From ben.boeckel at kitware.com Fri Jul 1 09:23:40 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Fri, 1 Jul 2016 09:23:40 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> Message-ID: <20160701132340.GA13101@megas.kitware.com> On Thu, Jun 30, 2016 at 17:09:18 -0600, David Gobbi wrote: > The scripts just modify the VTK source files. They don't call any git > commands so I don't think they can, by themselves, be idempotent. Do you > have any insights into what we can do to avoid conflicts? A function f is idempotent iff f(f(x)) == f(x). If the formatter is idempotent, it can be safely run over any branch during a rebase to make it not add any style violations. --Ben From marcus.hanwell at kitware.com Fri Jul 1 09:49:29 2016 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Fri, 1 Jul 2016 09:49:29 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: <20160630203713.1114207107@mail.rogue-research.com> References: <20160630203713.1114207107@mail.rogue-research.com> Message-ID: On Thu, Jun 30, 2016 at 4:37 PM, Sean McBride wrote: > On Thu, 30 Jun 2016 15:30:53 -0400, Marcus D. Hanwell said: > >>Both of these changes would be stylistic, and both would likely cause >>merge conflicts. Ideally we could combine the two, and update our >>guidance on coding style at the same time. We would like to coordinate >>timing, but I think these changes are both well overdue. >> >>Objections, blockers, or thoughts? > > I'd really like to get this in first: > > Seems reasonable, to be clear I wasn't saying we would do it tomorrow (or even before the hackathon), but wanted to check there were no remaining blockers. I will see if we can come up with a concrete date so that people can plan accordingly. From bill.lorensen at gmail.com Fri Jul 1 09:49:44 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Fri, 1 Jul 2016 09:49:44 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: <20160701132340.GA13101@megas.kitware.com> References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: Once the files are changed, I suggest that a KWStyle expert https://kitware.github.io/KWStyle/ take a stab at creating a VTK style file. That way we can enforce the style in the future. We use it in ITK and it is very effective, although it can take some effort to correct all of the existing defects. On Fri, Jul 1, 2016 at 9:23 AM, Ben Boeckel wrote: > On Thu, Jun 30, 2016 at 17:09:18 -0600, David Gobbi wrote: >> The scripts just modify the VTK source files. They don't call any git >> commands so I don't think they can, by themselves, be idempotent. Do you >> have any insights into what we can do to avoid conflicts? > > A function f is idempotent iff f(f(x)) == f(x). If the formatter is > idempotent, it can be safely run over any branch during a rebase to make > it not add any style violations. > > --Ben > _______________________________________________ > 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 > -- Unpaid intern in BillsBasement at noware dot com From utkarsh.ayachit at kitware.com Fri Jul 1 11:59:57 2016 From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit) Date: Fri, 1 Jul 2016 11:59:57 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: Maybe a little off topic, but has anyone looked at "Artistic Style" ( http://astyle.sourceforge.net/astyle.html). Seems like a project especially designed for changing formatting. Utkarsh On Fri, Jul 1, 2016 at 9:49 AM, Bill Lorensen wrote: > Once the files are changed, I suggest that a KWStyle expert > https://kitware.github.io/KWStyle/ take a stab at creating a VTK style > file. That way we can enforce the style in the future. We use it in > ITK and it is very effective, although it can take some effort to > correct all of the existing defects. > > > On Fri, Jul 1, 2016 at 9:23 AM, Ben Boeckel > wrote: > > On Thu, Jun 30, 2016 at 17:09:18 -0600, David Gobbi wrote: > >> The scripts just modify the VTK source files. They don't call any git > >> commands so I don't think they can, by themselves, be idempotent. Do > you > >> have any insights into what we can do to avoid conflicts? > > > > A function f is idempotent iff f(f(x)) == f(x). If the formatter is > > idempotent, it can be safely run over any branch during a rebase to make > > it not add any style violations. > > > > --Ben > > _______________________________________________ > > 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 > > > > > > -- > Unpaid intern in BillsBasement at noware dot 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=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 david.gobbi at gmail.com Fri Jul 1 13:32:16 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Fri, 1 Jul 2016 11:32:16 -0600 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: I've used it. It has an unforgivable flaw: it doesn't count the whitespace at the beginning of the line as part of the line length when it does its maximum line length check. My experience with style checkers has been bad. I find that people use them as a crutch, and write messy code because "the style checker can fix it before I commit". In any case, my main concern is that people write tidy code, I'm less concerned about strictly following a specific style as long as each module is internally self-consistent. - David On Fri, Jul 1, 2016 at 9:59 AM, Utkarsh Ayachit wrote: > Maybe a little off topic, but has anyone looked at "Artistic Style" ( > http://astyle.sourceforge.net/astyle.html). > > Seems like a project especially designed for changing formatting. > > Utkarsh > > > On Fri, Jul 1, 2016 at 9:49 AM, Bill Lorensen > wrote: > >> Once the files are changed, I suggest that a KWStyle expert >> https://kitware.github.io/KWStyle/ take a stab at creating a VTK style >> file. That way we can enforce the style in the future. We use it in >> ITK and it is very effective, although it can take some effort to >> correct all of the existing defects. >> >> >> On Fri, Jul 1, 2016 at 9:23 AM, Ben Boeckel >> wrote: >> > On Thu, Jun 30, 2016 at 17:09:18 -0600, David Gobbi wrote: >> >> The scripts just modify the VTK source files. They don't call any git >> >> commands so I don't think they can, by themselves, be idempotent. Do >> you >> >> have any insights into what we can do to avoid conflicts? >> > >> > A function f is idempotent iff f(f(x)) == f(x). If the formatter is >> > idempotent, it can be safely run over any branch during a rebase to make >> > it not add any style violations. >> > >> > --Ben >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.lorensen at gmail.com Fri Jul 1 13:36:44 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Fri, 1 Jul 2016 13:36:44 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: KWStyle has worked pretty good for ITK. On Fri, Jul 1, 2016 at 1:32 PM, David Gobbi wrote: > I've used it. It has an unforgivable flaw: it doesn't count the whitespace > at the beginning of the line as part of the line length when it does its > maximum line length check. > > My experience with style checkers has been bad. I find that people use them > as a crutch, and write messy code because "the style checker can fix it > before I commit". In any case, my main concern is that people write tidy > code, I'm less concerned about strictly following a specific style as long > as each module is internally self-consistent. > > - David > > On Fri, Jul 1, 2016 at 9:59 AM, Utkarsh Ayachit > wrote: >> >> Maybe a little off topic, but has anyone looked at "Artistic Style" >> (http://astyle.sourceforge.net/astyle.html). >> >> Seems like a project especially designed for changing formatting. >> >> Utkarsh >> >> >> On Fri, Jul 1, 2016 at 9:49 AM, Bill Lorensen >> wrote: >>> >>> Once the files are changed, I suggest that a KWStyle expert >>> https://kitware.github.io/KWStyle/ take a stab at creating a VTK style >>> file. That way we can enforce the style in the future. We use it in >>> ITK and it is very effective, although it can take some effort to >>> correct all of the existing defects. >>> >>> >>> On Fri, Jul 1, 2016 at 9:23 AM, Ben Boeckel >>> wrote: >>> > On Thu, Jun 30, 2016 at 17:09:18 -0600, David Gobbi wrote: >>> >> The scripts just modify the VTK source files. They don't call any git >>> >> commands so I don't think they can, by themselves, be idempotent. Do >>> >> you >>> >> have any insights into what we can do to avoid conflicts? >>> > >>> > A function f is idempotent iff f(f(x)) == f(x). If the formatter is >>> > idempotent, it can be safely run over any branch during a rebase to >>> > make >>> > it not add any style violations. >>> > >>> > --Ben > > > > _______________________________________________ > 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 > > -- Unpaid intern in BillsBasement at noware dot com From david.gobbi at gmail.com Fri Jul 1 14:14:06 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Fri, 1 Jul 2016 12:14:06 -0600 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: On Fri, Jul 1, 2016 at 11:44 AM, Will Schroeder wrote: > > On Fri, Jul 1, 2016 at 1:32 PM, David Gobbi wrote: > >> I'm less concerned about strictly following a specific style as long as >> each module is internally self-consistent. > > > I understand what you're saying, I agree that some of the religious style > arguments are dubious, but if the self-consistent style is a complete and > utter mess this is bad :-) VTK has a long, enforced history of a reasonable > style and I'd like to preserve this. Simple things like using camel-case, > naming variables in verbose fashion, and this-> pointers etc. go a long way > towards grokking code quickly IMO. > Hi Will, you're correct, I was being too dismissive. My main worry is that automatic re-styling of the VTK code can cause "git blame" to give useless results. Thats why, when I wrote "vtk_reindent_code.py", I made sure that it only modifies brace indentation and leaves all other code lines alone. - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.lorensen at gmail.com Fri Jul 1 15:23:47 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Fri, 1 Jul 2016 15:23:47 -0400 Subject: [vtk-developers] Indentation style and converting to Doxygen style docs In-Reply-To: References: <20160630203713.1114207107@mail.rogue-research.com> <20160630210457.GA4150@rotor.kitware.com> <20160701132340.GA13101@megas.kitware.com> Message-ID: For new code, I will definitely use astyle with these options: astyle --indent=spaces=2 --style=vtk If the indentation style changes, I will change the --style to conform If does a great job with braces and indentation. I will also use KWStyle to check headers, #ifndef guards, typedef indentations, extraneous ;'s and more. I agree with Will, that massive changes just for the sake of changing indentation, etc. adds little value. I also agree with D Gobbi's approach to retain the usefulness of git blame. Bill On Fri, Jul 1, 2016 at 2:14 PM, David Gobbi wrote: > On Fri, Jul 1, 2016 at 11:44 AM, Will Schroeder > wrote: >> >> >> On Fri, Jul 1, 2016 at 1:32 PM, David Gobbi wrote: >>> >>> I'm less concerned about strictly following a specific style as long as >>> each module is internally self-consistent. >> >> >> I understand what you're saying, I agree that some of the religious style >> arguments are dubious, but if the self-consistent style is a complete and >> utter mess this is bad :-) VTK has a long, enforced history of a reasonable >> style and I'd like to preserve this. Simple things like using camel-case, >> naming variables in verbose fashion, and this-> pointers etc. go a long way >> towards grokking code quickly IMO. > > > Hi Will, you're correct, I was being too dismissive. My main worry is that > automatic re-styling of the VTK code can cause "git blame" to give useless > results. Thats why, when I wrote "vtk_reindent_code.py", I made sure that > it only modifies brace indentation and leaves all other code lines alone. > > - David > > _______________________________________________ > 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 > > -- Unpaid intern in BillsBasement at noware dot com From matthew.brett at gmail.com Sat Jul 2 04:39:52 2016 From: matthew.brett at gmail.com (Matthew Brett) Date: Sat, 2 Jul 2016 09:39:52 +0100 Subject: [vtk-developers] Some progress on binary wheel for OSX Message-ID: Hi, Just to say, I did a few hours of hacking and investigation, which got me reasonably close to a copy of VTK that could be used for a very useful OSX binary installer - full description at: https://mail.python.org/pipermail/pythonmac-sig/2016-July/024137.html As y'all know, the current OSX installer isn't much use for doing real work, because it ships its own copy of Python, making it difficult to install other Python packages for doing standard things like array manipulation, statistics and so on. The ideal would be to be able to build a binary wheel [1] for OSX, so the user could do: pip install vtk to install a copy of vtk into the Python environment they are using - such as Python.org Python, homebrew, etc. This has already been done for nearly all the main scientific Python projects such as numpy, scipy, matplotlib etc. The big hold-up for VTK, is the current behavior of VTK / cmake, to link the Python library by absolute path directly into the VTK libs, so the build is tied to the path of the Python binary [2, 3]. I see there's a pull request already in from the homebrew folks to try and fix that [4] - which is excellent news. I believe that, once this is merged and released, it should be fairly easy to build an OSX wheel that would work in any standard OSX Python environment, which would be a huge step forward in terms of ease of the install process. Cheers, Matthew [1] https://www.python.org/dev/peps/pep-0427/ [2] http://blog.tim-smith.us/2015/09/python-extension-modules-os-x/ [3] https://github.com/Homebrew/homebrew-science/issues/3401 [4] https://gitlab.kitware.com/vtk/vtk/merge_requests/1511 From bill.lorensen at gmail.com Sat Jul 2 12:53:49 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sat, 2 Jul 2016 12:53:49 -0400 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py Message-ID: David, vtk_reindent_code.py has trouble with for loops after if blocks that do not have braces. See the attached file, around lines 3269 and 3290. For example, Original: int vtkReebGraph::Implementation::GetNumberOfNodes() { if(!this->NodeNumber) for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) { // check if node is cleared if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) this->NodeNumber++; } return this->NodeNumber; } after vtk_reindent_code.py int vtkReebGraph::Implementation::GetNumberOfNodes() { if(!this->NodeNumber) for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) { // check if node is cleared if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) this->NodeNumber++; } return this->NodeNumber; } There are there are other files that exhibit the same behavior. Bill -- Unpaid intern in BillsBasement at noware dot com -------------- next part -------------- A non-text attachment was scrubbed... Name: vtkReebGraph.cxx Type: text/x-c++src Size: 104419 bytes Desc: not available URL: From bill.lorensen at gmail.com Sat Jul 2 13:21:48 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sat, 2 Jul 2016 13:21:48 -0400 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: References: Message-ID: Also seems to have trouble with case statements... --- a/IO/Export/vtkX3DExporterXMLWriter.cxx +++ b/IO/Export/vtkX3DExporterXMLWriter.cxx @@ -211,17 +211,17 @@ void vtkX3DExporterXMLWriter::SetField(int attributeID, int type, vtkDataArray* { case(MFVEC3F): for (vtkIdType i = 0; i < a->GetNumberOfTuples(); i++) - { + { double* d = a->GetTuple(i); *this->OutputStream << this->ActTab << d[0] << " " << d[1] << " " << d[2] << "," << this->GetNewline(); - } + } break; case(MFVEC2F): for (vtkIdType i = 0; i < a->GetNumberOfTuples(); i++) - { + { double* d = a->GetTuple(i); *this->OutputStream << this->ActTab << d[0] << " " << d[1] << "," << this->GetNewline(); - } + } break; default: *this->OutputStream << "UNKNOWN DATATYPE"; @@ -289,14 +289,14 @@ void vtkX3DExporterXMLWriter::SetField(int attributeID, const int* values, size_ } else while (i < size) - { + { *this->OutputStream << values[i] << " "; if (values[i] == -1) { *this->OutputStream << this->GetNewline() << this->ActTab; } i++; - } + } *this->OutputStream << "\""; } On Sat, Jul 2, 2016 at 12:53 PM, Bill Lorensen wrote: > David, > > vtk_reindent_code.py has trouble with for loops after if blocks that > do not have braces. > > See the attached file, around lines 3269 and 3290. > > For example, Original: > int vtkReebGraph::Implementation::GetNumberOfNodes() > { > if(!this->NodeNumber) > for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) > { > // check if node is cleared > if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) > this->NodeNumber++; > } > > return this->NodeNumber; > } > > after vtk_reindent_code.py > int vtkReebGraph::Implementation::GetNumberOfNodes() > { > if(!this->NodeNumber) > for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) > { > // check if node is cleared > if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) > this->NodeNumber++; > } > > return this->NodeNumber; > } > > > There are there are other files that exhibit the same behavior. > > Bill > > -- > Unpaid intern in BillsBasement at noware dot com -- Unpaid intern in BillsBasement at noware dot com From david.gobbi at gmail.com Sat Jul 2 16:21:22 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Sat, 2 Jul 2016 14:21:22 -0600 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: References: Message-ID: Thanks, Bill. My case-statement regex was confused by the parentheses in "case(MFVEC3F):", and I completely missed the check for "if, else, switch" etc. because they're almost always followed by "{" in VTK. Here's an MR with the fixes: https://gitlab.kitware.com/vtk/vtk/merge_requests/1613/diffs Let me know if you find anything else. - David On Sat, Jul 2, 2016 at 11:21 AM, Bill Lorensen wrote: > Also seems to have trouble with case statements... > --- a/IO/Export/vtkX3DExporterXMLWriter.cxx > +++ b/IO/Export/vtkX3DExporterXMLWriter.cxx > @@ -211,17 +211,17 @@ void vtkX3DExporterXMLWriter::SetField(int > attributeID, int type, vtkDataArray* > { > case(MFVEC3F): > for (vtkIdType i = 0; i < a->GetNumberOfTuples(); i++) > - { > + { > > On Sat, Jul 2, 2016 at 12:53 PM, Bill Lorensen > wrote: > > David, > > > > vtk_reindent_code.py has trouble with for loops after if blocks that > > do not have braces. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhangjiang.dudu at gmail.com Sun Jul 3 23:57:23 2016 From: zhangjiang.dudu at gmail.com (=?utf-8?B?5byg5rGf?=) Date: Sun, 3 Jul 2016 22:57:23 -0500 Subject: [vtk-developers] unstructured data partition Message-ID: Hi, I would like to partition the unstructured data using vtkUnstructuredGridReader for further parallel application. The thought is to partition and assign one part of the data (mesh) to each process. The code is at below. However, the generated result of each process is somehow not very correct. For example, for one data partition of a process, the number of points is 3, and the number of cells is 5. But these 5 cells are the same (actually one cell contains 3 points). So how to avoid this? Can this code reach to my goal of data partitioning? vtkMPIController *contr = vtkMPIController::New(); contr->Initialize(&argc, &argv); vtkMultiProcessController::SetGlobalController(contr); int numProcs = contr->GetNumberOfProcesses(); int me = contr->GetLocalProcessId(); vtkUnstructuredGridReader *reader = vtkUnstructuredGridReader::New(); vtkUnstructuredGrid *ds = vtkUnstructuredGrid::New(); reader->SetFileName(?data.vtk"); reader->Update(); ds = reader->GetOutput(); vtkDistributedDataFilter *dd = vtkDistributedDataFilter::New(); dd->SetInputData(ds); dd->SetController(contr); dd->UseMinimalMemoryOff(); dd->SetBoundaryModeToAssignToOneRegion(); dd->Update(); -------------- next part -------------- An HTML attachment was scrubbed... URL: From sean at rogue-research.com Mon Jul 4 15:05:49 2016 From: sean at rogue-research.com (Sean McBride) Date: Mon, 4 Jul 2016 15:05:49 -0400 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: References: Message-ID: <20160704190549.480247096@mail.rogue-research.com> On Sat, 2 Jul 2016 12:53:49 -0400, Bill Lorensen said: >vtk_reindent_code.py has trouble with for loops after if blocks that >do not have braces. > >See the attached file, around lines 3269 and 3290. > >For example, Original: >int vtkReebGraph::Implementation::GetNumberOfNodes() >{ > if(!this->NodeNumber) > for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) > { > // check if node is cleared > if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) > this->NodeNumber++; > } clang-tidy can find ifs and loops with missing braces and even add them for you: After this re-indentation change lands, I'll give it a try... Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From paolo.quadrani at gmail.com Mon Jul 4 15:48:44 2016 From: paolo.quadrani at gmail.com (Paolo Quadrani) Date: Mon, 4 Jul 2016 21:48:44 +0200 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: <20160704190549.480247096@mail.rogue-research.com> References: <20160704190549.480247096@mail.rogue-research.com> Message-ID: Hi Bill, David, maybe Uncrustify can be of some help on managing the indentation braces check. It makes use of a configuration file with which you can specify which are the rules you want to apply to the source file. http://uncrustify.sourceforge.net Best Regards --- .: Paolo Quadrani :: [nEraLab ] - [LinkedIN ] - [Twitter ] - [SlideShare ] "It always seems impossible until it's done. (N. Mandela)" > On 04 lug 2016, at 21:05, Sean McBride wrote: > > On Sat, 2 Jul 2016 12:53:49 -0400, Bill Lorensen said: > >> vtk_reindent_code.py has trouble with for loops after if blocks that >> do not have braces. >> >> See the attached file, around lines 3269 and 3290. >> >> For example, Original: >> int vtkReebGraph::Implementation::GetNumberOfNodes() >> { >> if(!this->NodeNumber) >> for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) >> { >> // check if node is cleared >> if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) >> this->NodeNumber++; >> } > > clang-tidy can find ifs and loops with missing braces and even add them for you: > > > After this re-indentation change lands, I'll give it a try... > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > _______________________________________________ > 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 bill.lorensen at gmail.com Mon Jul 4 15:57:46 2016 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Mon, 4 Jul 2016 15:57:46 -0400 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: References: <20160704190549.480247096@mail.rogue-research.com> Message-ID: First, I have mixed emotions about massive changes to fix brace spacing. Who benefits? Some developers, yes, the customer, no. There are lots of tools to check indentation and fix it. If we make vtk-wide changes, I think it is important that the tool only change whitespace. That means that a git blame -w will retain the last author's changes. As far as I know, David's python script is the only tool that does this. Bill Bill On Mon, Jul 4, 2016 at 3:48 PM, Paolo Quadrani wrote: > Hi Bill, David, > maybe Uncrustify can be of some help on managing the indentation braces > check. It makes use of a configuration file with which you can specify which > are the rules you want to apply to the source file. > > http://uncrustify.sourceforge.net > > Best Regards > --- > .: Paolo Quadrani > > :: [nEraLab] - [LinkedIN] - [Twitter] - [SlideShare] > > "It always seems impossible until it's done. (N. Mandela)" > > On 04 lug 2016, at 21:05, Sean McBride wrote: > > On Sat, 2 Jul 2016 12:53:49 -0400, Bill Lorensen said: > > vtk_reindent_code.py has trouble with for loops after if blocks that > do not have braces. > > See the attached file, around lines 3269 and 3290. > > For example, Original: > int vtkReebGraph::Implementation::GetNumberOfNodes() > { > if(!this->NodeNumber) > for(vtkIdType nodeId = 1; nodeId < this->MainNodeTable.Size; nodeId++) > { > // check if node is cleared > if(!(this->GetNode(nodeId)->ArcUpId == ((int)-2))) > this->NodeNumber++; > } > > > clang-tidy can find ifs and loops with missing braces and even add them for > you: > > > After this re-indentation change lands, I'll give it a try... > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > _______________________________________________ > 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 > > -- Unpaid intern in BillsBasement at noware dot com From sean at rogue-research.com Mon Jul 4 17:41:53 2016 From: sean at rogue-research.com (Sean McBride) Date: Mon, 4 Jul 2016 17:41:53 -0400 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: References: <20160704190549.480247096@mail.rogue-research.com> Message-ID: <20160704214153.1332569143@mail.rogue-research.com> On Mon, 4 Jul 2016 15:57:46 -0400, Bill Lorensen said: >First, I have mixed emotions about massive changes to fix brace >spacing. Who benefits? Some developers, yes, the customer, no. Bill, I take your point, but missing braces and screwed up whitespace can cause real bugs, the most famous recent example being the goto fail bug: VTK's Whitesmiths style isn't supported by modern tools, and once we get away from it we can do things like autoformatting, git hooks that reject incorrect formatting, dashboard check of formatting, etc. See you guys Wednesday! Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From andrew.amaclean at gmail.com Mon Jul 4 19:07:15 2016 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Tue, 5 Jul 2016 09:07:15 +1000 Subject: [vtk-developers] vtk-developers Digest, Vol 147, Issue 6 In-Reply-To: References: Message-ID: I had a look at that article and two things really jump out (in quotes here): ------------------------------------------------------------- "The first thing that leapt out at me when looking at the full source of sslKeyExchange.c was a lack of consistency. It bears all the hallmarks of a file which has been worked on by multiple developers who were not bound by a single set of comprehensive coding standards. Immediately noticeable when the file is opened in an ordinary text editor (not a code editor or IDE), is that the indentation inconsistently mixes tabs and spaces. The second thing I realised (relevant given the nature of this bug) is the inconsistency with regard to single-line conditional blocks. There are places where a conditional statement is enclosed in braces even when it is only a single line, and other places where it is not." ------------------------------------------------------------- So yes, it is probably worthwhile looking at brace spacing etc. One of the strengths of VTK is its consistent coding style which does make it easy to spot bugs. So I agree with you Sean, lets adopt a coding style consistent with modern tools and then git hooks can be implemented supporting this. vtk_reindent_code.py goes a long way towards doing this. It is frustrating at present that I have to manually modify spacing etc. since my main editors are Visual Studio, Sublime text3 using SublimeAStyleFormatter and sometimes Eclipse and QTCreator. Have a great Wednesday! Wish I lived closer. Regards Andrew ---------- Forwarded message ---------- > From: Sean McBride > To: Bill Lorensen , Paolo Quadrani < > paolo.quadrani at gmail.com> > Cc: VTK Developers > Date: Mon, 4 Jul 2016 17:41:53 -0400 > Subject: Re: [vtk-developers] Incorrect indentation by vtk_reindent_code.py > On Mon, 4 Jul 2016 15:57:46 -0400, Bill Lorensen said: > > >First, I have mixed emotions about massive changes to fix brace > >spacing. Who benefits? Some developers, yes, the customer, no. > > Bill, I take your point, but missing braces and screwed up whitespace can > cause real bugs, the most famous recent example being the goto fail bug: > > > VTK's Whitesmiths style isn't supported by modern tools, and once we get > away from it we can do things like autoformatting, git hooks that reject > incorrect formatting, dashboard check of formatting, etc. > > See you guys Wednesday! > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Mon Jul 4 19:34:48 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Mon, 4 Jul 2016 17:34:48 -0600 Subject: [vtk-developers] Incorrect indentation by vtk_reindent_code.py In-Reply-To: <20160704214153.1332569143@mail.rogue-research.com> References: <20160704190549.480247096@mail.rogue-research.com> <20160704214153.1332569143@mail.rogue-research.com> Message-ID: On Mon, Jul 4, 2016 at 3:41 PM, Sean McBride wrote: > > VTK's Whitesmiths style isn't supported by modern tools, and once we get > away from it we can do things like autoformatting, Yup, this nicely summarizes the original reason for the proposal. git hooks that reject incorrect formatting, dashboard check of formatting, > etc. We could do this regardless of what style we choose :) My python script already has a "--test" option. Checking that braces are present and that conditional blocks are correctly indented is easy. We could run this check alongside the current HeaderTesting test (but unlike HeaderTesting, it'll run on .cxx files, too). - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 07:00:12 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 14:00:12 +0300 Subject: [vtk-developers] Rendering hangs in viewport Message-ID: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Hello, I am part of the team that is working on project that uses VTK 7.0 for rendering mesh grid (not big, about 400k triangles) couple point clouds (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very simple meshes. Our app can work fine for 12-15h for then main viewport hangs. When I do some mouse interaction (click + move or scroll) some meshes became invisible for couple seconds and that became visible again. Also when I resize viewport window scene aspect ratio became wrong. It works like aspect ratio does not depends of viewport size. Also we noticed other strange thing. Our point clouds update dynamically during app work. We create 2 classes: class CameraPointCloudSource : public vtkPolyDataAlgorithm and class PointCloudColoring : public vtkPolyDataAlgorithm. I pass data (vtkPolyData with initialized points and vertexes) from CameraPointCloudSource to PointCloudColoring. I setup pipeline that output of the CameraPointCloudSource is input of the PointCloudColoring. Sometimes (can?t always reproduce this) CameraPointCloudSource:: RequestData stops to be called and CameraPointCloudSource::RequestData always get the same input data (the same point cloud the same with the same points). As result I always see in viewport the same point cloud visualization. I checked, code that responsible for point cloud generation works fine and produce correct point clouds. For test we also tried to use new VTK PointCloud filters. There we also noticed similar behavior. We set filters chain. Alter some long time (> 12h of work) sometimes some filters star producing the same output that not depends of input point cloud. Interesting that some filters work fine and some start producing the same output. This issue can be in combination with viewport rendering hangs. Sometimes can happen before hang. Sometimes can happen after rendering hang. Our app uses many threads for generating data. We have many mutexes to protect data. But we are pretty sure that there are no deadlocks. Also we checked couple times that we not share Actors and Mappers between threads. We have couple viewports. But I was able to reproduce bug even when only one viewport is active. Interesting that when one viewport can hang, other viewports that uses the same data can work fine. We are using Windows 7/10 x64 and MS VS 2013. Any ideas what can lead to this problem? P.S. Sorry for my English, I am not a native speaker. Thanks, Taras -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Tue Jul 5 07:17:32 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 07:17:32 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: Taras- My initial reaction is that there is a memory leak. If I understand you correctly, you are running 12-15 hours and then problems start occurring. Even a small leak can add up over that length of time and corrupt memory. I would first instrument your classes and see if they are leaking; if not it could be a VTK problem although we try and test for this kind of behavior. Best, W On Tue, Jul 5, 2016 at 7:00 AM, Taras Shchehelskyi wrote: > > > > > Hello, > > > > I am part of the team that is working on project that uses VTK 7.0 for > rendering mesh grid (not big, about 400k triangles) couple point clouds > (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very simple > meshes. Our app can work fine for 12-15h for then main viewport hangs. When > I do some mouse interaction (click + move or scroll) some meshes became > invisible for couple seconds and that became visible again. Also when I > resize viewport window scene aspect ratio became wrong. It works like > aspect ratio does not depends of viewport size. > > > > Also we noticed other strange thing. Our point clouds update dynamically > during app work. We create 2 classes: > > class CameraPointCloudSource : public vtkPolyDataAlgorithm > > and > > class PointCloudColoring : public vtkPolyDataAlgorithm. > > > > I pass data (vtkPolyData with initialized points and vertexes) from > CameraPointCloudSource to PointCloudColoring. I setup pipeline that output > of the CameraPointCloudSource is input of the PointCloudColoring. > > > > Sometimes (can?t always reproduce this) *CameraPointCloudSource:: > RequestData* stops to be called and *CameraPointCloudSource::RequestData* > always get the same input data (the same point cloud the same with the same > points). As result I always see in viewport the same point cloud > visualization. I checked, code that responsible for point cloud generation > works fine and produce correct point clouds. > > > > For test we also tried to use new VTK PointCloud filters. There we also > noticed similar behavior. We set filters chain. Alter some long time (> > 12h of work) sometimes some filters star producing the same output that not > depends of input point cloud. Interesting that some filters work fine and > some start producing the same output. This issue can be in combination with > viewport rendering hangs. Sometimes can happen before hang. Sometimes can > happen after rendering hang. > > > > Our app uses many threads for generating data. We have many mutexes to > protect data. But we are pretty sure that there are no deadlocks. Also we > checked couple times that we not share Actors and Mappers between threads. > We have couple viewports. But I was able to reproduce bug even when only > one viewport is active. Interesting that when one viewport can hang, other > viewports that uses the same data can work fine. > > > > We are using Windows 7/10 x64 and MS VS 2013. > > > > > > Any ideas what can lead to this problem? > > > > P.S. Sorry for my English, I am not a native speaker. > > > > > > Thanks, > > Taras > > _______________________________________________ > 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 > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 07:29:25 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 14:29:25 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: Hello Will, Yes, you are right. Problem starts occurring only after long run time. Thank you for your suggestion about memory leak. Will check this. Thanks, Taras 5 ???. 2016 14:17 "Will Schroeder" ????: > Taras- > > My initial reaction is that there is a memory leak. If I understand you > correctly, you are running 12-15 hours and then problems start occurring. > Even a small leak can add up over that length of time and corrupt memory. I > would first instrument your classes and see if they are leaking; if not it > could be a VTK problem although we try and test for this kind of behavior. > > Best, > W > > On Tue, Jul 5, 2016 at 7:00 AM, Taras Shchehelskyi > wrote: > >> >> >> >> >> Hello, >> >> >> >> I am part of the team that is working on project that uses VTK 7.0 for >> rendering mesh grid (not big, about 400k triangles) couple point clouds >> (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very simple >> meshes. Our app can work fine for 12-15h for then main viewport hangs. When >> I do some mouse interaction (click + move or scroll) some meshes became >> invisible for couple seconds and that became visible again. Also when I >> resize viewport window scene aspect ratio became wrong. It works like >> aspect ratio does not depends of viewport size. >> >> >> >> Also we noticed other strange thing. Our point clouds update dynamically >> during app work. We create 2 classes: >> >> class CameraPointCloudSource : public vtkPolyDataAlgorithm >> >> and >> >> class PointCloudColoring : public vtkPolyDataAlgorithm. >> >> >> >> I pass data (vtkPolyData with initialized points and vertexes) from >> CameraPointCloudSource to PointCloudColoring. I setup pipeline that output >> of the CameraPointCloudSource is input of the PointCloudColoring. >> >> >> >> Sometimes (can?t always reproduce this) *CameraPointCloudSource:: >> RequestData* stops to be called and *CameraPointCloudSource::RequestData* >> always get the same input data (the same point cloud the same with the same >> points). As result I always see in viewport the same point cloud >> visualization. I checked, code that responsible for point cloud generation >> works fine and produce correct point clouds. >> >> >> >> For test we also tried to use new VTK PointCloud filters. There we also >> noticed similar behavior. We set filters chain. Alter some long time (> >> 12h of work) sometimes some filters star producing the same output that not >> depends of input point cloud. Interesting that some filters work fine and >> some start producing the same output. This issue can be in combination with >> viewport rendering hangs. Sometimes can happen before hang. Sometimes can >> happen after rendering hang. >> >> >> >> Our app uses many threads for generating data. We have many mutexes to >> protect data. But we are pretty sure that there are no deadlocks. Also we >> checked couple times that we not share Actors and Mappers between threads. >> We have couple viewports. But I was able to reproduce bug even when only >> one viewport is active. Interesting that when one viewport can hang, other >> viewports that uses the same data can work fine. >> >> >> >> We are using Windows 7/10 x64 and MS VS 2013. >> >> >> >> >> >> Any ideas what can lead to this problem? >> >> >> >> P.S. Sorry for my English, I am not a native speaker. >> >> >> >> >> >> Thanks, >> >> Taras >> >> _______________________________________________ >> 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 >> >> >> > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Tue Jul 5 08:00:16 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 08:00:16 -0400 Subject: [vtk-developers] vtk-developers Digest, Vol 147, Issue 6 In-Reply-To: References: Message-ID: On Mon, Jul 4, 2016 at 7:07 PM, Andrew Maclean wrote: > Have a great Wednesday! Wish I lived closer. Andrew you have an open invitation to come and stay with me, either in Maine or the NY office or both :-) We've got to find a way to get you out here at some point. Best, W -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From mathieu.malaterre at gmail.com Tue Jul 5 08:04:01 2016 From: mathieu.malaterre at gmail.com (Mathieu Malaterre) Date: Tue, 5 Jul 2016 14:04:01 +0200 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: Definitely not up-to-speed these days, but I would have bet for a unsigned long wrap in vtkTimeStamp instead... On Tue, Jul 5, 2016 at 1:17 PM, Will Schroeder wrote: > Taras- > > My initial reaction is that there is a memory leak. If I understand you > correctly, you are running 12-15 hours and then problems start occurring. > Even a small leak can add up over that length of time and corrupt memory. I > would first instrument your classes and see if they are leaking; if not it > could be a VTK problem although we try and test for this kind of behavior. > > Best, > W > > On Tue, Jul 5, 2016 at 7:00 AM, Taras Shchehelskyi > wrote: >> >> >> >> >> >> Hello, >> >> >> >> I am part of the team that is working on project that uses VTK 7.0 for >> rendering mesh grid (not big, about 400k triangles) couple point clouds >> (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very simple >> meshes. Our app can work fine for 12-15h for then main viewport hangs. When >> I do some mouse interaction (click + move or scroll) some meshes became >> invisible for couple seconds and that became visible again. Also when I >> resize viewport window scene aspect ratio became wrong. It works like aspect >> ratio does not depends of viewport size. >> >> >> >> Also we noticed other strange thing. Our point clouds update dynamically >> during app work. We create 2 classes: >> >> class CameraPointCloudSource : public vtkPolyDataAlgorithm >> >> and >> >> class PointCloudColoring : public vtkPolyDataAlgorithm. >> >> >> >> I pass data (vtkPolyData with initialized points and vertexes) from >> CameraPointCloudSource to PointCloudColoring. I setup pipeline that output >> of the CameraPointCloudSource is input of the PointCloudColoring. >> >> >> >> Sometimes (can?t always reproduce this) CameraPointCloudSource:: >> RequestData stops to be called and CameraPointCloudSource::RequestData >> always get the same input data (the same point cloud the same with the same >> points). As result I always see in viewport the same point cloud >> visualization. I checked, code that responsible for point cloud generation >> works fine and produce correct point clouds. >> >> >> >> For test we also tried to use new VTK PointCloud filters. There we also >> noticed similar behavior. We set filters chain. Alter some long time (> 12h >> of work) sometimes some filters star producing the same output that not >> depends of input point cloud. Interesting that some filters work fine and >> some start producing the same output. This issue can be in combination with >> viewport rendering hangs. Sometimes can happen before hang. Sometimes can >> happen after rendering hang. >> >> >> >> Our app uses many threads for generating data. We have many mutexes to >> protect data. But we are pretty sure that there are no deadlocks. Also we >> checked couple times that we not share Actors and Mappers between threads. >> We have couple viewports. But I was able to reproduce bug even when only one >> viewport is active. Interesting that when one viewport can hang, other >> viewports that uses the same data can work fine. >> >> >> >> We are using Windows 7/10 x64 and MS VS 2013. >> >> >> >> >> >> Any ideas what can lead to this problem? >> >> >> >> P.S. Sorry for my English, I am not a native speaker. >> >> >> >> >> >> Thanks, >> >> Taras >> >> >> _______________________________________________ >> 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 >> >> > > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > _______________________________________________ > 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 > > -- Mathieu From will.schroeder at kitware.com Tue Jul 5 08:16:46 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 08:16:46 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: I like this theory! It's hard to imagine an unsigned long wrapping around, but there are a lot of mtime updates happening in VTK especially over 12 hrs.... On Tue, Jul 5, 2016 at 8:04 AM, Mathieu Malaterre < mathieu.malaterre at gmail.com> wrote: > Definitely not up-to-speed these days, but I would have bet for a > unsigned long wrap in vtkTimeStamp instead... > > On Tue, Jul 5, 2016 at 1:17 PM, Will Schroeder > wrote: > > Taras- > > > > My initial reaction is that there is a memory leak. If I understand you > > correctly, you are running 12-15 hours and then problems start occurring. > > Even a small leak can add up over that length of time and corrupt > memory. I > > would first instrument your classes and see if they are leaking; if not > it > > could be a VTK problem although we try and test for this kind of > behavior. > > > > Best, > > W > > > > On Tue, Jul 5, 2016 at 7:00 AM, Taras Shchehelskyi < > shchegelskij at gmail.com> > > wrote: > >> > >> > >> > >> > >> > >> Hello, > >> > >> > >> > >> I am part of the team that is working on project that uses VTK 7.0 for > >> rendering mesh grid (not big, about 400k triangles) couple point clouds > >> (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very > simple > >> meshes. Our app can work fine for 12-15h for then main viewport hangs. > When > >> I do some mouse interaction (click + move or scroll) some meshes became > >> invisible for couple seconds and that became visible again. Also when I > >> resize viewport window scene aspect ratio became wrong. It works like > aspect > >> ratio does not depends of viewport size. > >> > >> > >> > >> Also we noticed other strange thing. Our point clouds update dynamically > >> during app work. We create 2 classes: > >> > >> class CameraPointCloudSource : public vtkPolyDataAlgorithm > >> > >> and > >> > >> class PointCloudColoring : public vtkPolyDataAlgorithm. > >> > >> > >> > >> I pass data (vtkPolyData with initialized points and vertexes) from > >> CameraPointCloudSource to PointCloudColoring. I setup pipeline that > output > >> of the CameraPointCloudSource is input of the PointCloudColoring. > >> > >> > >> > >> Sometimes (can?t always reproduce this) CameraPointCloudSource:: > >> RequestData stops to be called and CameraPointCloudSource::RequestData > >> always get the same input data (the same point cloud the same with the > same > >> points). As result I always see in viewport the same point cloud > >> visualization. I checked, code that responsible for point cloud > generation > >> works fine and produce correct point clouds. > >> > >> > >> > >> For test we also tried to use new VTK PointCloud filters. There we also > >> noticed similar behavior. We set filters chain. Alter some long time > (> 12h > >> of work) sometimes some filters star producing the same output that not > >> depends of input point cloud. Interesting that some filters work fine > and > >> some start producing the same output. This issue can be in combination > with > >> viewport rendering hangs. Sometimes can happen before hang. Sometimes > can > >> happen after rendering hang. > >> > >> > >> > >> Our app uses many threads for generating data. We have many mutexes to > >> protect data. But we are pretty sure that there are no deadlocks. Also > we > >> checked couple times that we not share Actors and Mappers between > threads. > >> We have couple viewports. But I was able to reproduce bug even when > only one > >> viewport is active. Interesting that when one viewport can hang, other > >> viewports that uses the same data can work fine. > >> > >> > >> > >> We are using Windows 7/10 x64 and MS VS 2013. > >> > >> > >> > >> > >> > >> Any ideas what can lead to this problem? > >> > >> > >> > >> P.S. Sorry for my English, I am not a native speaker. > >> > >> > >> > >> > >> > >> Thanks, > >> > >> Taras > >> > >> > >> _______________________________________________ > >> 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 > >> > >> > > > > > > > > -- > > William J. Schroeder, PhD > > Kitware, Inc. - Building the World's Technical Computing Software > > 28 Corporate Drive > > Clifton Park, NY 12065 > > will.schroeder at kitware.com > > http://www.kitware.com > > (518) 881-4902 > > > > _______________________________________________ > > 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 > > > > > > > > -- > Mathieu > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 08:19:34 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 15:19:34 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: <577ba5d7.0a4c190a.b997f.fffff159@mx.google.com> Hello, Can I somehow check this theory? Thanks, Taras From: Will Schroeder -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Tue Jul 5 08:36:53 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 08:36:53 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <577ba5d7.0a4c190a.b997f.fffff159@mx.google.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <577ba5d7.0a4c190a.b997f.fffff159@mx.google.com> Message-ID: A simple way is to modify VTK/Common/Core/vtkTimeStamp.cxx, the Modified() method. Insert the appropriate logic to conditionally print out if an integral wrap occurs. There may be a better way but this should do the trick. Let us know how this works out, you've got my curiosity piqued :-) Best, W On Tue, Jul 5, 2016 at 8:19 AM, Taras Shchehelskyi wrote: > Hello, > > > > Can I somehow check this theory? > > > > Thanks, > > Taras > > > > *From: *Will Schroeder > *Sent: *5 ????? 2016 ?. 15:17 > *To: *Mathieu Malaterre > *Cc: *Taras Shchehelskyi ; vtk-developers at vtk.org > *Subject: *Re: [vtk-developers] Rendering hangs in viewport > > > > I like this theory! It's hard to imagine an unsigned long wrapping around, > but there are a lot of mtime updates happening in VTK especially over 12 > hrs.... > > > > On Tue, Jul 5, 2016 at 8:04 AM, Mathieu Malaterre < > mathieu.malaterre at gmail.com> wrote: > > Definitely not up-to-speed these days, but I would have bet for a > unsigned long wrap in vtkTimeStamp instead... > > > On Tue, Jul 5, 2016 at 1:17 PM, Will Schroeder > wrote: > > Taras- > > > > My initial reaction is that there is a memory leak. If I understand you > > correctly, you are running 12-15 hours and then problems start occurring. > > Even a small leak can add up over that length of time and corrupt > memory. I > > would first instrument your classes and see if they are leaking; if not > it > > could be a VTK problem although we try and test for this kind of > behavior. > > > > Best, > > W > > > > On Tue, Jul 5, 2016 at 7:00 AM, Taras Shchehelskyi < > shchegelskij at gmail.com> > > wrote: > >> > >> > >> > >> > >> > >> Hello, > >> > >> > >> > >> I am part of the team that is working on project that uses VTK 7.0 for > >> rendering mesh grid (not big, about 400k triangles) couple point clouds > >> (about 8000 ? 30000 points per cloud, up to 4 clouds) + couple very > simple > >> meshes. Our app can work fine for 12-15h for then main viewport hangs. > When > >> I do some mouse interaction (click + move or scroll) some meshes became > >> invisible for couple seconds and that became visible again. Also when I > >> resize viewport window scene aspect ratio became wrong. It works like > aspect > >> ratio does not depends of viewport size. > >> > >> > >> > >> Also we noticed other strange thing. Our point clouds update dynamically > >> during app work. We create 2 classes: > >> > >> class CameraPointCloudSource : public vtkPolyDataAlgorithm > >> > >> and > >> > >> class PointCloudColoring : public vtkPolyDataAlgorithm. > >> > >> > >> > >> I pass data (vtkPolyData with initialized points and vertexes) from > >> CameraPointCloudSource to PointCloudColoring. I setup pipeline that > output > >> of the CameraPointCloudSource is input of the PointCloudColoring. > >> > >> > >> > >> Sometimes (can?t always reproduce this) CameraPointCloudSource:: > >> RequestData stops to be called and CameraPointCloudSource::RequestData > >> always get the same input data (the same point cloud the same with the > same > >> points). As result I always see in viewport the same point cloud > >> visualization. I checked, code that responsible for point cloud > generation > >> works fine and produce correct point clouds. > >> > >> > >> > >> For test we also tried to use new VTK PointCloud filters. There we also > >> noticed similar behavior. We set filters chain. Alter some long time > (> 12h > >> of work) sometimes some filters star producing the same output that not > >> depends of input point cloud. Interesting that some filters work fine > and > >> some start producing the same output. This issue can be in combination > with > >> viewport rendering hangs. Sometimes can happen before hang. Sometimes > can > >> happen after rendering hang. > >> > >> > >> > >> Our app uses many threads for generating data. We have many mutexes to > >> protect data. But we are pretty sure that there are no deadlocks. Also > we > >> checked couple times that we not share Actors and Mappers between > threads. > >> We have couple viewports. But I was able to reproduce bug even when > only one > >> viewport is active. Interesting that when one viewport can hang, other > >> viewports that uses the same data can work fine. > >> > >> > >> > >> We are using Windows 7/10 x64 and MS VS 2013. > >> > >> > >> > >> > >> > >> Any ideas what can lead to this problem? > >> > >> > >> > >> P.S. Sorry for my English, I am not a native speaker. > >> > >> > >> > >> > >> > >> Thanks, > >> > >> Taras > >> > >> > >> _______________________________________________ > >> 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 > >> > >> > > > > > > > > -- > > William J. Schroeder, PhD > > Kitware, Inc. - Building the World's Technical Computing Software > > 28 Corporate Drive > > Clifton Park, NY 12065 > > will.schroeder at kitware.com > > http://www.kitware.com > > (518) 881-4902 > > > > _______________________________________________ > > 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 > > > > > > > > -- > > Mathieu > > > > > > -- > > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 08:40:56 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 15:40:56 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <577ba5d7.0a4c190a.b997f.fffff159@mx.google.com> Message-ID: <577baad9.13472e0a.c8445.ffff9d65@mx.google.com> Thank you. Will try today. From: Will Schroeder -------------- next part -------------- An HTML attachment was scrubbed... URL: From sean at rogue-research.com Tue Jul 5 11:02:32 2016 From: sean at rogue-research.com (Sean McBride) Date: Tue, 5 Jul 2016 11:02:32 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> Message-ID: <20160705150232.2016783904@mail.rogue-research.com> On Tue, 5 Jul 2016 08:16:46 -0400, Will Schroeder said: >I like this theory! It's hard to imagine an unsigned long wrapping around, Not on Windows, where unsigned long is only 32 bits. :( Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From zhangjiang.dudu at gmail.com Tue Jul 5 11:33:24 2016 From: zhangjiang.dudu at gmail.com (=?utf-8?B?5byg5rGf?=) Date: Tue, 5 Jul 2016 10:33:24 -0500 Subject: [vtk-developers] A problem of data partitioning Message-ID: Hi, In the below code, I was trying to read and partition an unstructured grid data (11.7 GB). However, it always came out an error that: terminate called after throwing an instance of 'std::bad_alloc' what(): std::bad_alloc =================================================================================== = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = PID 14658 RUNNING AT compute001 = EXIT CODE: 134 = CLEANING UP REMAINING PROCESSES = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES =================================================================================== Actually the memory of my computer is about 1.5 TB (96*16 GB modules). I think the memory is enough. Can anyone help me about this problem? Any operation in the program is wrong? int main(int argc, char** argv) { vtkMPIController *contr = vtkMPIController::New(); contr->Initialize(&argc, &argv); vtkMultiProcessController::SetGlobalController(contr); int numProcs = contr->GetNumberOfProcesses(); int me = contr->GetLocalProcessId(); vtkUnstructuredGridReader *reader = vtkUnstructuredGridReader::New(); vtkUnstructuredGrid *ds = NULL; if (me == 0) { reader->SetFileName(?data.vtk"); ds = reader->GetOutput(); reader->Update(); } else { ds = vtkUnstructuredGrid::New(); } vtkDistributedDataFilter *dd = vtkDistributedDataFilter::New(); dd->SetInputData(ds); dd->SetController(contr); dd->UseMinimalMemoryOff(); dd->SetBoundaryModeToAssignToOneRegion(); dd->Update(); int ncells = vtkUnstructuredGrid::SafeDownCast(dd->GetOutput())->GetNumberOfCells(); fprintf(stderr, "rank = %d, number of cells = %d\n", me, ncells); return 0; } -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 12:31:29 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 19:31:29 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <20160705150232.2016783904@mail.rogue-research.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> Message-ID: <577be0e2.ca52190a.58dc5.14ec@mx.google.com> Hello, Just checked. Our app does enough vtkTimeStamp::Modified() calls to wrapping 32 bit unsigned long in about 15-20h (depending on how many other processes I have on my PC). When enable more features ? less time necessary. I am new to VTK. Can you please give me advice how hard (how many changes necessary) to move this->ModifiedTime to int64 type in windows? If move to int64 type, from my rough estimate app should work fine for 400000+ years. And this more than enough for any use case ? Thanks, Taras From: Sean McBride -------------- next part -------------- An HTML attachment was scrubbed... URL: From berk.geveci at kitware.com Tue Jul 5 13:52:04 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Tue, 5 Jul 2016 13:52:04 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <577be0e2.ca52190a.58dc5.14ec@mx.google.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> <577be0e2.ca52190a.58dc5.14ec@mx.google.com> Message-ID: We use 64 bit atomic integer for timestamp on 64 bit Windows builds. We can't use 64 bit on 32 bit builds because not all versions of Windows support 64 bit atomic ints on 32 bit systems. It is also not possible to check this at compile time as it is a runtime issue. The simplest fix is to just compile a 64 bit binary. Best, -berk On Tue, Jul 5, 2016 at 12:31 PM, Taras Shchehelskyi wrote: > Hello, > > > > Just checked. Our app does enough vtkTimeStamp::Modified() calls to > wrapping 32 bit unsigned long in about 15-20h (depending on how many > other processes I have on my PC). When enable more features ? less time > necessary. > > > > I am new to VTK. Can you please give me advice how hard (how many changes > necessary) to move this->ModifiedTime to int64 type in windows? > > > > If move to int64 type, from my rough estimate app should work fine for > 400000+ years. And this more than enough for any use case ? > > > > Thanks, > > Taras > > > > *From: *Sean McBride > *Sent: *5 ????? 2016 ?. 18:02 > *To: *Will Schroeder ; Mathieu Malaterre > > *Cc: *vtk-developers at vtk.org > *Subject: *Re: [vtk-developers] Rendering hangs in viewport > > > > On Tue, 5 Jul 2016 08:16:46 -0400, Will Schroeder said: > > > > >I like this theory! It's hard to imagine an unsigned long wrapping around, > > > > Not on Windows, where unsigned long is only 32 bits. :( > > > > Cheers, > > > > -- > > ____________________________________________________________ > > Sean McBride, B. Eng sean at rogue-research.com > > Rogue Research www.rogue-research.com > > Mac Software Developer Montr?al, Qu?bec, Canada > > > > > > _______________________________________________ > > 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 will.schroeder at kitware.com Tue Jul 5 13:53:49 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 13:53:49 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <577be0e2.ca52190a.58dc5.14ec@mx.google.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> <577be0e2.ca52190a.58dc5.14ec@mx.google.com> Message-ID: Okay Taras, I talked to someone smarter than me and it may be that some versions of Windows support vtkAtomicInt64 ..... and at one point there were some versions that did not, I'm not sure what the current status is. So for a simple fix I would try using vtkAtomicInt64 on your machine and recompiling. If that doesn't work we'll have to scratch our heads a little more. On Tue, Jul 5, 2016 at 12:31 PM, Taras Shchehelskyi wrote: > Hello, > > > > Just checked. Our app does enough vtkTimeStamp::Modified() calls to > wrapping 32 bit unsigned long in about 15-20h (depending on how many > other processes I have on my PC). When enable more features ? less time > necessary. > > > > I am new to VTK. Can you please give me advice how hard (how many changes > necessary) to move this->ModifiedTime to int64 type in windows? > > > > If move to int64 type, from my rough estimate app should work fine for > 400000+ years. And this more than enough for any use case ? > > > > Thanks, > > Taras > > > > *From: *Sean McBride > *Sent: *5 ????? 2016 ?. 18:02 > *To: *Will Schroeder ; Mathieu Malaterre > > *Cc: *vtk-developers at vtk.org > *Subject: *Re: [vtk-developers] Rendering hangs in viewport > > > > On Tue, 5 Jul 2016 08:16:46 -0400, Will Schroeder said: > > > > >I like this theory! It's hard to imagine an unsigned long wrapping around, > > > > Not on Windows, where unsigned long is only 32 bits. :( > > > > Cheers, > > > > -- > > ____________________________________________________________ > > Sean McBride, B. Eng sean at rogue-research.com > > Rogue Research www.rogue-research.com > > Mac Software Developer Montr?al, Qu?bec, Canada > > > > > > _______________________________________________ > > 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 > > > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Tue Jul 5 14:34:14 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 5 Jul 2016 14:34:14 -0400 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: <577bf9b5.d5052e0a.d3d9a.ffffce07@mx.google.com> References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> <577be0e2.ca52190a.58dc5.14ec@mx.google.com> <577bf9b5.d5052e0a.d3d9a.ffffce07@mx.google.com> Message-ID: Okay we have a VTK hackathon tomorrow, I'll talk with some of the folks there.... Best, W On Tue, Jul 5, 2016 at 2:17 PM, Taras Shchehelskyi wrote: > I did. Out app work only in 64bit mode. I see that VTK really uses > vtkAtomicInt64 for GlobalTimeStamp. But what about next line > > *this->ModifiedTime = (unsigned long)++GlobalTimeStamp;* (in > vtkTimeStamp::Modified()) > > > > Also I checked and everywhere in code timestamp is unsigned long. In > windows x64 unsigned long is 32bit type. So despite 64bit vtkAtomicInt64 we > still have conversion to unsigned long. > > > > Thanks, > > Taras > > > > > > *From: *Will Schroeder > *Sent: *5 ????? 2016 ?. 21:00 > *To: *Taras Shchehelskyi > > *Subject: *Re: [vtk-developers] Rendering hangs in viewport > > > > Taras why aren't you building 64-bit? That's the easiest solution... > > > > On Tue, Jul 5, 2016 at 1:53 PM, Will Schroeder > wrote: > > Okay Taras, I talked to someone smarter than me and it may be that some > versions of Windows support vtkAtomicInt64 ..... and at one point there > were some versions that did not, I'm not sure what the current status is. > So for a simple fix I would try using vtkAtomicInt64 on your machine and > recompiling. If that doesn't work we'll have to scratch our heads a little > more. > > > > > > On Tue, Jul 5, 2016 at 12:31 PM, Taras Shchehelskyi < > shchegelskij at gmail.com> wrote: > > Hello, > > > > Just checked. Our app does enough vtkTimeStamp::Modified() calls to > wrapping 32 bit unsigned long in about 15-20h (depending on how many > other processes I have on my PC). When enable more features ? less time > necessary. > > > > I am new to VTK. Can you please give me advice how hard (how many changes > necessary) to move this->ModifiedTime to int64 type in windows? > > > > If move to int64 type, from my rough estimate app should work fine for > 400000+ years. And this more than enough for any use case ? > > > > Thanks, > > Taras > > > > *From: *Sean McBride > *Sent: *5 ????? 2016 ?. 18:02 > *To: *Will Schroeder ; Mathieu Malaterre > > *Cc: *vtk-developers at vtk.org > *Subject: *Re: [vtk-developers] Rendering hangs in viewport > > > > On Tue, 5 Jul 2016 08:16:46 -0400, Will Schroeder said: > > > > >I like this theory! It's hard to imagine an unsigned long wrapping around, > > > > Not on Windows, where unsigned long is only 32 bits. :( > > > > Cheers, > > > > -- > > ____________________________________________________________ > > Sean McBride, B. Eng sean at rogue-research.com > > Rogue Research www.rogue-research.com > > Mac Software Developer Montr?al, Qu?bec, Canada > > > > > > _______________________________________________ > > 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 > > > > > > > > > > -- > > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > > > > > -- > > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Tue Jul 5 14:37:54 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Tue, 5 Jul 2016 21:37:54 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> <577be0e2.ca52190a.58dc5.14ec@mx.google.com> <577bf9b5.d5052e0a.d3d9a.ffffce07@mx.google.com> Message-ID: <577bfe83.c1cd190a.55ad7.ffffbfb8@mx.google.com> Thank you very much for your support. From: Will Schroeder -------------- next part -------------- An HTML attachment was scrubbed... URL: From sankhesh.jhaveri at kitware.com Tue Jul 5 17:00:06 2016 From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri) Date: Tue, 5 Jul 2016 17:00:06 -0400 Subject: [vtk-developers] Orientation marker widget SetViewport call Message-ID: Hello devs, The orientation marker widget has a SetViewport call that confines the widget to the supplied viewport coordinates in the render window. This works fine as long as the renderer (that the widget is added to) occupies the whole render window. The widget treats the supplied viewport coordinates as absolute and does not scale them according to the viewport of the vtkInteractorObserver::CurrentRenderer(). While this is a special case, all other widget representations are scaled with respect to the viewport of the current renderer. I have a merge request (https://gitlab.kitware.com/vtk/vtk/merge_requests/1607) that addresses this issue. Since, this alters the class behavior, I'd see if anyone has issues/comments/objections to making the change. ?Thanks! Sankhesh -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Tue Jul 5 18:34:30 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Tue, 5 Jul 2016 16:34:30 -0600 Subject: [vtk-developers] [vtkusers] VTK bug hackathon In-Reply-To: References: Message-ID: Looks like I might only be in the hangout for the first hour, but I'll be available by email most of the day. - David On Tue, Jun 28, 2016 at 7:46 AM, Berk Geveci wrote: > Hi folks, > > We are all set with July 6, 9-5 EDT. I reserved a large conference room at > Kitware Headquarters in Clifton Park, NY. If you are interested in > attending physically, please let me know. We will also have a Google > Hangout running. Here is the link: > > https://hangouts.google.com/hangouts/_/kitware.com/vtk-hackathon > > I will check email regularly in the morning so please send me a note then > if you have issues joining. > > Did I forget anything? > > Best, > -berk > > > > On Mon, Jun 20, 2016 at 4:31 PM, Berk Geveci > wrote: > >> Hi folks, >> >> We are planning a VTK bug hackathon on July 6. Going forward, we would >> like this to be a monthly event during which the community comes together >> to address bugs reported to the tracker. It will be an all day event that >> we will host physically at Kitware Headquarters in Clifton Park, NY as well >> as online via Google Hangout. All are welcome to attend as long as they >> know the code base enough to fix bugs. >> >> Any questions? Comments? Any potential issue with the date? >> >> Best, >> -berk >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhangjiang.dudu at gmail.com Wed Jul 6 00:45:08 2016 From: zhangjiang.dudu at gmail.com (=?utf-8?B?5byg5rGf?=) Date: Tue, 5 Jul 2016 23:45:08 -0500 Subject: [vtk-developers] point to process id after data partitioning Message-ID: Hi, I use vtkDistributedDataFilter to partition the unstructured data. And my question is, after the partitioning, given a point, how can I know which process owns it? Thanks. Best regards, From mathieu.westphal at kitware.com Wed Jul 6 10:49:09 2016 From: mathieu.westphal at kitware.com (Mathieu Westphal) Date: Wed, 6 Jul 2016 16:49:09 +0200 Subject: [vtk-developers] VTK/Paraview Courses in October Message-ID: Hello Kitware will be holding a 2-day VTK and ParaView course on October 11th and 12th 2016 in Lyon, France. Please visit our web site for more information and registration details at VTK (English) : http://training.kitware.fr/browse/130 VTK (French) : http://formations.kitware.fr/browse/130 ParaView (English) : http://training.kitware.fr/browse/131 ParaView (French) : http://formations.kitware.fr/browse/131 Note that the course will be taught in English. If you have any question, please contact us at formations at http://www.kitware.fr Thank you, Mathieu Westphal -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Thu Jul 7 08:56:14 2016 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 7 Jul 2016 08:56:14 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well Message-ID: Hey Folks, As Robert just pointed out, the +1 state in the review process is nothing but trouble. In practice, it seems to help us avoid responsibility and delays merging, contributing to the accumulation of stale merge requests and newbie frustration. How about we say "+1 means I approve this code and assert that it is ready to merge". Merge requests from a authorized developers should be promptly merged after the +1 by the original author. Merge requests from unauthorized developers should be merged more or less immediately by the authorized reviewer after the +1. In either case both authors and reviewers are responsible for watching the dashboards and addressing issues that come up afterward. 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 david.lonie at kitware.com Thu Jul 7 09:25:45 2016 From: david.lonie at kitware.com (David Lonie) Date: Thu, 7 Jul 2016 09:25:45 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well In-Reply-To: References: Message-ID: On Thu, Jul 7, 2016 at 8:56 AM, David E DeMarle wrote: > As Robert just pointed out, the +1 state in the review process is nothing > but trouble. In practice, it seems to help us avoid responsibility and > delays merging, contributing to the accumulation of stale merge requests > and newbie frustration. > > How about we say "+1 means I approve this code and assert that it is ready > to merge". Merge requests from a authorized developers should be promptly > merged after the +1 by the original author. Merge requests from > unauthorized developers should be merged more or less immediately by the > authorized reviewer after the +1. > +1 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brad.king at kitware.com Thu Jul 7 09:56:09 2016 From: brad.king at kitware.com (Brad King) Date: Thu, 7 Jul 2016 09:56:09 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: Message-ID: <577E5F79.9000806@kitware.com> On 07/07/2016 08:56 AM, David E DeMarle wrote: > In either case both authors and reviewers are responsible for > watching the dashboards and addressing issues that come up afterward. This is the real reason people are hesitant to +2. If we strengthen the meaning of +1 then people will just say "LGTM" or something else to "approve" without taking responsibility. Posting +1 is a common convention for voting and should not be given stronger meaning. The real problem we'd like to address is stagnation of MRs that are ready but not merged. The syntax for approving is not very important. We should identify reasons MRs stagnate and address them directly: * It may be a governance problem. No one has responsibility to ensure everything that is ready gets merged. * It may be a workflow problem, like waiting for buildbot results to approve and then forgetting. The new workflow:buildbots label may help here. -Brad From will.schroeder at kitware.com Thu Jul 7 10:10:00 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Thu, 7 Jul 2016 10:10:00 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: <577E5F79.9000806@kitware.com> References: <577E5F79.9000806@kitware.com> Message-ID: It's a time problem: I see chunks of code which at a superficial level seem fine and I'm willing to give it a +1. The problem with a +2 is that issuing such a score has implications that I've jumped into the code at a deep level and can vouch for it. This can occasionally mean hours or days of work hence the delay. Maybe it's because I'm a chicken and should have faith (or otherwise ensure) that there is enough testing to proof it out, but I feel reluctant in some cases to give a +2 without enough due diligence which takes time I often don't have. Besides more testing, one thing that might help would be more "documentation" providing me with the confidence that important issues have been considered and accounted for..... Best, W On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: > On 07/07/2016 08:56 AM, David E DeMarle wrote: > > In either case both authors and reviewers are responsible for > > watching the dashboards and addressing issues that come up afterward. > > This is the real reason people are hesitant to +2. If we strengthen > the meaning of +1 then people will just say "LGTM" or something else > to "approve" without taking responsibility. Posting +1 is a common > convention for voting and should not be given stronger meaning. > > The real problem we'd like to address is stagnation of MRs that > are ready but not merged. The syntax for approving is not very > important. We should identify reasons MRs stagnate and address > them directly: > > * It may be a governance problem. No one has responsibility > to ensure everything that is ready gets merged. > > * It may be a workflow problem, like waiting for buildbot results > to approve and then forgetting. The new workflow:buildbots > label may help here. > > -Brad > > _______________________________________________ > 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 > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.maynard at kitware.com Thu Jul 7 10:18:08 2016 From: robert.maynard at kitware.com (Robert Maynard) Date: Thu, 7 Jul 2016 10:18:08 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: Currently I believe the issue is that the reviewer is equally at responsible for watching the dashboard for regressions caused by MR. Moving to a simple Yes/No +1/No system where the responsibility for fixing failures falls to the author would solve the hesitation of reviewers to give +2. On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder wrote: > It's a time problem: I see chunks of code which at a superficial level seem > fine and I'm willing to give it a +1. The problem with a +2 is that issuing > such a score has implications that I've jumped into the code at a deep level > and can vouch for it. This can occasionally mean hours or days of work hence > the delay. Maybe it's because I'm a chicken and should have faith (or > otherwise ensure) that there is enough testing to proof it out, but I feel > reluctant in some cases to give a +2 without enough due diligence which > takes time I often don't have. Besides more testing, one thing that might > help would be more "documentation" providing me with the confidence that > important issues have been considered and accounted for..... > > Best, > W > > > On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: >> >> On 07/07/2016 08:56 AM, David E DeMarle wrote: >> > In either case both authors and reviewers are responsible for >> > watching the dashboards and addressing issues that come up afterward. >> >> This is the real reason people are hesitant to +2. If we strengthen >> the meaning of +1 then people will just say "LGTM" or something else >> to "approve" without taking responsibility. Posting +1 is a common >> convention for voting and should not be given stronger meaning. >> >> The real problem we'd like to address is stagnation of MRs that >> are ready but not merged. The syntax for approving is not very >> important. We should identify reasons MRs stagnate and address >> them directly: >> >> * It may be a governance problem. No one has responsibility >> to ensure everything that is ready gets merged. >> >> * It may be a workflow problem, like waiting for buildbot results >> to approve and then forgetting. The new workflow:buildbots >> label may help here. >> >> -Brad >> >> _______________________________________________ >> 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 >> > > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > _______________________________________________ > 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 > > From ken.martin at kitware.com Thu Jul 7 11:18:30 2016 From: ken.martin at kitware.com (Ken Martin) Date: Thu, 7 Jul 2016 11:18:30 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: For me I try to avoid +1 as it still leaves the main developer stuck. For me +2 means the buildbots look green and I looked at the changes and didn't see anything obvious that looks bad (mainly in terms of approach). It does not mean I downloaded the topic, compiled it, tested it, understood every line, etc. Maybe that is wrong but I feel like the main responsibility still lies with the topic author and I am just doing a sanity check. On Thu, Jul 7, 2016 at 10:18 AM, Robert Maynard wrote: > Currently I believe the issue is that the reviewer is equally at > responsible for watching the dashboard for regressions caused by MR. > > Moving to a simple Yes/No +1/No system where the responsibility for > fixing failures falls to the author would solve the hesitation of > reviewers to give +2. > > On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder > wrote: > > It's a time problem: I see chunks of code which at a superficial level > seem > > fine and I'm willing to give it a +1. The problem with a +2 is that > issuing > > such a score has implications that I've jumped into the code at a deep > level > > and can vouch for it. This can occasionally mean hours or days of work > hence > > the delay. Maybe it's because I'm a chicken and should have faith (or > > otherwise ensure) that there is enough testing to proof it out, but I > feel > > reluctant in some cases to give a +2 without enough due diligence which > > takes time I often don't have. Besides more testing, one thing that might > > help would be more "documentation" providing me with the confidence that > > important issues have been considered and accounted for..... > > > > Best, > > W > > > > > > On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: > >> > >> On 07/07/2016 08:56 AM, David E DeMarle wrote: > >> > In either case both authors and reviewers are responsible for > >> > watching the dashboards and addressing issues that come up afterward. > >> > >> This is the real reason people are hesitant to +2. If we strengthen > >> the meaning of +1 then people will just say "LGTM" or something else > >> to "approve" without taking responsibility. Posting +1 is a common > >> convention for voting and should not be given stronger meaning. > >> > >> The real problem we'd like to address is stagnation of MRs that > >> are ready but not merged. The syntax for approving is not very > >> important. We should identify reasons MRs stagnate and address > >> them directly: > >> > >> * It may be a governance problem. No one has responsibility > >> to ensure everything that is ready gets merged. > >> > >> * It may be a workflow problem, like waiting for buildbot results > >> to approve and then forgetting. The new workflow:buildbots > >> label may help here. > >> > >> -Brad > >> > >> _______________________________________________ > >> 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 > >> > > > > > > > > -- > > William J. Schroeder, PhD > > Kitware, Inc. - Building the World's Technical Computing Software > > 28 Corporate Drive > > Clifton Park, NY 12065 > > will.schroeder at kitware.com > > http://www.kitware.com > > (518) 881-4902 > > > > _______________________________________________ > > 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 > > -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From berk.geveci at kitware.com Thu Jul 7 13:27:09 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 13:27:09 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: > Maybe it's because I'm a chicken +1 On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder wrote: > It's a time problem: I see chunks of code which at a superficial level > seem fine and I'm willing to give it a +1. The problem with a +2 is that > issuing such a score has implications that I've jumped into the code at a > deep level and can vouch for it. This can occasionally mean hours or days > of work hence the delay. Maybe it's because I'm a chicken and should have > faith (or otherwise ensure) that there is enough testing to proof it out, > but I feel reluctant in some cases to give a +2 without enough due > diligence which takes time I often don't have. Besides more testing, one > thing that might help would be more "documentation" providing me with the > confidence that important issues have been considered and accounted for..... > > Best, > W > > > On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: > >> On 07/07/2016 08:56 AM, David E DeMarle wrote: >> > In either case both authors and reviewers are responsible for >> > watching the dashboards and addressing issues that come up afterward. >> >> This is the real reason people are hesitant to +2. If we strengthen >> the meaning of +1 then people will just say "LGTM" or something else >> to "approve" without taking responsibility. Posting +1 is a common >> convention for voting and should not be given stronger meaning. >> >> The real problem we'd like to address is stagnation of MRs that >> are ready but not merged. The syntax for approving is not very >> important. We should identify reasons MRs stagnate and address >> them directly: >> >> * It may be a governance problem. No one has responsibility >> to ensure everything that is ready gets merged. >> >> * It may be a workflow problem, like waiting for buildbot results >> to approve and then forgetting. The new workflow:buildbots >> label may help here. >> >> -Brad >> >> _______________________________________________ >> 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 >> >> > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > _______________________________________________ > 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 berk.geveci at kitware.com Thu Jul 7 13:38:06 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 13:38:06 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: Who said that the reviewer is responsible for watching the dashboard? And why would changing the meaning of +1 change anything? I regularly +2 MRs and don't follow the dashboard. If some developer starts regularly breaking things, I can always stop reviewing their code. On Thu, Jul 7, 2016 at 10:18 AM, Robert Maynard wrote: > Currently I believe the issue is that the reviewer is equally at > responsible for watching the dashboard for regressions caused by MR. > > Moving to a simple Yes/No +1/No system where the responsibility for > fixing failures falls to the author would solve the hesitation of > reviewers to give +2. > > On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder > wrote: > > It's a time problem: I see chunks of code which at a superficial level > seem > > fine and I'm willing to give it a +1. The problem with a +2 is that > issuing > > such a score has implications that I've jumped into the code at a deep > level > > and can vouch for it. This can occasionally mean hours or days of work > hence > > the delay. Maybe it's because I'm a chicken and should have faith (or > > otherwise ensure) that there is enough testing to proof it out, but I > feel > > reluctant in some cases to give a +2 without enough due diligence which > > takes time I often don't have. Besides more testing, one thing that might > > help would be more "documentation" providing me with the confidence that > > important issues have been considered and accounted for..... > > > > Best, > > W > > > > > > On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: > >> > >> On 07/07/2016 08:56 AM, David E DeMarle wrote: > >> > In either case both authors and reviewers are responsible for > >> > watching the dashboards and addressing issues that come up afterward. > >> > >> This is the real reason people are hesitant to +2. If we strengthen > >> the meaning of +1 then people will just say "LGTM" or something else > >> to "approve" without taking responsibility. Posting +1 is a common > >> convention for voting and should not be given stronger meaning. > >> > >> The real problem we'd like to address is stagnation of MRs that > >> are ready but not merged. The syntax for approving is not very > >> important. We should identify reasons MRs stagnate and address > >> them directly: > >> > >> * It may be a governance problem. No one has responsibility > >> to ensure everything that is ready gets merged. > >> > >> * It may be a workflow problem, like waiting for buildbot results > >> to approve and then forgetting. The new workflow:buildbots > >> label may help here. > >> > >> -Brad > >> > >> _______________________________________________ > >> 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 > >> > > > > > > > > -- > > William J. Schroeder, PhD > > Kitware, Inc. - Building the World's Technical Computing Software > > 28 Corporate Drive > > Clifton Park, NY 12065 > > will.schroeder at kitware.com > > http://www.kitware.com > > (518) 881-4902 > > > > _______________________________________________ > > 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 berk.geveci at kitware.com Thu Jul 7 13:38:31 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 13:38:31 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: +2 On Thu, Jul 7, 2016 at 11:18 AM, Ken Martin wrote: > For me I try to avoid +1 as it still leaves the main developer stuck. For > me +2 means the buildbots look green and I looked at the changes and didn't > see anything obvious that looks bad (mainly in terms of approach). It does > not mean I downloaded the topic, compiled it, tested it, understood every > line, etc. Maybe that is wrong but I feel like the main responsibility > still lies with the topic author and I am just doing a sanity check. > > On Thu, Jul 7, 2016 at 10:18 AM, Robert Maynard < > robert.maynard at kitware.com> wrote: > >> Currently I believe the issue is that the reviewer is equally at >> responsible for watching the dashboard for regressions caused by MR. >> >> Moving to a simple Yes/No +1/No system where the responsibility for >> fixing failures falls to the author would solve the hesitation of >> reviewers to give +2. >> >> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder >> wrote: >> > It's a time problem: I see chunks of code which at a superficial level >> seem >> > fine and I'm willing to give it a +1. The problem with a +2 is that >> issuing >> > such a score has implications that I've jumped into the code at a deep >> level >> > and can vouch for it. This can occasionally mean hours or days of work >> hence >> > the delay. Maybe it's because I'm a chicken and should have faith (or >> > otherwise ensure) that there is enough testing to proof it out, but I >> feel >> > reluctant in some cases to give a +2 without enough due diligence which >> > takes time I often don't have. Besides more testing, one thing that >> might >> > help would be more "documentation" providing me with the confidence that >> > important issues have been considered and accounted for..... >> > >> > Best, >> > W >> > >> > >> > On Thu, Jul 7, 2016 at 9:56 AM, Brad King >> wrote: >> >> >> >> On 07/07/2016 08:56 AM, David E DeMarle wrote: >> >> > In either case both authors and reviewers are responsible for >> >> > watching the dashboards and addressing issues that come up afterward. >> >> >> >> This is the real reason people are hesitant to +2. If we strengthen >> >> the meaning of +1 then people will just say "LGTM" or something else >> >> to "approve" without taking responsibility. Posting +1 is a common >> >> convention for voting and should not be given stronger meaning. >> >> >> >> The real problem we'd like to address is stagnation of MRs that >> >> are ready but not merged. The syntax for approving is not very >> >> important. We should identify reasons MRs stagnate and address >> >> them directly: >> >> >> >> * It may be a governance problem. No one has responsibility >> >> to ensure everything that is ready gets merged. >> >> >> >> * It may be a workflow problem, like waiting for buildbot results >> >> to approve and then forgetting. The new workflow:buildbots >> >> label may help here. >> >> >> >> -Brad >> >> >> >> _______________________________________________ >> >> 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 >> >> >> > >> > >> > >> > -- >> > William J. Schroeder, PhD >> > Kitware, Inc. - Building the World's Technical Computing Software >> > 28 Corporate Drive >> > Clifton Park, NY 12065 >> > will.schroeder at kitware.com >> > http://www.kitware.com >> > (518) 881-4902 >> > >> > _______________________________________________ >> > 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 >> >> > > > -- > Ken Martin PhD > Chairman & CFO > Kitware Inc. > 28 Corporate Drive > Clifton Park NY 12065 > 518 371 3971 > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you are > not the intended recipient, any disclosure, copying, distribution or any > action taken in reliance on it is prohibited and may be unlawful. If you > received this communication in error please notify us immediately and > destroy the original message. Thank you. > > _______________________________________________ > 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 will.schroeder at kitware.com Thu Jul 7 13:38:45 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Thu, 7 Jul 2016 13:38:45 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: :-) okay then in order to cross the road, I'll insist more strongly on good tests and documentation On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci wrote: > > Maybe it's because I'm a chicken > > +1 > > On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder < > will.schroeder at kitware.com> wrote: > >> It's a time problem: I see chunks of code which at a superficial level >> seem fine and I'm willing to give it a +1. The problem with a +2 is that >> issuing such a score has implications that I've jumped into the code at a >> deep level and can vouch for it. This can occasionally mean hours or days >> of work hence the delay. Maybe it's because I'm a chicken and should have >> faith (or otherwise ensure) that there is enough testing to proof it out, >> but I feel reluctant in some cases to give a +2 without enough due >> diligence which takes time I often don't have. Besides more testing, one >> thing that might help would be more "documentation" providing me with the >> confidence that important issues have been considered and accounted for..... >> >> Best, >> W >> >> >> On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: >> >>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>> > In either case both authors and reviewers are responsible for >>> > watching the dashboards and addressing issues that come up afterward. >>> >>> This is the real reason people are hesitant to +2. If we strengthen >>> the meaning of +1 then people will just say "LGTM" or something else >>> to "approve" without taking responsibility. Posting +1 is a common >>> convention for voting and should not be given stronger meaning. >>> >>> The real problem we'd like to address is stagnation of MRs that >>> are ready but not merged. The syntax for approving is not very >>> important. We should identify reasons MRs stagnate and address >>> them directly: >>> >>> * It may be a governance problem. No one has responsibility >>> to ensure everything that is ready gets merged. >>> >>> * It may be a workflow problem, like waiting for buildbot results >>> to approve and then forgetting. The new workflow:buildbots >>> label may help here. >>> >>> -Brad >>> >>> _______________________________________________ >>> 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 >>> >>> >> >> >> -- >> William J. Schroeder, PhD >> Kitware, Inc. - Building the World's Technical Computing Software >> 28 Corporate Drive >> Clifton Park, NY 12065 >> will.schroeder at kitware.com >> http://www.kitware.com >> (518) 881-4902 >> >> _______________________________________________ >> 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 >> >> >> > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From berk.geveci at kitware.com Thu Jul 7 13:41:02 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 13:41:02 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: If there is new functionality, I do not approve without testing & documentation. If it's a bug fix, I may ask if there is a test that would fail without the fix. On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder wrote: > :-) okay then in order to cross the road, I'll insist more strongly on > good tests and documentation > > On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci > wrote: > >> > Maybe it's because I'm a chicken >> >> +1 >> >> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder < >> will.schroeder at kitware.com> wrote: >> >>> It's a time problem: I see chunks of code which at a superficial level >>> seem fine and I'm willing to give it a +1. The problem with a +2 is that >>> issuing such a score has implications that I've jumped into the code at a >>> deep level and can vouch for it. This can occasionally mean hours or days >>> of work hence the delay. Maybe it's because I'm a chicken and should have >>> faith (or otherwise ensure) that there is enough testing to proof it out, >>> but I feel reluctant in some cases to give a +2 without enough due >>> diligence which takes time I often don't have. Besides more testing, one >>> thing that might help would be more "documentation" providing me with the >>> confidence that important issues have been considered and accounted for..... >>> >>> Best, >>> W >>> >>> >>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King wrote: >>> >>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>>> > In either case both authors and reviewers are responsible for >>>> > watching the dashboards and addressing issues that come up afterward. >>>> >>>> This is the real reason people are hesitant to +2. If we strengthen >>>> the meaning of +1 then people will just say "LGTM" or something else >>>> to "approve" without taking responsibility. Posting +1 is a common >>>> convention for voting and should not be given stronger meaning. >>>> >>>> The real problem we'd like to address is stagnation of MRs that >>>> are ready but not merged. The syntax for approving is not very >>>> important. We should identify reasons MRs stagnate and address >>>> them directly: >>>> >>>> * It may be a governance problem. No one has responsibility >>>> to ensure everything that is ready gets merged. >>>> >>>> * It may be a workflow problem, like waiting for buildbot results >>>> to approve and then forgetting. The new workflow:buildbots >>>> label may help here. >>>> >>>> -Brad >>>> >>>> _______________________________________________ >>>> 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 >>>> >>>> >>> >>> >>> -- >>> William J. Schroeder, PhD >>> Kitware, Inc. - Building the World's Technical Computing Software >>> 28 Corporate Drive >>> Clifton Park, NY 12065 >>> will.schroeder at kitware.com >>> http://www.kitware.com >>> (518) 881-4902 >>> >>> _______________________________________________ >>> 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 >>> >>> >>> >> > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Thu Jul 7 14:33:26 2016 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 7 Jul 2016 14:33:26 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: +2 on Ken's interpretation. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Thu, Jul 7, 2016 at 2:32 PM, David E DeMarle wrote: > > This is the real reason people are hesitant to +2. If we strengthen >> the meaning of +1 then people will just say "LGTM" or something else >> to "approve" without taking responsibility. Posting +1 is a common >> convention for voting and should not be given stronger meaning. > > > How widespread is the non-binary and easier to misinterpret +1/+2 system? > I had to explain it to someone recently and they looked at me really funny. > > >> The real problem we'd like to address is stagnation of MRs that >> are ready but not merged. The syntax for approving is not very >> important. > > > Yes it won't solve the problem, but I think this cultural change will help > a tad. > > We should identify reasons MRs stagnate and address >> them directly: > > > Agree. > > >> * It may be a governance problem. No one has responsibility >> to ensure everything that is ready gets merged. > > > Agree. > > > * I think the big thing is the time problem Will mentioned. Each merge > request adds work to uninvolved developers. If our standard is that we only > give approval after hours to days of thorough review in the general case > then something is broken. If a cursory review says that probably isn't > necessary than just give the approval and trust the author to follow > through. If it looks like that is necessary than just day so rather than > giving a whishy washy +1. > > The +1=="LGTM but don't merge" adds nothing of value. > > * I've also been convinced that the reviewer should not be responsible for > monitoring the dashboards at all. Except in the case of newbie authors who > don't know how to monitor the dashboards, we should trust each others > capabilities and make the original author solely responsible. The important > exception is when the original author is a newbie who probably isn't > familiar with the dashboards. > > > > David E DeMarle > Kitware, Inc. > R&D Engineer > 21 Corporate Drive > Clifton Park, NY 12065-8662 > Phone: 518-881-4909 > > On Thu, Jul 7, 2016 at 1:41 PM, Berk Geveci > wrote: > >> If there is new functionality, I do not approve without testing & >> documentation. If it's a bug fix, I may ask if there is a test that would >> fail without the fix. >> >> On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder < >> will.schroeder at kitware.com> wrote: >> >>> :-) okay then in order to cross the road, I'll insist more strongly on >>> good tests and documentation >>> >>> On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci >>> wrote: >>> >>>> > Maybe it's because I'm a chicken >>>> >>>> +1 >>>> >>>> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder < >>>> will.schroeder at kitware.com> wrote: >>>> >>>>> It's a time problem: I see chunks of code which at a superficial level >>>>> seem fine and I'm willing to give it a +1. The problem with a +2 is that >>>>> issuing such a score has implications that I've jumped into the code at a >>>>> deep level and can vouch for it. This can occasionally mean hours or days >>>>> of work hence the delay. Maybe it's because I'm a chicken and should have >>>>> faith (or otherwise ensure) that there is enough testing to proof it out, >>>>> but I feel reluctant in some cases to give a +2 without enough due >>>>> diligence which takes time I often don't have. Besides more testing, one >>>>> thing that might help would be more "documentation" providing me with the >>>>> confidence that important issues have been considered and accounted for..... >>>>> >>>>> Best, >>>>> W >>>>> >>>>> >>>>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King >>>>> wrote: >>>>> >>>>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>>>>> > In either case both authors and reviewers are responsible for >>>>>> > watching the dashboards and addressing issues that come up >>>>>> afterward. >>>>>> >>>>>> This is the real reason people are hesitant to +2. If we strengthen >>>>>> the meaning of +1 then people will just say "LGTM" or something else >>>>>> to "approve" without taking responsibility. Posting +1 is a common >>>>>> convention for voting and should not be given stronger meaning. >>>>>> >>>>>> The real problem we'd like to address is stagnation of MRs that >>>>>> are ready but not merged. The syntax for approving is not very >>>>>> important. We should identify reasons MRs stagnate and address >>>>>> them directly: >>>>>> >>>>>> * It may be a governance problem. No one has responsibility >>>>>> to ensure everything that is ready gets merged. >>>>>> >>>>>> * It may be a workflow problem, like waiting for buildbot results >>>>>> to approve and then forgetting. The new workflow:buildbots >>>>>> label may help here. >>>>>> >>>>>> -Brad >>>>>> >>>>>> _______________________________________________ >>>>>> 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 >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> William J. Schroeder, PhD >>>>> Kitware, Inc. - Building the World's Technical Computing Software >>>>> 28 Corporate Drive >>>>> Clifton Park, NY 12065 >>>>> will.schroeder at kitware.com >>>>> http://www.kitware.com >>>>> (518) 881-4902 >>>>> >>>>> _______________________________________________ >>>>> 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 >>>>> >>>>> >>>>> >>>> >>> >>> >>> -- >>> William J. Schroeder, PhD >>> Kitware, Inc. - Building the World's Technical Computing Software >>> 28 Corporate Drive >>> Clifton Park, NY 12065 >>> will.schroeder at kitware.com >>> http://www.kitware.com >>> (518) 881-4902 >>> >> >> >> _______________________________________________ >> 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 dave.demarle at kitware.com Thu Jul 7 14:32:43 2016 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 7 Jul 2016 14:32:43 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: > This is the real reason people are hesitant to +2. If we strengthen > the meaning of +1 then people will just say "LGTM" or something else > to "approve" without taking responsibility. Posting +1 is a common > convention for voting and should not be given stronger meaning. How widespread is the non-binary and easier to misinterpret +1/+2 system? I had to explain it to someone recently and they looked at me really funny. > The real problem we'd like to address is stagnation of MRs that > are ready but not merged. The syntax for approving is not very > important. Yes it won't solve the problem, but I think this cultural change will help a tad. We should identify reasons MRs stagnate and address > them directly: Agree. > * It may be a governance problem. No one has responsibility > to ensure everything that is ready gets merged. Agree. * I think the big thing is the time problem Will mentioned. Each merge request adds work to uninvolved developers. If our standard is that we only give approval after hours to days of thorough review in the general case then something is broken. If a cursory review says that probably isn't necessary than just give the approval and trust the author to follow through. If it looks like that is necessary than just day so rather than giving a whishy washy +1. The +1=="LGTM but don't merge" adds nothing of value. * I've also been convinced that the reviewer should not be responsible for monitoring the dashboards at all. Except in the case of newbie authors who don't know how to monitor the dashboards, we should trust each others capabilities and make the original author solely responsible. The important exception is when the original author is a newbie who probably isn't familiar with the dashboards. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Thu, Jul 7, 2016 at 1:41 PM, Berk Geveci wrote: > If there is new functionality, I do not approve without testing & > documentation. If it's a bug fix, I may ask if there is a test that would > fail without the fix. > > On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder > wrote: > >> :-) okay then in order to cross the road, I'll insist more strongly on >> good tests and documentation >> >> On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci >> wrote: >> >>> > Maybe it's because I'm a chicken >>> >>> +1 >>> >>> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder < >>> will.schroeder at kitware.com> wrote: >>> >>>> It's a time problem: I see chunks of code which at a superficial level >>>> seem fine and I'm willing to give it a +1. The problem with a +2 is that >>>> issuing such a score has implications that I've jumped into the code at a >>>> deep level and can vouch for it. This can occasionally mean hours or days >>>> of work hence the delay. Maybe it's because I'm a chicken and should have >>>> faith (or otherwise ensure) that there is enough testing to proof it out, >>>> but I feel reluctant in some cases to give a +2 without enough due >>>> diligence which takes time I often don't have. Besides more testing, one >>>> thing that might help would be more "documentation" providing me with the >>>> confidence that important issues have been considered and accounted for..... >>>> >>>> Best, >>>> W >>>> >>>> >>>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King >>>> wrote: >>>> >>>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>>>> > In either case both authors and reviewers are responsible for >>>>> > watching the dashboards and addressing issues that come up afterward. >>>>> >>>>> This is the real reason people are hesitant to +2. If we strengthen >>>>> the meaning of +1 then people will just say "LGTM" or something else >>>>> to "approve" without taking responsibility. Posting +1 is a common >>>>> convention for voting and should not be given stronger meaning. >>>>> >>>>> The real problem we'd like to address is stagnation of MRs that >>>>> are ready but not merged. The syntax for approving is not very >>>>> important. We should identify reasons MRs stagnate and address >>>>> them directly: >>>>> >>>>> * It may be a governance problem. No one has responsibility >>>>> to ensure everything that is ready gets merged. >>>>> >>>>> * It may be a workflow problem, like waiting for buildbot results >>>>> to approve and then forgetting. The new workflow:buildbots >>>>> label may help here. >>>>> >>>>> -Brad >>>>> >>>>> _______________________________________________ >>>>> 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 >>>>> >>>>> >>>> >>>> >>>> -- >>>> William J. Schroeder, PhD >>>> Kitware, Inc. - Building the World's Technical Computing Software >>>> 28 Corporate Drive >>>> Clifton Park, NY 12065 >>>> will.schroeder at kitware.com >>>> http://www.kitware.com >>>> (518) 881-4902 >>>> >>>> _______________________________________________ >>>> 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 >>>> >>>> >>>> >>> >> >> >> -- >> William J. Schroeder, PhD >> Kitware, Inc. - Building the World's Technical Computing Software >> 28 Corporate Drive >> Clifton Park, NY 12065 >> will.schroeder at kitware.com >> http://www.kitware.com >> (518) 881-4902 >> > > > _______________________________________________ > 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 berk.geveci at kitware.com Thu Jul 7 14:46:25 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 14:46:25 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: I fully agree with Dave & Ken. One minor (not passionate) defense for +1/+2 is that I may review a piece of code as a non-expert and want to express my support but not in full confidence of its correctness. But at the end, the process doesn't matter since I can always write instead "I like this code but I am not an expert in this." By the way, if we stick with +1, I suggest that multiple +1s mean +2. On Thu, Jul 7, 2016 at 2:32 PM, David E DeMarle wrote: > > This is the real reason people are hesitant to +2. If we strengthen >> the meaning of +1 then people will just say "LGTM" or something else >> to "approve" without taking responsibility. Posting +1 is a common >> convention for voting and should not be given stronger meaning. > > > How widespread is the non-binary and easier to misinterpret +1/+2 system? > I had to explain it to someone recently and they looked at me really funny. > > >> The real problem we'd like to address is stagnation of MRs that >> are ready but not merged. The syntax for approving is not very >> important. > > > Yes it won't solve the problem, but I think this cultural change will help > a tad. > > We should identify reasons MRs stagnate and address >> them directly: > > > Agree. > > >> * It may be a governance problem. No one has responsibility >> to ensure everything that is ready gets merged. > > > Agree. > > > * I think the big thing is the time problem Will mentioned. Each merge > request adds work to uninvolved developers. If our standard is that we only > give approval after hours to days of thorough review in the general case > then something is broken. If a cursory review says that probably isn't > necessary than just give the approval and trust the author to follow > through. If it looks like that is necessary than just day so rather than > giving a whishy washy +1. > > The +1=="LGTM but don't merge" adds nothing of value. > > * I've also been convinced that the reviewer should not be responsible for > monitoring the dashboards at all. Except in the case of newbie authors who > don't know how to monitor the dashboards, we should trust each others > capabilities and make the original author solely responsible. The important > exception is when the original author is a newbie who probably isn't > familiar with the dashboards. > > > > David E DeMarle > Kitware, Inc. > R&D Engineer > 21 Corporate Drive > Clifton Park, NY 12065-8662 > Phone: 518-881-4909 > > On Thu, Jul 7, 2016 at 1:41 PM, Berk Geveci > wrote: > >> If there is new functionality, I do not approve without testing & >> documentation. If it's a bug fix, I may ask if there is a test that would >> fail without the fix. >> >> On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder < >> will.schroeder at kitware.com> wrote: >> >>> :-) okay then in order to cross the road, I'll insist more strongly on >>> good tests and documentation >>> >>> On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci >>> wrote: >>> >>>> > Maybe it's because I'm a chicken >>>> >>>> +1 >>>> >>>> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder < >>>> will.schroeder at kitware.com> wrote: >>>> >>>>> It's a time problem: I see chunks of code which at a superficial level >>>>> seem fine and I'm willing to give it a +1. The problem with a +2 is that >>>>> issuing such a score has implications that I've jumped into the code at a >>>>> deep level and can vouch for it. This can occasionally mean hours or days >>>>> of work hence the delay. Maybe it's because I'm a chicken and should have >>>>> faith (or otherwise ensure) that there is enough testing to proof it out, >>>>> but I feel reluctant in some cases to give a +2 without enough due >>>>> diligence which takes time I often don't have. Besides more testing, one >>>>> thing that might help would be more "documentation" providing me with the >>>>> confidence that important issues have been considered and accounted for..... >>>>> >>>>> Best, >>>>> W >>>>> >>>>> >>>>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King >>>>> wrote: >>>>> >>>>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>>>>> > In either case both authors and reviewers are responsible for >>>>>> > watching the dashboards and addressing issues that come up >>>>>> afterward. >>>>>> >>>>>> This is the real reason people are hesitant to +2. If we strengthen >>>>>> the meaning of +1 then people will just say "LGTM" or something else >>>>>> to "approve" without taking responsibility. Posting +1 is a common >>>>>> convention for voting and should not be given stronger meaning. >>>>>> >>>>>> The real problem we'd like to address is stagnation of MRs that >>>>>> are ready but not merged. The syntax for approving is not very >>>>>> important. We should identify reasons MRs stagnate and address >>>>>> them directly: >>>>>> >>>>>> * It may be a governance problem. No one has responsibility >>>>>> to ensure everything that is ready gets merged. >>>>>> >>>>>> * It may be a workflow problem, like waiting for buildbot results >>>>>> to approve and then forgetting. The new workflow:buildbots >>>>>> label may help here. >>>>>> >>>>>> -Brad >>>>>> >>>>>> _______________________________________________ >>>>>> 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 >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> William J. Schroeder, PhD >>>>> Kitware, Inc. - Building the World's Technical Computing Software >>>>> 28 Corporate Drive >>>>> Clifton Park, NY 12065 >>>>> will.schroeder at kitware.com >>>>> http://www.kitware.com >>>>> (518) 881-4902 >>>>> >>>>> _______________________________________________ >>>>> 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 >>>>> >>>>> >>>>> >>>> >>> >>> >>> -- >>> William J. Schroeder, PhD >>> Kitware, Inc. - Building the World's Technical Computing Software >>> 28 Corporate Drive >>> Clifton Park, NY 12065 >>> will.schroeder at kitware.com >>> http://www.kitware.com >>> (518) 881-4902 >>> >> >> >> _______________________________________________ >> 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 marcus.hanwell at kitware.com Thu Jul 7 15:00:42 2016 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Thu, 7 Jul 2016 15:00:42 -0400 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: For what it's worth I have never thought I assumed responsibility for a merge/pull request I approved/merged for another contributor with merge rights. I think for new contributors you need to be willing to mentor/assume a little of that responsibility, but that is part of building trust, and if they fail to follow up you are less likely to take their patches in future. This is version control, and if you accept something and it totally backfires we can revert changes, and on occasion I have. Simplifying sounds good to me, and others can say looks OK but someone who knows it better should take a look. In my opinion one of the other things lacking in VTK is someone owning a particular module, such that if things are broken/in need of review the buck stops with them. This makes things a little nebulous at times, Qt, the Linux kernel and countless others tend to distribute responsibility for modules, subsystems, etc. I think the buildbots have been very slow when I have worked on changes, and it can be an issue that by the time results are back you have switched on to other tasks. We had lower coverage but faster response when only 3 dashboards were required (continuous, and later Gerrit with CDash at Home). Hopefully this can be improved in future, such that we could request a branch be tested and see the results within an hour or so. It is tough balancing having enough process that things are managed, and not so much that it is just too hard to get anything in. I agree with Berk that if we stick with +1/+2 then they should be additive and two +1s = a +2. On Thu, Jul 7, 2016 at 2:46 PM, Berk Geveci wrote: > I fully agree with Dave & Ken. One minor (not passionate) defense for +1/+2 > is that I may review a piece of code as a non-expert and want to express my > support but not in full confidence of its correctness. But at the end, the > process doesn't matter since I can always write instead "I like this code > but I am not an expert in this." > > By the way, if we stick with +1, I suggest that multiple +1s mean +2. > > On Thu, Jul 7, 2016 at 2:32 PM, David E DeMarle > wrote: >> >> >>> This is the real reason people are hesitant to +2. If we strengthen >>> the meaning of +1 then people will just say "LGTM" or something else >>> to "approve" without taking responsibility. Posting +1 is a common >>> convention for voting and should not be given stronger meaning. >> >> >> How widespread is the non-binary and easier to misinterpret +1/+2 system? >> I had to explain it to someone recently and they looked at me really funny. >> >>> >>> The real problem we'd like to address is stagnation of MRs that >>> are ready but not merged. The syntax for approving is not very >>> important. >> >> >> Yes it won't solve the problem, but I think this cultural change will help >> a tad. >> >>> We should identify reasons MRs stagnate and address >>> them directly: >> >> >> Agree. >> >>> >>> * It may be a governance problem. No one has responsibility >>> to ensure everything that is ready gets merged. >> >> >> Agree. >> >> >> * I think the big thing is the time problem Will mentioned. Each merge >> request adds work to uninvolved developers. If our standard is that we only >> give approval after hours to days of thorough review in the general case >> then something is broken. If a cursory review says that probably isn't >> necessary than just give the approval and trust the author to follow >> through. If it looks like that is necessary than just day so rather than >> giving a whishy washy +1. >> >> The +1=="LGTM but don't merge" adds nothing of value. >> >> * I've also been convinced that the reviewer should not be responsible for >> monitoring the dashboards at all. Except in the case of newbie authors who >> don't know how to monitor the dashboards, we should trust each others >> capabilities and make the original author solely responsible. The important >> exception is when the original author is a newbie who probably isn't >> familiar with the dashboards. >> >> >> >> David E DeMarle >> Kitware, Inc. >> R&D Engineer >> 21 Corporate Drive >> Clifton Park, NY 12065-8662 >> Phone: 518-881-4909 >> >> On Thu, Jul 7, 2016 at 1:41 PM, Berk Geveci >> wrote: >>> >>> If there is new functionality, I do not approve without testing & >>> documentation. If it's a bug fix, I may ask if there is a test that would >>> fail without the fix. >>> >>> On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder >>> wrote: >>>> >>>> :-) okay then in order to cross the road, I'll insist more strongly on >>>> good tests and documentation >>>> >>>> On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci >>>> wrote: >>>>> >>>>> > Maybe it's because I'm a chicken >>>>> >>>>> +1 >>>>> >>>>> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder >>>>> wrote: >>>>>> >>>>>> It's a time problem: I see chunks of code which at a superficial level >>>>>> seem fine and I'm willing to give it a +1. The problem with a +2 is that >>>>>> issuing such a score has implications that I've jumped into the code at a >>>>>> deep level and can vouch for it. This can occasionally mean hours or days of >>>>>> work hence the delay. Maybe it's because I'm a chicken and should have faith >>>>>> (or otherwise ensure) that there is enough testing to proof it out, but I >>>>>> feel reluctant in some cases to give a +2 without enough due diligence which >>>>>> takes time I often don't have. Besides more testing, one thing that might >>>>>> help would be more "documentation" providing me with the confidence that >>>>>> important issues have been considered and accounted for..... >>>>>> >>>>>> Best, >>>>>> W >>>>>> >>>>>> >>>>>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King >>>>>> wrote: >>>>>>> >>>>>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: >>>>>>> > In either case both authors and reviewers are responsible for >>>>>>> > watching the dashboards and addressing issues that come up >>>>>>> > afterward. >>>>>>> >>>>>>> This is the real reason people are hesitant to +2. If we strengthen >>>>>>> the meaning of +1 then people will just say "LGTM" or something else >>>>>>> to "approve" without taking responsibility. Posting +1 is a common >>>>>>> convention for voting and should not be given stronger meaning. >>>>>>> >>>>>>> The real problem we'd like to address is stagnation of MRs that >>>>>>> are ready but not merged. The syntax for approving is not very >>>>>>> important. We should identify reasons MRs stagnate and address >>>>>>> them directly: >>>>>>> >>>>>>> * It may be a governance problem. No one has responsibility >>>>>>> to ensure everything that is ready gets merged. >>>>>>> >>>>>>> * It may be a workflow problem, like waiting for buildbot results >>>>>>> to approve and then forgetting. The new workflow:buildbots >>>>>>> label may help here. >>>>>>> >>>>>>> -Brad >>>>>>> >>>>>>> _______________________________________________ >>>>>>> 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 >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> William J. Schroeder, PhD >>>>>> Kitware, Inc. - Building the World's Technical Computing Software >>>>>> 28 Corporate Drive >>>>>> Clifton Park, NY 12065 >>>>>> will.schroeder at kitware.com >>>>>> http://www.kitware.com >>>>>> (518) 881-4902 >>>>>> >>>>>> _______________________________________________ >>>>>> 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 >>>>>> >>>>>> >>>>> >>>> >>>> >>>> >>>> -- >>>> William J. Schroeder, PhD >>>> Kitware, Inc. - Building the World's Technical Computing Software >>>> 28 Corporate Drive >>>> Clifton Park, NY 12065 >>>> will.schroeder at kitware.com >>>> http://www.kitware.com >>>> (518) 881-4902 >>> >>> >>> >>> _______________________________________________ >>> 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 > > From berk.geveci at kitware.com Thu Jul 7 15:06:48 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Thu, 7 Jul 2016 15:06:48 -0400 Subject: [vtk-developers] The VTK hackathon Message-ID: Yesterday, we had a great hackathon. It was a great turnout. See below for a full list of attendees. This time, our focus was on cleaning up the issue tracker ? for maintenance as well as in anticipation of a move to the Gitlab tracker. As of today, 55 bugs have been closed! There are also merge requests for a few more. In addition, we worked on resolving merge requests. 31 merged or closed! Not bad. We plan on holding these hackathon monthly going forward. Attendees: * Utkarsh Ayachit* Ben Boeckel* Shawn Walden* Dave DeMarle* Andy Bauer* Chuck Atkins* Rob Maynard* Marcus Hanwell* Dan Lipsa* Seb Jourdain* Alvaro Sanchez* Berk Geveci* Will Schroeder* Dave Thompson* Cory Quammen* Tim Thirion* David Gobbi* Betsy McPhail* Sujin Philip* Sean McBride -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Thu Jul 7 19:03:20 2016 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 7 Jul 2016 19:03:20 -0400 Subject: [vtk-developers] Some progress on binary wheel for OSX In-Reply-To: References: Message-ID: nice! David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Sat, Jul 2, 2016 at 4:39 AM, Matthew Brett wrote: > Hi, > > Just to say, I did a few hours of hacking and investigation, which got > me reasonably close to a copy of VTK that could be used for a very > useful OSX binary installer - full description at: > > https://mail.python.org/pipermail/pythonmac-sig/2016-July/024137.html > > As y'all know, the current OSX installer isn't much use for doing real > work, because it ships its own copy of Python, making it difficult to > install other Python packages for doing standard things like array > manipulation, statistics and so on. > > The ideal would be to be able to build a binary wheel [1] for OSX, so > the user could do: > > pip install vtk > > to install a copy of vtk into the Python environment they are using - > such as Python.org Python, homebrew, etc. This has already been done > for nearly all the main scientific Python projects such as numpy, > scipy, matplotlib etc. > > The big hold-up for VTK, is the current behavior of VTK / cmake, to > link the Python library by absolute path directly into the VTK libs, > so the build is tied to the path of the Python binary [2, 3]. > > I see there's a pull request already in from the homebrew folks to try > and fix that [4] - which is excellent news. I believe that, once > this is merged and released, it should be fairly easy to build an OSX > wheel that would work in any standard OSX Python environment, which > would be a huge step forward in terms of ease of the install process. > > Cheers, > > Matthew > > [1] https://www.python.org/dev/peps/pep-0427/ > [2] http://blog.tim-smith.us/2015/09/python-extension-modules-os-x/ > [3] https://github.com/Homebrew/homebrew-science/issues/3401 > [4] https://gitlab.kitware.com/vtk/vtk/merge_requests/1511 > _______________________________________________ > 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 mathieu.westphal at kitware.com Fri Jul 8 03:06:21 2016 From: mathieu.westphal at kitware.com (Mathieu Westphal) Date: Fri, 8 Jul 2016 09:06:21 +0200 Subject: [vtk-developers] Avoiding MR stagnation (was: rules proposal drop "+1") In-Reply-To: References: <577E5F79.9000806@kitware.com> Message-ID: Hello Considering the Testing delay, I was never part of the old system, and I always had to wait for the whole buildbot suite to finish. My work flow is I begin my day by checking cdash results and taking action upen these results. No more forgotten MR this way. Mathieu Westphal On Thu, Jul 7, 2016 at 9:00 PM, Marcus D. Hanwell < marcus.hanwell at kitware.com> wrote: > For what it's worth I have never thought I assumed responsibility for > a merge/pull request I approved/merged for another contributor with > merge rights. I think for new contributors you need to be willing to > mentor/assume a little of that responsibility, but that is part of > building trust, and if they fail to follow up you are less likely to > take their patches in future. > > This is version control, and if you accept something and it totally > backfires we can revert changes, and on occasion I have. Simplifying > sounds good to me, and others can say looks OK but someone who knows > it better should take a look. > > In my opinion one of the other things lacking in VTK is someone owning > a particular module, such that if things are broken/in need of review > the buck stops with them. This makes things a little nebulous at > times, Qt, the Linux kernel and countless others tend to distribute > responsibility for modules, subsystems, etc. > > I think the buildbots have been very slow when I have worked on > changes, and it can be an issue that by the time results are back you > have switched on to other tasks. We had lower coverage but faster > response when only 3 dashboards were required (continuous, and later > Gerrit with CDash at Home). Hopefully this can be improved in future, > such that we could request a branch be tested and see the results > within an hour or so. > > It is tough balancing having enough process that things are managed, > and not so much that it is just too hard to get anything in. I agree > with Berk that if we stick with +1/+2 then they should be additive and > two +1s = a +2. > > On Thu, Jul 7, 2016 at 2:46 PM, Berk Geveci > wrote: > > I fully agree with Dave & Ken. One minor (not passionate) defense for > +1/+2 > > is that I may review a piece of code as a non-expert and want to express > my > > support but not in full confidence of its correctness. But at the end, > the > > process doesn't matter since I can always write instead "I like this code > > but I am not an expert in this." > > > > By the way, if we stick with +1, I suggest that multiple +1s mean +2. > > > > On Thu, Jul 7, 2016 at 2:32 PM, David E DeMarle < > dave.demarle at kitware.com> > > wrote: > >> > >> > >>> This is the real reason people are hesitant to +2. If we strengthen > >>> the meaning of +1 then people will just say "LGTM" or something else > >>> to "approve" without taking responsibility. Posting +1 is a common > >>> convention for voting and should not be given stronger meaning. > >> > >> > >> How widespread is the non-binary and easier to misinterpret +1/+2 > system? > >> I had to explain it to someone recently and they looked at me really > funny. > >> > >>> > >>> The real problem we'd like to address is stagnation of MRs that > >>> are ready but not merged. The syntax for approving is not very > >>> important. > >> > >> > >> Yes it won't solve the problem, but I think this cultural change will > help > >> a tad. > >> > >>> We should identify reasons MRs stagnate and address > >>> them directly: > >> > >> > >> Agree. > >> > >>> > >>> * It may be a governance problem. No one has responsibility > >>> to ensure everything that is ready gets merged. > >> > >> > >> Agree. > >> > >> > >> * I think the big thing is the time problem Will mentioned. Each merge > >> request adds work to uninvolved developers. If our standard is that we > only > >> give approval after hours to days of thorough review in the general case > >> then something is broken. If a cursory review says that probably isn't > >> necessary than just give the approval and trust the author to follow > >> through. If it looks like that is necessary than just day so rather than > >> giving a whishy washy +1. > >> > >> The +1=="LGTM but don't merge" adds nothing of value. > >> > >> * I've also been convinced that the reviewer should not be responsible > for > >> monitoring the dashboards at all. Except in the case of newbie authors > who > >> don't know how to monitor the dashboards, we should trust each others > >> capabilities and make the original author solely responsible. The > important > >> exception is when the original author is a newbie who probably isn't > >> familiar with the dashboards. > >> > >> > >> > >> David E DeMarle > >> Kitware, Inc. > >> R&D Engineer > >> 21 Corporate Drive > >> Clifton Park, NY 12065-8662 > >> Phone: 518-881-4909 > >> > >> On Thu, Jul 7, 2016 at 1:41 PM, Berk Geveci > >> wrote: > >>> > >>> If there is new functionality, I do not approve without testing & > >>> documentation. If it's a bug fix, I may ask if there is a test that > would > >>> fail without the fix. > >>> > >>> On Thu, Jul 7, 2016 at 1:38 PM, Will Schroeder > >>> wrote: > >>>> > >>>> :-) okay then in order to cross the road, I'll insist more strongly on > >>>> good tests and documentation > >>>> > >>>> On Thu, Jul 7, 2016 at 1:27 PM, Berk Geveci > >>>> wrote: > >>>>> > >>>>> > Maybe it's because I'm a chicken > >>>>> > >>>>> +1 > >>>>> > >>>>> On Thu, Jul 7, 2016 at 10:10 AM, Will Schroeder > >>>>> wrote: > >>>>>> > >>>>>> It's a time problem: I see chunks of code which at a superficial > level > >>>>>> seem fine and I'm willing to give it a +1. The problem with a +2 is > that > >>>>>> issuing such a score has implications that I've jumped into the > code at a > >>>>>> deep level and can vouch for it. This can occasionally mean hours > or days of > >>>>>> work hence the delay. Maybe it's because I'm a chicken and should > have faith > >>>>>> (or otherwise ensure) that there is enough testing to proof it out, > but I > >>>>>> feel reluctant in some cases to give a +2 without enough due > diligence which > >>>>>> takes time I often don't have. Besides more testing, one thing that > might > >>>>>> help would be more "documentation" providing me with the confidence > that > >>>>>> important issues have been considered and accounted for..... > >>>>>> > >>>>>> Best, > >>>>>> W > >>>>>> > >>>>>> > >>>>>> On Thu, Jul 7, 2016 at 9:56 AM, Brad King > >>>>>> wrote: > >>>>>>> > >>>>>>> On 07/07/2016 08:56 AM, David E DeMarle wrote: > >>>>>>> > In either case both authors and reviewers are responsible for > >>>>>>> > watching the dashboards and addressing issues that come up > >>>>>>> > afterward. > >>>>>>> > >>>>>>> This is the real reason people are hesitant to +2. If we > strengthen > >>>>>>> the meaning of +1 then people will just say "LGTM" or something > else > >>>>>>> to "approve" without taking responsibility. Posting +1 is a common > >>>>>>> convention for voting and should not be given stronger meaning. > >>>>>>> > >>>>>>> The real problem we'd like to address is stagnation of MRs that > >>>>>>> are ready but not merged. The syntax for approving is not very > >>>>>>> important. We should identify reasons MRs stagnate and address > >>>>>>> them directly: > >>>>>>> > >>>>>>> * It may be a governance problem. No one has responsibility > >>>>>>> to ensure everything that is ready gets merged. > >>>>>>> > >>>>>>> * It may be a workflow problem, like waiting for buildbot results > >>>>>>> to approve and then forgetting. The new workflow:buildbots > >>>>>>> label may help here. > >>>>>>> > >>>>>>> -Brad > >>>>>>> > >>>>>>> _______________________________________________ > >>>>>>> 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 > >>>>>>> > >>>>>> > >>>>>> > >>>>>> > >>>>>> -- > >>>>>> William J. Schroeder, PhD > >>>>>> Kitware, Inc. - Building the World's Technical Computing Software > >>>>>> 28 Corporate Drive > >>>>>> Clifton Park, NY 12065 > >>>>>> will.schroeder at kitware.com > >>>>>> http://www.kitware.com > >>>>>> (518) 881-4902 > >>>>>> > >>>>>> _______________________________________________ > >>>>>> 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 > >>>>>> > >>>>>> > >>>>> > >>>> > >>>> > >>>> > >>>> -- > >>>> William J. Schroeder, PhD > >>>> Kitware, Inc. - Building the World's Technical Computing Software > >>>> 28 Corporate Drive > >>>> Clifton Park, NY 12065 > >>>> will.schroeder at kitware.com > >>>> http://www.kitware.com > >>>> (518) 881-4902 > >>> > >>> > >>> > >>> _______________________________________________ > >>> 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 > > > > > _______________________________________________ > 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 ben.boeckel at kitware.com Fri Jul 8 10:44:14 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Fri, 8 Jul 2016 10:44:14 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well In-Reply-To: References: Message-ID: <20160708144414.GB26651@megas.kitware.com> On Thu, Jul 07, 2016 at 08:56:14 -0400, David E DeMarle wrote: > As Robert just pointed out, the +1 state in the review process is nothing > but trouble. In practice, it seems to help us avoid responsibility and > delays merging, contributing to the accumulation of stale merge requests > and newbie frustration. > > How about we say "+1 means I approve this code and assert that it is ready > to merge". Merge requests from a authorized developers should be promptly > merged after the +1 by the original author. Merge requests from > unauthorized developers should be merged more or less immediately by the > authorized reviewer after the +1. The problem here is that for people (like me) with more domain-specific knowledge might be able to +1 the CMake side of things, but without a "+1", what do I do? Responding to other bits from the thread: On Thu, Jul 07, 2016 at 09:56:09 -0400, Brad King wrote: > The real problem we'd like to address is stagnation of MRs that > are ready but not merged. The syntax for approving is not very > important. We should identify reasons MRs stagnate and address > them directly: An idea: have kwrobot scan MRs every Monday looking for +1 and +2 comments on unmerged MRs. It adds a comment asking "what's the holdup?" pinging any reviewers. It could also use the `workflow:buildbot` tag to see if the buildbots have reported back successfully (would need some bugs fixed in status submission from buildbot). On Thu, Jul 07, 2016 at 10:10:00 -0400, Will Schroeder wrote: > It's a time problem: I see chunks of code which at a superficial level seem > fine and I'm willing to give it a +1. The problem with a +2 is that issuing > such a score has implications that I've jumped into the code at a deep Copying from an earlier email: On Wed, Mar 25, 2015 at 4:41 PM, Ben Boeckel wrote: > Acked-by - "I see nothing wrong with the change" (which is all > kwrobot's checks can do) +1 > Reviewed-by - "I reviewed the change and agree with it" (we should > probably use this rather than Acked-by for humans) +2 > Tested-by - "I tested the change and approve that it works" +3 What you're describing is probably closer to +3 here. On Thu, Jul 07, 2016 at 15:00:42 -0400, Marcus D. Hanwell wrote: > In my opinion one of the other things lacking in VTK is someone owning > a particular module, such that if things are broken/in need of review > the buck stops with them. This makes things a little nebulous at > times, Qt, the Linux kernel and countless others tend to distribute > responsibility for modules, subsystems, etc. One idea I've batted around is to have kwrobot auto-assign MRs randomly (being in the rotation would be voluntary). That user would at least know who the right person to assign it to would be. The kwrobot could also be made aware of areas maintained by certain users and assign/mention based on that. Gitlab upstream uses assignment in a slightly more involved way: when an MR is waiting on the submitter, it is assigned back to the submitter to indicate who is blocking it. Assigned MRs (and issues) increase the number in the left sidebar in gitlab, so people should be able to use that as a todo list of sorts. The problem with this is that non-developers cannot be assigned anything, but it would work for the majority of contributors. > I think the buildbots have been very slow when I have worked on > changes, and it can be an issue that by the time results are back you > have switched on to other tasks. We had lower coverage but faster > response when only 3 dashboards were required (continuous, and later > Gerrit with CDash at Home). Hopefully this can be improved in future, > such that we could request a branch be tested and see the results > within an hour or so. I think if we could have more targeted testing to start, that would improve things. For example, working on a Windows change? Only test on the Windows buildbots (`Do: test -i windows`) until those are green and *then* do a full test request (`Do: test`). Fixing warnings on megas? Use `Do: test -i megas`. And so on. This would help keep machines from doing unnecessary work. Another (probably additive) way to do this would be to have --oneshot be the default so that new builds are not queued automatically on push (I asked the list about this policy change a while ago and got ~no response). > It is tough balancing having enough process that things are managed, > and not so much that it is just too hard to get anything in. I agree > with Berk that if we stick with +1/+2 then they should be additive and > two +1s = a +2. For the most part, yes (two reviewers for CMake changes and none for code is probably not enough). --Ben From ken.martin at kitware.com Tue Jul 12 09:24:59 2016 From: ken.martin at kitware.com (Ken Martin) Date: Tue, 12 Jul 2016 09:24:59 -0400 Subject: [vtk-developers] agora Message-ID: Could someone familiar with agora please fix the X server error, maybe stick in a nightly cron job to reboot the machine so it doesn't reoccur if that makes sense -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From chuck.atkins at kitware.com Tue Jul 12 11:50:49 2016 From: chuck.atkins at kitware.com (Chuck Atkins) Date: Tue, 12 Jul 2016 11:50:49 -0400 Subject: [vtk-developers] agora In-Reply-To: References: Message-ID: It needed more than a reboot. Something went squirely with the NVidia kernel module and needed to be reinstalled. It should be good to go now. - Chuck On Tue, Jul 12, 2016 at 9:24 AM, Ken Martin wrote: > Could someone familiar with agora please fix the X server error, maybe > stick in a nightly cron job to reboot the machine so it doesn't reoccur if > that makes sense > > > > > -- > Ken Martin PhD > Chairman & CFO > Kitware Inc. > 28 Corporate Drive > Clifton Park NY 12065 > 518 371 3971 > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you are > not the intended recipient, any disclosure, copying, distribution or any > action taken in reliance on it is prohibited and may be unlawful. If you > received this communication in error please notify us immediately and > destroy the original message. Thank you. > > _______________________________________________ > 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 ken.martin at kitware.com Tue Jul 12 11:51:50 2016 From: ken.martin at kitware.com (Ken Martin) Date: Tue, 12 Jul 2016 11:51:50 -0400 Subject: [vtk-developers] agora In-Reply-To: References: Message-ID: Mucho Thanks Chuck! On Tue, Jul 12, 2016 at 11:50 AM, Chuck Atkins wrote: > It needed more than a reboot. Something went squirely with the NVidia > kernel module and needed to be reinstalled. It should be good to go now. > > - Chuck > > On Tue, Jul 12, 2016 at 9:24 AM, Ken Martin > wrote: > >> Could someone familiar with agora please fix the X server error, maybe >> stick in a nightly cron job to reboot the machine so it doesn't reoccur if >> that makes sense >> >> >> >> >> -- >> Ken Martin PhD >> Chairman & CFO >> Kitware Inc. >> 28 Corporate Drive >> Clifton Park NY 12065 >> 518 371 3971 >> >> This communication, including all attachments, contains confidential and >> legally privileged information, and it is intended only for the use of the >> addressee. Access to this email by anyone else is unauthorized. If you are >> not the intended recipient, any disclosure, copying, distribution or any >> action taken in reliance on it is prohibited and may be unlawful. If you >> received this communication in error please notify us immediately and >> destroy the original message. Thank you. >> >> _______________________________________________ >> 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 >> >> >> > -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From chuck.atkins at kitware.com Tue Jul 12 16:22:00 2016 From: chuck.atkins at kitware.com (Chuck Atkins) Date: Tue, 12 Jul 2016 16:22:00 -0400 Subject: [vtk-developers] External HDF5 testers wanted Message-ID: I've been doing some rather invasive tinkering on CMake's FindHDF5 module lately and it seems to have broken for a lot of people. I believe I've got all the various issues addressed but would greatly appreciate a few additional testers. If you would like to help, please check out the merge request here: https://gitlab.kitware.com/vtk/vtk/merge_requests/1679 . I would appreciate any feedback if it is still broken during configure for anybody when using an external HDF5, i.e. VTK_USE_SYSTEM_HDF5=On. Note the MR title is specifically to fix issues with home-brew on Mac but it addressed a few other issues as well on Linux and Windows. Thank you. - Chuck -------------- next part -------------- An HTML attachment was scrubbed... URL: From berk.geveci at kitware.com Wed Jul 13 10:37:27 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Wed, 13 Jul 2016 10:37:27 -0400 Subject: [vtk-developers] Mingw build on amber12 Message-ID: Who maintains the Mingw build on amber12? The nightly dashboard is failing there with a compile error in the PDB reader. -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.amaclean at gmail.com Fri Jul 15 02:02:02 2016 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Fri, 15 Jul 2016 16:02:02 +1000 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? Message-ID: Hi All, I have been looking through the documentation and user groups and can't find any solution to needless duplication of ExternalData in the build directories when building different versions of VTK. Also does VTK_DATA_STORE now replace VTK_DATA? According to my testing, they seem to be interchangeable. Is it possible to have only one ExternalData directory available to all builds of VTK? In my current configuration I have: (the . represents the rest of the absolute path) .\Kitware\src\VTK\ .\Kitware\src\VTKExternalData\ With VTK_DATA_STORE set to .\Kitware\src\VTKExternalData\ (the default and sibling setting) and VTK_DATA_EXCLUDE_FROM_ALL is configured as OFF. When I build a Debug and a Release version of VTK I get the following: .\Kitware\build\VTK-Debug\ExternalData .\Kitware\build\VTK-Release\ExternalData Is it possible to have one ExternalData directory in say for example: .\Kitware\build\ExternalData Just to stop needless duplication? FYI: .\Kitware\src\VTKExternalData\ just contains a single folder called MD5. .\Kitware\build\VTK-Debug\ExternalData\ contains 15 subfolders with the test files and their md5-stamp. Regards Andrew -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Fri Jul 15 06:09:15 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Fri, 15 Jul 2016 06:09:15 -0400 Subject: [vtk-developers] vtkDepthImageToPointCloud Message-ID: FYI- especially for those of you interested in point clouds- with the help of some friends I added a threaded (vtkSMPTools) filter to transform a depth map plus optional color information into a point cloud. There are some nice features like the ability to cull points on the back clipping plane (which is typically the less interesting part of the image). The major intent is to provide simulated data for point cloud processing and to test point algorithms. However I'm sure that there are other clever things we'll come up with like using it to "seed" streamline generation or perform sampling for statistical analysis. Have fun, W -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brad.king at kitware.com Fri Jul 15 07:52:11 2016 From: brad.king at kitware.com (Brad King) Date: Fri, 15 Jul 2016 07:52:11 -0400 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: References: Message-ID: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> On 07/15/2016 02:02 AM, Andrew Maclean wrote: > I have been looking through the documentation and user groups and can't > find any solution to needless duplication of ExternalData in the build > directories when building different versions of VTK. [snip] > .\Kitware\src\VTKExternalData\ just contains a single folder called MD5. > .\Kitware\build\VTK-Debug\ExternalData\ contains 15 subfolders with the > test files and their md5-stamp. The former is the "object store" and the latter is the directory where the data files with names expected by the tests appear. On UNIX the latter is actually just full of symbolic links to the former, but on Windows we have to make copies instead. Unfortunately there is no way to avoid the per-build-tree copies on Windows because we need to make the file content available at specific paths in the build tree. The object store is content-addressed and so can be shared among many VTK versions and even separate projects (e.g. ITK uses this too). There is a VTK_DATA_STORE cache entry that can be set to point at a common object store directory. See CMake/vtkExternalData.cmake for more. It has a few defaults. One is to just create a VTKExternalData directory next to your VTK source tree. That will be shared among all build trees. Your Kitware\src\VTKExternalData is already playing this role, so there may not be much room for improvement over your current setup. I personally keep ExternalData_OBJECT_STORES set in my environment to point to a single directory outside of all my build and source trees to hold objects persistently. This avoids needing to re-download them from fresh builds or fresh source trees. It won't help with per-build-tree copies on Windows though. -Brad From brad.king at kitware.com Fri Jul 15 09:46:11 2016 From: brad.king at kitware.com (Brad King) Date: Fri, 15 Jul 2016 09:46:11 -0400 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> Message-ID: <415332dd-2be6-eb14-6603-ff6384607ac8@kitware.com> On 07/15/2016 09:41 AM, Andras Lasso wrote: > You can create symlinks in Windows Yes, but as you point out it is basically useless for applications in general due to lack of permissions. > you should be able to set up shared external directory manually. Every individual file needs to be linked with a specific name in the build tree that is unrelated to its name in the object store. This is not practical to do manually. -Brad From lasso at queensu.ca Fri Jul 15 09:41:34 2016 From: lasso at queensu.ca (Andras Lasso) Date: Fri, 15 Jul 2016 13:41:34 +0000 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> Message-ID: You can create symlinks in Windows (the feature was added a 10+ years ago): https://technet.microsoft.com/en-us/library/cc753194(v=ws.11).aspx The only difficulty may be that by default you may not have the security privileges to run mklink (admin users need to run elevated command prompt; for other users "Create symbolic links" security policy has to be enabled). Permission issues may make it tricky to add symlink creation to CMake/ExternalData scripts, but you should be able to set up shared external directory manually. Andras -----Original Message----- From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of Brad King Sent: July 15, 2016 7:52 To: Andrew Maclean Cc: VTK Developers Subject: Re: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? On 07/15/2016 02:02 AM, Andrew Maclean wrote: > I have been looking through the documentation and user groups and > can't find any solution to needless duplication of ExternalData in the > build directories when building different versions of VTK. [snip] > .\Kitware\src\VTKExternalData\ just contains a single folder called MD5. > .\Kitware\build\VTK-Debug\ExternalData\ contains 15 subfolders with > the test files and their md5-stamp. The former is the "object store" and the latter is the directory where the data files with names expected by the tests appear. On UNIX the latter is actually just full of symbolic links to the former, but on Windows we have to make copies instead. Unfortunately there is no way to avoid the per-build-tree copies on Windows because we need to make the file content available at specific paths in the build tree. The object store is content-addressed and so can be shared among many VTK versions and even separate projects (e.g. ITK uses this too). There is a VTK_DATA_STORE cache entry that can be set to point at a common object store directory. See CMake/vtkExternalData.cmake for more. It has a few defaults. One is to just create a VTKExternalData directory next to your VTK source tree. That will be shared among all build trees. Your Kitware\src\VTKExternalData is already playing this role, so there may not be much room for improvement over your current setup. I personally keep ExternalData_OBJECT_STORES set in my environment to point to a single directory outside of all my build and source trees to hold objects persistently. This avoids needing to re-download them from fresh builds or fresh source trees. It won't help with per-build-tree copies on Windows though. -Brad _______________________________________________ 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 From david.lonie at kitware.com Fri Jul 15 10:31:20 2016 From: david.lonie at kitware.com (David Lonie) Date: Fri, 15 Jul 2016 10:31:20 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well In-Reply-To: <20160708144414.GB26651@megas.kitware.com> References: <20160708144414.GB26651@megas.kitware.com> Message-ID: On Fri, Jul 8, 2016 at 10:44 AM, Ben Boeckel wrote: > Another (probably additive) way to do this would be to have --oneshot be > the default so that new builds are not queued automatically on push (I > asked the list about this policy change a while ago and got ~no > response). Interesting, I wasn't aware of the --oneshot argument. I must have missed that discussion. I'd be ok with using that as the default, and either way I'll try to start using it on my patches to cut down on unneeded builds. Often times I don't bother to do more targeted tests because I tend to push first, then go to gitlab to monitor results -- and by that time, the full test suite has already started. +1 to the idea of a default --oneshot. I think it'd definitely encourage more targeted testing and help reduce buildslave contention. On a related note, would it be possible to add a link to the kwrobot command documentation from the gitlab MR discussion page, or maybe just embed a "kwrobot help" link in the "Do: check" reply? I know that I'm pretty bad at remembering all of the options kwrobot supports and having a link to the docs shoved in my face when I go to request testing would be an effective reminder. Dave From brad.king at kitware.com Fri Jul 15 10:44:08 2016 From: brad.king at kitware.com (Brad King) Date: Fri, 15 Jul 2016 10:44:08 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well In-Reply-To: References: <20160708144414.GB26651@megas.kitware.com> Message-ID: <8c5ed963-fb91-4655-873b-51ef20179ef3@kitware.com> On 07/15/2016 10:31 AM, David Lonie wrote: > +1 to the idea of a default --oneshot. I think it'd definitely > encourage more targeted testing and help reduce buildslave contention. Another +1 to making --oneshot default (and maybe not even having a way to make it repeat automatically on update). Choosing to test the current version should be an explicit step. Sometimes after tests more rounds of manual review are needed before testing again. Also one may want to reduce the set of platforms on which tests occur after an update to the topic. -Brad From brad.king at kitware.com Fri Jul 15 10:57:39 2016 From: brad.king at kitware.com (Brad King) Date: Fri, 15 Jul 2016 10:57:39 -0400 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> <415332dd-2be6-eb14-6603-ff6384607ac8@kitware.com> Message-ID: <36d1853d-519e-aeba-3b28-fc9ce91f336a@kitware.com> On 07/15/2016 10:52 AM, Andras Lasso wrote: > 1. In kwsys implement SystemTools::CreateSymlink for Windows > (this would be useful in general) If you're interested in working on that, please see KWSys development instructions here: http://public.kitware.com/Wiki/KWSys/Git/Develop In order to avoid requiring a CreateSymbolicLink symbol to be available (which it may not be pre-Vista) please use GetProcAddress to look up a pointer to the function at runtime. Thanks, -Brad From lasso at queensu.ca Fri Jul 15 10:52:18 2016 From: lasso at queensu.ca (Andras Lasso) Date: Fri, 15 Jul 2016 14:52:18 +0000 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: <415332dd-2be6-eb14-6603-ff6384607ac8@kitware.com> References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> <415332dd-2be6-eb14-6603-ff6384607ac8@kitware.com> Message-ID: OK, then probably there is no simple solution without CMake changes. However, with a few simple improvements it can be all made to work in Windows: 1. In kwsys implement SystemTools::CreateSymlink for Windows (this would be useful in general) 2. In ExternalData.cmake remove "UNIX AND" from "if(UNIX AND NOT ExternalData_NO_SYMLINKS)..." in _ExternalData_link_or_copy 3. In ExternalData.cmake set ExternalData_NO_SYMLINKS to TRUE by default on Windows (or maybe do a build-time check for SystemTools::CreateSymlink and set ExternalData_NO_SYMLINKS to TRUE by default only if the check fails) This would allow symlinking on Windows for anybody who is concerned about disk space. Andras -----Original Message----- From: Brad King [mailto:brad.king at kitware.com] Sent: July 15, 2016 9:46 To: Andras Lasso ; Andrew Maclean Cc: VTK Developers Subject: Re: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? On 07/15/2016 09:41 AM, Andras Lasso wrote: > You can create symlinks in Windows Yes, but as you point out it is basically useless for applications in general due to lack of permissions. > you should be able to set up shared external directory manually. Every individual file needs to be linked with a specific name in the build tree that is unrelated to its name in the object store. This is not practical to do manually. -Brad From cory.quammen at kitware.com Fri Jul 15 11:30:27 2016 From: cory.quammen at kitware.com (Cory Quammen) Date: Fri, 15 Jul 2016 11:30:27 -0400 Subject: [vtk-developers] rules proposal drop "+1" ie, looks ok to me but someone else needs to review as well In-Reply-To: <8c5ed963-fb91-4655-873b-51ef20179ef3@kitware.com> References: <20160708144414.GB26651@megas.kitware.com> <8c5ed963-fb91-4655-873b-51ef20179ef3@kitware.com> Message-ID: +1 for --oneshot On Fri, Jul 15, 2016 at 10:44 AM, Brad King wrote: > On 07/15/2016 10:31 AM, David Lonie wrote: >> +1 to the idea of a default --oneshot. I think it'd definitely >> encourage more targeted testing and help reduce buildslave contention. > > Another +1 to making --oneshot default (and maybe not even having a > way to make it repeat automatically on update). Choosing to test the > current version should be an explicit step. Sometimes after tests > more rounds of manual review are needed before testing again. Also > one may want to reduce the set of platforms on which tests occur after > an update to the topic. > > -Brad > > _______________________________________________ > 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 > -- Cory Quammen R&D Engineer Kitware, Inc. From ben.boeckel at kitware.com Fri Jul 15 12:47:49 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Fri, 15 Jul 2016 12:47:49 -0400 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> Message-ID: <20160715164749.GA29477@megas.kitware.com> On Fri, Jul 15, 2016 at 13:41:34 +0000, Andras Lasso wrote: > You can create symlinks in Windows (the feature was added a 10+ years > ago): > https://technet.microsoft.com/en-us/library/cc753194(v=ws.11).aspx Hard links are probably more useful on Windows since (I believe), they do not require elevated privs: https://msdn.microsoft.com/en-us/library/windows/desktop/aa365006(v=vs.85).aspx I belive symlinks require privs because there ~all Windows programs do not guard against symlink redirection attacks (usually TOCTOU). And since Windows doesn't really have atomic filesystem operations, there's ~no way avoid TOCTOU. --Ben From andrew.amaclean at gmail.com Sat Jul 16 05:43:18 2016 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Sat, 16 Jul 2016 19:43:18 +1000 Subject: [vtk-developers] A single ExternalData folder for multiple variants of VTK. Is it possible? In-Reply-To: References: <32a36eb6-a67d-ab8d-ce96-85a9242c4b5e@kitware.com> <20160715164749.GA29477@megas.kitware.com> Message-ID: Thankyou Brad for your very clear explanation. This nicely explains the difference between the Windows and Linux systems. Andrew Maclean On 16 Jul 2016 2:47 am, "Ben Boeckel" wrote: On Fri, Jul 15, 2016 at 13:41:34 +0000, Andras Lasso wrote: > You can create symlinks in Windows (the feature was added a 10+ years > ago): > https://technet.microsoft.com/en-us/library/cc753194(v=ws.11).aspx Hard links are probably more useful on Windows since (I believe), they do not require elevated privs: https://msdn.microsoft.com/en-us/library/windows/desktop/aa365006(v=vs.85).aspx I belive symlinks require privs because there ~all Windows programs do not guard against symlink redirection attacks (usually TOCTOU). And since Windows doesn't really have atomic filesystem operations, there's ~no way avoid TOCTOU. --Ben -------------- next part -------------- An HTML attachment was scrubbed... URL: From berk.geveci at kitware.com Tue Jul 19 09:30:34 2016 From: berk.geveci at kitware.com (Berk Geveci) Date: Tue, 19 Jul 2016 09:30:34 -0400 Subject: [vtk-developers] Kitware is hiring Message-ID: Apologies for duplicate postings ======================== Hi folks, Kitware is seeking to hire highly skilled research and development engineers (R&D Engineers) with strong software engineering experience to work on biomechanical modelling and simulation projects. Kitware is developing surgical simulation and computational tools for patient-specific structural and fluid dynamics models that would allow surgeons to plan surgeries and to practice and advance their surgical skills in a risk-free environment. The target clinical applications include open and laparoscopic surgical procedures such as cardiovascular interventions, neurosurgery, nasal surgery, and orthopedic surgery. For the full posting see: https://goo.gl/2VQ5v1 JOB DESCRIPTION Kitware is seeking to hire highly skilled Research and Development Engineers (R&D Engineers) with strong software engineering experience to work on biomechanical modelling and simulation projects. Kitware is developing surgical simulation and computational tools for patient-specific structural and fluid dynamics models that would allow surgeons to plan surgeries and to practice and advance their surgical skills in a risk-free environment. The target clinical applications include open and laparoscopic surgical procedures such as cardiovascular interventions, neurosurgery, nasal surgery, and orthopedic surgery. Qualifications * Masters degree or PhD. in mechanical engineering, applied mathematics or a related field with emphasis in biomedical applications. * Experience in simulation workflow i.e tuning simulations by varying their discretization, constitutive equations, and solver parameters to obtain convergence with minimal computational effort is strongly desirable. * Knowledge of inner-workings of numerical methods such as finite elements, finite difference and numerical solvers is extremely desirable. * Specific knowledge of biomechanical modeling of bone and soft tissue as well as the interaction between these tissues and other materials. * Experience in programming a simulation software and participation in software development workflows is desirable. * Prior experience with other open source physics-based medical simulation software is a plus. * Prior experience developing algorithms to perform fluid-structure interaction and computational fluid dynamics simulations is a plus. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ken.martin at kitware.com Tue Jul 19 10:53:11 2016 From: ken.martin at kitware.com (Ken Martin) Date: Tue, 19 Jul 2016 10:53:11 -0400 Subject: [vtk-developers] karego-at Message-ID: Can someone with access to karego update the cmake version? It seems to be failing nightly expected builds due to an old cmake. Thanks Ken -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sheng.fu at intel.com Tue Jul 19 14:09:01 2016 From: sheng.fu at intel.com (Fu, Sheng) Date: Tue, 19 Jul 2016 18:09:01 +0000 Subject: [vtk-developers] How to build Utilities/Benchmarks Message-ID: Hi, I am new to VTK. I am working on improve OpenGL performance for VTK use cases. I want to build Utilities/Benchmarks. However, it seems run ccmake for the root directory of VTK source code does not include Utilities/Benchmarks. What should I do to make it built? Thanks Sheng -------------- next part -------------- An HTML attachment was scrubbed... URL: From ken.martin at kitware.com Tue Jul 19 14:28:03 2016 From: ken.martin at kitware.com (Ken Martin) Date: Tue, 19 Jul 2016 14:28:03 -0400 Subject: [vtk-developers] How to build Utilities/Benchmarks In-Reply-To: References: Message-ID: Turn on utiltiies/benchmarks in the cmake gui. It may be an advanced option so you may have to show advanced options. - Ken On Tue, Jul 19, 2016 at 2:09 PM, Fu, Sheng wrote: > Hi, > > > > I am new to VTK. I am working on improve OpenGL performance for VTK use > cases. I want to build Utilities/Benchmarks. However, it seems run ccmake > for the root directory of VTK source code does not include > Utilities/Benchmarks. What should I do to make it built? > > > > Thanks > > > > Sheng > > > > _______________________________________________ > 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 > > > -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sheng.fu at intel.com Tue Jul 19 14:51:05 2016 From: sheng.fu at intel.com (Fu, Sheng) Date: Tue, 19 Jul 2016 18:51:05 +0000 Subject: [vtk-developers] How to build Utilities/Benchmarks In-Reply-To: References: Message-ID: Thanks for your quick reply, I do not realize there are advanced options. However, after I built it and run ?make install?, I still only see vtkEncodeString-7.1 and vtkHashSource-7.1 in bin directory. Where is the executable? Thanks Sheng From: Ken Martin [mailto:ken.martin at kitware.com] Sent: Tuesday, July 19, 2016 11:28 AM To: Fu, Sheng Cc: vtk-developers at vtk.org Subject: Re: [vtk-developers] How to build Utilities/Benchmarks Turn on utiltiies/benchmarks in the cmake gui. It may be an advanced option so you may have to show advanced options. - Ken On Tue, Jul 19, 2016 at 2:09 PM, Fu, Sheng > wrote: Hi, I am new to VTK. I am working on improve OpenGL performance for VTK use cases. I want to build Utilities/Benchmarks. However, it seems run ccmake for the root directory of VTK source code does not include Utilities/Benchmarks. What should I do to make it built? Thanks Sheng _______________________________________________ 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 -- Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 518 371 3971 This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Wed Jul 20 09:08:16 2016 From: david.gobbi at gmail.com (David Gobbi) Date: Wed, 20 Jul 2016 07:08:16 -0600 Subject: [vtk-developers] Rogue7 vtktiff dashboard build failure Message-ID: Looks like the tiff CMakeLists.txt (starting at line 184) is doing some linker stuff that doesn't belong in vtktiff: -Wl,--version-script=/.../VTK/ThirdParty/tiff/vtktiff/libtiff/libtiff.map https://open.cdash.org/viewBuildError.php?buildid=4465228 - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerald.Lodron at joanneum.at Fri Jul 22 06:51:08 2016 From: Gerald.Lodron at joanneum.at (Lodron, Gerald) Date: Fri, 22 Jul 2016 10:51:08 +0000 Subject: [vtk-developers] Number of neighbours for each cell Message-ID: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> Hi I have a mesh (vtkPolyData with all triangles) where i want to visualize the number of neighbouring cells/triangles for each cell/triangle. (I later may want to threshold/filter it with that number to remove "noisy" triangles on border with low number of neighbours) Is there a filter for that available? Best regards ------------------------------------------------------------------------------------ Gerald Lodron Researcher of Machine Vision Applications Group DIGITAL - Institute for Information and Communication Technologies JOANNEUM RESEARCH Forschungsgesellschaft mbH Steyrergasse 17, 8010 Graz, AUSTRIA phone: +43-316-876-1751 general fax: +43-316-876-1751 web: http://www.joanneum.at/digital e-mail: gerald.lodron at joanneum.at -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Fri Jul 22 08:08:39 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Fri, 22 Jul 2016 08:08:39 -0400 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> References: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> Message-ID: I don't know of such a filter but it would be easy to put together. Certainly it would be another useful tool in the topological analysis toolbox. Question: what is your definition of a neighbor? Connected at a vertex or along an edge? I suppose if we were to write such a filter the definition of a neighbor would be an optional parameter. I also suggest that this be a general filter so the same analysis could be applied to unstructured grids, etc. or to polygonal meshes with other than triangle cells. Best, W On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald wrote: > > > Hi > > I have a mesh (vtkPolyData with all triangles) where i want to visualize > the number of neighbouring cells/triangles for each cell/triangle. (I later > may want to threshold/filter it with that number to remove ?noisy? > triangles on border with low number of neighbours) > > Is there a filter for that available? > > Best regards > > > *------------------------------------------------------------------------------------* > *Gerald Lodron* > > Researcher of Machine Vision Applications Group > DIGITAL - Institute for Information and Communication Technologies > > JOANNEUM RESEARCH Forschungsgesellschaft mbH > Steyrergasse 17, 8010 Graz, AUSTRIA > > phone: +43-316-876-1751 > general fax: +43-316-876-1751 > web: *http://www.joanneum.at/digital* > e-mail: *gerald.lodron at joanneum.at* > > > > > _______________________________________________ > 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 > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerald.Lodron at joanneum.at Fri Jul 22 08:25:58 2016 From: Gerald.Lodron at joanneum.at (Lodron, Gerald) Date: Fri, 22 Jul 2016 12:25:58 +0000 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: References: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> Message-ID: <3f97047034494c71b107fad7b010d76f@RZJMBX2.jr1.local> In my case the definition would be an edge, but yes it would be useful to define that as parameter?. I already found a maybe not beautiful algorithm for vertices, not edges, may you want to reuse it as initial code: vtkSmartPointer neighbours = vtkSmartPointer::New(); neighbours ->SetName("NumberOfNeighbours"); neighbours ->SetNumberOfValues(numberOfCells); vtkSmartPointer cellPointIds = vtkSmartPointer::New(); vtkSmartPointer ptids = vtkSmartPointer::New(); vtkSmartPointer cellids = vtkSmartPointer::New(); for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) { input->GetCellPoints(cellId, ptids); /* does not work input ->GetCellNeighbors(cellId, oPtIds, oCellIds); neighbours ->SetValue(cellId, static_cast(oCellIds->GetNumberOfIds()));*/ std::set id_set; for (vtkIdType ptid = 0; ptid < ptids ->GetNumberOfIds();ptid++) { poInput1->GetPointCells(ptids ->GetId(ptid), cellids); for (vtkIdType cid = 0; cid < cellids ->GetNumberOfIds();cid++) { vtkIdType id = cellids ->GetId(cid); if (id != cellId) { id_set.insert(cellids->GetId(cid)); } } } neighbours ->SetValue(cellId, static_cast( id_set.size())); } output->GetCellData()->AddArray(neighbours); Von: Will Schroeder [mailto:will.schroeder at kitware.com] Gesendet: Freitag, 22. Juli 2016 14:09 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org) Betreff: Re: [vtk-developers] Number of neighbours for each cell I don't know of such a filter but it would be easy to put together. Certainly it would be another useful tool in the topological analysis toolbox. Question: what is your definition of a neighbor? Connected at a vertex or along an edge? I suppose if we were to write such a filter the definition of a neighbor would be an optional parameter. I also suggest that this be a general filter so the same analysis could be applied to unstructured grids, etc. or to polygonal meshes with other than triangle cells. Best, W On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald wrote: ? ? Hi ? I have a mesh (vtkPolyData with all triangles) where i want to visualize the number of neighbouring cells/triangles for each cell/triangle. (I later may want to threshold/filter it with that number to remove ?noisy? triangles on border with low number of neighbours) ? Is there a filter for that available? ? Best regards ? ------------------------------------------------------------------------------------ Gerald Lodron ? Researcher of Machine Vision Applications Group DIGITAL - Institute for Information and Communication Technologies ? JOANNEUM RESEARCH Forschungsgesellschaft mbH Steyrergasse 17, 8010 Graz, AUSTRIA ? phone:?? +43-316-876-1751??? general fax: +43-316-876-1751 web: http://www.joanneum.at/digital e-mail: gerald.lodron at joanneum.at ? ? ? _______________________________________________ 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 -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 From will.schroeder at kitware.com Fri Jul 22 08:36:54 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Fri, 22 Jul 2016 08:36:54 -0400 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: <3f97047034494c71b107fad7b010d76f@RZJMBX2.jr1.local> References: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> <3f97047034494c71b107fad7b010d76f@RZJMBX2.jr1.local> Message-ID: Yes this is a great start. Since I am on a performance kick, I would probably use (under the hood) vtkStaticCellLocator which builds the links between points and cells much faster (threaded using vtkSMPTools and a parallel sort). The actual "counting" of the neighbors could then be sped up using vtkSMPTools as well. It's probably overkill, but I'd like VTK to knock off the socks of anyone using it, and also to support our HPC friends who have this humbling habit of dumping really big data on top of VTK :-) Best, W On Fri, Jul 22, 2016 at 8:25 AM, Lodron, Gerald wrote: > In my case the definition would be an edge, but yes it would be useful to > define that as parameter?. > > I already found a maybe not beautiful algorithm for vertices, not edges, > may you want to reuse it as initial code: > > vtkSmartPointer neighbours = > vtkSmartPointer::New(); > neighbours ->SetName("NumberOfNeighbours"); > neighbours ->SetNumberOfValues(numberOfCells); > > vtkSmartPointer cellPointIds = > vtkSmartPointer::New(); > > vtkSmartPointer ptids = > vtkSmartPointer::New(); > vtkSmartPointer cellids = > vtkSmartPointer::New(); > > for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) > { > input->GetCellPoints(cellId, ptids); > /* > does not work > input ->GetCellNeighbors(cellId, oPtIds, oCellIds); > neighbours ->SetValue(cellId, static_cast int>(oCellIds->GetNumberOfIds()));*/ > std::set id_set; > > for (vtkIdType ptid = 0; ptid < ptids > ->GetNumberOfIds();ptid++) > { > poInput1->GetPointCells(ptids ->GetId(ptid), cellids); > for (vtkIdType cid = 0; cid < cellids > ->GetNumberOfIds();cid++) > { > vtkIdType id = cellids ->GetId(cid); > if (id != cellId) > { > id_set.insert(cellids->GetId(cid)); > } > } > } > neighbours ->SetValue(cellId, static_cast( > id_set.size())); > } > output->GetCellData()->AddArray(neighbours); > > Von: Will Schroeder [mailto:will.schroeder at kitware.com] > Gesendet: Freitag, 22. Juli 2016 14:09 > An: Lodron, Gerald > Cc: VTK Developer (vtk-developers at vtk.org) > Betreff: Re: [vtk-developers] Number of neighbours for each cell > > I don't know of such a filter but it would be easy to put together. > Certainly it would be another useful tool in the topological analysis > toolbox. > > Question: what is your definition of a neighbor? Connected at a vertex or > along an edge? I suppose if we were to write such a filter the definition > of a neighbor would be an optional parameter. I also suggest that this be a > general filter so the same analysis could be applied to unstructured grids, > etc. or to polygonal meshes with other than triangle cells. > > Best, > W > > On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald > wrote: > > > Hi > > I have a mesh (vtkPolyData with all triangles) where i want to visualize > the number of neighbouring cells/triangles for each cell/triangle. (I later > may want to threshold/filter it with that number to remove ?noisy? > triangles on border with low number of neighbours) > > Is there a filter for that available? > > Best regards > > > ------------------------------------------------------------------------------------ > Gerald Lodron > > Researcher of Machine Vision Applications Group > DIGITAL - Institute for Information and Communication Technologies > > JOANNEUM RESEARCH Forschungsgesellschaft mbH > Steyrergasse 17, 8010 Graz, AUSTRIA > > phone: +43-316-876-1751 > general fax: +43-316-876-1751 > web: http://www.joanneum.at/digital > e-mail: gerald.lodron at joanneum.at > > > > > _______________________________________________ > 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 > > > > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerald.Lodron at joanneum.at Fri Jul 22 08:44:28 2016 From: Gerald.Lodron at joanneum.at (Lodron, Gerald) Date: Fri, 22 Jul 2016 12:44:28 +0000 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: References: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> <3f97047034494c71b107fad7b010d76f@RZJMBX2.jr1.local> Message-ID: <230f792aa0354ef99551e5cdf5a73ba0@RZJMBX2.jr1.local> Cool For my first test that implementation below worked and i found out that this kind of algorithm does not help me on my problem, arg Can I send you an image/screenshot of the ?geometric? stuff I want to filter? May you as an expert has a better idea for solving my problem. Short text description (but is really hard to explain): I have a mesh of a surface, scanned with a 3d scanner (triangles, vtkPolydata as explained). On the border of that mesh I have single ?noisy? (or fringy) structures which I want to filter. These noise structures are minimal connected to the main mesh, so I can not use connected component filter (which I also used apriory to remove noise in the air). Von: Will Schroeder [mailto:will.schroeder at kitware.com] Gesendet: Freitag, 22. Juli 2016 14:37 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org) Betreff: Re: [vtk-developers] Number of neighbours for each cell Yes this is a great start. Since I am on a performance kick, I would probably use (under the hood) vtkStaticCellLocator which builds the links between points and cells much faster (threaded using vtkSMPTools and a parallel sort). The actual "counting" of the neighbors could then be sped up using vtkSMPTools as well. It's probably overkill, but I'd like VTK to knock off the socks of anyone using it, and also to support our HPC friends who have this humbling habit of dumping really big data on top of VTK :-) Best, W On Fri, Jul 22, 2016 at 8:25 AM, Lodron, Gerald > wrote: In my case the definition would be an edge, but yes it would be useful to define that as parameter?. I already found a maybe not beautiful algorithm for vertices, not edges, may you want to reuse it as initial code: vtkSmartPointer neighbours = vtkSmartPointer::New(); neighbours ->SetName("NumberOfNeighbours"); neighbours ->SetNumberOfValues(numberOfCells); vtkSmartPointer cellPointIds = vtkSmartPointer::New(); vtkSmartPointer ptids = vtkSmartPointer::New(); vtkSmartPointer cellids = vtkSmartPointer::New(); for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) { input->GetCellPoints(cellId, ptids); /* does not work input ->GetCellNeighbors(cellId, oPtIds, oCellIds); neighbours ->SetValue(cellId, static_cast(oCellIds->GetNumberOfIds()));*/ std::set id_set; for (vtkIdType ptid = 0; ptid < ptids ->GetNumberOfIds();ptid++) { poInput1->GetPointCells(ptids ->GetId(ptid), cellids); for (vtkIdType cid = 0; cid < cellids ->GetNumberOfIds();cid++) { vtkIdType id = cellids ->GetId(cid); if (id != cellId) { id_set.insert(cellids->GetId(cid)); } } } neighbours ->SetValue(cellId, static_cast( id_set.size())); } output->GetCellData()->AddArray(neighbours); Von: Will Schroeder [mailto:will.schroeder at kitware.com] Gesendet: Freitag, 22. Juli 2016 14:09 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org) Betreff: Re: [vtk-developers] Number of neighbours for each cell I don't know of such a filter but it would be easy to put together. Certainly it would be another useful tool in the topological analysis toolbox. Question: what is your definition of a neighbor? Connected at a vertex or along an edge? I suppose if we were to write such a filter the definition of a neighbor would be an optional parameter. I also suggest that this be a general filter so the same analysis could be applied to unstructured grids, etc. or to polygonal meshes with other than triangle cells. Best, W On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald > wrote: Hi I have a mesh (vtkPolyData with all triangles) where i want to visualize the number of neighbouring cells/triangles for each cell/triangle. (I later may want to threshold/filter it with that number to remove ?noisy? triangles on border with low number of neighbours) Is there a filter for that available? Best regards ------------------------------------------------------------------------------------ Gerald Lodron Researcher of Machine Vision Applications Group DIGITAL - Institute for Information and Communication Technologies JOANNEUM RESEARCH Forschungsgesellschaft mbH Steyrergasse 17, 8010 Graz, AUSTRIA phone: +43-316-876-1751 general fax: +43-316-876-1751 web: http://www.joanneum.at/digital e-mail: gerald.lodron at joanneum.at _______________________________________________ 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 -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Fri Jul 22 08:51:58 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Fri, 22 Jul 2016 08:51:58 -0400 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: <230f792aa0354ef99551e5cdf5a73ba0@RZJMBX2.jr1.local> References: <4ea3d63c508e4f6c9d52cbf1feb6ff04@RZJMBX2.jr1.local> <3f97047034494c71b107fad7b010d76f@RZJMBX2.jr1.local> <230f792aa0354ef99551e5cdf5a73ba0@RZJMBX2.jr1.local> Message-ID: Sure send the image we always love a challenge :-) Obviously if it becomes more involved we can talk about a support product , etc. Best, W On Fri, Jul 22, 2016 at 8:44 AM, Lodron, Gerald wrote: > Cool > > > > For my first test that implementation below worked and i found out that > this kind of algorithm does not help me on my problem, arg > > > > Can I send you an image/screenshot of the ?geometric? stuff I want to > filter? May you as an expert has a better idea for solving my problem. > Short text description (but is really hard to explain): > > > > I have a mesh of a surface, scanned with a 3d scanner (triangles, > vtkPolydata as explained). On the border of that mesh I have single ?noisy? > (or fringy) structures which I want to filter. These noise structures are > minimal connected to the main mesh, so I can not use connected component > filter (which I also used apriory to remove noise in the air). > > > > *Von:* Will Schroeder [mailto:will.schroeder at kitware.com] > *Gesend**et:* Freitag, 22. Juli 2016 14:37 > > *An:* Lodron, Gerald > *Cc:* VTK Developer (vtk-developers at vtk.org) > *Betreff:* Re: [vtk-developers] Number of neighbours for each cell > > > > Yes this is a great start. Since I am on a performance kick, I would > probably use (under the hood) vtkStaticCellLocator which builds the links > between points and cells much faster (threaded using vtkSMPTools and a > parallel sort). The actual "counting" of the neighbors could then be sped > up using vtkSMPTools as well. It's probably overkill, but I'd like VTK to > knock off the socks of anyone using it, and also to support our HPC friends > who have this humbling habit of dumping really big data on top of VTK :-) > > > > Best, > W > > > > On Fri, Jul 22, 2016 at 8:25 AM, Lodron, Gerald > wrote: > > In my case the definition would be an edge, but yes it would be useful to > define that as parameter?. > > I already found a maybe not beautiful algorithm for vertices, not edges, > may you want to reuse it as initial code: > > vtkSmartPointer neighbours = > vtkSmartPointer::New(); > neighbours ->SetName("NumberOfNeighbours"); > neighbours ->SetNumberOfValues(numberOfCells); > > vtkSmartPointer cellPointIds = > vtkSmartPointer::New(); > > vtkSmartPointer ptids = > vtkSmartPointer::New(); > vtkSmartPointer cellids = > vtkSmartPointer::New(); > > for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) > { > input->GetCellPoints(cellId, ptids); > /* > does not work > input ->GetCellNeighbors(cellId, oPtIds, oCellIds); > neighbours ->SetValue(cellId, static_cast int>(oCellIds->GetNumberOfIds()));*/ > std::set id_set; > > for (vtkIdType ptid = 0; ptid < ptids > ->GetNumberOfIds();ptid++) > { > poInput1->GetPointCells(ptids ->GetId(ptid), cellids); > for (vtkIdType cid = 0; cid < cellids > ->GetNumberOfIds();cid++) > { > vtkIdType id = cellids ->GetId(cid); > if (id != cellId) > { > id_set.insert(cellids->GetId(cid)); > } > } > } > neighbours ->SetValue(cellId, static_cast( > id_set.size())); > } > output->GetCellData()->AddArray(neighbours); > > Von: Will Schroeder [mailto:will.schroeder at kitware.com] > Gesendet: Freitag, 22. Juli 2016 14:09 > An: Lodron, Gerald > Cc: VTK Developer (vtk-developers at vtk.org) > Betreff: Re: [vtk-developers] Number of neighbours for each cell > > > I don't know of such a filter but it would be easy to put together. > Certainly it would be another useful tool in the topological analysis > toolbox. > > Question: what is your definition of a neighbor? Connected at a vertex or > along an edge? I suppose if we were to write such a filter the definition > of a neighbor would be an optional parameter. I also suggest that this be a > general filter so the same analysis could be applied to unstructured grids, > etc. or to polygonal meshes with other than triangle cells. > > Best, > W > > On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald > wrote: > > > Hi > > I have a mesh (vtkPolyData with all triangles) where i want to visualize > the number of neighbouring cells/triangles for each cell/triangle. (I later > may want to threshold/filter it with that number to remove ?noisy? > triangles on border with low number of neighbours) > > Is there a filter for that available? > > Best regards > > > ------------------------------------------------------------------------------------ > Gerald Lodron > > Researcher of Machine Vision Applications Group > DIGITAL - Institute for Information and Communication Technologies > > JOANNEUM RESEARCH Forschungsgesellschaft mbH > Steyrergasse 17, 8010 Graz, AUSTRIA > > phone: +43-316-876-1751 > general fax: +43-316-876-1751 > web: http://www.joanneum.at/digital > e-mail: gerald.lodron at joanneum.at > > > > > _______________________________________________ > 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 > > > > > > -- > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > > > > > -- > > William J. Schroeder, PhD > Kitware, Inc. - Building the World's Technical Computing Software > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From brad.king at kitware.com Fri Jul 22 09:14:25 2016 From: brad.king at kitware.com (Brad King) Date: Fri, 22 Jul 2016 09:14:25 -0400 Subject: [vtk-developers] [ITK-dev] c:\.git\hooks In-Reply-To: References: Message-ID: On 07/22/2016 09:00 AM, D?enan Zuki? wrote: > occasionally, a folder "C:\.git\hooks" appears on my computer. [snip] > but support people think that git hooks in some of my projects > (almost all instances of ITK, VTK and Slicer builds) are causing > this behavior. Did anyone else encounter this or something similar? Dig through our Utilities/SetupForDevelopment.sh scripts (or equivalent in each project) and the scripts it invokes. See if running any of them causes this to happen. Otherwise check whether running CMake on each project causes it in case there is any attempt to install hooks that way. You'll likely have to diligently check for the folder after each step of working with each project until you figure out which one causes it. -Brad From h.moelle at googlemail.com Mon Jul 25 04:14:21 2016 From: h.moelle at googlemail.com (=?UTF-8?Q?Hagen_M=c3=b6lle?=) Date: Mon, 25 Jul 2016 10:14:21 +0200 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis Message-ID: Hi, Due to spam reasons the bug tracker's user login is disabled right now. One should send bug reports to the vtk's user mailing list, which I did twice. My bug report was totally ignored. That's why I send it here again. Hopefully someone can finally create a ticket in the bug tracker. Thanks. I want to report a bug for vtkChartXY. So I already send this Email two days ago to the vtkUsers mailing list without receiving any answer. The bug is not visible in the vtk bug tracker too. Thus I send the Email again. Bug environment: a) I am using VTK 6.2. I checked my test code on machines running Windows 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 (gcc). On all these systems the bug is visible. b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu 16.04 machine and compiled VTK 7.0 using OpenGL2 interface). Bug description: I am using vtkChartXY to display one line. I also enabled logarithmic x axis (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing more special. In the compiled application I start zooming into the chart using the mouse wheel, making sure I always have a line displayed in the view. At some zoom level this line just disappears. At the zoom level the line disappears I did the following: 1. Zooming more into the chart the line will not appear again. 2. Zooming out twice (with mouse wheel) the line will reappear. How to reproduce the bug: I basically used the LinePlot example for vtkChartXY (see http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). I added one line to the example code marked with *** (line was added after line 54 in example code). I added the full code to the end of the Email. --------------- // Add multiple line plots, setting the colors etc vtkSmartPointer chart= vtkSmartPointer::New(); *** chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); view->GetScene()->AddItem(chart); --------------- Compile and execute the program. See my bug description above to visualize the bug. Please contact me if you need more information. I can also provide images of the bug if needed. I hope the bug can be fixed. Hagen. #include #include #include #include #include #include #include #include #include #include #include #include #include int main(int, char *[]) { // Create a table with some points in it vtkSmartPointer table = vtkSmartPointer::New(); vtkSmartPointer arrX = vtkSmartPointer::New(); arrX->SetName("X Axis"); table->AddColumn(arrX); vtkSmartPointer arrC = vtkSmartPointer::New(); arrC->SetName("Cosine"); table->AddColumn(arrC); vtkSmartPointer arrS = vtkSmartPointer::New(); arrS->SetName("Sine"); table->AddColumn(arrS); // Fill in the table with some example values int numPoints = 69; float inc = 7.5 / (numPoints-1); table->SetNumberOfRows(numPoints); for (int i = 0; i < numPoints; ++i) { table->SetValue(i, 0, i * inc); table->SetValue(i, 1, cos(i * inc)); table->SetValue(i, 2, sin(i * inc)); } // Set up the view vtkSmartPointer view = vtkSmartPointer::New(); view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); // Add multiple line plots, setting the colors etc vtkSmartPointer chart = vtkSmartPointer::New(); chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); view->GetScene()->AddItem(chart); vtkPlot *line = chart->AddPlot(vtkChart::LINE); #if VTK_MAJOR_VERSION <= 5 line->SetInput(table, 0, 1); #else line->SetInputData(table, 0, 1); #endif line->SetColor(0, 255, 0, 255); line->SetWidth(1.0); line = chart->AddPlot(vtkChart::LINE); #if VTK_MAJOR_VERSION <= 5 line->SetInput(table, 0, 2); #else line->SetInputData(table, 0, 2); #endif line->SetColor(255, 0, 0, 255); line->SetWidth(5.0); // For dotted line, the line type can be from 2 to 5 for different dash/dot // patterns (see enum in vtkPen containing DASH_LINE, value 2): #ifndef WIN32 line->GetPen()->SetLineType(vtkPen::DASH_LINE); #endif // (ifdef-ed out on Windows because DASH_LINE does not work on Windows // machines with built-in Intel HD graphics card...) //view->GetRenderWindow()->SetMultiSamples(0); // Start interactor view->GetInteractor()->Initialize(); view->GetInteractor()->Start(); return EXIT_SUCCESS; } From andrew.amaclean at gmail.com Mon Jul 25 19:52:41 2016 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Tue, 26 Jul 2016 09:52:41 +1000 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis Message-ID: Hi Hagen, I just tested your example and I get similar behaviou using an NVIDIA Video card, OpenGL2, VS2015 and the VTK Master. I can zoom in and, at some point the line disappears, but zooming out immediately restores the line. It seeme to me that once a certain zoom level is reached the line is not drawn, however zooming out restores it. Attached are two pngs showing the effect.When the line disappears it should still be visible as the scales on the axes have only changed slightly. Andrew ---------- Forwarded message ---------- > From: "Hagen M?lle" > To: vtk-developers at vtk.org > Cc: > Date: Mon, 25 Jul 2016 10:14:21 +0200 > Subject: [vtk-developers] vtkChartXY lines not visible when using > logarithmic x axis > Hi, > > Due to spam reasons the bug tracker's user login is disabled right now. > One should send bug reports to the vtk's user mailing list, which I did > twice. My bug report was totally ignored. That's why I send it here again. > Hopefully someone can finally create a ticket in the bug tracker. Thanks. > > I want to report a bug for vtkChartXY. So I already send this Email two > days ago to the vtkUsers mailing list without receiving any answer. The bug > is not visible in the vtk bug tracker too. Thus I send the Email again. > > Bug environment: > a) I am using VTK 6.2. I checked my test code on machines running Windows > 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 (gcc). On all > these systems the bug is visible. > b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu 16.04 > machine and compiled VTK 7.0 using OpenGL2 interface). > > Bug description: > > I am using vtkChartXY to display one line. I also enabled logarithmic x > axis (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing more > special. In the compiled application I start zooming into the chart using > the mouse wheel, making sure I always have a line displayed in the view. At > some zoom level this line just disappears. At the zoom level the line > disappears I did the following: > > 1. Zooming more into the chart the line will not appear again. > 2. Zooming out twice (with mouse wheel) the line will reappear. > > How to reproduce the bug: > > I basically used the LinePlot example for vtkChartXY (see > http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). > > I added one line to the example code marked with *** (line was added after > line 54 in example code). > I added the full code to the end of the Email. > --------------- > > // Add multiple line plots, setting the colors etc > vtkSmartPointer chart= vtkSmartPointer >::New(); *** chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); > view->GetScene()->AddItem(chart); > > --------------- > > Compile and execute the program. See my bug description above to visualize > the bug. > > > Please contact me if you need more information. I can also provide images > of the bug if needed. > I hope the bug can be fixed. > > Hagen. > > > > > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > > int main(int, char *[]) > { > // Create a table with some points in it > vtkSmartPointer table = > vtkSmartPointer::New(); > > vtkSmartPointer arrX = > vtkSmartPointer::New(); > arrX->SetName("X Axis"); > table->AddColumn(arrX); > > vtkSmartPointer arrC = > vtkSmartPointer::New(); > arrC->SetName("Cosine"); > table->AddColumn(arrC); > > vtkSmartPointer arrS = > vtkSmartPointer::New(); > arrS->SetName("Sine"); > table->AddColumn(arrS); > > // Fill in the table with some example values > int numPoints = 69; > float inc = 7.5 / (numPoints-1); > table->SetNumberOfRows(numPoints); > for (int i = 0; i < numPoints; ++i) > { > table->SetValue(i, 0, i * inc); > table->SetValue(i, 1, cos(i * inc)); > table->SetValue(i, 2, sin(i * inc)); > } > > // Set up the view > vtkSmartPointer view = > vtkSmartPointer::New(); > view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); > > // Add multiple line plots, setting the colors etc > vtkSmartPointer chart = > vtkSmartPointer::New(); > chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); > view->GetScene()->AddItem(chart); > vtkPlot *line = chart->AddPlot(vtkChart::LINE); > #if VTK_MAJOR_VERSION <= 5 > line->SetInput(table, 0, 1); > #else > line->SetInputData(table, 0, 1); > #endif > line->SetColor(0, 255, 0, 255); > line->SetWidth(1.0); > line = chart->AddPlot(vtkChart::LINE); > #if VTK_MAJOR_VERSION <= 5 > line->SetInput(table, 0, 2); > #else > line->SetInputData(table, 0, 2); > #endif > line->SetColor(255, 0, 0, 255); > line->SetWidth(5.0); > > // For dotted line, the line type can be from 2 to 5 for different > dash/dot > // patterns (see enum in vtkPen containing DASH_LINE, value 2): > #ifndef WIN32 > line->GetPen()->SetLineType(vtkPen::DASH_LINE); > #endif > // (ifdef-ed out on Windows because DASH_LINE does not work on Windows > // machines with built-in Intel HD graphics card...) > > //view->GetRenderWindow()->SetMultiSamples(0); > > // Start interactor > view->GetInteractor()->Initialize(); > view->GetInteractor()->Start(); > > return EXIT_SUCCESS; > } > > > _______________________________________________ > 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 > -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: JustAfterLineDisappearance.PNG Type: image/png Size: 27648 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: JustBeforeLineDisappearance.PNG Type: image/png Size: 35481 bytes Desc: not available URL: From sean at rogue-research.com Mon Jul 25 21:15:46 2016 From: sean at rogue-research.com (Sean McBride) Date: Mon, 25 Jul 2016 21:15:46 -0400 Subject: [vtk-developers] Rogue7 vtktiff dashboard build failure In-Reply-To: References: Message-ID: <20160726011546.2016066190@mail.rogue-research.com> On Wed, 20 Jul 2016 07:08:16 -0600, David Gobbi said: >Looks like the tiff CMakeLists.txt (starting at line 184) is doing some >linker stuff that doesn't belong in vtktiff: > >-Wl,--version-script=/.../VTK/ThirdParty/tiff/vtktiff/libtiff/libtiff.map >https://open.cdash.org/viewBuildError.php?buildid=4465228 The build failure is still on the dashboard... I guess the unique thing about that submission is that it's the only one (I believe) that uses CMake's Xcode generator and not the Unix Makefiles generator. Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From h.moelle at googlemail.com Tue Jul 26 03:02:01 2016 From: h.moelle at googlemail.com (=?UTF-8?Q?Hagen_M=c3=b6lle?=) Date: Tue, 26 Jul 2016 09:02:01 +0200 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis In-Reply-To: References: Message-ID: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> Hi Andrew, Hi Mathieu, That's exactly the behaviour I see. By the way I am using ATI/AMD video cards in my Windows/Linux machine. !!!! UPDATE: The error also occurs when logarithmic x axis is not enabled. But you have to zoom in a lot more. See attached pictures. I am sorry for not realizing this earlier. I am not that familiar with OpenGL. But I can imagine two things: (1) Line is clipped (2) If shaders are used to draw the line thickness it might be a floating point precision problem. Actually this could be the problem because the line thickness starts fluctuating at some point of zoom level before it finally disappears. See attached picture. The line thickness is about the half the original line thickness. I hope this helps to fix the bug. One might also contact Marcus D. Hanwell. He is the developer who mainly implemented vtkCharts. I am not sure if he is still responsible. Hagen On 26.07.2016 01:52, Andrew Maclean wrote: > Hi Hagen, > I just tested your example and I get similar behaviou using an > NVIDIA Video card, OpenGL2, VS2015 and the VTK Master. > I can zoom in and, at some point the line disappears, but zooming out > immediately restores the line. > > It seeme to me that once a certain zoom level is reached the line is > not drawn, however zooming out restores it. > > Attached are two pngs showing the effect.When the line disappears it > should still be visible as the scales on the axes have only changed > slightly. > > Andrew > > ---------- Forwarded message ---------- > From: "Hagen M?lle" > > To: vtk-developers at vtk.org > Cc: > Date: Mon, 25 Jul 2016 10:14:21 +0200 > Subject: [vtk-developers] vtkChartXY lines not visible when using > logarithmic x axis > Hi, > > Due to spam reasons the bug tracker's user login is disabled right > now. One should send bug reports to the vtk's user mailing list, > which I did twice. My bug report was totally ignored. That's why I > send it here again. Hopefully someone can finally create a ticket > in the bug tracker. Thanks. > > I want to report a bug for vtkChartXY. So I already send this > Email two days ago to the vtkUsers mailing list without receiving > any answer. The bug is not visible in the vtk bug tracker too. > Thus I send the Email again. > > Bug environment: > a) I am using VTK 6.2. I checked my test code on machines running > Windows 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 > (gcc). On all these systems the bug is visible. > b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu > 16.04 machine and compiled VTK 7.0 using OpenGL2 interface). > > Bug description: > > I am using vtkChartXY to display one line. I also enabled > logarithmic x axis > (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing > more special. In the compiled application I start zooming into the > chart using the mouse wheel, making sure I always have a line > displayed in the view. At some zoom level this line just > disappears. At the zoom level the line disappears I did the following: > > 1. Zooming more into the chart the line will not appear again. > 2. Zooming out twice (with mouse wheel) the line will reappear. > > How to reproduce the bug: > > I basically used the LinePlot example for vtkChartXY (see > http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). > > I added one line to the example code marked with *** (line was > added after line 54 in example code). > I added the full code to the end of the Email. > --------------- > > // Add multiple line plots, setting the colors etc > vtkSmartPointer chart= > vtkSmartPointer::New(); *** > chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); > view->GetScene()->AddItem(chart); > > --------------- > > Compile and execute the program. See my bug description above to > visualize the bug. > > > Please contact me if you need more information. I can also provide > images of the bug if needed. > I hope the bug can be fixed. > > Hagen. > > > > > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > > int main(int, char *[]) > { > // Create a table with some points in it > vtkSmartPointer table = > vtkSmartPointer::New(); > > vtkSmartPointer arrX = > vtkSmartPointer::New(); > arrX->SetName("X Axis"); > table->AddColumn(arrX); > > vtkSmartPointer arrC = > vtkSmartPointer::New(); > arrC->SetName("Cosine"); > table->AddColumn(arrC); > > vtkSmartPointer arrS = > vtkSmartPointer::New(); > arrS->SetName("Sine"); > table->AddColumn(arrS); > > // Fill in the table with some example values > int numPoints = 69; > float inc = 7.5 / (numPoints-1); > table->SetNumberOfRows(numPoints); > for (int i = 0; i < numPoints; ++i) > { > table->SetValue(i, 0, i * inc); > table->SetValue(i, 1, cos(i * inc)); > table->SetValue(i, 2, sin(i * inc)); > } > > // Set up the view > vtkSmartPointer view = > vtkSmartPointer::New(); > view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); > > // Add multiple line plots, setting the colors etc > vtkSmartPointer chart = > vtkSmartPointer::New(); > chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); > view->GetScene()->AddItem(chart); > vtkPlot *line = chart->AddPlot(vtkChart::LINE); > #if VTK_MAJOR_VERSION <= 5 > line->SetInput(table, 0, 1); > #else > line->SetInputData(table, 0, 1); > #endif > line->SetColor(0, 255, 0, 255); > line->SetWidth(1.0); > line = chart->AddPlot(vtkChart::LINE); > #if VTK_MAJOR_VERSION <= 5 > line->SetInput(table, 0, 2); > #else > line->SetInputData(table, 0, 2); > #endif > line->SetColor(255, 0, 0, 255); > line->SetWidth(5.0); > > // For dotted line, the line type can be from 2 to 5 for > different dash/dot > // patterns (see enum in vtkPen containing DASH_LINE, value 2): > #ifndef WIN32 > line->GetPen()->SetLineType(vtkPen::DASH_LINE); > #endif > // (ifdef-ed out on Windows because DASH_LINE does not work on > Windows > // machines with built-in Intel HD graphics card...) > > //view->GetRenderWindow()->SetMultiSamples(0); > > // Start interactor > view->GetInteractor()->Initialize(); > view->GetInteractor()->Start(); > > return EXIT_SUCCESS; > } > > > _______________________________________________ > 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 > > > > > -- > ___________________________________________ > Andrew J. P. Maclean > > ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: nonlogarihmic_justbeforelinedisappears.png Type: image/png Size: 48659 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: nonlogarithmic_justafterlinedisappears.png Type: image/png Size: 35388 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: linethickness_fluctuates.png Type: image/png Size: 44196 bytes Desc: not available URL: From andrew.amaclean at gmail.com Tue Jul 26 04:57:43 2016 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Tue, 26 Jul 2016 18:57:43 +1000 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis In-Reply-To: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> References: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> Message-ID: Hi Hagen, Mathieu, I agree with you, if I zoom in when the logarithmic axis is disabled, the thick line vanishes, however if you look carefully there is a very faint thin line. It may be (2) as you outline below. Regards Andrew Andrew Maclean On 26 Jul 2016 5:02 pm, "Hagen M?lle" wrote: > Hi Andrew, Hi Mathieu, > > That's exactly the behaviour I see. By the way I am using ATI/AMD video > cards in my Windows/Linux machine. > > !!!! UPDATE: The error also occurs when logarithmic x axis is not enabled. > But you have to zoom in a lot more. See attached pictures. I am sorry for > not realizing this earlier. > > > I am not that familiar with OpenGL. But I can imagine two things: > > (1) Line is clipped > (2) If shaders are used to draw the line thickness it might be a floating > point precision problem. Actually this could be the problem because the > line thickness starts fluctuating at some point of zoom level before it > finally disappears. See attached picture. The line thickness is about the > half the original line thickness. > > > I hope this helps to fix the bug. One might also contact Marcus D. > Hanwell. He is the developer who mainly implemented vtkCharts. I am not > sure if he is still responsible. > > Hagen > > > On 26.07.2016 01:52, Andrew Maclean wrote: > > Hi Hagen, > I just tested your example and I get similar behaviou using an NVIDIA > Video card, OpenGL2, VS2015 and the VTK Master. > > I can zoom in and, at some point the line disappears, but zooming out > immediately restores the line. > > It seeme to me that once a certain zoom level is reached the line is not > drawn, however zooming out restores it. > > Attached are two pngs showing the effect.When the line disappears it > should still be visible as the scales on the axes have only changed > slightly. > > Andrew > > ---------- Forwarded message ---------- >> From: "Hagen M?lle" >> To: vtk-developers at vtk.org >> Cc: >> Date: Mon, 25 Jul 2016 10:14:21 +0200 >> Subject: [vtk-developers] vtkChartXY lines not visible when using >> logarithmic x axis >> Hi, >> >> Due to spam reasons the bug tracker's user login is disabled right now. >> One should send bug reports to the vtk's user mailing list, which I did >> twice. My bug report was totally ignored. That's why I send it here again. >> Hopefully someone can finally create a ticket in the bug tracker. Thanks. >> >> I want to report a bug for vtkChartXY. So I already send this Email two >> days ago to the vtkUsers mailing list without receiving any answer. The bug >> is not visible in the vtk bug tracker too. Thus I send the Email again. >> >> Bug environment: >> a) I am using VTK 6.2. I checked my test code on machines running Windows >> 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 (gcc). On all >> these systems the bug is visible. >> b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu 16.04 >> machine and compiled VTK 7.0 using OpenGL2 interface). >> >> Bug description: >> >> I am using vtkChartXY to display one line. I also enabled logarithmic x >> axis (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing more >> special. In the compiled application I start zooming into the chart using >> the mouse wheel, making sure I always have a line displayed in the view. At >> some zoom level this line just disappears. At the zoom level the line >> disappears I did the following: >> >> 1. Zooming more into the chart the line will not appear again. >> 2. Zooming out twice (with mouse wheel) the line will reappear. >> >> How to reproduce the bug: >> >> I basically used the LinePlot example for vtkChartXY (see >> http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). >> >> I added one line to the example code marked with *** (line was added >> after line 54 in example code). >> I added the full code to the end of the Email. >> --------------- >> >> // Add multiple line plots, setting the colors etc >> vtkSmartPointer chart= vtkSmartPointer> >::New(); *** chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >> view->GetScene()->AddItem(chart); >> >> --------------- >> >> Compile and execute the program. See my bug description above to >> visualize the bug. >> >> >> Please contact me if you need more information. I can also provide images >> of the bug if needed. >> I hope the bug can be fixed. >> >> Hagen. >> >> >> >> >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> >> int main(int, char *[]) >> { >> // Create a table with some points in it >> vtkSmartPointer table = >> vtkSmartPointer::New(); >> >> vtkSmartPointer arrX = >> vtkSmartPointer::New(); >> arrX->SetName("X Axis"); >> table->AddColumn(arrX); >> >> vtkSmartPointer arrC = >> vtkSmartPointer::New(); >> arrC->SetName("Cosine"); >> table->AddColumn(arrC); >> >> vtkSmartPointer arrS = >> vtkSmartPointer::New(); >> arrS->SetName("Sine"); >> table->AddColumn(arrS); >> >> // Fill in the table with some example values >> int numPoints = 69; >> float inc = 7.5 / (numPoints-1); >> table->SetNumberOfRows(numPoints); >> for (int i = 0; i < numPoints; ++i) >> { >> table->SetValue(i, 0, i * inc); >> table->SetValue(i, 1, cos(i * inc)); >> table->SetValue(i, 2, sin(i * inc)); >> } >> >> // Set up the view >> vtkSmartPointer view = >> vtkSmartPointer::New(); >> view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); >> >> // Add multiple line plots, setting the colors etc >> vtkSmartPointer chart = >> vtkSmartPointer::New(); >> chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >> view->GetScene()->AddItem(chart); >> vtkPlot *line = chart->AddPlot(vtkChart::LINE); >> #if VTK_MAJOR_VERSION <= 5 >> line->SetInput(table, 0, 1); >> #else >> line->SetInputData(table, 0, 1); >> #endif >> line->SetColor(0, 255, 0, 255); >> line->SetWidth(1.0); >> line = chart->AddPlot(vtkChart::LINE); >> #if VTK_MAJOR_VERSION <= 5 >> line->SetInput(table, 0, 2); >> #else >> line->SetInputData(table, 0, 2); >> #endif >> line->SetColor(255, 0, 0, 255); >> line->SetWidth(5.0); >> >> // For dotted line, the line type can be from 2 to 5 for different >> dash/dot >> // patterns (see enum in vtkPen containing DASH_LINE, value 2): >> #ifndef WIN32 >> line->GetPen()->SetLineType(vtkPen::DASH_LINE); >> #endif >> // (ifdef-ed out on Windows because DASH_LINE does not work on Windows >> // machines with built-in Intel HD graphics card...) >> >> //view->GetRenderWindow()->SetMultiSamples(0); >> >> // Start interactor >> view->GetInteractor()->Initialize(); >> view->GetInteractor()->Start(); >> >> return EXIT_SUCCESS; >> } >> >> >> _______________________________________________ >> 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 >> > > > > -- > ___________________________________________ > Andrew J. P. Maclean > > ___________________________________________ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From h.moelle at googlemail.com Tue Jul 26 05:19:53 2016 From: h.moelle at googlemail.com (=?UTF-8?Q?Hagen_M=c3=b6lle?=) Date: Tue, 26 Jul 2016 11:19:53 +0200 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis In-Reply-To: References: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> Message-ID: <85b2f3dc-f769-79ae-3af2-d8ab587cc7da@googlemail.com> Hi Andrew, Hi Mathieu, I can not see the line after it disappears. This might be a special feature of the open source radeon graphics driver under Linux. Just for completeness one can easily see the line thickness fluctuation by zooming into the scene just right before the line disappears. At this zoom level just drag the scene using the left mouse button and move the mouse around. In fact it would be very nice if this bug could be fixed at least when logarithmic x axis is enabled. Because when logarithmic x axis is enabled the error is visible after zooming in only a few steps. Hagen On 26.07.2016 10:57, Andrew Maclean wrote: > > Hi Hagen, Mathieu, > > I agree with you, if I zoom in when the logarithmic axis is disabled, > the thick line vanishes, however if you look carefully there is a very > faint thin line. It may be (2) as you outline below. > > Regards > Andrew > > Andrew Maclean > > > On 26 Jul 2016 5:02 pm, "Hagen M?lle" > wrote: > > Hi Andrew, Hi Mathieu, > > That's exactly the behaviour I see. By the way I am using ATI/AMD > video cards in my Windows/Linux machine. > > !!!! UPDATE: The error also occurs when logarithmic x axis is not > enabled. But you have to zoom in a lot more. See attached > pictures. I am sorry for not realizing this earlier. > > > I am not that familiar with OpenGL. But I can imagine two things: > > (1) Line is clipped > (2) If shaders are used to draw the line thickness it might be a > floating point precision problem. Actually this could be the > problem because the line thickness starts fluctuating at some > point of zoom level before it finally disappears. See attached > picture. The line thickness is about the half the original line > thickness. > > > I hope this helps to fix the bug. One might also contact Marcus D. > Hanwell. He is the developer who mainly implemented vtkCharts. I > am not sure if he is still responsible. > > Hagen > > > On 26.07.2016 01:52, Andrew Maclean wrote: >> Hi Hagen, >> I just tested your example and I get similar behaviou using an >> NVIDIA Video card, OpenGL2, VS2015 and the VTK Master. >> I can zoom in and, at some point the line disappears, but zooming >> out immediately restores the line. >> >> It seeme to me that once a certain zoom level is reached the line >> is not drawn, however zooming out restores it. >> >> Attached are two pngs showing the effect.When the line disappears >> it should still be visible as the scales on the axes have only >> changed slightly. >> >> Andrew >> >> ---------- Forwarded message ---------- >> From: "Hagen M?lle" > > >> To: vtk-developers at vtk.org >> Cc: >> Date: Mon, 25 Jul 2016 10:14:21 +0200 >> Subject: [vtk-developers] vtkChartXY lines not visible when >> using logarithmic x axis >> Hi, >> >> Due to spam reasons the bug tracker's user login is disabled >> right now. One should send bug reports to the vtk's user >> mailing list, which I did twice. My bug report was totally >> ignored. That's why I send it here again. Hopefully someone >> can finally create a ticket in the bug tracker. Thanks. >> >> I want to report a bug for vtkChartXY. So I already send this >> Email two days ago to the vtkUsers mailing list without >> receiving any answer. The bug is not visible in the vtk bug >> tracker too. Thus I send the Email again. >> >> Bug environment: >> a) I am using VTK 6.2. I checked my test code on machines >> running Windows 7 (Visual Studio 2012/2015) and Ubuntu >> 15.10/Ubuntu 16.04 (gcc). On all these systems the bug is >> visible. >> b) I can also reproduce the bug in VTK 7.0 (here I used an >> Ubuntu 16.04 machine and compiled VTK 7.0 using OpenGL2 >> interface). >> >> Bug description: >> >> I am using vtkChartXY to display one line. I also enabled >> logarithmic x axis >> (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). >> Nothing more special. In the compiled application I start >> zooming into the chart using the mouse wheel, making sure I >> always have a line displayed in the view. At some zoom level >> this line just disappears. At the zoom level the line >> disappears I did the following: >> >> 1. Zooming more into the chart the line will not appear again. >> 2. Zooming out twice (with mouse wheel) the line will reappear. >> >> How to reproduce the bug: >> >> I basically used the LinePlot example for vtkChartXY (see >> http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). >> >> I added one line to the example code marked with *** (line >> was added after line 54 in example code). >> I added the full code to the end of the Email. >> --------------- >> >> // Add multiple line plots, setting the colors etc >> vtkSmartPointer chart= >> vtkSmartPointer::New(); *** >> chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >> view->GetScene()->AddItem(chart); >> >> --------------- >> >> Compile and execute the program. See my bug description above >> to visualize the bug. >> >> >> Please contact me if you need more information. I can also >> provide images of the bug if needed. >> I hope the bug can be fixed. >> >> Hagen. >> >> >> >> >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> #include >> >> int main(int, char *[]) >> { >> // Create a table with some points in it >> vtkSmartPointer table = >> vtkSmartPointer::New(); >> >> vtkSmartPointer arrX = >> vtkSmartPointer::New(); >> arrX->SetName("X Axis"); >> table->AddColumn(arrX); >> >> vtkSmartPointer arrC = >> vtkSmartPointer::New(); >> arrC->SetName("Cosine"); >> table->AddColumn(arrC); >> >> vtkSmartPointer arrS = >> vtkSmartPointer::New(); >> arrS->SetName("Sine"); >> table->AddColumn(arrS); >> >> // Fill in the table with some example values >> int numPoints = 69; >> float inc = 7.5 / (numPoints-1); >> table->SetNumberOfRows(numPoints); >> for (int i = 0; i < numPoints; ++i) >> { >> table->SetValue(i, 0, i * inc); >> table->SetValue(i, 1, cos(i * inc)); >> table->SetValue(i, 2, sin(i * inc)); >> } >> >> // Set up the view >> vtkSmartPointer view = >> vtkSmartPointer::New(); >> view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); >> >> // Add multiple line plots, setting the colors etc >> vtkSmartPointer chart = >> vtkSmartPointer::New(); >> chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >> view->GetScene()->AddItem(chart); >> vtkPlot *line = chart->AddPlot(vtkChart::LINE); >> #if VTK_MAJOR_VERSION <= 5 >> line->SetInput(table, 0, 1); >> #else >> line->SetInputData(table, 0, 1); >> #endif >> line->SetColor(0, 255, 0, 255); >> line->SetWidth(1.0); >> line = chart->AddPlot(vtkChart::LINE); >> #if VTK_MAJOR_VERSION <= 5 >> line->SetInput(table, 0, 2); >> #else >> line->SetInputData(table, 0, 2); >> #endif >> line->SetColor(255, 0, 0, 255); >> line->SetWidth(5.0); >> >> // For dotted line, the line type can be from 2 to 5 for >> different dash/dot >> // patterns (see enum in vtkPen containing DASH_LINE, >> value 2): >> #ifndef WIN32 >> line->GetPen()->SetLineType(vtkPen::DASH_LINE); >> #endif >> // (ifdef-ed out on Windows because DASH_LINE does not >> work on Windows >> // machines with built-in Intel HD graphics card...) >> >> //view->GetRenderWindow()->SetMultiSamples(0); >> >> // Start interactor >> view->GetInteractor()->Initialize(); >> view->GetInteractor()->Start(); >> >> return EXIT_SUCCESS; >> } >> >> >> _______________________________________________ >> 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 >> >> >> >> >> -- >> ___________________________________________ >> Andrew J. P. Maclean >> >> ___________________________________________ > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mathieu.westphal at kitware.com Tue Jul 26 05:56:15 2016 From: mathieu.westphal at kitware.com (Mathieu Westphal) Date: Tue, 26 Jul 2016 11:56:15 +0200 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis In-Reply-To: <85b2f3dc-f769-79ae-3af2-d8ab587cc7da@googlemail.com> References: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> <85b2f3dc-f769-79ae-3af2-d8ab587cc7da@googlemail.com> Message-ID: Hello I indeed reproduce the problem, I will submit a bug report and ping Marcus to see if he knows what is going on. http://www.paraview.org/Bug/view.php?id=16797 Regards, Mathieu Westphal On Tue, Jul 26, 2016 at 11:19 AM, Hagen M?lle wrote: > Hi Andrew, Hi Mathieu, > > I can not see the line after it disappears. This might be a special > feature of the open source radeon graphics driver under Linux. > > Just for completeness one can easily see the line thickness fluctuation by > zooming into the scene just right before the line disappears. At this zoom > level just drag the scene using the left mouse button and move the mouse > around. > > In fact it would be very nice if this bug could be fixed at least when > logarithmic x axis is enabled. Because when logarithmic x axis is enabled > the error is visible after zooming in only a few steps. > > Hagen > > > On 26.07.2016 10:57, Andrew Maclean wrote: > > Hi Hagen, Mathieu, > > I agree with you, if I zoom in when the logarithmic axis is disabled, the > thick line vanishes, however if you look carefully there is a very faint > thin line. It may be (2) as you outline below. > > Regards > Andrew > > Andrew Maclean > > On 26 Jul 2016 5:02 pm, "Hagen M?lle" wrote: > >> Hi Andrew, Hi Mathieu, >> >> That's exactly the behaviour I see. By the way I am using ATI/AMD video >> cards in my Windows/Linux machine. >> >> !!!! UPDATE: The error also occurs when logarithmic x axis is not >> enabled. But you have to zoom in a lot more. See attached pictures. I am >> sorry for not realizing this earlier. >> >> >> I am not that familiar with OpenGL. But I can imagine two things: >> >> (1) Line is clipped >> (2) If shaders are used to draw the line thickness it might be a floating >> point precision problem. Actually this could be the problem because the >> line thickness starts fluctuating at some point of zoom level before it >> finally disappears. See attached picture. The line thickness is about the >> half the original line thickness. >> >> >> I hope this helps to fix the bug. One might also contact Marcus D. >> Hanwell. He is the developer who mainly implemented vtkCharts. I am not >> sure if he is still responsible. >> >> Hagen >> >> >> On 26.07.2016 01:52, Andrew Maclean wrote: >> >> Hi Hagen, >> I just tested your example and I get similar behaviou using an NVIDIA >> Video card, OpenGL2, VS2015 and the VTK Master. >> >> I can zoom in and, at some point the line disappears, but zooming out >> immediately restores the line. >> >> It seeme to me that once a certain zoom level is reached the line is not >> drawn, however zooming out restores it. >> >> Attached are two pngs showing the effect.When the line disappears it >> should still be visible as the scales on the axes have only changed >> slightly. >> >> Andrew >> >> ---------- Forwarded message ---------- >>> From: "Hagen M?lle" >>> To: vtk-developers at vtk.org >>> Cc: >>> Date: Mon, 25 Jul 2016 10:14:21 +0200 >>> Subject: [vtk-developers] vtkChartXY lines not visible when using >>> logarithmic x axis >>> Hi, >>> >>> Due to spam reasons the bug tracker's user login is disabled right now. >>> One should send bug reports to the vtk's user mailing list, which I did >>> twice. My bug report was totally ignored. That's why I send it here again. >>> Hopefully someone can finally create a ticket in the bug tracker. Thanks. >>> >>> I want to report a bug for vtkChartXY. So I already send this Email two >>> days ago to the vtkUsers mailing list without receiving any answer. The bug >>> is not visible in the vtk bug tracker too. Thus I send the Email again. >>> >>> Bug environment: >>> a) I am using VTK 6.2. I checked my test code on machines running >>> Windows 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 (gcc). On >>> all these systems the bug is visible. >>> b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu 16.04 >>> machine and compiled VTK 7.0 using OpenGL2 interface). >>> >>> Bug description: >>> >>> I am using vtkChartXY to display one line. I also enabled logarithmic x >>> axis (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing more >>> special. In the compiled application I start zooming into the chart using >>> the mouse wheel, making sure I always have a line displayed in the view. At >>> some zoom level this line just disappears. At the zoom level the line >>> disappears I did the following: >>> >>> 1. Zooming more into the chart the line will not appear again. >>> 2. Zooming out twice (with mouse wheel) the line will reappear. >>> >>> How to reproduce the bug: >>> >>> I basically used the LinePlot example for vtkChartXY (see >>> http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). >>> >>> I added one line to the example code marked with *** (line was added >>> after line 54 in example code). >>> I added the full code to the end of the Email. >>> --------------- >>> >>> // Add multiple line plots, setting the colors etc >>> vtkSmartPointer chart= vtkSmartPointer>> >::New(); *** chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >>> view->GetScene()->AddItem(chart); >>> >>> --------------- >>> >>> Compile and execute the program. See my bug description above to >>> visualize the bug. >>> >>> >>> Please contact me if you need more information. I can also provide >>> images of the bug if needed. >>> I hope the bug can be fixed. >>> >>> Hagen. >>> >>> >>> >>> >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> #include >>> >>> int main(int, char *[]) >>> { >>> // Create a table with some points in it >>> vtkSmartPointer table = >>> vtkSmartPointer::New(); >>> >>> vtkSmartPointer arrX = >>> vtkSmartPointer::New(); >>> arrX->SetName("X Axis"); >>> table->AddColumn(arrX); >>> >>> vtkSmartPointer arrC = >>> vtkSmartPointer::New(); >>> arrC->SetName("Cosine"); >>> table->AddColumn(arrC); >>> >>> vtkSmartPointer arrS = >>> vtkSmartPointer::New(); >>> arrS->SetName("Sine"); >>> table->AddColumn(arrS); >>> >>> // Fill in the table with some example values >>> int numPoints = 69; >>> float inc = 7.5 / (numPoints-1); >>> table->SetNumberOfRows(numPoints); >>> for (int i = 0; i < numPoints; ++i) >>> { >>> table->SetValue(i, 0, i * inc); >>> table->SetValue(i, 1, cos(i * inc)); >>> table->SetValue(i, 2, sin(i * inc)); >>> } >>> >>> // Set up the view >>> vtkSmartPointer view = >>> vtkSmartPointer::New(); >>> view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); >>> >>> // Add multiple line plots, setting the colors etc >>> vtkSmartPointer chart = >>> vtkSmartPointer::New(); >>> chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >>> view->GetScene()->AddItem(chart); >>> vtkPlot *line = chart->AddPlot(vtkChart::LINE); >>> #if VTK_MAJOR_VERSION <= 5 >>> line->SetInput(table, 0, 1); >>> #else >>> line->SetInputData(table, 0, 1); >>> #endif >>> line->SetColor(0, 255, 0, 255); >>> line->SetWidth(1.0); >>> line = chart->AddPlot(vtkChart::LINE); >>> #if VTK_MAJOR_VERSION <= 5 >>> line->SetInput(table, 0, 2); >>> #else >>> line->SetInputData(table, 0, 2); >>> #endif >>> line->SetColor(255, 0, 0, 255); >>> line->SetWidth(5.0); >>> >>> // For dotted line, the line type can be from 2 to 5 for different >>> dash/dot >>> // patterns (see enum in vtkPen containing DASH_LINE, value 2): >>> #ifndef WIN32 >>> line->GetPen()->SetLineType(vtkPen::DASH_LINE); >>> #endif >>> // (ifdef-ed out on Windows because DASH_LINE does not work on >>> Windows >>> // machines with built-in Intel HD graphics card...) >>> >>> //view->GetRenderWindow()->SetMultiSamples(0); >>> >>> // Start interactor >>> view->GetInteractor()->Initialize(); >>> view->GetInteractor()->Start(); >>> >>> return EXIT_SUCCESS; >>> } >>> >>> >>> _______________________________________________ >>> 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 >>> >> >> >> >> -- >> ___________________________________________ >> Andrew J. P. Maclean >> >> ___________________________________________ >> >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Tue Jul 26 07:43:44 2016 From: will.schroeder at kitware.com (Will Schroeder) Date: Tue, 26 Jul 2016 07:43:44 -0400 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: <5aa0228f-05cf-4813-9428-dc288df8b465@email.android.com> References: <5aa0228f-05cf-4813-9428-dc288df8b465@email.android.com> Message-ID: Gerald- I knocked out a vtkPointConnectivityFilter and it's going through the VTK merge process. Not that this will help in your case but it's an interesting filter to have around :-) I think the solution should be something like: remove small surface areas > which are only minimal connected to the main Cloud/Mesh. In principal I agree although there are nasty exceptions. E.g. there are going to be triangles at the corners of a perfectly tessellated plane that are small, and are connected to only one other triangle (minimally connected). So I think what you are really looking for are "isthmus" which are features narrowly connected to larger regions, where the feature size is bigger than the connection "cross section" if that makes any sense. To analyze the connection you'd have to do some sort of graph analysis, effectively performing a graph cut. Size would simply use triangle area. Best, W On Fri, Jul 22, 2016 at 4:08 PM, Lodron, Gerald wrote: > I an not sure what you mean with manifold but the noise/the structure to > remove is definitely only on the boundary. > > I only want to remove those structures, i don't want to add any triangles. > Those structures are caused from wrong Stereo Matches from our Stereo > Matches above edges into Infinity distances. > > I think the solution should be something like: remove small surface areas > which are only minimal connected to the main Cloud/Mesh. > > However I am currently not sure what mathematically "minimal connected" > means by Definition.... > > Am 22.07.2016 3:42 nachm. schrieb Will Schroeder < > will.schroeder at kitware.com>: > Gerald- > > Away from the boundaries, is the surface manifold? (i.e., is each edge is > used twice?). > > Also, can we add triangles as well as delete triangles? Sort of like hole > filling but to fill in boundary "gaps". > > It seems to me that taking a cue from the imaging computing people we can > use some sort of erosion / dilation to clean up noise on the boundary. I'm > going to mull a little more... > > Best, > W > > On Fri, Jul 22, 2016 at 9:02 AM, Lodron, Gerald > wrote: > > Thats the surface: > > [cid:image001.png at 01D1E42A.02D5F230] > > Zoom on structures to remove: > > [cid:image002.png at 01D1E42A.02D5F230] > > > > > > Von: Will Schroeder [mailto:will.schroeder at kitware.com will.schroeder at kitware.com>] > Gesendet: Freitag, 22. Juli 2016 14:52 > > An: Lodron, Gerald > Cc: VTK Developer (vtk-developers at vtk.org) > Betreff: Re: [vtk-developers] Number of neighbours for each cell > > > > Sure send the image we always love a challenge :-) Obviously if it becomes > more involved we can talk about a support product< > http://www.kitware.com/products/support.html>, etc. > > Best, > > W > > > > On Fri, Jul 22, 2016 at 8:44 AM, Lodron, Gerald > wrote: > > Cool > > > > For my first test that implementation below worked and i found out that > this kind of algorithm does not help me on my problem, arg > > > > Can I send you an image/screenshot of the ?geometric? stuff I want to > filter? May you as an expert has a better idea for solving my problem. > Short text description (but is really hard to explain): > > > > I have a mesh of a surface, scanned with a 3d scanner (triangles, > vtkPolydata as explained). On the border of that mesh I have single ?noisy? > (or fringy) structures which I want to filter. These noise structures are > minimal connected to the main mesh, so I can not use connected component > filter (which I also used apriory to remove noise in the air). > > > > Von: Will Schroeder [mailto:will.schroeder at kitware.com will.schroeder at kitware.com>] > Gesendet: Freitag, 22. Juli 2016 14:37 > > An: Lodron, Gerald > Cc: VTK Developer (vtk-developers at vtk.org) > Betreff: Re: [vtk-developers] Number of neighbours for each cell > > > > Yes this is a great start. Since I am on a performance kick, I would > probably use (under the hood) vtkStaticCellLocator which builds the links > between points and cells much faster (threaded using vtkSMPTools and a > parallel sort). The actual "counting" of the neighbors could then be sped > up using vtkSMPTools as well. It's probably overkill, but I'd like VTK to > knock off the socks of anyone using it, and also to support our HPC friends > who have this humbling habit of dumping really big data on top of VTK :-) > > > > Best, > W > > > > On Fri, Jul 22, 2016 at 8:25 AM, Lodron, Gerald > wrote: > > In my case the definition would be an edge, but yes it would be useful to > define that as parameter?. > > I already found a maybe not beautiful algorithm for vertices, not edges, > may you want to reuse it as initial code: > > vtkSmartPointer neighbours = > vtkSmartPointer::New(); > neighbours ->SetName("NumberOfNeighbours"); > neighbours ->SetNumberOfValues(numberOfCells); > > vtkSmartPointer cellPointIds = > vtkSmartPointer::New(); > > vtkSmartPointer ptids = > vtkSmartPointer::New(); > vtkSmartPointer cellids = > vtkSmartPointer::New(); > > for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) > { > input->GetCellPoints(cellId, ptids); > /* > does not work > input ->GetCellNeighbors(cellId, oPtIds, oCellIds); > neighbours ->SetValue(cellId, static_cast int>(oCellIds->GetNumberOfIds()));*/ > std::set id_set; > > for (vtkIdType ptid = 0; ptid < ptids > ->GetNumberOfIds();ptid++) > { > poInput1->GetPointCells(ptids ->GetId(ptid), cellids); > for (vtkIdType cid = 0; cid < cellids > ->GetNumberOfIds();cid++) > { > vtkIdType id = cellids ->GetId(cid); > if (id != cellId) > { > id_set.insert(cellids->GetId(cid)); > } > } > } > neighbours ->SetValue(cellId, static_cast( > id_set.size())); > } > output->GetCellData()->AddArray(neighbours); > > Von: Will Schroeder [mailto:will.schroeder at kitware.com will.schroeder at kitware.com>] > Gesendet: Freitag, 22. Juli 2016 14:09 > An: Lodron, Gerald > Cc: VTK Developer (vtk-developers at vtk.org) > Betreff: Re: [vtk-developers] Number of neighbours for each cell > > I don't know of such a filter but it would be easy to put together. > Certainly it would be another useful tool in the topological analysis > toolbox. > > Question: what is your definition of a neighbor? Connected at a vertex or > along an edge? I suppose if we were to write such a filter the definition > of a neighbor would be an optional parameter. I also suggest that this be a > general filter so the same analysis could be applied to unstructured grids, > etc. or to polygonal meshes with other than triangle cells. > > Best, > W > > On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald > wrote: > > > Hi > > I have a mesh (vtkPolyData with all triangles) where i want to visualize > the number of neighbouring cells/triangles for each cell/triangle. (I later > may want to threshold/filter it with that number to remove ?noisy? > triangles on border with low number of neighbours) > > Is there a filter for that available? > > Best regards > > > ------------------------------------------------------------------------------------ > Gerald Lodron > > Researcher of Machine Vision Applications Group > DIGITAL - Institute for Information and Communication Technologies > > JOANNEUM RESEARCH Forschungsgesellschaft mbH > Steyrergasse 17, 8010 Graz, AUSTRIA > > phone: +43-316-876-1751 > general fax: +43-316-876-1751 > web: http://www.joanneum.at/digital > e-mail: gerald.lodron at joanneum.at > > > > > _______________________________________________ > 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 > > > > > > -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben.boeckel at kitware.com Tue Jul 26 10:05:45 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Tue, 26 Jul 2016 10:05:45 -0400 Subject: [vtk-developers] Rogue7 vtktiff dashboard build failure In-Reply-To: <20160726011546.2016066190@mail.rogue-research.com> References: <20160726011546.2016066190@mail.rogue-research.com> Message-ID: <20160726140545.GA14126@rotor> On Mon, Jul 25, 2016 at 21:15:46 -0400, Sean McBride wrote: > The build failure is still on the dashboard... > > I guess the unique thing about that submission is that it's the only > one (I believe) that uses CMake's Xcode generator and not the Unix > Makefiles generator. Oops, dropped off my TODO list. I'll submit an MR here soon. --Ben From ben.boeckel at kitware.com Tue Jul 26 10:22:52 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Tue, 26 Jul 2016 10:22:52 -0400 Subject: [vtk-developers] Rogue7 vtktiff dashboard build failure In-Reply-To: <20160726140545.GA14126@rotor> References: <20160726011546.2016066190@mail.rogue-research.com> <20160726140545.GA14126@rotor> Message-ID: <20160726142252.GA10822@rotor.kitware.com> On Tue, Jul 26, 2016 at 10:05:45 -0400, Ben Boeckel wrote: > Oops, dropped off my TODO list. I'll submit an MR here soon. Reproduced the issue: Xcode: -- Performing Test HAVE_LD_VERSION_SCRIPT - Success Ninja: -- Performing Test HAVE_LD_VERSION_SCRIPT - Failed Now to figure out why the difference happens? --Ben From ben.boeckel at kitware.com Tue Jul 26 10:42:35 2016 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Tue, 26 Jul 2016 10:42:35 -0400 Subject: [vtk-developers] Rogue7 vtktiff dashboard build failure In-Reply-To: <20160726142252.GA10822@rotor.kitware.com> References: <20160726011546.2016066190@mail.rogue-research.com> <20160726140545.GA14126@rotor> <20160726142252.GA10822@rotor.kitware.com> Message-ID: <20160726144235.GB14126@rotor> On Tue, Jul 26, 2016 at 10:22:52 -0400, Ben Boeckel wrote: > Now to figure out why the difference happens? tiff MR up: https://gitlab.kitware.com/third-party/tiff/merge_requests/1/commits --Ben From joaolsvieira at gmail.com Tue Jul 26 10:49:15 2016 From: joaolsvieira at gmail.com (=?UTF-8?Q?Jo=C3=A3o_Luis?=) Date: Tue, 26 Jul 2016 10:49:15 -0400 Subject: [vtk-developers] vtkPlaneSource by vtkPlaneWidget Message-ID: Hello vtkdevelopers, I am trying to draw a vtkPlaneSource from a vtkPlaneWidget after some user interactions. I had extracted Point1, Point2, Normal and Center from my Widget as well. However, applying against a vtkPlaneSource the result does not reflect the vtkPlaneWidget's definitions chosen by the user. It means, basically size and orientation of the final plane. Many thanks for any help. following my snapshot. planeWidget = vtkSmartPointer::New(); planeWidget->SetInteractor(mouseClickCallback->renderWindowInteractor); //iren is vtkRenderWindowInteractor planeWidget->SetInputConnection(mydataset); planeWidget->SetPlaceFactor(1.0); planeWidget->SetHandleSize(0.03); planeWidget->SetResolution(10); planeWidget->GetHandleProperty()->SetColor(0, 0.6, 0.1); planeWidget->GetPlaneProperty()->SetColor(0, 0.3, 1); planeWidget->GetPlaneProperty()->SetEdgeColor(0, 0.9, 1); planeWidget->GetPlaneProperty()->SetEdgeVisibility(1); planeWidget->PlaceWidget(); planeWidget->SetRepresentationToSurface(); planeWidget->GetPolyData(planeW); planeWidget->AddObserver(vtkCommand::InteractionEvent, mouseClickCallback); planeWidget->SetNormal(normalVector[0], normalVector[1], normalVector[2]); // normal of my target plane planeWidget->SetCenter(center[0], center[1], center[2]); // origin of my target plane planeWidget->UpdatePlacement(); planeWidget->EnabledOn(); Callback Interactions _plane = vtkPlane::New(); vtkPolyData *polydata = vtkPolyData::New(); vtkSmartPointer extractEdges = vtkSmartPointer::New(); planeWidget->GetPlane(_plane); planeWidget->GetPolyData(polydata); planeWidget->GetPoint1(Point1); planeWidget->GetPoint2(Point2); planeWidget->GetCenter(Center); extractEdges->SetInputData(0, polydata); extractEdges->Update(); vtkSmartPointer newPlane = vtkSmartPointer::New(); newPlane->SetNormal(planeWidget->GetNormal()); newPlane->SetOrigin(planeWidget->GetOrigin()); VtkPlaneSource vtkSmartPointer planeSource = vtkSmartPointer::New(); planeSource->SetPoint1(Point1[0], Point1[1], Point1[2]);//Point1 planeSource->SetPoint2(Point2[0], Point2[1], Point2[2]);//Point2 //planeSource->SetOrigin(newPlane->Origin[0], newPlane->Origin[1], newPlane->Origin[2]);// planeSource->SetCenter(Center[0], Center[1], Center[2]); planeSource->SetNormal(newPlane->GetNormal()[0], newPlane->GetNormal()[1], newPlane->GetNormal()[2]); planeSource->Update(); Anybody here knows whats is wrong with my solution? Once, again many thanks for any help. Luis -------------- next part -------------- An HTML attachment was scrubbed... URL: From luis.vieira at vektore.com Tue Jul 26 10:51:51 2016 From: luis.vieira at vektore.com (Luis Vieira) Date: Tue, 26 Jul 2016 10:51:51 -0400 Subject: [vtk-developers] vtkPlaneSource from a vtkPlaneWidget Message-ID: <0be401d1e74d$3efd70f0$bcf852d0$@vektore.com> Hello vtkdevelopers, I am trying to draw a vtkPlaneSource from a vtkPlaneWidget after some user interactions. I had extracted Point1, Point2, Normal and Center from my Widget as well. However, applying against a vtkPlaneSource the result does not reflect the vtkPlaneWidget's definitions chosen by the user. It means, basically size and orientation of the final plane. Many thanks for any help. following my snapshot. planeWidget = vtkSmartPointer::New(); planeWidget->SetInteractor(mouseClickCallback->renderWindowInteractor); //iren is vtkRenderWindowInteractor planeWidget->SetInputConnection(mydataset); planeWidget->SetPlaceFactor(1.0); planeWidget->SetHandleSize(0.03); planeWidget->SetResolution(10); planeWidget->GetHandleProperty()->SetColor(0, 0.6, 0.1); planeWidget->GetPlaneProperty()->SetColor(0, 0.3, 1); planeWidget->GetPlaneProperty()->SetEdgeColor(0, 0.9, 1); planeWidget->GetPlaneProperty()->SetEdgeVisibility(1); planeWidget->PlaceWidget(); planeWidget->SetRepresentationToSurface(); planeWidget->GetPolyData(planeW); planeWidget->AddObserver(vtkCommand::InteractionEvent, mouseClickCallback); planeWidget->SetNormal(normalVector[0], normalVector[1], normalVector[2]); // normal of my target plane planeWidget->SetCenter(center[0], center[1], center[2]); // origin of my target plane planeWidget->UpdatePlacement(); planeWidget->EnabledOn(); Callback Interactions _plane = vtkPlane::New(); vtkPolyData *polydata = vtkPolyData::New(); vtkSmartPointer extractEdges = vtkSmartPointer::New(); planeWidget->GetPlane(_plane); planeWidget->GetPolyData(polydata); planeWidget->GetPoint1(Point1); planeWidget->GetPoint2(Point2); planeWidget->GetCenter(Center); extractEdges->SetInputData(0, polydata); extractEdges->Update(); vtkSmartPointer newPlane = vtkSmartPointer::New(); newPlane->SetNormal(planeWidget->GetNormal()); newPlane->SetOrigin(planeWidget->GetOrigin()); VtkPlaneSource vtkSmartPointer planeSource = vtkSmartPointer::New(); planeSource->SetPoint1(Point1[0], Point1[1], Point1[2]);//Point1 planeSource->SetPoint2(Point2[0], Point2[1], Point2[2]);//Point2 //planeSource->SetOrigin(newPlane->Origin[0], newPlane->Origin[1], newPlane->Origin[2]);// planeSource->SetCenter(Center[0], Center[1], Center[2]); planeSource->SetNormal(newPlane->GetNormal()[0], newPlane->GetNormal()[1], newPlane->GetNormal()[2]); planeSource->Update(); Anybody here knows what is wrong with my solution? Once, again many thanks for any help. Luis -------------- next part -------------- An HTML attachment was scrubbed... URL: From luis.vieira at vektore.com Tue Jul 26 12:46:47 2016 From: luis.vieira at vektore.com (Luis Vieira) Date: Tue, 26 Jul 2016 12:46:47 -0400 Subject: [vtk-developers] vtk-developers Digest, Vol 147, Issue 36 In-Reply-To: References: Message-ID: <281e01d1e75d$4d1e9640$e75bc2c0$@vektore.com> Worked, Thank you very much Mike For vtkPlaneSource, you have to set the Origin (you have this commented out), Point1, and Point2. You shouldn't have to set the PlaneSource's normal unless you want to change it's orientation as defined by the origin and the two points. On Tue, Jul 26, 2016 at 9:52 AM, Luis Vieira wrote: > Hello vtkusers, > > > > I am trying to draw a vtkPlaneSource from a vtkPlaneWidget after some user > interactions. I had extracted Point1, Point2, Normal and Center from my > Widget as well. However, applying against a vtkPlaneSource the result does > not reflect the vtkPlaneWidget's definitions chosen by the user. It means, > basically size and orientation of the final plane. Many thanks for any help. > > > > following my snapshot. > > > > planeWidget = vtkSmartPointer::New(); > > planeWidget->SetInteractor(mouseClickCallback->renderWindowInteractor); > //iren is vtkRenderWindowInteractor > > planeWidget->SetInputConnection(mydataset); > > planeWidget->SetPlaceFactor(1.0); > > planeWidget->SetHandleSize(0.03); > > planeWidget->SetResolution(10); > > planeWidget->GetHandleProperty()->SetColor(0, 0.6, 0.1); > > planeWidget->GetPlaneProperty()->SetColor(0, 0.3, 1); > > planeWidget->GetPlaneProperty()->SetEdgeColor(0, 0.9, 1); > > planeWidget->GetPlaneProperty()->SetEdgeVisibility(1); > > planeWidget->PlaceWidget(); > > planeWidget->SetRepresentationToSurface(); > > planeWidget->GetPolyData(planeW); > > planeWidget->AddObserver(vtkCommand::InteractionEvent, mouseClickCallback); > > planeWidget->SetNormal(normalVector[0], normalVector[1], normalVector[2]); > // normal of my target plane > > planeWidget->SetCenter(center[0], center[1], center[2]); // origin of my > target plane > > planeWidget->UpdatePlacement(); > > planeWidget->EnabledOn(); > > > > Callback Interactions > > > > _plane = vtkPlane::New(); > > vtkPolyData *polydata = vtkPolyData::New(); > > vtkSmartPointer extractEdges = > > vtkSmartPointer::New(); > > planeWidget->GetPlane(_plane); > > planeWidget->GetPolyData(polydata); > > planeWidget->GetPoint1(Point1); > > planeWidget->GetPoint2(Point2); > > planeWidget->GetCenter(Center); > > extractEdges->SetInputData(0, polydata); > > extractEdges->Update(); > > > > vtkSmartPointer newPlane = vtkSmartPointer::New(); > > newPlane->SetNormal(planeWidget->GetNormal()); > > newPlane->SetOrigin(planeWidget->GetOrigin()); > > > > VtkPlaneSource > > > > vtkSmartPointer planeSource = > > vtkSmartPointer::New(); > > planeSource->SetPoint1(Point1[0], Point1[1], Point1[2]);//Point1 > > planeSource->SetPoint2(Point2[0], Point2[1], Point2[2]);//Point2 > > //planeSource->SetOrigin(newPlane->Origin[0], newPlane->Origin[1], > newPlane->Origin[2]);// > > planeSource->SetCenter(Center[0], Center[1], Center[2]); > > planeSource->SetNormal(newPlane->GetNormal()[0], newPlane->GetNormal()[1], > newPlane->GetNormal()[2]); > > planeSource->Update(); > > > > > > Anybody here knows whats is wrong with my solution? Once, again many > thanks for any help. > > > > > > *Luis * -----Original Message----- From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of vtk-developers-request at vtk.org Sent: July 26, 2016 10:59 AM To: vtk-developers at vtk.org Subject: vtk-developers Digest, Vol 147, Issue 36 Send vtk-developers mailing list submissions to vtk-developers at vtk.org To subscribe or unsubscribe via the World Wide Web, visit http://public.kitware.com/mailman/listinfo/vtk-developers or, via email, send a message with subject or body 'help' to vtk-developers-request at vtk.org You can reach the person managing the list at vtk-developers-owner at vtk.org When replying, please edit your Subject line so it is more specific than "Re: Contents of vtk-developers digest..." Today's Topics: 1. Re: Rogue7 vtktiff dashboard build failure (Ben Boeckel) 2. vtkPlaneSource by vtkPlaneWidget (Jo?o Luis) 3. vtkPlaneSource from a vtkPlaneWidget (Luis Vieira) ---------------------------------------------------------------------- Message: 1 Date: Tue, 26 Jul 2016 10:42:35 -0400 From: Ben Boeckel To: Sean McBride Cc: VTK Developers , David Gobbi Subject: Re: [vtk-developers] Rogue7 vtktiff dashboard build failure Message-ID: <20160726144235.GB14126 at rotor> Content-Type: text/plain; charset=utf-8 On Tue, Jul 26, 2016 at 10:22:52 -0400, Ben Boeckel wrote: > Now to figure out why the difference happens? tiff MR up: https://gitlab.kitware.com/third-party/tiff/merge_requests/1/commits --Ben ------------------------------ Message: 2 Date: Tue, 26 Jul 2016 10:49:15 -0400 From: Jo?o Luis To: vtk-developers at vtk.org Subject: [vtk-developers] vtkPlaneSource by vtkPlaneWidget Message-ID: Content-Type: text/plain; charset="utf-8" Hello vtkdevelopers, I am trying to draw a vtkPlaneSource from a vtkPlaneWidget after some user interactions. I had extracted Point1, Point2, Normal and Center from my Widget as well. However, applying against a vtkPlaneSource the result does not reflect the vtkPlaneWidget's definitions chosen by the user. It means, basically size and orientation of the final plane. Many thanks for any help. following my snapshot. planeWidget = vtkSmartPointer::New(); planeWidget->SetInteractor(mouseClickCallback->renderWindowInteractor); //iren is vtkRenderWindowInteractor planeWidget->SetInputConnection(mydataset); planeWidget->SetPlaceFactor(1.0); planeWidget->SetHandleSize(0.03); planeWidget->SetResolution(10); planeWidget->GetHandleProperty()->SetColor(0, 0.6, 0.1); planeWidget->GetPlaneProperty()->SetColor(0, 0.3, 1); planeWidget->GetPlaneProperty()->SetEdgeColor(0, 0.9, 1); planeWidget->GetPlaneProperty()->SetEdgeVisibility(1); planeWidget->PlaceWidget(); planeWidget->SetRepresentationToSurface(); planeWidget->GetPolyData(planeW); planeWidget->AddObserver(vtkCommand::InteractionEvent, planeWidget->mouseClickCallback); SetNormal(normalVector[0], planeWidget->normalVector[1], normalVector[2]); // normal of my target plane planeWidget->SetCenter(center[0], center[1], center[2]); // origin of my target plane planeWidget->UpdatePlacement(); planeWidget->EnabledOn(); Callback Interactions _plane = vtkPlane::New(); vtkPolyData *polydata = vtkPolyData::New(); vtkSmartPointer extractEdges = vtkSmartPointer::New(); planeWidget->GetPlane(_plane); planeWidget->GetPolyData(polydata); planeWidget->GetPoint1(Point1); planeWidget->GetPoint2(Point2); planeWidget->GetCenter(Center); extractEdges->SetInputData(0, polydata); Update(); vtkSmartPointer newPlane = vtkSmartPointer::New(); newPlane->SetNormal(planeWidget->GetNormal()); newPlane->SetOrigin(planeWidget->GetOrigin()); VtkPlaneSource vtkSmartPointer planeSource = vtkSmartPointer::New(); planeSource->SetPoint1(Point1[0], Point1[1], Point1[2]);//Point1 planeSource->SetPoint2(Point2[0], Point2[1], Point2[2]);//Point2 //planeSource->SetOrigin(newPlane->Origin[0], newPlane->Origin[1], newPlane->Origin[2]);// planeSource->SetCenter(Center[0], Center[1], Center[2]); planeSource->SetNormal(newPlane->GetNormal()[0], planeSource->newPlane->GetNormal()[1], newPlane->GetNormal()[2]); planeSource->Update(); Anybody here knows whats is wrong with my solution? Once, again many thanks for any help. Luis -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 3 Date: Tue, 26 Jul 2016 10:51:51 -0400 From: "Luis Vieira" To: Subject: [vtk-developers] vtkPlaneSource from a vtkPlaneWidget Message-ID: <0be401d1e74d$3efd70f0$bcf852d0$@vektore.com> Content-Type: text/plain; charset="us-ascii" Hello vtkdevelopers, I am trying to draw a vtkPlaneSource from a vtkPlaneWidget after some user interactions. I had extracted Point1, Point2, Normal and Center from my Widget as well. However, applying against a vtkPlaneSource the result does not reflect the vtkPlaneWidget's definitions chosen by the user. It means, basically size and orientation of the final plane. Many thanks for any help. following my snapshot. planeWidget = vtkSmartPointer::New(); planeWidget->SetInteractor(mouseClickCallback->renderWindowInteractor); //iren is vtkRenderWindowInteractor planeWidget->SetInputConnection(mydataset); planeWidget->SetPlaceFactor(1.0); planeWidget->SetHandleSize(0.03); planeWidget->SetResolution(10); planeWidget->GetHandleProperty()->SetColor(0, 0.6, 0.1); planeWidget->GetPlaneProperty()->SetColor(0, 0.3, 1); planeWidget->GetPlaneProperty()->SetEdgeColor(0, 0.9, 1); planeWidget->GetPlaneProperty()->SetEdgeVisibility(1); planeWidget->PlaceWidget(); planeWidget->SetRepresentationToSurface(); planeWidget->GetPolyData(planeW); planeWidget->AddObserver(vtkCommand::InteractionEvent, planeWidget->mouseClickCallback); planeWidget->SetNormal(normalVector[0], normalVector[1], planeWidget->normalVector[2]); // normal of my target plane planeWidget->SetCenter(center[0], center[1], center[2]); // origin of my target plane planeWidget->UpdatePlacement(); planeWidget->EnabledOn(); Callback Interactions _plane = vtkPlane::New(); vtkPolyData *polydata = vtkPolyData::New(); vtkSmartPointer extractEdges = vtkSmartPointer::New(); planeWidget->GetPlane(_plane); planeWidget->GetPolyData(polydata); planeWidget->GetPoint1(Point1); planeWidget->GetPoint2(Point2); planeWidget->GetCenter(Center); extractEdges->SetInputData(0, polydata); extractEdges->Update(); vtkSmartPointer newPlane = vtkSmartPointer::New(); newPlane->SetNormal(planeWidget->GetNormal()); newPlane->SetOrigin(planeWidget->GetOrigin()); VtkPlaneSource vtkSmartPointer planeSource = vtkSmartPointer::New(); planeSource->SetPoint1(Point1[0], Point1[1], Point1[2]);//Point1 planeSource->SetPoint2(Point2[0], Point2[1], Point2[2]);//Point2 //planeSource->SetOrigin(newPlane->Origin[0], newPlane->Origin[1], newPlane->Origin[2]);// planeSource->SetCenter(Center[0], Center[1], Center[2]); planeSource->SetNormal(newPlane->GetNormal()[0], planeSource->newPlane->GetNormal()[1], newPlane->GetNormal()[2]); planeSource->Update(); Anybody here knows what is wrong with my solution? Once, again many thanks for any help. Luis -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Subject: Digest Footer _______________________________________________ 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 ------------------------------ End of vtk-developers Digest, Vol 147, Issue 36 *********************************************** From Gerald.Lodron at joanneum.at Wed Jul 27 00:45:29 2016 From: Gerald.Lodron at joanneum.at (Lodron, Gerald) Date: Wed, 27 Jul 2016 04:45:29 +0000 Subject: [vtk-developers] Number of neighbours for each cell In-Reply-To: References: <5aa0228f-05cf-4813-9428-dc288df8b465@email.android.com> Message-ID: <339a84e0b3e64e7da74d365985e43b74@RZJMBX2.jr1.local> Hi Thanks, i will check it out as a paraview filter if it gets released sometime?. Do you have any idea how to detect such ?isthmus?es with graph cuts? I don?t have any clue, but I am not a graph expert?. But in the meantime I got another idea solving that problem over computation of the 3d mesh in a higher resolution at edge areas, but that?s not really a vtk solving task. Thanks for your suggestions, it was a great pleasure for me to discuss with you ;-) Best regards Von: Will Schroeder [mailto:will.schroeder at kitware.com] Gesendet: Dienstag, 26. Juli 2016 13:44 An: Lodron, Gerald Cc: vtk-developers Betreff: Re: [vtk-developers] Number of neighbours for each cell Gerald- I knocked out a vtkPointConnectivityFilter and it's going through the VTK merge process. Not that this will help in your case but it's an interesting filter to have around :-) I think the solution should be something like: remove small surface areas which are only minimal connected to the main Cloud/Mesh. In principal I agree although there are nasty exceptions. E.g. there are going to be triangles at the corners of a perfectly tessellated plane that are small, and are connected to only one other triangle (minimally connected). So I think what you are really looking for are "isthmus" which are features narrowly connected to larger regions, where the feature size is bigger than the connection "cross section" if that makes any sense. To analyze the connection you'd have to do some sort of graph analysis, effectively performing a graph cut. Size would simply use triangle area. Best, W On Fri, Jul 22, 2016 at 4:08 PM, Lodron, Gerald > wrote: I an not sure what you mean with manifold but the noise/the structure to remove is definitely only on the boundary. I only want to remove those structures, i don't want to add any triangles. Those structures are caused from wrong Stereo Matches from our Stereo Matches above edges into Infinity distances. I think the solution should be something like: remove small surface areas which are only minimal connected to the main Cloud/Mesh. However I am currently not sure what mathematically "minimal connected" means by Definition.... Am 22.07.2016 3:42 nachm. schrieb Will Schroeder >: Gerald- Away from the boundaries, is the surface manifold? (i.e., is each edge is used twice?). Also, can we add triangles as well as delete triangles? Sort of like hole filling but to fill in boundary "gaps". It seems to me that taking a cue from the imaging computing people we can use some sort of erosion / dilation to clean up noise on the boundary. I'm going to mull a little more... Best, W On Fri, Jul 22, 2016 at 9:02 AM, Lodron, Gerald >> wrote: Thats the surface: [cid:image001.png at 01D1E42A.02D5F230] Zoom on structures to remove: [cid:image002.png at 01D1E42A.02D5F230] Von: Will Schroeder [mailto:will.schroeder at kitware.com>] Gesendet: Freitag, 22. Juli 2016 14:52 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org>) Betreff: Re: [vtk-developers] Number of neighbours for each cell Sure send the image we always love a challenge :-) Obviously if it becomes more involved we can talk about a support product, etc. Best, W On Fri, Jul 22, 2016 at 8:44 AM, Lodron, Gerald >> wrote: Cool For my first test that implementation below worked and i found out that this kind of algorithm does not help me on my problem, arg Can I send you an image/screenshot of the ?geometric? stuff I want to filter? May you as an expert has a better idea for solving my problem. Short text description (but is really hard to explain): I have a mesh of a surface, scanned with a 3d scanner (triangles, vtkPolydata as explained). On the border of that mesh I have single ?noisy? (or fringy) structures which I want to filter. These noise structures are minimal connected to the main mesh, so I can not use connected component filter (which I also used apriory to remove noise in the air). Von: Will Schroeder [mailto:will.schroeder at kitware.com>] Gesendet: Freitag, 22. Juli 2016 14:37 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org>) Betreff: Re: [vtk-developers] Number of neighbours for each cell Yes this is a great start. Since I am on a performance kick, I would probably use (under the hood) vtkStaticCellLocator which builds the links between points and cells much faster (threaded using vtkSMPTools and a parallel sort). The actual "counting" of the neighbors could then be sped up using vtkSMPTools as well. It's probably overkill, but I'd like VTK to knock off the socks of anyone using it, and also to support our HPC friends who have this humbling habit of dumping really big data on top of VTK :-) Best, W On Fri, Jul 22, 2016 at 8:25 AM, Lodron, Gerald >> wrote: In my case the definition would be an edge, but yes it would be useful to define that as parameter?. I already found a maybe not beautiful algorithm for vertices, not edges, may you want to reuse it as initial code: vtkSmartPointer neighbours = vtkSmartPointer::New(); neighbours ->SetName("NumberOfNeighbours"); neighbours ->SetNumberOfValues(numberOfCells); vtkSmartPointer cellPointIds = vtkSmartPointer::New(); vtkSmartPointer ptids = vtkSmartPointer::New(); vtkSmartPointer cellids = vtkSmartPointer::New(); for (vtkIdType cellId = 0; cellId < numberOfCells; ++cellId) { input->GetCellPoints(cellId, ptids); /* does not work input ->GetCellNeighbors(cellId, oPtIds, oCellIds); neighbours ->SetValue(cellId, static_cast(oCellIds->GetNumberOfIds()));*/ std::set id_set; for (vtkIdType ptid = 0; ptid < ptids ->GetNumberOfIds();ptid++) { poInput1->GetPointCells(ptids ->GetId(ptid), cellids); for (vtkIdType cid = 0; cid < cellids ->GetNumberOfIds();cid++) { vtkIdType id = cellids ->GetId(cid); if (id != cellId) { id_set.insert(cellids->GetId(cid)); } } } neighbours ->SetValue(cellId, static_cast( id_set.size())); } output->GetCellData()->AddArray(neighbours); Von: Will Schroeder [mailto:will.schroeder at kitware.com>] Gesendet: Freitag, 22. Juli 2016 14:09 An: Lodron, Gerald Cc: VTK Developer (vtk-developers at vtk.org>) Betreff: Re: [vtk-developers] Number of neighbours for each cell I don't know of such a filter but it would be easy to put together. Certainly it would be another useful tool in the topological analysis toolbox. Question: what is your definition of a neighbor? Connected at a vertex or along an edge? I suppose if we were to write such a filter the definition of a neighbor would be an optional parameter. I also suggest that this be a general filter so the same analysis could be applied to unstructured grids, etc. or to polygonal meshes with other than triangle cells. Best, W On Fri, Jul 22, 2016 at 6:51 AM, Lodron, Gerald >> wrote: Hi I have a mesh (vtkPolyData with all triangles) where i want to visualize the number of neighbouring cells/triangles for each cell/triangle. (I later may want to threshold/filter it with that number to remove ?noisy? triangles on border with low number of neighbours) Is there a filter for that available? Best regards ------------------------------------------------------------------------------------ Gerald Lodron Researcher of Machine Vision Applications Group DIGITAL - Institute for Information and Communication Technologies JOANNEUM RESEARCH Forschungsgesellschaft mbH Steyrergasse 17, 8010 Graz, AUSTRIA phone: +43-316-876-1751 general fax: +43-316-876-1751 web: http://www.joanneum.at/digital e-mail: gerald.lodron at joanneum.at> _______________________________________________ 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 -- William J. Schroeder, PhD Kitware, Inc. - Building the World's Technical Computing Software 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmattern at arcor.de Wed Jul 27 08:44:02 2016 From: pmattern at arcor.de (Peter Mattern) Date: Wed, 27 Jul 2016 12:44:02 +0000 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? Message-ID: <9b3e74b62b23af345a7da83976adbab2@arcor.de> Hello. Issue #14126 was closed as fixed but the problem which was addressed persists. Similarly, #15549 was tagged "expired" which suggests attempts to fix will no longer be made. But IMO both problems are definitely worth having another look at. So I wonder whether something can be done to have those two issues treated as regular ones again. Regards, Peter Mattern From dan.lipsa at kitware.com Wed Jul 27 09:20:09 2016 From: dan.lipsa at kitware.com (Dan Lipsa) Date: Wed, 27 Jul 2016 09:20:09 -0400 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: <9b3e74b62b23af345a7da83976adbab2@arcor.de> References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> Message-ID: Hi Peter, In VTK we include many external library which are used by default by the build system. The symbols in these libraries are mangled, so no problem is created if a system uses both VTK and a system library included by VTK. We do offer the option for VTK to use a system library instead of the included one. However, the version of that external library has to be close to the version included in VTK so that the API is identical. Sometimes we try to support several versions - especially if this support is driven by paying customers. For #14126, can you detail what is the problem that persists? Regarding 'expired issues' you should be able to edit an expired issue if you think that bug is still valid. If you are not registered in the bug tracker you can ask on the mailing list for someone to do so. Hope it helps, Dan On Wed, Jul 27, 2016 at 8:44 AM, Peter Mattern wrote: > Hello. > > Issue #14126 was closed as fixed but the problem which was addressed > persists. Similarly, #15549 was tagged "expired" which suggests attempts to > fix will no longer be made. > > But IMO both problems are definitely worth having another look at. So I > wonder whether something can be done to have those two issues treated as > regular ones again. > > Regards, > > Peter Mattern > _______________________________________________ > 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 pmattern at arcor.de Wed Jul 27 10:20:18 2016 From: pmattern at arcor.de (Peter Mattern) Date: Wed, 27 Jul 2016 14:20:18 +0000 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> Message-ID: <68c3c4b4ed31aac7aa18339d86843809@arcor.de> Hi Dan. Thanks for your quick response! I was aware that you're using "in-tree" copies of a lot of dependencies. And note I'm assuming you're doing this for a reason so I certainly don't criticise. It's just that as an old Linuxer I think using system libraries wherever feasible is basically the better solution... ad 14126 The problem is that CMake just doesn't find the system-wide PROJ.4 instance and hence fails > [...] > -- Found TIFF: /usr/lib64/libtiff.so (found version "4.0.6") > CMake Error at CMake/vtkModuleMacros.cmake:864 (message): > VTK_USE_SYSTEM_LIBPROJ4 is ON but LIBPROJ4 is not found! > Call Stack (most recent call first): > ThirdParty/libproj4/CMakeLists.txt:1 (vtk_module_third_party) > > > -- Configuring incomplete, errors occurred! The findings are the very same compiling VTK commit ddcd09c on Arch Linux, Debian testing (stretch) and Fedora 24 using either PROJ.4 4.9.2 as provided by those distributions or PROJ.4 commit 06a708d compiled on the respective system. The PROJ.4 commit was compiled either by running autogen, configure [--prefix=/usr], make and make install or cmake -DCMAKE_INSTALL_PREFIX=/usr, make and make install as compiling yields slightly different results right now, see https://github.com/OSGeo/proj.4/issues/378, but this didn't make a difference regarding VTK. ad 15549 I've added a comment describing the current situation but don't seem to be able to change field "Status", maybe as I didn't open the issue? From marcus.hanwell at kitware.com Wed Jul 27 11:47:06 2016 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Wed, 27 Jul 2016 11:47:06 -0400 Subject: [vtk-developers] vtkChartXY lines not visible when using logarithmic x axis In-Reply-To: References: <389b6adf-b84a-9f7c-6c37-f0387c2001e5@googlemail.com> <85b2f3dc-f769-79ae-3af2-d8ab587cc7da@googlemail.com> Message-ID: I can take a look when I return, I am currently out on travel and have limited connectivity. On Tue, Jul 26, 2016 at 5:56 AM, Mathieu Westphal wrote: > Hello > > I indeed reproduce the problem, I will submit a bug report and ping Marcus > to see if he knows what is going on. > http://www.paraview.org/Bug/view.php?id=16797 > > Regards, > > Mathieu Westphal > > On Tue, Jul 26, 2016 at 11:19 AM, Hagen M?lle > wrote: >> >> Hi Andrew, Hi Mathieu, >> >> I can not see the line after it disappears. This might be a special >> feature of the open source radeon graphics driver under Linux. >> >> Just for completeness one can easily see the line thickness fluctuation by >> zooming into the scene just right before the line disappears. At this zoom >> level just drag the scene using the left mouse button and move the mouse >> around. >> >> In fact it would be very nice if this bug could be fixed at least when >> logarithmic x axis is enabled. Because when logarithmic x axis is enabled >> the error is visible after zooming in only a few steps. >> >> Hagen >> >> >> On 26.07.2016 10:57, Andrew Maclean wrote: >> >> Hi Hagen, Mathieu, >> >> I agree with you, if I zoom in when the logarithmic axis is disabled, the >> thick line vanishes, however if you look carefully there is a very faint >> thin line. It may be (2) as you outline below. >> >> Regards >> Andrew >> >> Andrew Maclean >> >> >> On 26 Jul 2016 5:02 pm, "Hagen M?lle" wrote: >>> >>> Hi Andrew, Hi Mathieu, >>> >>> That's exactly the behaviour I see. By the way I am using ATI/AMD video >>> cards in my Windows/Linux machine. >>> >>> !!!! UPDATE: The error also occurs when logarithmic x axis is not >>> enabled. But you have to zoom in a lot more. See attached pictures. I am >>> sorry for not realizing this earlier. >>> >>> >>> I am not that familiar with OpenGL. But I can imagine two things: >>> >>> (1) Line is clipped >>> (2) If shaders are used to draw the line thickness it might be a floating >>> point precision problem. Actually this could be the problem because the line >>> thickness starts fluctuating at some point of zoom level before it finally >>> disappears. See attached picture. The line thickness is about the half the >>> original line thickness. >>> >>> >>> I hope this helps to fix the bug. One might also contact Marcus D. >>> Hanwell. He is the developer who mainly implemented vtkCharts. I am not sure >>> if he is still responsible. >>> >>> Hagen >>> >>> >>> On 26.07.2016 01:52, Andrew Maclean wrote: >>> >>> Hi Hagen, >>> I just tested your example and I get similar behaviou using an NVIDIA >>> Video card, OpenGL2, VS2015 and the VTK Master. >>> >>> I can zoom in and, at some point the line disappears, but zooming out >>> immediately restores the line. >>> >>> It seeme to me that once a certain zoom level is reached the line is not >>> drawn, however zooming out restores it. >>> >>> Attached are two pngs showing the effect.When the line disappears it >>> should still be visible as the scales on the axes have only changed >>> slightly. >>> >>> Andrew >>> >>>> ---------- Forwarded message ---------- >>>> From: "Hagen M?lle" >>>> To: vtk-developers at vtk.org >>>> Cc: >>>> Date: Mon, 25 Jul 2016 10:14:21 +0200 >>>> Subject: [vtk-developers] vtkChartXY lines not visible when using >>>> logarithmic x axis >>>> Hi, >>>> >>>> Due to spam reasons the bug tracker's user login is disabled right now. >>>> One should send bug reports to the vtk's user mailing list, which I did >>>> twice. My bug report was totally ignored. That's why I send it here again. >>>> Hopefully someone can finally create a ticket in the bug tracker. Thanks. >>>> >>>> I want to report a bug for vtkChartXY. So I already send this Email two >>>> days ago to the vtkUsers mailing list without receiving any answer. The bug >>>> is not visible in the vtk bug tracker too. Thus I send the Email again. >>>> >>>> Bug environment: >>>> a) I am using VTK 6.2. I checked my test code on machines running >>>> Windows 7 (Visual Studio 2012/2015) and Ubuntu 15.10/Ubuntu 16.04 (gcc). On >>>> all these systems the bug is visible. >>>> b) I can also reproduce the bug in VTK 7.0 (here I used an Ubuntu 16.04 >>>> machine and compiled VTK 7.0 using OpenGL2 interface). >>>> >>>> Bug description: >>>> >>>> I am using vtkChartXY to display one line. I also enabled logarithmic x >>>> axis (chartXY->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true);). Nothing more >>>> special. In the compiled application I start zooming into the chart using >>>> the mouse wheel, making sure I always have a line displayed in the view. At >>>> some zoom level this line just disappears. At the zoom level the line >>>> disappears I did the following: >>>> >>>> 1. Zooming more into the chart the line will not appear again. >>>> 2. Zooming out twice (with mouse wheel) the line will reappear. >>>> >>>> How to reproduce the bug: >>>> >>>> I basically used the LinePlot example for vtkChartXY (see >>>> http://www.vtk.org/Wiki/VTK/Examples/Cxx/Plotting/LinePlot). >>>> >>>> I added one line to the example code marked with *** (line was added >>>> after line 54 in example code). >>>> I added the full code to the end of the Email. >>>> --------------- >>>> >>>> // Add multiple line plots, setting the colors etc >>>> vtkSmartPointer chart= vtkSmartPointer>>> >::New(); *** chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >>>> view->GetScene()->AddItem(chart); >>>> >>>> --------------- >>>> >>>> Compile and execute the program. See my bug description above to >>>> visualize the bug. >>>> >>>> >>>> Please contact me if you need more information. I can also provide >>>> images of the bug if needed. >>>> I hope the bug can be fixed. >>>> >>>> Hagen. >>>> >>>> >>>> >>>> >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> #include >>>> >>>> int main(int, char *[]) >>>> { >>>> // Create a table with some points in it >>>> vtkSmartPointer table = >>>> vtkSmartPointer::New(); >>>> >>>> vtkSmartPointer arrX = >>>> vtkSmartPointer::New(); >>>> arrX->SetName("X Axis"); >>>> table->AddColumn(arrX); >>>> >>>> vtkSmartPointer arrC = >>>> vtkSmartPointer::New(); >>>> arrC->SetName("Cosine"); >>>> table->AddColumn(arrC); >>>> >>>> vtkSmartPointer arrS = >>>> vtkSmartPointer::New(); >>>> arrS->SetName("Sine"); >>>> table->AddColumn(arrS); >>>> >>>> // Fill in the table with some example values >>>> int numPoints = 69; >>>> float inc = 7.5 / (numPoints-1); >>>> table->SetNumberOfRows(numPoints); >>>> for (int i = 0; i < numPoints; ++i) >>>> { >>>> table->SetValue(i, 0, i * inc); >>>> table->SetValue(i, 1, cos(i * inc)); >>>> table->SetValue(i, 2, sin(i * inc)); >>>> } >>>> >>>> // Set up the view >>>> vtkSmartPointer view = >>>> vtkSmartPointer::New(); >>>> view->GetRenderer()->SetBackground(1.0, 1.0, 1.0); >>>> >>>> // Add multiple line plots, setting the colors etc >>>> vtkSmartPointer chart = >>>> vtkSmartPointer::New(); >>>> chart.Get()->GetAxis(vtkAxis::BOTTOM)->SetLogScale(true); >>>> view->GetScene()->AddItem(chart); >>>> vtkPlot *line = chart->AddPlot(vtkChart::LINE); >>>> #if VTK_MAJOR_VERSION <= 5 >>>> line->SetInput(table, 0, 1); >>>> #else >>>> line->SetInputData(table, 0, 1); >>>> #endif >>>> line->SetColor(0, 255, 0, 255); >>>> line->SetWidth(1.0); >>>> line = chart->AddPlot(vtkChart::LINE); >>>> #if VTK_MAJOR_VERSION <= 5 >>>> line->SetInput(table, 0, 2); >>>> #else >>>> line->SetInputData(table, 0, 2); >>>> #endif >>>> line->SetColor(255, 0, 0, 255); >>>> line->SetWidth(5.0); >>>> >>>> // For dotted line, the line type can be from 2 to 5 for different >>>> dash/dot >>>> // patterns (see enum in vtkPen containing DASH_LINE, value 2): >>>> #ifndef WIN32 >>>> line->GetPen()->SetLineType(vtkPen::DASH_LINE); >>>> #endif >>>> // (ifdef-ed out on Windows because DASH_LINE does not work on >>>> Windows >>>> // machines with built-in Intel HD graphics card...) >>>> >>>> //view->GetRenderWindow()->SetMultiSamples(0); >>>> >>>> // Start interactor >>>> view->GetInteractor()->Initialize(); >>>> view->GetInteractor()->Start(); >>>> >>>> return EXIT_SUCCESS; >>>> } >>>> >>>> >>>> _______________________________________________ >>>> 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 >>> >>> >>> >>> >>> -- >>> ___________________________________________ >>> Andrew J. P. Maclean >>> >>> ___________________________________________ >>> >>> >> > From dan.lipsa at kitware.com Wed Jul 27 13:24:04 2016 From: dan.lipsa at kitware.com (Dan Lipsa) Date: Wed, 27 Jul 2016 13:24:04 -0400 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: <68c3c4b4ed31aac7aa18339d86843809@arcor.de> References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> <68c3c4b4ed31aac7aa18339d86843809@arcor.de> Message-ID: On Wed, Jul 27, 2016 at 10:20 AM, Peter Mattern wrote: > Hi Dan. > > Thanks for your quick response! > > I was aware that you're using "in-tree" copies of a lot of dependencies. > And note I'm assuming you're doing this for a reason so I certainly don't > criticise. > It's just that as an old Linuxer I think using system libraries wherever > feasible is basically the better solution... > The problem we try to solve with in-tree copies is version differences and missing libraries on various platforms. > > ad 14126 > The problem is that CMake just doesn't find the system-wide PROJ.4 > instance and hence fails > >> [...] >> -- Found TIFF: /usr/lib64/libtiff.so (found version "4.0.6") >> CMake Error at CMake/vtkModuleMacros.cmake:864 (message): >> VTK_USE_SYSTEM_LIBPROJ4 is ON but LIBPROJ4 is not found! >> Call Stack (most recent call first): >> ThirdParty/libproj4/CMakeLists.txt:1 (vtk_module_third_party) >> >> >> -- Configuring incomplete, errors occurred! >> > The findings are the very same compiling VTK commit ddcd09c on Arch Linux, > Debian testing (stretch) and Fedora 24 using either PROJ.4 4.9.2 as > provided by those distributions or PROJ.4 commit 06a708d compiled on the > respective system. > The PROJ.4 commit was compiled either by running autogen, configure > [--prefix=/usr], make and make install or cmake > -DCMAKE_INSTALL_PREFIX=/usr, make and make install as compiling yields > slightly different results right now, see > https://github.com/OSGeo/proj.4/issues/378, but this didn't make a > difference regarding VTK. > > Finding a library can be a tricky issue. Unless the library/headers are in standard locations it won't find them. If headers/libraries are in a standard location this warrants a bug report. However you can manually point the configuration variables to the right locations. In the cmake configuration program (ccmake or cmake-gui) switch to advanced variables and make the following two variables point to the correct location on your system: LIBPROJ4_INCLUDE_DIR LIBPROJ4_LIBRARIES > ad 15549 > I've added a comment describing the current situation but don't seem to be > able to change field "Status", maybe as I didn't open the issue? > I've changed this to backlog. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmattern at arcor.de Fri Jul 29 07:57:09 2016 From: pmattern at arcor.de (Peter Mattern) Date: Fri, 29 Jul 2016 11:57:09 +0000 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> <68c3c4b4ed31aac7aa18339d86843809@arcor.de> Message-ID: ? PROJ.4 / #14126 After setting -DLIBPROJ4_INCLUDE_DIR=/usr/include and -DLIBPROJ4_LIBRARIES=/usr/lib/libproj.so.9.1.0 (or to the linker name ./libproj.so only) VTK commit 981a1e1 compiled flawlessly against both PROJ.4 release 4.9.2 and commit 06a708d. Both headers and shared libraries of various other dependencies like netCDF-C++ legacy, GDAL or GL2PS are found automatically at the very same locations, though. Some of them where used by path /usr/lib64 which is a symbolic link to /usr/lib on Arch Linux which shouldn't matter with regards to PROJ.4, IMO. So all in all I still think a but in VTK regarding PROJ.4 remains and the question is whether a new one should be opened or #14126 should be continued. ? netCDF-C++ / #15549 > I've changed this to backlog. That's pretty good. The dependency on that legacy version should be removed in the long run. It doesn't offer all features of the current API besides this probably wasn't relevant for VTK yet. From dan.lipsa at kitware.com Fri Jul 29 08:23:44 2016 From: dan.lipsa at kitware.com (Dan Lipsa) Date: Fri, 29 Jul 2016 08:23:44 -0400 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> <68c3c4b4ed31aac7aa18339d86843809@arcor.de> Message-ID: Peter, I would file a new bug as this is FindProj4 specific, rather than the fact that VTK used in the past a very old proj4 which was not compatible with recent versions. Thanks, Dan On Fri, Jul 29, 2016 at 7:57 AM, Peter Mattern wrote: > ? PROJ.4 / #14126 > After setting -DLIBPROJ4_INCLUDE_DIR=/usr/include and > -DLIBPROJ4_LIBRARIES=/usr/lib/libproj.so.9.1.0 (or to the linker name > ./libproj.so only) VTK commit 981a1e1 compiled flawlessly against both > PROJ.4 release 4.9.2 and commit 06a708d. > Both headers and shared libraries of various other dependencies like > netCDF-C++ legacy, GDAL or GL2PS are found automatically at the very same > locations, though. Some of them where used by path /usr/lib64 which is a > symbolic link to /usr/lib on Arch Linux which shouldn't matter with regards > to PROJ.4, IMO. > So all in all I still think a but in VTK regarding PROJ.4 remains and the > question is whether a new one should be opened or #14126 should be > continued. > > ? netCDF-C++ / #15549 > >> I've changed this to backlog. >> > That's pretty good. The dependency on that legacy version should be > removed in the long run. It doesn't offer all features of the current API > besides this probably wasn't relevant for VTK yet. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From shchegelskij at gmail.com Fri Jul 29 10:27:22 2016 From: shchegelskij at gmail.com (Taras Shchehelskyi) Date: Fri, 29 Jul 2016 17:27:22 +0300 Subject: [vtk-developers] Rendering hangs in viewport In-Reply-To: References: <577b933d.98012e0a.c47e1.ffffaff2@mx.google.com> <20160705150232.2016783904@mail.rogue-research.com> <577be0e2.ca52190a.58dc5.14ec@mx.google.com> <577bf9b5.d5052e0a.d3d9a.ffffce07@mx.google.com> Message-ID: <579b67ca.0fcd190a.d06aa.cd6b@mx.google.com> Hello, Just want to check if there are some news about this issue. Maybe somebody have idea how to fix this bug? Thanks, Taras From: Will Schroeder -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmattern at arcor.de Sun Jul 31 17:40:19 2016 From: pmattern at arcor.de (Peter Mattern) Date: Sun, 31 Jul 2016 21:40:19 +0000 Subject: [vtk-developers] Bug Tracker: How to deal with issues closed in error? In-Reply-To: References: <9b3e74b62b23af345a7da83976adbab2@arcor.de> <68c3c4b4ed31aac7aa18339d86843809@arcor.de> Message-ID: <7484b983085e867e8d5e97c38491b92a@arcor.de> > I would file a new bug as this is FindProj4 specific, rather than the > fact that VTK used in the past a very old proj4 which > was not compatible with recent versions. Done: http://www.vtk.org/Bug/view.php?id=16803