. I may try passing the real and imaginary separate. I may also look into row/column order issues although this grid is 100x100x100 so it shouldn't affect anything. Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:12 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Andy and Berk, Here is the error that I am receiving. I know that it is initializing ok and failing at the first call to the co-processor. I have attached my python script output as well as my test code and adaptors. CoProcessingExample(19247,0x7fff77470180) malloc: *** mmap(size=3739605888335872) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Memory allocation failed Thanks for the help, Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Wednesday, March 20, 2013 9:08 AM To: Pettey . Lucas Cc: Berk Geveci; paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Lucas, Can you share your co-processing script and any run output? There may be warnings in the output or info in the script that will help figure that out. Thanks, Andy On Wed, Mar 20, 2013 at 10:04 AM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Berk, Thanks for the help. I found
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:45 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 So after some troubleshooting it seems that C++ doesn't like the complex data type the way Fortran is sending it. The code runs when I just send a real value. Images appear, but of course they are meaningless since it isn't the correct data. I may try passing the real and imaginary separate. I may also look into row/column order issues although this grid is 100x100x100 so it shouldn't affect anything. Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:12 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Andy and Berk, Here is the error that I am receiving. I know that it is initializing ok and failing at the first call to the co-processor. I have attached my python script output as well as my test code and adaptors. CoProcessingExample(19247,0x7fff77470180) malloc: *** mmap(size=3739605888335872) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Operating system error: Cannot allocate memory Memory allocation failed Thanks for the help, Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Wednesday, March 20, 2013 9:08 AM To: Pettey . Lucas Cc: Berk Geveci; paraview [paraview@paraview.orgmailto: paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hi Lucas, Can you share your co-processing script and any run output? There may
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:45 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 So after some troubleshooting it seems that C++ doesn't like the complex data type the way Fortran is sending it. The code runs when I just send a real value. Images appear, but of course they are meaningless since it isn't the correct data. I may try passing the real and imaginary separate. I may also look into row/column order issues although this grid is 100x100x100 so it shouldn't affect anything. Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:12 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com ] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun...@paraview.orgmailto: paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun...@paraview.org] on behalf of Pettey . Lucas [ lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto: lpet...@drc.com] Sent: Wednesday, March 20, 2013 11:45 AM To: Andy Bauer Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto: paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 So after some troubleshooting it seems that C++ doesn't like the complex data type the way Fortran is sending it. The code runs when I just send a real value. Images appear, but of course they are meaningless since it isn't the correct data. I may try passing the real and imaginary separate. I may also look into row/column order issues although this grid is 100x100x100 so it shouldn't affect anything. Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun...@paraview.orgmailto: paraview-boun...@paraview.org
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Sorry, Andy, that didn't seem to work either. Let me know what information would be useful to print. Thanks Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:36 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
How about trying: psi01_.UpdatePipeline() print psi01_.GetDataInformation().DataInformation.GetNumberOfPoints() right about where the writer was created. This will give print out the number of points in the data set that you have. Andy On Thu, Mar 21, 2013 at 3:47 PM, Pettey . Lucas lpet...@drc.com wrote: Sorry, Andy, that didn't seem to work either. Let me know what information would be useful to print. Thanks Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:36 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun...@paraview.orgmailto: paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [ andy.ba...@kitware.commailto:andy.ba...@kitware.commailto: andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto: paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work for Fortran77? That's a neat little trick! Andy On Wed, Mar 20, 2013 at 5:29 PM, Pettey . Lucas lpet...@drc.commailto: lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto: lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto: lpet...@drc.com wrote: I have solved the C++ to Fortran array issues, but I am still not getting a meaningful visualization. I have attached the latest files, including a screenshot of the exported state and the resulting image that is produced by the coprocessor. The coprocessor is generating an error: ERROR: In /Users/lucaspettey/ParaView-git/ParaView/VTK/Filters/General/vtkTableBasedClipDataSet.cxx, line 1956 vtkPVClipDataSet (0x7feaa4998550): no input scalars. This error is not appearing at the time=0 co-process, but it is appearing each time step the co-processor is being asked to write an image. All of the images are the same blank axes as in the attached image0.png. I'm not sure what to try next so any help is appreciated. Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org mailto:paraview-boun
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Nothing printed at all. Does anything special have to be done during the compile/link steps for the python script to work properly? Thanks, Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 3:00 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 How about trying: psi01_.UpdatePipeline() print psi01_.GetDataInformation().DataInformation.GetNumberOfPoints() right about where the writer was created. This will give print out the number of points in the data set that you have. Andy On Thu, Mar 21, 2013 at 3:47 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: Sorry, Andy, that didn't seem to work either. Let me know what information would be useful to print. Thanks Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:36 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 I think the problem is that: Clip1 = Clip( guiName=Clip1, InsideOut=0, Scalars=['POINTS', ''], Value=95.0, UseValueAsOffset=0, Crinkleclip=0, ClipType=Scalar ) Should probably have the name of the point data field set in it. Do this with (assuming psi01 is the name of the field you want to clip with respect to: Scalars=['POINTS', 'psi01'] My guess is that the scalar field isn't getting set properly. Try adding in: w = XMLPImageDataWriter(Input=psi01_, FileName = 'fullgrid.pvti') w.UpdatePipeline() right after the CreateProducer() method to see what's in your grid. By the way, does adding in the //char(0) work for converting Fortran strings to C/C++ strings (e.g. in call addfield(psi,psi01//char(0))) ? If it does, will it also work
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Also something weird that I have noted. The Initialize Helper Script line prints twice when I run the Fortran code and no print statements anywhere in my Python script execute (not even if they are at the top). When I run the C++ example located here: http://www.itk.org/Wiki/Coprocessing_example, the Initialize Helper Script only appears once and print statements in the Python script execute normally. I'm not sure if this helps or makes things worse. Thanks, Lucas From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf of Pettey . Lucas [lpet...@drc.com] Sent: Thursday, March 21, 2013 3:12 PM To: Andy Bauer Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Nothing printed at all. Does anything special have to be done during the compile/link steps for the python script to work properly? Thanks, Lucas From: Andy Bauer [andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 3:00 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 How about trying: psi01_.UpdatePipeline() print psi01_.GetDataInformation().DataInformation.GetNumberOfPoints() right about where the writer was created. This will give print out the number of points in the data set that you have. Andy On Thu, Mar 21, 2013 at 3:47 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.com wrote: Sorry, Andy, that didn't seem to work either. Let me know what information would be useful to print. Thanks Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:36 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Hmm, how about trying replacing those two lines with: ParallelImageDataWriter1 = CreateCPWriter( XMLPImageDataWriter, fullgrid_%t.pvti, 1, cp_writers ) Let me know how that goes. If not, I think we'll need to put some print statements in there. Andy On Thu, Mar 21, 2013 at 3:31 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Yes I did. I have attached the latest script. Thank you for your valiant efforts in trying to solve this problem. Lucas From: Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 2:22 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98 Did you also include the w.UpdatePipeline() part after the writer was created? It won't create the file unless that's in there. Andy On Thu, Mar 21, 2013 at 3:17 PM, Pettey . Lucas lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com wrote: Hi Andy, Changing to Scalars=['POINTS','psi01'] did not change anything. I still get the same error and blank image. I tried adding the XMLPImageDataWriter command, but no file 'fullgrid.pvti' was produced so I can't tell what is being populated into the grid. I hardwired the C++ adaptor to 'psi01' rather than the character array name in case their was a problem with the way the string was passed. I had checked that name contained the correct characters, but I figured hardwiring would be safer at this stage. Anything else that you think I should try? Thanks, Lucas From: paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org [paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.orgmailto:paraview-boun...@paraview.org] on behalf of Andy Bauer [andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.commailto:andy.ba...@kitware.com] Sent: Thursday, March 21, 2013 12:49 PM To: Pettey . Lucas Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.orgmailto:paraview@paraview.org] Subject: Re: [Paraview
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Berk,
Thanks for the help.
I found that this example is written based on an earlier version of VTK so some
of the function calls have changed. Based on your advice and some of the
examples contained in the ParaView wiki, I have gotten it to compile and run.
Now I just have to figure out why it isn't generating any VTK objects
Probably something wrong with my python co-processing script.
Thanks
Lucas
From: Berk Geveci [berk.gev...@kitware.com]
Sent: Tuesday, March 19, 2013 6:48 PM
To: Pettey . Lucas
Cc: paraview [paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
You can simply delete this line:
Grid-SetNumberOfScalarComponents(2);
It is redundant given the following later:
field-SetNumberOfComponents(2);
You also need to #include vtkSmartPointer.h.
Best,
-berk
On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas
lpet...@drc.commailto:lpet...@drc.com wrote:
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex data
type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualization_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary, opened
that in ParaView, output the coprocessing script and then written the adaptor
following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz)
{
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client script.
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but
// omeg probably passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name)
{
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from the Makefile generated by Cmake:
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:40:38: error: too few arguments to
function call, expected 2, have 1
Grid-SetNumberOfScalarComponents(2);
~ ^
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/DataModel/vtkImageData.h:298:3:
note: 'SetNumberOfScalarComponents' declared here
static void SetNumberOfScalarComponents( int n, vtkInformation* meta_data);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:5: error: unknown type name
'vtkSmartPointer'; did you mean 'vtkSmartPointerBase'?
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^~~
vtkSmartPointerBase
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/Core/vtkOStreamWrapper.h:36:7:
note: 'vtkSmartPointerBase' declared here
class vtkSmartPointerBase;
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:20: error: expected unqualified-id
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:68:5: error: use of undeclared
identifier 'field'
field-SetNumberOfComponents(2);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:69:5: error: use of undeclared
identifier 'field'
field-SetName(name);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:70:5: error: use of undeclared
identifier 'field'
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:71:37: error: use of undeclared
identifier 'field'
Image-GetPointData()-AddArray(field);
^
Any help is appreciated.
Thanks,
Lucas
This electronic message transmission and any attachments that accompany it
contain information from DRC® (Dynamics Research Corporation) or its
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Lucas,
Can you share your co-processing script and any run output? There may be
warnings in the output or info in the script that will help figure that out.
Thanks,
Andy
On Wed, Mar 20, 2013 at 10:04 AM, Pettey . Lucas lpet...@drc.com wrote:
Hi Berk,
Thanks for the help.
I found that this example is written based on an earlier version of VTK so
some of the function calls have changed. Based on your advice and some of
the examples contained in the ParaView wiki, I have gotten it to compile
and run.
Now I just have to figure out why it isn't generating any VTK objects
Probably something wrong with my python co-processing script.
Thanks
Lucas
From: Berk Geveci [berk.gev...@kitware.com]
Sent: Tuesday, March 19, 2013 6:48 PM
To: Pettey . Lucas
Cc: paraview [paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
You can simply delete this line:
Grid-SetNumberOfScalarComponents(2);
It is redundant given the following later:
field-SetNumberOfComponents(2);
You also need to #include vtkSmartPointer.h.
Best,
-berk
On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas lpet...@drc.commailto:
lpet...@drc.com wrote:
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex
data type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualization_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary,
opened that in ParaView, output the coprocessing script and then written
the adaptor following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz)
{
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client
script.
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but
// omeg probably passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name)
{
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from the Makefile generated by Cmake:
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:40:38: error: too few arguments
to function call, expected 2, have 1
Grid-SetNumberOfScalarComponents(2);
~ ^
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/DataModel/vtkImageData.h:298:3:
note: 'SetNumberOfScalarComponents' declared here
static void SetNumberOfScalarComponents( int n, vtkInformation*
meta_data);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:5: error: unknown type name
'vtkSmartPointer'; did you mean 'vtkSmartPointerBase'?
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^~~
vtkSmartPointerBase
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/Core/vtkOStreamWrapper.h:36:7:
note: 'vtkSmartPointerBase' declared here
class vtkSmartPointerBase;
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:20: error: expected
unqualified-id
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:68:5: error: use of undeclared
identifier 'field'
field-SetNumberOfComponents(2);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:69:5: error: use of undeclared
identifier 'field'
field-SetName(name);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:70:5: error: use of undeclared
identifier 'field'
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:71:37
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Andy and Berk,
Here is the error that I am receiving.
I know that it is initializing ok and failing at the first call to the
co-processor. I have attached my python script output as well as my test code
and adaptors.
CoProcessingExample(19247,0x7fff77470180) malloc: ***
mmap(size=3739605888335872) failed (error code=12)
*** error: can't allocate region
*** set a breakpoint in malloc_error_break to debug
Operating system error: Cannot allocate memory
Memory allocation failed
Thanks for the help,
Lucas
From: Andy Bauer [andy.ba...@kitware.com]
Sent: Wednesday, March 20, 2013 9:08 AM
To: Pettey . Lucas
Cc: Berk Geveci; paraview [paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Lucas,
Can you share your co-processing script and any run output? There may be
warnings in the output or info in the script that will help figure that out.
Thanks,
Andy
On Wed, Mar 20, 2013 at 10:04 AM, Pettey . Lucas
lpet...@drc.commailto:lpet...@drc.com wrote:
Hi Berk,
Thanks for the help.
I found that this example is written based on an earlier version of VTK so some
of the function calls have changed. Based on your advice and some of the
examples contained in the ParaView wiki, I have gotten it to compile and run.
Now I just have to figure out why it isn't generating any VTK objects
Probably something wrong with my python co-processing script.
Thanks
Lucas
From: Berk Geveci [berk.gev...@kitware.commailto:berk.gev...@kitware.com]
Sent: Tuesday, March 19, 2013 6:48 PM
To: Pettey . Lucas
Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
You can simply delete this line:
Grid-SetNumberOfScalarComponents(2);
It is redundant given the following later:
field-SetNumberOfComponents(2);
You also need to #include vtkSmartPointer.h.
Best,
-berk
On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas
lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com
wrote:
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex data
type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualization_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary, opened
that in ParaView, output the coprocessing script and then written the adaptor
following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz)
{
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client script.
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but
// omeg probably passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name)
{
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from the Makefile generated by Cmake:
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:40:38: error: too few arguments to
function call, expected 2, have 1
Grid-SetNumberOfScalarComponents(2);
~ ^
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/DataModel/vtkImageData.h:298:3:
note: 'SetNumberOfScalarComponents' declared here
static void SetNumberOfScalarComponents( int n, vtkInformation* meta_data);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:5: error: unknown type name
'vtkSmartPointer'; did you mean 'vtkSmartPointerBase'?
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^~~
vtkSmartPointerBase
/Users/lucaspettey/ParaView-git/ParaView/VTK
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
So after some troubleshooting it seems that C++ doesn't like the complex data
type the way Fortran is sending it.
The code runs when I just send a real value. Images appear, but of course they
are meaningless since it isn't the correct data. I may try passing the real and
imaginary separate. I may also look into row/column order issues although this
grid is 100x100x100 so it shouldn't affect anything.
Lucas
From: paraview-boun...@paraview.org [paraview-boun...@paraview.org] on behalf
of Pettey . Lucas [lpet...@drc.com]
Sent: Wednesday, March 20, 2013 11:12 AM
To: Andy Bauer
Cc: paraview [paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Andy and Berk,
Here is the error that I am receiving.
I know that it is initializing ok and failing at the first call to the
co-processor. I have attached my python script output as well as my test code
and adaptors.
CoProcessingExample(19247,0x7fff77470180) malloc: ***
mmap(size=3739605888335872) failed (error code=12)
*** error: can't allocate region
*** set a breakpoint in malloc_error_break to debug
Operating system error: Cannot allocate memory
Memory allocation failed
Thanks for the help,
Lucas
From: Andy Bauer [andy.ba...@kitware.com]
Sent: Wednesday, March 20, 2013 9:08 AM
To: Pettey . Lucas
Cc: Berk Geveci; paraview [paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Hi Lucas,
Can you share your co-processing script and any run output? There may be
warnings in the output or info in the script that will help figure that out.
Thanks,
Andy
On Wed, Mar 20, 2013 at 10:04 AM, Pettey . Lucas
lpet...@drc.commailto:lpet...@drc.com wrote:
Hi Berk,
Thanks for the help.
I found that this example is written based on an earlier version of VTK so some
of the function calls have changed. Based on your advice and some of the
examples contained in the ParaView wiki, I have gotten it to compile and run.
Now I just have to figure out why it isn't generating any VTK objects
Probably something wrong with my python co-processing script.
Thanks
Lucas
From: Berk Geveci [berk.gev...@kitware.commailto:berk.gev...@kitware.com]
Sent: Tuesday, March 19, 2013 6:48 PM
To: Pettey . Lucas
Cc: paraview [paraview@paraview.orgmailto:paraview@paraview.org]
Subject: Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
You can simply delete this line:
Grid-SetNumberOfScalarComponents(2);
It is redundant given the following later:
field-SetNumberOfComponents(2);
You also need to #include vtkSmartPointer.h.
Best,
-berk
On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas
lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.commailto:lpet...@drc.com
wrote:
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex data
type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualization_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary, opened
that in ParaView, output the coprocessing script and then written the adaptor
following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz)
{
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client script.
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but
// omeg probably passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name)
{
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
Message: 5
Date: Tue, 19 Mar 2013 18:31:28 +
From: Pettey . Lucas lpet...@drc.com
Subject: [Paraview] CoProcessing with Fortran and ParaView 3.98
To: paraview ?[paraview@paraview.org]?paraview@paraview.org
Message-ID:
c7acf6f2c9304a4c81eac2e2c96b5e165f1e0...@exmb01.drc.com
Content-Type: text/plain; charset=windows-1256
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex data
type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualiz
ation_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary, opened
that in ParaView, output the coprocessing script and then written the
adaptor following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz) {
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client
script.
ParaViewCoProcessing::GetCoProcessorData()-
GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but // omeg probably
passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name) {
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-
GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from the Makefile generated by Cmake:
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:40:38: error: too few
arguments to function call, expected 2, have 1
Grid-SetNumberOfScalarComponents(2);
~ ^
/Users/lucaspettey/ParaView-
git/ParaView/VTK/Common/DataModel/vtkImageData.h:298:3: note:
'SetNumberOfScalarComponents' declared here
static void SetNumberOfScalarComponents( int n, vtkInformation*
meta_data);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:5: error: unknown type
name 'vtkSmartPointer'; did you mean 'vtkSmartPointerBase'?
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^~~
vtkSmartPointerBase
/Users/lucaspettey/ParaView-
git/ParaView/VTK/Common/Core/vtkOStreamWrapper.h:36:7: note:
'vtkSmartPointerBase' declared here class vtkSmartPointerBase;
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:20: error: expected
unqualified-id
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:68:5: error: use of undeclared
identifier 'field'
field-SetNumberOfComponents(2);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:69:5: error: use of undeclared
identifier 'field'
field-SetName(name);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:70:5: error: use of undeclared
identifier 'field'
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:71:37: error: use of
undeclared identifier 'field'
Image-GetPointData()-AddArray(field);
^
Any help is appreciated.
Thanks,
Lucas
Hello Lucas,
I helped write those notes for a project through XSEDE last year. They were
based on Nautilus down at NICS. The ParaView on Nautilus used VTK 5.x, it looks
like the API has changed for VTK 6, so some of those calls will have to be
changed. For example, the SetNumberOfScalarComponents method is now handled
completely differently.
I haven't studied VTK 6 yet, hopefully someone with VTK 6 knowledge can add
more.
Mark
___
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please keep messages on-topic and check the ParaView Wiki at:
Re: [Paraview] CoProcessing with Fortran and ParaView 3.98
You can simply delete this line:
Grid-SetNumberOfScalarComponents(2);
It is redundant given the following later:
field-SetNumberOfComponents(2);
You also need to #include vtkSmartPointer.h.
Best,
-berk
On Tue, Mar 19, 2013 at 2:31 PM, Pettey . Lucas lpet...@drc.com wrote:
Hello everyone,
I am trying to run a simple Fortran coprocessing example with a complex
data type. I am following the example here:
http://en.wikibooks.org/wiki/Parallel_Spectral_Numerical_Methods/Visualization_with_ParaView_CoProcessing#ParaView_CoProcessing_Resources
I have written a basic Fortran code that outputs the data in binary,
opened that in ParaView, output the coprocessing script and then written
the adaptor following the example in the above link. My adaptor looks like:
extern C void createcpimagedata_(int* nx, int* ny, int* nz)
{
if (!ParaViewCoProcessing::GetCoProcessorData()) {
vtkGenericWarningMacro(Unable to access CoProcessorData.);
return;
}
// The simulation grid is a 2-dimensional topologically and geometrically
// regular grid. In VTK/ParaView, this is considered an image data set.
vtkImageData* Grid = vtkImageData::New();
// assuming dimZ == 1 for now
Grid-SetDimensions(*nx, *ny, *nz);
Grid-SetNumberOfScalarComponents(2);
// Setting the Origin and Spacing are also options.
// Name should be consistent between here, Fortran and Python client
script.
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input)-SetGrid(Grid);
}
// Add field(s) to the data container.
// Separate from above because this will be dynamic, grid is static.
// Might be an issue, VTK probably assumes row major, but
// omeg probably passed column major...
// by hand name mangling for fortran
extern C void addfield_(double* scalars, char* name)
{
vtkCPInputDataDescription *idd =
ParaViewCoProcessing::GetCoProcessorData()-GetInputDescriptionByName(input);
vtkImageData* Image = vtkImageData::SafeDownCast(idd-GetGrid());
if (!Image) {
vtkGenericWarningMacro(No adaptor grid to attach field data to.);
return;
}
// field name must match that in the fortran code.
if (idd-IsFieldNeeded(name)) {
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
field-SetNumberOfComponents(2);
field-SetName(name);
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
Image-GetPointData()-AddArray(field);
}
}
I am getting some errors from the Makefile generated by Cmake:
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:40:38: error: too few arguments
to function call, expected 2, have 1
Grid-SetNumberOfScalarComponents(2);
~ ^
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/DataModel/vtkImageData.h:298:3:
note: 'SetNumberOfScalarComponents' declared here
static void SetNumberOfScalarComponents( int n, vtkInformation*
meta_data);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:5: error: unknown type name
'vtkSmartPointer'; did you mean 'vtkSmartPointerBase'?
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^~~
vtkSmartPointerBase
/Users/lucaspettey/ParaView-git/ParaView/VTK/Common/Core/vtkOStreamWrapper.h:36:7:
note: 'vtkSmartPointerBase' declared here
class vtkSmartPointerBase;
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:67:20: error: expected
unqualified-id
vtkSmartPointervtkDoubleArray field =
vtkSmartPointervtkDoubleArray::New();
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:68:5: error: use of undeclared
identifier 'field'
field-SetNumberOfComponents(2);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:69:5: error: use of undeclared
identifier 'field'
field-SetName(name);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:70:5: error: use of undeclared
identifier 'field'
field-SetArray(scalars, 2* Image-GetNumberOfPoints(), 1);
^
/Users/lucaspettey/codes/BEC/coprocVTK.cxx:71:37: error: use of undeclared
identifier 'field'
Image-GetPointData()-AddArray(field);
^
Any help is appreciated.
Thanks,
Lucas
This electronic message transmission and any attachments that accompany it
contain information from DRC® (Dynamics Research Corporation) or its
subsidiaries, or the intended recipient, which is privileged, proprietary,
business confidential, or otherwise protected from disclosure and is the
exclusive property of DRC and/or the intended recipient. The information in
this email is solely intended for the use of the individual or entity that
is the intended recipient. If you are not the intended recipient, any use,
dissemination, distribution, retention, or copying of this communication,
attachments, or substance is prohibited. If you have received this
electronic transmission in error, please immediately reply to the author
via
