From houssen at ipgp.fr  Wed Jul  1 16:49:37 2015
From: houssen at ipgp.fr (houssen)
Date: Wed, 01 Jul 2015 22:49:37 +0200
Subject: [Paraview-developers]
 =?utf-8?q?=5BParaview=5D_Fwd=3A_How_to_plug?=
 =?utf-8?q?_into_paraview_his_own_writer_=3F?=
In-Reply-To: <CAB5Fpx6hLd4Of0iopNU-zEpVV9X=w8-2tPjGChm=sp1OPaM9fA@mail.gmail.com>
References: <86347d91c723cd90b081325c08b678ec@imap.ipgp.fr>
 <CAB5Fpx6hLd4Of0iopNU-zEpVV9X=w8-2tPjGChm=sp1OPaM9fA@mail.gmail.com>
Message-ID: <8fc449b95eac010b5d9af85b836c351e@imap.ipgp.fr>

Hi Cory,

I had to spread my time over several topics these days... So I had not 
much time to get back to the paraview writer.

I believe I have done pretty much everything for the writer to work but 
at save time, I still get this error :
~> ERROR: In 
/home/houssen/Programs/ParaView/ParaView-v4.3.1-source/ParaViewCore/ServerImplementation/Core/vtkSIProxy.cxx, 
line 307
vtkSIWriterProxy (0x5cc38e0): Failed to create myWriter. Aborting for 
debugging purposes.
The writer instance is not created for some reason.

At this point, I run paraview, I load the plugin with "tools / manage 
plugins" (load seems OK), I open some data, I save with "my writer" (I 
see my writer in the save GUI + I am asked for a file-to-save name), I 
click OK and then it crahses. I believe the xml file is wrong and / or 
something is missing... But I didn't yet figured it out. It seems I have 
no pvserver (see in the mail) : can this trigger the crash ? Do I need a 
pvserver for the writer to work ?

I answered your question in the mail.

Franck

Le 2015-06-30 6:08, Cory Quammen a ?crit?:
> Hi Franck,
>
> It looks like you've been making progress, so I'll answer your
> questions from your second email.
>
> ?
>
>> Question 1 : how to write myWriter.xml ?
>>
>> 1.1. seems I must set <InputProperty command="SetInputConnection"
>> name="Input"> in myWriter.xml. Does this means I will also have to
>> implement myWriter::SetInputConnection ?
>
> No. Your writer will inherit from subclass of vtkAlgorithm, the class
> that defines "SetInputConnection", and you should not need to 
> override
> it. XdmfWriter inherits from vtkDataObjectAlgorithm, which itself
> inherits vtkAlgorithm.
> ?
>
>> 1.2. seems I must set <StringVectorProperty name="FileName"
>> command="SetFileName"> in myWriter.xml. Does this means I will also
>> have to implement myWriter::SetFileName ?
>
> Yes, like vtkXdmfWriter.
>
>  2.2 in myWriter.hpp, I need to derive from a vtkObject this way
> (like XdmfWriter.cpp does)
>
>> ew (); // Needed to be plugged into paraview
>> ? ? vtkTypeMacro ( myWriter, vtkDataObjectAlgorithm ); // Needed
>> to be plugged into paraview
>> + suppress copy constructor and operator=
>>
>> Am I correct ?
>>
>> Yes.
>>
>> Question 3 : in myWriter.cpp, on
> I am supposed to be called back (to get focus back) when I click the
> "save data" button ? Here, how can I get data to be saved (data that 
> I
> can see in paraview viewer befo
>
>> n ParaView, when you choose, File -> Save Data, the data from the
>> selected Pipeline Browser object will be saved. This will have some
>> kind of data type, e.g., vtkImageData, vtkPolyData, etc. What kind
>> of data type would you like your writer to support?

Structured and likely also unstructured grids

>>
>> The data should
> in the member function myWriter::RequestData(). This is where you
> should write your data.?

OK (I suppose I'll get vtkStructuredGrid / vtkUnstructuredGrid as 
vtkObject in myWriter::RequestData according to 
http://www.vtk.org/Wiki/VTK/Tutorials/New_Pipeline#Typical_Pipeline_Execution)

>
>  Finally, myWriter.xml seems OK. myWriter.*pp compile OK. Paraview
> finds myWriter.xml at run time (exporting PV_PLUGIN_PATH) and loads
> libmywriter.so without problem. I see the writer in the "save data"
> GUI (drop down list with different save formats). I save with
> myWriter, I get a (basic) GUI with file name only (no extra 
> parameter)
> : I crash when I click "OK, save" with this error message :
>  ERROR: In /.../ParaView-v4.3.1-sourc
>
>> ailed to create myWriter. Aborting for debugging purposes.
>>
>> vtkSIWriterProxy does this at line 303 :
>> vtkObjectBase* obj = this->Interpreter->NewInstance(className);
>> but obj is NULL.
>>
>> I get obj is NULL because I missed something in myWriter
>> implementation : what did I miss ?
>>
>> Are you running pvserver separately and connecting to it with the
>> client? If so, make sure PV_PLUGIN_PATH is defined in the
>> environment in which pvserver is run. Have you tried to load your
>> plugin through the Tools -> Manage Plugins... dialog?
>>

No I just run a client. I don't need the server : is it necessary for 
the writer to work ?

~> paraview &
~> ps -aux | grep pvserver
houssen   5315  0.0  0.0  14544   948 pts/0    S+   22:38   0:00 grep 
pvserver
~> ps -aux | grep paraview
houssen   5313 16.0  1.1 1175692 179424 pts/0  S    22:38   0:01 
/home/houssen/Programs/ParaView/ParaView-v4.3.1-source/local/lib/paraview-4.3/paraview
houssen   5317  0.0  0.0  14544   944 pts/0    S+   22:38   0:00 grep 
paraview


>> I hope that helps get you further in writing your data files.
>>
>> Thanks,
>> Cory
>> ?
>> Can somebody help
> t me some clue ?
>
>  FH
>
>  -------- Message original --------
>  Objet: How to plug into paraview his own writer ?
>  Date: 2015-06-12 15:54
>  De: houssen <houssen at imap.ipgp.fr [6]>
>  ?: <paraview at paraview.org [7]>
>
>  I followed 2 tutorials (
>
> Links:
> ------
> [1] http://www.kitware.com
> [2] http://www.kitware.com/opensource/opensource.html
> [3] http://paraview.org/Wiki/ParaView
> [4] http://markmail.org/search/?q=ParaView
> [5] http://public.kitware.com/mailman/listinfo/paraview
> [6] mailto:houssen at imap.ipgp.fr
> [7] mailto:paraview at paraview.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: myWriter.tar.gz
Type: application/x-gzip
Size: 1591 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150701/8116c72b/attachment.bin>

From m.schlottke-lakemper at aia.rwth-aachen.de  Sun Jul  5 04:37:39 2015
From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael)
Date: Sun, 5 Jul 2015 08:37:39 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
Message-ID: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>

Dear all,

this (http://www.kitware.com/blog/home/post/856) excellent blog post details how to use the vtkGhostLevels with the next ParaView release. However, it focuses on the changes in comparison to the previous way of doing things. Thus I am wondering if there is a comprehensive guide online that describes in a little more depth how to properly use ghost levels in the ?old? (or current as of PV 4.3.1) way?

Thanks a lot in advance

Michael


--
Michael Schlottke-Lakemper

SimLab Highly Scalable Fluids & Solids Engineering
J?lich Aachen Research Alliance (JARA-HPC)
RWTH Aachen University
W?llnerstra?e 5a
52062 Aachen
Germany

Phone: +49 (241) 80 95188
Fax: +49 (241) 80 92257
Mail: m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>
Web: http://www.jara.org/jara-hpc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150705/64ce6fc4/attachment.html>

From houssen at ipgp.fr  Sun Jul  5 12:09:16 2015
From: houssen at ipgp.fr (houssen)
Date: Sun, 05 Jul 2015 18:09:16 +0200
Subject: [Paraview-developers] =?utf-8?q?paraview_=3A_how_to_get_the_GUI_n?=
 =?utf-8?q?ame_from_a_vtkObject_=3F?=
Message-ID: <7767480b495b9facdd845eaf5ad23563@imap.ipgp.fr>

 From a vtkObject, how to get the name (alias) one can see / modify in 
the paraview GUI ? (name in the pipeline browser)

Franck

From houssen at ipgp.fr  Sun Jul  5 12:20:32 2015
From: houssen at ipgp.fr (houssen)
Date: Sun, 05 Jul 2015 18:20:32 +0200
Subject: [Paraview-developers] =?utf-8?q?paraview_=3A_how_to_add_message_t?=
 =?utf-8?q?o_the_output_message_window_=3F?=
Message-ID: <ef656560a43ee4444fbf9770edcca7c7@imap.ipgp.fr>

How to add message to the output message window ?

For now, I use : vtkDebugMacro + DebugOn + GlobalWarningDisplayOn.
This is a lot verbose. Is there a way to print only "info message" but 
not "all debug message" ?

Franck

From cory.quammen at kitware.com  Sun Jul  5 14:23:02 2015
From: cory.quammen at kitware.com (Cory Quammen)
Date: Sun, 5 Jul 2015 14:23:02 -0400
Subject: [Paraview-developers] paraview : how to add message to the
 output message window ?
In-Reply-To: <ef656560a43ee4444fbf9770edcca7c7@imap.ipgp.fr>
References: <ef656560a43ee4444fbf9770edcca7c7@imap.ipgp.fr>
Message-ID: <CAB5Fpx58rK5kDcjEi064Yn-YXBoOSqsAqt+inn4RWiRH8jfQSA@mail.gmail.com>

Hi Franck,

If you are developing your own plugin, then you could temporarily use
vtkWarningMacro("Informational message"); to write info messages to the
output window. These messages will be output no matter what. There is,
unfortunately, no vtkInfoMacro, but using the vtkWarningMacro will at least
let you avoid seeing all the other Debug messages that you might not want
to see.

HTH,
Cory




On Sun, Jul 5, 2015 at 12:20 PM, houssen <houssen at ipgp.fr> wrote:

> How to add message to the output message window ?
>
> For now, I use : vtkDebugMacro + DebugOn + GlobalWarningDisplayOn.
> This is a lot verbose. Is there a way to print only "info message" but not
> "all debug message" ?
>
> Franck
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>



-- 
Cory Quammen
R&D Engineer
Kitware, Inc.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150705/c65c88fb/attachment.html>

From dan.lipsa at kitware.com  Mon Jul  6 11:10:43 2015
From: dan.lipsa at kitware.com (Dan Lipsa)
Date: Mon, 06 Jul 2015 15:10:43 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
Message-ID: <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>

Hi Michael,
I am not aware of a guide to the old way of storing/accessing ghost cells
but I'll be happy to try to answer any specific questions you may have.
I can also add a link in the blog to the document if someone points it out.

Dan


On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <
m.schlottke-lakemper at aia.rwth-aachen.de> wrote:

>  Dear all,
>
>  this (http://www.kitware.com/blog/home/post/856) excellent blog post
> details how to use the vtkGhostLevels with the next ParaView release.
> However, it focuses on the changes in comparison to the previous way of
> doing things. Thus I am wondering if there is a comprehensive guide online
> that describes in a little more depth how to properly use ghost levels in
> the ?old? (or current as of PV 4.3.1) way?
>
>  Thanks a lot in advance
>
>  Michael
>
>
>             --
> Michael Schlottke-Lakemper
>
>  SimLab Highly Scalable Fluids & Solids Engineering
> J?lich Aachen Research Alliance (JARA-HPC)
> RWTH Aachen University
> W?llnerstra?e 5a
> 52062 Aachen
> Germany
>
>  Phone: +49 (241) 80 95188
> Fax: +49 (241) 80 92257
> Mail: m.schlottke-lakemper at aia.rwth-aachen.de
> Web: http://www.jara.org/jara-hpc
>
>  _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150706/1181eaea/attachment.html>

From m.schlottke-lakemper at aia.rwth-aachen.de  Mon Jul  6 11:42:06 2015
From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael)
Date: Mon, 6 Jul 2015 15:42:06 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
 <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
Message-ID: <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>

Hi Dan,

Thank you for your answer. From a first impression, it seems like the only really open question right now is how to generate the ?vtkGhostLevels? array if I am working with an vtkUnstructuredGrid in my custom reader plugin? ?GenerateGhostLevelArray? in vtkDataSet seems to be only relevant for structured grids. Maybe I just need to create a vtkIntArray myself, add it as a cell array, and name it ?vtkGhostLevels??

Regards,

Michael


On 06 Jul 2015, at 17:10 , Dan Lipsa <dan.lipsa at kitware.com<mailto:dan.lipsa at kitware.com>> wrote:

Hi Michael,
I am not aware of a guide to the old way of storing/accessing ghost cells but I'll be happy to try to answer any specific questions you may have.
I can also add a link in the blog to the document if someone points it out.

Dan


On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>> wrote:
Dear all,

this (http://www.kitware.com/blog/home/post/856) excellent blog post details how to use the vtkGhostLevels with the next ParaView release. However, it focuses on the changes in comparison to the previous way of doing things. Thus I am wondering if there is a comprehensive guide online that describes in a little more depth how to properly use ghost levels in the ?old? (or current as of PV 4.3.1) way?

Thanks a lot in advance

Michael


--
Michael Schlottke-Lakemper

SimLab Highly Scalable Fluids & Solids Engineering
J?lich Aachen Research Alliance (JARA-HPC)
RWTH Aachen University
W?llnerstra?e 5a
52062 Aachen
Germany

Phone: +49 (241) 80 95188
Fax: +49 (241) 80 92257
Mail: m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>
Web: http://www.jara.org/jara-hpc

_______________________________________________
Powered by www.kitware.com<http://www.kitware.com/>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Search the list archives at: http://markmail.org/search/?q=Paraview-developers

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview-developers

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150706/4925fad4/attachment-0001.html>

From dan.lipsa at kitware.com  Mon Jul  6 12:16:40 2015
From: dan.lipsa at kitware.com (Dan Lipsa)
Date: Mon, 06 Jul 2015 16:16:40 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
 <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
 <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>
Message-ID: <CADwGrVqJnuuS-dw3G1bku3rPczKY-FuOb0QiDmYzraLw4xgFnQ@mail.gmail.com>

Hi Michael,
Yes, indeed this is how you would do it. I believe the array has to be
unsigned char array though. On the first layer of ghost cells you store a
1, on the second layer you store a 2 and so on.
(we just set a bit for the new version, so we don't store information about
how many layers we have)

Also note that ghost cells are only useful if your reader reads different
parts of the dataset on different mpi processors. If you split your dataset
on different mpi nodes latter on, this is when you need to add ghost cells
(because this is when you know how your data is split). If you run on a
single processor ghosts cells are not used.

Dan


On Mon, Jul 6, 2015 at 11:43 AM Schlottke-Lakemper, Michael <
m.schlottke-lakemper at aia.rwth-aachen.de> wrote:

>  Hi Dan,
>
>  Thank you for your answer. From a first impression, it seems like the
> only really open question right now is how to generate the ?vtkGhostLevels?
> array if I am working with an vtkUnstructuredGrid in my custom reader
> plugin? ?GenerateGhostLevelArray? in vtkDataSet seems to be only relevant
> for structured grids. Maybe I just need to create a vtkIntArray myself, add
> it as a cell array, and name it ?vtkGhostLevels??
>
>  Regards,
>
>  Michael
>
>
>  On 06 Jul 2015, at 17:10 , Dan Lipsa <dan.lipsa at kitware.com> wrote:
>
>  Hi Michael,
> I am not aware of a guide to the old way of storing/accessing ghost cells
> but I'll be happy to try to answer any specific questions you may have.
> I can also add a link in the blog to the document if someone points it out.
>
>  Dan
>
>
>  On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <
> m.schlottke-lakemper at aia.rwth-aachen.de> wrote:
>
>> Dear all,
>>
>>  this (http://www.kitware.com/blog/home/post/856) excellent blog post
>> details how to use the vtkGhostLevels with the next ParaView release.
>> However, it focuses on the changes in comparison to the previous way of
>> doing things. Thus I am wondering if there is a comprehensive guide online
>> that describes in a little more depth how to properly use ghost levels in
>> the ?old? (or current as of PV 4.3.1) way?
>>
>>  Thanks a lot in advance
>>
>>  Michael
>>
>>
>>        --
>> Michael Schlottke-Lakemper
>>
>>  SimLab Highly Scalable Fluids & Solids Engineering
>> J?lich Aachen Research Alliance (JARA-HPC)
>> RWTH Aachen University
>> W?llnerstra?e 5a
>> 52062 Aachen
>> Germany
>>
>>  Phone: +49 (241) 80 95188
>> Fax: +49 (241) 80 92257
>> Mail: m.schlottke-lakemper at aia.rwth-aachen.de
>> Web: http://www.jara.org/jara-hpc
>>
>>  _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150706/b1e1ff74/attachment.html>

From m.schlottke-lakemper at aia.rwth-aachen.de  Mon Jul  6 14:36:00 2015
From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael)
Date: Mon, 6 Jul 2015 18:36:00 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <CADwGrVqJnuuS-dw3G1bku3rPczKY-FuOb0QiDmYzraLw4xgFnQ@mail.gmail.com>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
 <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
 <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>
 <CADwGrVqJnuuS-dw3G1bku3rPczKY-FuOb0QiDmYzraLw4xgFnQ@mail.gmail.com>
Message-ID: <f04f9084-7221-4b2e-abd4-f78db7fdaf43@HUB2.rwth-ad.de>

Hi Dan,

Thanks a lot for this clarification - this should be enough information for us to get started.

Indeed, our reader plugin is designed to be used in a parallel fashion. Due to the size of the grid/data files, we typically use O(10)-O(1000) MPI ranks for pvserver. Currently we use the D3 filter if we need smooth gradients/statistics across domain boundaries, but the D3 filter does not scale well beyond a hundred or so cores.

Michael

On 06 Jul 2015, at 18:16 , Dan Lipsa <dan.lipsa at kitware.com<mailto:dan.lipsa at kitware.com>> wrote:

Hi Michael,
Yes, indeed this is how you would do it. I believe the array has to be unsigned char array though. On the first layer of ghost cells you store a 1, on the second layer you store a 2 and so on.
(we just set a bit for the new version, so we don't store information about how many layers we have)

Also note that ghost cells are only useful if your reader reads different parts of the dataset on different mpi processors. If you split your dataset on different mpi nodes latter on, this is when you need to add ghost cells (because this is when you know how your data is split). If you run on a single processor ghosts cells are not used.

Dan


On Mon, Jul 6, 2015 at 11:43 AM Schlottke-Lakemper, Michael <m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>> wrote:
Hi Dan,

Thank you for your answer. From a first impression, it seems like the only really open question right now is how to generate the ?vtkGhostLevels? array if I am working with an vtkUnstructuredGrid in my custom reader plugin? ?GenerateGhostLevelArray? in vtkDataSet seems to be only relevant for structured grids. Maybe I just need to create a vtkIntArray myself, add it as a cell array, and name it ?vtkGhostLevels??

Regards,

Michael


On 06 Jul 2015, at 17:10 , Dan Lipsa <dan.lipsa at kitware.com<mailto:dan.lipsa at kitware.com>> wrote:

Hi Michael,
I am not aware of a guide to the old way of storing/accessing ghost cells but I'll be happy to try to answer any specific questions you may have.
I can also add a link in the blog to the document if someone points it out.

Dan


On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>> wrote:
Dear all,

this (http://www.kitware.com/blog/home/post/856) excellent blog post details how to use the vtkGhostLevels with the next ParaView release. However, it focuses on the changes in comparison to the previous way of doing things. Thus I am wondering if there is a comprehensive guide online that describes in a little more depth how to properly use ghost levels in the ?old? (or current as of PV 4.3.1) way?

Thanks a lot in advance

Michael


--
Michael Schlottke-Lakemper

SimLab Highly Scalable Fluids & Solids Engineering
J?lich Aachen Research Alliance (JARA-HPC)
RWTH Aachen University
W?llnerstra?e 5a
52062 Aachen
Germany

Phone: +49 (241) 80 95188
Fax: +49 (241) 80 92257
Mail: m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>
Web: http://www.jara.org/jara-hpc

_______________________________________________
Powered by www.kitware.com<http://www.kitware.com/>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Search the list archives at: http://markmail.org/search/?q=Paraview-developers

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview-developers


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150706/ebbcd9e8/attachment-0001.html>

From dan.lipsa at kitware.com  Mon Jul  6 14:56:10 2015
From: dan.lipsa at kitware.com (Dan Lipsa)
Date: Mon, 06 Jul 2015 18:56:10 +0000
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <f04f9084-7221-4b2e-abd4-f78db7fdaf43@HUB2.rwth-ad.de>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
 <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
 <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>
 <CADwGrVqJnuuS-dw3G1bku3rPczKY-FuOb0QiDmYzraLw4xgFnQ@mail.gmail.com>
 <f04f9084-7221-4b2e-abd4-f78db7fdaf43@HUB2.rwth-ad.de>
Message-ID: <CADwGrVrnWEvy4XRG=4-C5zso67WrWP2byktMud9AYimQEJG2fA@mail.gmail.com>

Great. Also note there is a UPDATE_NUMBER_OF_GHOST_LEVELS information that
tells you how many ghost levels you need to read and/or create - that
depends on your pipeline.
I would also look at one of the parallel ParaView/VTK readers to see what
they do.

Dan


On Mon, Jul 6, 2015 at 2:36 PM Schlottke-Lakemper, Michael <
m.schlottke-lakemper at aia.rwth-aachen.de> wrote:

>  Hi Dan,
>
>  Thanks a lot for this clarification - this should be enough information
> for us to get started.
>
>  Indeed, our reader plugin is designed to be used in a parallel fashion.
> Due to the size of the grid/data files, we typically use O(10)-O(1000) MPI
> ranks for pvserver. Currently we use the D3 filter if we need smooth
> gradients/statistics across domain boundaries, but the D3 filter does not
> scale well beyond a hundred or so cores.
>
>  Michael
>
>   On 06 Jul 2015, at 18:16 , Dan Lipsa <dan.lipsa at kitware.com> wrote:
>
>  Hi Michael,
> Yes, indeed this is how you would do it. I believe the array has to be
> unsigned char array though. On the first layer of ghost cells you store a
> 1, on the second layer you store a 2 and so on.
> (we just set a bit for the new version, so we don't store information
> about how many layers we have)
>
>  Also note that ghost cells are only useful if your reader reads
> different parts of the dataset on different mpi processors. If you split
> your dataset on different mpi nodes latter on, this is when you need to add
> ghost cells (because this is when you know how your data is split). If you
> run on a single processor ghosts cells are not used.
>
>  Dan
>
>
>  On Mon, Jul 6, 2015 at 11:43 AM Schlottke-Lakemper, Michael <
> m.schlottke-lakemper at aia.rwth-aachen.de> wrote:
>
>> Hi Dan,
>>
>>  Thank you for your answer. From a first impression, it seems like the
>> only really open question right now is how to generate the ?vtkGhostLevels?
>> array if I am working with an vtkUnstructuredGrid in my custom reader
>> plugin? ?GenerateGhostLevelArray? in vtkDataSet seems to be only relevant
>> for structured grids. Maybe I just need to create a vtkIntArray myself, add
>> it as a cell array, and name it ?vtkGhostLevels??
>>
>>  Regards,
>>
>>  Michael
>>
>>
>>  On 06 Jul 2015, at 17:10 , Dan Lipsa <dan.lipsa at kitware.com> wrote:
>>
>>  Hi Michael,
>> I am not aware of a guide to the old way of storing/accessing ghost cells
>> but I'll be happy to try to answer any specific questions you may have.
>> I can also add a link in the blog to the document if someone points it
>> out.
>>
>>  Dan
>>
>>
>>  On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <
>> m.schlottke-lakemper at aia.rwth-aachen.de> wrote:
>>
>>> Dear all,
>>>
>>>  this (http://www.kitware.com/blog/home/post/856) excellent blog post
>>> details how to use the vtkGhostLevels with the next ParaView release.
>>> However, it focuses on the changes in comparison to the previous way of
>>> doing things. Thus I am wondering if there is a comprehensive guide online
>>> that describes in a little more depth how to properly use ghost levels in
>>> the ?old? (or current as of PV 4.3.1) way?
>>>
>>>  Thanks a lot in advance
>>>
>>>  Michael
>>>
>>>
>>>        --
>>> Michael Schlottke-Lakemper
>>>
>>>  SimLab Highly Scalable Fluids & Solids Engineering
>>> J?lich Aachen Research Alliance (JARA-HPC)
>>> RWTH Aachen University
>>> W?llnerstra?e 5a
>>> 52062 Aachen
>>> Germany
>>>
>>>  Phone: +49 (241) 80 95188
>>> Fax: +49 (241) 80 92257
>>> Mail: m.schlottke-lakemper at aia.rwth-aachen.de
>>> Web: http://www.jara.org/jara-hpc
>>>
>>>  _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150706/f8562577/attachment.html>

From prashant.v at quest-global.com  Tue Jul  7 06:28:18 2015
From: prashant.v at quest-global.com (Prashant V)
Date: Tue, 7 Jul 2015 10:28:18 +0000
Subject: [Paraview-developers] Paraview runtime tcl error: Can't find a
	usable init.tcl
Message-ID: <E493A24A2D2E32408F6F053DB5082EDD423F86BC@INBLCORMX02.quest-global.com>

Hi,



I am new to Paraview and developing a paraview plugin. From plugin, I am executing a python script using pqPythonManager.

        pqPythonManager *pm = pqPVApplicationCore::instance()->pythonManager();
        pm->executeScript("button.py");



>From script, I am trying to create gui button. Scripy snippet is below:



        import os
        import sys
        import Tkinter
        import tkMessageBox



        top = Tkinter.Tk()

        def helloCallBack():
           tkMessageBox.showinfo( "Hello Python", "Hello World")



        B = Tkinter.Button(top, text ="Hello", command = helloCallBack)

        B.pack()
        top.mainloop()



During run, I am getting below error (at top = Tkinter.Tk())



----------------------------------------------------

Traceback (most recent call last):
  File "<string>", line 6, in <module>
  File "<install-dir>/bin/Python/lib/python2.7/lib-tk/Tkinter.py", line 1764, in __init__
    self.tk = _tkinter.create(screenName, baseName, className, interactive, wantobjects, useTk, sync, use)
_tkinter.TclError: Can't find a usable init.tcl in the following directories:
    /opt/anaconda1anaconda2anaconda3/lib/tcl8.5 <install-dir>/paraview/lib/lib/tcl8.5 <install-dir>/paraview/lib/tcl8.5 <install-dir>/paraview/lib/library <install-dir>/paraview/library <install-dir>/paraview/tcl8.5.15/library <install-dir>/paraview/tcl8.5.15/library



This probably means that Tcl wasn't installed properly.

-----------------------------------------------



I could find init.tcl in directory <install-dir>/bin/Python/lib/python2.7/lib/tcl8.5, where system is not looking for.



Can somebody please help with this error? Do I need to set tcl path somewhere?

Thanks a lot in advance.


-Prashant

---Disclaimer------------------------------ This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended solely for the use of the addressee(s). If you are not the intended recipient, please notify the sender by e-mail and delete the original message. Opinions, conclusions and other information in this transmission that do not relate to the official business of QuEST Global and/or its subsidiaries, shall be understood as neither given nor endorsed by it. Any statements made herein that are tantamount to contractual obligations, promises, claims or commitments shall not be binding on the Company unless followed by written confirmation by an authorized signatory of the Company. -----------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150707/9b3c5da1/attachment.html>

From mathieu.westphal at kitware.com  Wed Jul  8 01:25:53 2015
From: mathieu.westphal at kitware.com (Mathieu Westphal)
Date: Wed, 8 Jul 2015 07:25:53 +0200
Subject: [Paraview-developers] Reload button
Message-ID: <CAC2t4UT3SP7PCBhxCbiF+XGinvT_uvc0Zgv5fTHDYnx12nWvjQ@mail.gmail.com>

Hello

A client is willing to have a reload button on it's filter panel, but we
think it would be woth it to add a reload button on the interface, wich
would basically Invoke Modified on a reader and apply the pipeline again.

It may not work with some caching reader, but other than that i think it
should be allright.
What do you think ?

Mathieu Westphal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150708/6dafac07/attachment.html>

From utkarsh.ayachit at kitware.com  Wed Jul  8 09:36:51 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 08 Jul 2015 13:36:51 +0000
Subject: [Paraview-developers] Reload button
In-Reply-To: <CAC2t4UT3SP7PCBhxCbiF+XGinvT_uvc0Zgv5fTHDYnx12nWvjQ@mail.gmail.com>
References: <CAC2t4UT3SP7PCBhxCbiF+XGinvT_uvc0Zgv5fTHDYnx12nWvjQ@mail.gmail.com>
Message-ID: <CADHSJjBWVq-0WPFHy1QOSd7uG8E0qcFLpjb18JnujjoX+s-btA@mail.gmail.com>

There's a mechanism to add a button to the autogenerated panel to call a
method on the reader already. No objection to adding it for readers where
this would work. In most cases, however, not that the readers check if the
filename has changed to re-read meta-data, hence do it case-by-case.

Utkarsh

On Wed, Jul 8, 2015 at 1:26 AM Mathieu Westphal <
mathieu.westphal at kitware.com> wrote:

> Hello
>
> A client is willing to have a reload button on it's filter panel, but we
> think it would be woth it to add a reload button on the interface, wich
> would basically Invoke Modified on a reader and apply the pipeline again.
>
> It may not work with some caching reader, but other than that i think it
> should be allright.
> What do you think ?
>
> Mathieu Westphal
>  _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150708/8a7396e7/attachment.html>

From berk.geveci at kitware.com  Thu Jul  9 10:06:17 2015
From: berk.geveci at kitware.com (Berk Geveci)
Date: Thu, 9 Jul 2015 10:06:17 -0400
Subject: [Paraview-developers] Guide to using ghost levels (old style)
In-Reply-To: <f04f9084-7221-4b2e-abd4-f78db7fdaf43@HUB2.rwth-ad.de>
References: <1ECE755A-84DA-42E1-B2CF-C44FB06DB5D5@aia.rwth-aachen.de>
 <CADwGrVqC=kUOUaDsa54r42VtroCbPE1X+-jgRPFzSw7vzz8Y6w@mail.gmail.com>
 <0c87e656-1c46-4333-8485-6178f35f463b@HUB2.rwth-ad.de>
 <CADwGrVqJnuuS-dw3G1bku3rPczKY-FuOb0QiDmYzraLw4xgFnQ@mail.gmail.com>
 <f04f9084-7221-4b2e-abd4-f78db7fdaf43@HUB2.rwth-ad.de>
Message-ID: <CAE32kpVcV6jLSN8SxyV2ny2YaHLZ7eVfKsbS+r6d49D4gAi1mA@mail.gmail.com>

Hi Michael,

If you already have the information handy, simply creating the
vtkGhostLevels array as Dan described is the way to go. We have plans on
writing a filter that generates ghost levels without redistributing the
entire dataset, which will hopefully scale better. Having said that if the
simulation has ghost levels, simply writing them out and marking them as
such will always scale better.

Best,
-berk

On Mon, Jul 6, 2015 at 2:36 PM, Schlottke-Lakemper, Michael <
m.schlottke-lakemper at aia.rwth-aachen.de> wrote:

>  Hi Dan,
>
>  Thanks a lot for this clarification - this should be enough information
> for us to get started.
>
>  Indeed, our reader plugin is designed to be used in a parallel fashion.
> Due to the size of the grid/data files, we typically use O(10)-O(1000) MPI
> ranks for pvserver. Currently we use the D3 filter if we need smooth
> gradients/statistics across domain boundaries, but the D3 filter does not
> scale well beyond a hundred or so cores.
>
>  Michael
>
>   On 06 Jul 2015, at 18:16 , Dan Lipsa <dan.lipsa at kitware.com> wrote:
>
>  Hi Michael,
> Yes, indeed this is how you would do it. I believe the array has to be
> unsigned char array though. On the first layer of ghost cells you store a
> 1, on the second layer you store a 2 and so on.
> (we just set a bit for the new version, so we don't store information
> about how many layers we have)
>
>  Also note that ghost cells are only useful if your reader reads
> different parts of the dataset on different mpi processors. If you split
> your dataset on different mpi nodes latter on, this is when you need to add
> ghost cells (because this is when you know how your data is split). If you
> run on a single processor ghosts cells are not used.
>
>  Dan
>
>
>  On Mon, Jul 6, 2015 at 11:43 AM Schlottke-Lakemper, Michael <
> m.schlottke-lakemper at aia.rwth-aachen.de> wrote:
>
>> Hi Dan,
>>
>>  Thank you for your answer. From a first impression, it seems like the
>> only really open question right now is how to generate the ?vtkGhostLevels?
>> array if I am working with an vtkUnstructuredGrid in my custom reader
>> plugin? ?GenerateGhostLevelArray? in vtkDataSet seems to be only relevant
>> for structured grids. Maybe I just need to create a vtkIntArray myself, add
>> it as a cell array, and name it ?vtkGhostLevels??
>>
>>  Regards,
>>
>>  Michael
>>
>>
>>  On 06 Jul 2015, at 17:10 , Dan Lipsa <dan.lipsa at kitware.com> wrote:
>>
>>  Hi Michael,
>> I am not aware of a guide to the old way of storing/accessing ghost cells
>> but I'll be happy to try to answer any specific questions you may have.
>> I can also add a link in the blog to the document if someone points it
>> out.
>>
>>  Dan
>>
>>
>>  On Sun, Jul 5, 2015 at 4:47 AM Schlottke-Lakemper, Michael <
>> m.schlottke-lakemper at aia.rwth-aachen.de> wrote:
>>
>>> Dear all,
>>>
>>>  this (http://www.kitware.com/blog/home/post/856) excellent blog post
>>> details how to use the vtkGhostLevels with the next ParaView release.
>>> However, it focuses on the changes in comparison to the previous way of
>>> doing things. Thus I am wondering if there is a comprehensive guide online
>>> that describes in a little more depth how to properly use ghost levels in
>>> the ?old? (or current as of PV 4.3.1) way?
>>>
>>>  Thanks a lot in advance
>>>
>>>  Michael
>>>
>>>
>>>        --
>>> Michael Schlottke-Lakemper
>>>
>>>  SimLab Highly Scalable Fluids & Solids Engineering
>>> J?lich Aachen Research Alliance (JARA-HPC)
>>> RWTH Aachen University
>>> W?llnerstra?e 5a
>>> 52062 Aachen
>>> Germany
>>>
>>>  Phone: +49 (241) 80 95188
>>> Fax: +49 (241) 80 92257
>>> Mail: m.schlottke-lakemper at aia.rwth-aachen.de
>>> Web: http://www.jara.org/jara-hpc
>>>
>>>  _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>
>>
>>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150709/0e17f53f/attachment.html>

From houssen at ipgp.fr  Fri Jul 10 10:55:02 2015
From: houssen at ipgp.fr (houssen)
Date: Fri, 10 Jul 2015 16:55:02 +0200
Subject: [Paraview-developers] How to add multiple inputs to a plugin
	(writer)
Message-ID: <e711e70f1ad39e5528a52408e34d7ad2@imap.ipgp.fr>

I have a plugin (writer which inherit from vtkWriter) with :
- SetNumberOfInputPorts(1) in the constructor
- vtkDataObject* data=GetInput() to get data to save (when I am called 
back on the save button click)
This works OK.

How to save "all what is in the pipeline browser" ? Say I opened 2 
files : I have 2 sets of data to save in the pipeline browser. How to 
save the 2 data set ?
I tried :
- SetNumberOfInputPorts(GetNumberOfInputPorts()) in the constructor
- for (int p = 0; p < GetNumberOfInputPorts() ...) vtkDataObject* 
data=GetInput(p) to get data to save
But this doesn't work : seems I always get one data to save (whatever 
what is hidden/shown or highlighted in the pipeline browser when I click 
on the save button)

FH

From cory.quammen at kitware.com  Fri Jul 10 11:35:37 2015
From: cory.quammen at kitware.com (Cory Quammen)
Date: Fri, 10 Jul 2015 10:35:37 -0500
Subject: [Paraview-developers] How to add multiple inputs to a plugin
	(writer)
In-Reply-To: <e711e70f1ad39e5528a52408e34d7ad2@imap.ipgp.fr>
References: <e711e70f1ad39e5528a52408e34d7ad2@imap.ipgp.fr>
Message-ID: <CAB5Fpx6_ntVcq-jZYVG_ynedXoJ81pvWonkWw1OYX=oGF+BZkA@mail.gmail.com>

Hi FH,

Do you want to save the entire ParaView state so that you can reload it
later? If so, you can use File -> Save State.../Load State...

I suspect that is not what you want, though.

Cory

On Fri, Jul 10, 2015 at 9:55 AM, houssen <houssen at ipgp.fr> wrote:

> I have a plugin (writer which inherit from vtkWriter) with :
> - SetNumberOfInputPorts(1) in the constructor
> - vtkDataObject* data=GetInput() to get data to save (when I am called
> back on the save button click)
> This works OK.
>
> How to save "all what is in the pipeline browser" ? Say I opened 2 files :
> I have 2 sets of data to save in the pipeline browser. How to save the 2
> data set ?
> I tried :
> - SetNumberOfInputPorts(GetNumberOfInputPorts()) in the constructor
> - for (int p = 0; p < GetNumberOfInputPorts() ...) vtkDataObject*
> data=GetInput(p) to get data to save
> But this doesn't work : seems I always get one data to save (whatever what
> is hidden/shown or highlighted in the pipeline browser when I click on the
> save button)
>
> FH
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>



-- 
Cory Quammen
R&D Engineer
Kitware, Inc.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150710/6882b0f8/attachment.html>

From cory.quammen at kitware.com  Fri Jul 10 11:47:48 2015
From: cory.quammen at kitware.com (Cory Quammen)
Date: Fri, 10 Jul 2015 10:47:48 -0500
Subject: [Paraview-developers] paraview : how to get the GUI name from a
 vtkObject ?
In-Reply-To: <7767480b495b9facdd845eaf5ad23563@imap.ipgp.fr>
References: <7767480b495b9facdd845eaf5ad23563@imap.ipgp.fr>
Message-ID: <CAB5Fpx7k+sGjrH9c+w02uyn9ixr=aVhfqN+PWDi_KR=T1L8Vzg@mail.gmail.com>

Hi Frank,

There is no direct way to do this. You would have to iterate over the all
the proxies in ParaView, call GetClientSideObject() on each proxy, and
compare the returned pointer to the pointer of the vtkObject you are
looking for.

Best,
Cory

On Sun, Jul 5, 2015 at 11:09 AM, houssen <houssen at ipgp.fr> wrote:

> From a vtkObject, how to get the name (alias) one can see / modify in the
> paraview GUI ? (name in the pipeline browser)
>
> Franck
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>



-- 
Cory Quammen
R&D Engineer
Kitware, Inc.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150710/43a6e0f7/attachment.html>

From houssen at ipgp.fr  Sat Jul 11 10:50:25 2015
From: houssen at ipgp.fr (houssen)
Date: Sat, 11 Jul 2015 16:50:25 +0200
Subject: [Paraview-developers] How to add multiple inputs to a plugin
 (writer)
In-Reply-To: <CAB5Fpx6_ntVcq-jZYVG_ynedXoJ81pvWonkWw1OYX=oGF+BZkA@mail.gmail.com>
References: <e711e70f1ad39e5528a52408e34d7ad2@imap.ipgp.fr>
 <CAB5Fpx6_ntVcq-jZYVG_ynedXoJ81pvWonkWw1OYX=oGF+BZkA@mail.gmail.com>
Message-ID: <42fd0ab6335ea099c94e416d6d732d87@imap.ipgp.fr>

Hi Cory,

My need is very basic (natural) : I'd like to save (with a plugin 
writer) "what I see in the ParaView 3D viewpoint".
Ideally, I'd would like to save "only shown items" in the pipeline 
browser (not the hidden items - gray eye associated to each pipeline 
item).

How to do this ? (use multiple input port ? get the pipeline and scan 
it piece by piece + write each piece only if it is in "show" mode) Is 
there an existing example of that ?

Franck

Le 2015-07-10 17:35, Cory Quammen a ?crit?:
> Hi FH,
>
> Do you want to save the entire ParaView state so that you can reload
> it later? If so, you can use File -> Save State.../Load State...
>
> I suspect that is not what you want, though.
>
> Cory
>
> On Fri, Jul 10, 2015 at 9:55 AM, houssen <houssen at ipgp.fr [5]> wrote:
>
>> I have a plugin (writer which inherit from vtkWriter) with :
>> - SetNumberOfInputPorts(1) in the constructor
>> - vtkDataObject* data=GetInput() to get data to save (when I am
>> called back on the save button click)
>> This works OK.
>>
>> How to save "all what is in the pipeline browser" ? Say I opened 2
>> files : I have 2 sets of data to save in the pipeline browser. How
>> to save the 2 data set ?
>> I tried :
>> - SetNumberOfInputPorts(GetNumberOfInputPorts()) in the
>> constructor
>> - for (int p = 0; p < GetNumberOfInputPorts() ...) vtkDataObject*
>> data=GetInput(p) to get data to save
>> But this doesn't work : seems I always get one data to save
>> (whatever what is hidden/shown or highlighted in the pipeline
>> browser when I click on the save button)
>>
>> FH
>> _______________________________________________
>> Powered by www.kitware.com [1]
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html [2]
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers [3]
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers [4]


From houssen at ipgp.fr  Sat Jul 11 11:18:08 2015
From: houssen at ipgp.fr (houssen)
Date: Sat, 11 Jul 2015 17:18:08 +0200
Subject: [Paraview-developers] ParaView plugin in MPI context
Message-ID: <f1a073f815c983db16f375aa0efc4336@imap.ipgp.fr>

I have a paraview plugin (writer) which works OK with a serial version 
of ParaView (built from source).
Is this code safe if used with a MPI version of Paraview ? If not, what 
kind of modifications (check, changes) do I have to do ? (I googled this 
but didn't really find clear answer)

FH

From cory.quammen at kitware.com  Sat Jul 11 17:19:41 2015
From: cory.quammen at kitware.com (Cory Quammen)
Date: Sat, 11 Jul 2015 17:19:41 -0400
Subject: [Paraview-developers] How to add multiple inputs to a plugin
	(writer)
In-Reply-To: <42fd0ab6335ea099c94e416d6d732d87@imap.ipgp.fr>
References: <e711e70f1ad39e5528a52408e34d7ad2@imap.ipgp.fr>
 <CAB5Fpx6_ntVcq-jZYVG_ynedXoJ81pvWonkWw1OYX=oGF+BZkA@mail.gmail.com>
 <42fd0ab6335ea099c94e416d6d732d87@imap.ipgp.fr>
Message-ID: <CAB5Fpx5eDYUu-p9nQXy_LAXbqgYAUr_mWjh6O7DT32Zx=rmkkA@mail.gmail.com>

Yes, look at File -> Export Scene.

You can export to WebGL, VRML, X3D, etc. One of these options should be a
good example for you to follow. The XML for the WebGL exporter proxy is in

ParaView/ParaViewCore/ServerManager/SMApplication/Resources/utilities.xml

and the exporter proxy is described here:

http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classvtkSMRenderViewExporterProxy.html

HTH,
Cory

On Sat, Jul 11, 2015 at 10:50 AM, houssen <houssen at ipgp.fr> wrote:

> Hi Cory,
>
> My need is very basic (natural) : I'd like to save (with a plugin writer)
> "what I see in the ParaView 3D viewpoint".
> Ideally, I'd would like to save "only shown items" in the pipeline browser
> (not the hidden items - gray eye associated to each pipeline item).
>
> How to do this ? (use multiple input port ? get the pipeline and scan it
> piece by piece + write each piece only if it is in "show" mode) Is there an
> existing example of that ?
>
> Franck
>
> Le 2015-07-10 17:35, Cory Quammen a ?crit :
>
>> Hi FH,
>>
>> Do you want to save the entire ParaView state so that you can reload
>> it later? If so, you can use File -> Save State.../Load State...
>>
>> I suspect that is not what you want, though.
>>
>> Cory
>>
>> On Fri, Jul 10, 2015 at 9:55 AM, houssen <houssen at ipgp.fr [5]> wrote:
>>
>>  I have a plugin (writer which inherit from vtkWriter) with :
>>> - SetNumberOfInputPorts(1) in the constructor
>>> - vtkDataObject* data=GetInput() to get data to save (when I am
>>> called back on the save button click)
>>> This works OK.
>>>
>>> How to save "all what is in the pipeline browser" ? Say I opened 2
>>> files : I have 2 sets of data to save in the pipeline browser. How
>>> to save the 2 data set ?
>>> I tried :
>>> - SetNumberOfInputPorts(GetNumberOfInputPorts()) in the
>>> constructor
>>> - for (int p = 0; p < GetNumberOfInputPorts() ...) vtkDataObject*
>>> data=GetInput(p) to get data to save
>>> But this doesn't work : seems I always get one data to save
>>> (whatever what is hidden/shown or highlighted in the pipeline
>>> browser when I click on the save button)
>>>
>>> FH
>>> _______________________________________________
>>> Powered by www.kitware.com [1]
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html [2]
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers [3]
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers [4]
>>>
>>
>


-- 
Cory Quammen
R&D Engineer
Kitware, Inc.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150711/a62ca47a/attachment.html>

From houssen at ipgp.fr  Sun Jul 12 06:08:55 2015
From: houssen at ipgp.fr (houssen)
Date: Sun, 12 Jul 2015 12:08:55 +0200
Subject: [Paraview-developers] Documenting plugin
Message-ID: <038c197758b4582ea7b5d4f6f21a1179@imap.ipgp.fr>

I have a plugin (writer). I added a documentation (index.thml + 
style.css) for it by adding this in the CMakeLists.txt :
ADD_PARAVIEW_PLUGIN(SurfaceLIC "1.0" DOCUMENTATION_DIR 
"${CMAKE_CURRENT_SOURCE_DIR}/doc" ...)
according to 
http://www.paraview.org/Wiki/ParaView/Plugin_HowTo#Adding_Documentation_for_Plugins.

When pressing F1, the documentation appears in the ParaView GUI but 
besides the "ParaView User Manual / writers" : is this expected ? Did I 
miss something to get the doc in the right place ?

Franck

From berk.geveci at kitware.com  Mon Jul 13 10:26:57 2015
From: berk.geveci at kitware.com (Berk Geveci)
Date: Mon, 13 Jul 2015 10:26:57 -0400
Subject: [Paraview-developers] Fwd: ANNOUNCE: Kitware is hiring
In-Reply-To: <CAE32kpUDB1uOzjKkv8nqNG4nK+YLE9eiz74A4Aejxz2gY7FO-A@mail.gmail.com>
References: <CAE32kpUDB1uOzjKkv8nqNG4nK+YLE9eiz74A4Aejxz2gY7FO-A@mail.gmail.com>
Message-ID: <CAE32kpUKyNkxvd1wZiqLak+idmTB4Dy6b1OQTdEGzKgJzpCWWw@mail.gmail.com>

Hi folks,

We are looking to hire visualization developers to our Scientific Computing
team. If you are a talented visualization researcher and developer with
strong C++ skills, please consider applying.  You will join a great team
and work on many interesting and challenging technical problems - always
aiming to deliver robust and widely used software solutions.

For the full posting see:
http://tinyurl.com/l8sgvzw

JOB DESCRIPTION
Kitware is seeking to hire highly skilled Research and Development
Engineers (R&D Engineers) to join our Scientific Computing team and
contribute to our scientific and information visualization efforts.
Candidates will work to develop and improve leading visualization software
solutions.

Kitware collaborates on a multitude of basic and applied research and
development projects. Our collaborators include the top universities from
around the world, national research labs, medical device manufacturers, car
manufacturers, oil and gas companies, financial institutes, and many
others. The projects range from extending our open source C++ libraries and
applications, such as VTK, ParaView, and CMake, to developing proprietary
domain-specific vertical applications for a wide array of platforms
including web and mobile devices.

By joining our team you will participate in a dynamic work environment with
exceptionally talented and friendly coworkers who are committed to
high-quality development practices.  You will collaborate with esteemed
researchers from around the world by:

* Designing and developing scalable data analysis and visualization tools
for use by  researchers and professionals from various domains;

* Solving a wide array of problems ranging from developing distributed
memory parallel algorithms for data analysis, optimizing distributed
parallel codes to compiling and maintaining software on supercomputers.

* Designing and developing tools to improve scientific data analysis
workflows;

* Contributing to and supporting our dynamic open source communities built
around several of our open source tools.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150713/acfd8c30/attachment.html>

From houssen at ipgp.fr  Mon Jul 13 10:58:43 2015
From: houssen at ipgp.fr (houssen)
Date: Mon, 13 Jul 2015 16:58:43 +0200
Subject: [Paraview-developers] Documenting plugin
In-Reply-To: <038c197758b4582ea7b5d4f6f21a1179@imap.ipgp.fr>
References: <038c197758b4582ea7b5d4f6f21a1179@imap.ipgp.fr>
Message-ID: <ba77f3fe8079fc99a28d1cc9fa969009@imap.ipgp.fr>

Seems the doc is OK and at the expected position.
How to tell to ADD_PARAVIEW_PLUGIN that my doc is only a couple of 
.html and .css file ? (I get messages about *.jpg and *.png that are not 
there)

Franck

Le 2015-07-12 12:08, houssen a ?crit?:
> I have a plugin (writer). I added a documentation (index.thml +
> style.css) for it by adding this in the CMakeLists.txt :
> ADD_PARAVIEW_PLUGIN(SurfaceLIC "1.0" DOCUMENTATION_DIR
> "${CMAKE_CURRENT_SOURCE_DIR}/doc" ...)
> according to
> 
> http://www.paraview.org/Wiki/ParaView/Plugin_HowTo#Adding_Documentation_for_Plugins.
>
> When pressing F1, the documentation appears in the ParaView GUI but
> besides the "ParaView User Manual / writers" : is this expected ? Did
> I miss something to get the doc in the right place ?
>
> Franck


From utkarsh.ayachit at kitware.com  Mon Jul 13 11:25:49 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 13 Jul 2015 15:25:49 +0000
Subject: [Paraview-developers] Documenting plugin
In-Reply-To: <ba77f3fe8079fc99a28d1cc9fa969009@imap.ipgp.fr>
References: <038c197758b4582ea7b5d4f6f21a1179@imap.ipgp.fr>
 <ba77f3fe8079fc99a28d1cc9fa969009@imap.ipgp.fr>
Message-ID: <CADHSJjC8L2FzQA95cxxOkPXykE9dwDhcweU+atmRcawVEx7S_w@mail.gmail.com>

Franck,

Those message should just be a warning. They can be ignored.

Utkarsh

On Mon, Jul 13, 2015 at 11:01 AM houssen <houssen at ipgp.fr> wrote:

> Seems the doc is OK and at the expected position.
> How to tell to ADD_PARAVIEW_PLUGIN that my doc is only a couple of
> .html and .css file ? (I get messages about *.jpg and *.png that are not
> there)
>
> Franck
>
> Le 2015-07-12 12:08, houssen a ?crit :
> > I have a plugin (writer). I added a documentation (index.thml +
> > style.css) for it by adding this in the CMakeLists.txt :
> > ADD_PARAVIEW_PLUGIN(SurfaceLIC "1.0" DOCUMENTATION_DIR
> > "${CMAKE_CURRENT_SOURCE_DIR}/doc" ...)
> > according to
> >
> >
> http://www.paraview.org/Wiki/ParaView/Plugin_HowTo#Adding_Documentation_for_Plugins
> .
> >
> > When pressing F1, the documentation appears in the ParaView GUI but
> > besides the "ParaView User Manual / writers" : is this expected ? Did
> > I miss something to get the doc in the right place ?
> >
> > Franck
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150713/6e81b81d/attachment.html>

From ben.boeckel at kitware.com  Tue Jul 14 09:22:50 2015
From: ben.boeckel at kitware.com (Ben Boeckel)
Date: Tue, 14 Jul 2015 09:22:50 -0400
Subject: [Paraview-developers] Buildbot now scanning for build warnings
Message-ID: <20150714132250.GA14351@megas.kitware.com>

Hi,

As you may have noticed, buildbot will now be coloring the grid yellow
where build warnings occur. Now it should be even easier to keep the
warnings column in CDash green :) .

--Ben

From houssen at ipgp.fr  Tue Jul 14 17:38:15 2015
From: houssen at ipgp.fr (houssen)
Date: Tue, 14 Jul 2015 23:38:15 +0200
Subject: [Paraview-developers] =?utf-8?q?How_to_implement_an_identity_filt?=
	=?utf-8?q?er_=3F?=
Message-ID: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>

How to write an identity filter ? (1 input port, 1 output port, output 
= input whatever input may be)

I need to write a custom filter. I am used to VTK "from outside" 
(creating/handling vtk pipelines/scene/object from a main program) that 
is basically without the need to understand all internal VTK mechanics 
(except the SetInputConnection concept). I am not yet used to VTK 
internal mechanics that is necessary for writing a ParaView plugin.

I need my filter to do "complex" stuffs. As I couldn't make it work, I 
simplified as much as possible to obtain finally an identity filter (!) 
: this is still not working. I feel I still miss something from VTK 
internal mechanics.

As far as I understand, in the constructor, I need :
myFilter::myFilter () { SetNumberOfInputPorts  ( 1 ); 
SetNumberOfOutputPorts ( 1 ); }

So then I need to implement the FillXXXPortInformation :
int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ), 
vtkInformation * ipInfo )
{
   if ( ipInfo ) ipInfo -> Set ( vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE 
(), "vtkDataObject" ); // Filter allowed data types
   return 1;
}
int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort ), 
vtkInformation * ipInfo )
{
   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (), 
"vtkDataObject" ); // Output data type to be created
   return 1;
}

So, now I only need to implement the ProcessRequest gateway:
int myFilter::ProcessRequest ( vtkInformation * ipRequest, 
vtkInformationVector ** ipInputVector, vtkInformationVector * 
opOutputVector )
{
   if ( ipRequest && ipRequest -> Has ( 
vtkDemandDrivenPipeline::REQUEST_DATA () ) )
     if ( ipInputVector && opOutputVector ) opOutputVector = 
ipInputVector[0];
    return Superclass::ProcessRequest ( ipRequest, ipInputVector, 
opOutputVector );
}

I believed that I would be the only steps to proceed to get an identity 
filter. When I use my identity filter in Paraview, I get errors like :
ERROR: In 
/.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx, 
line 810
vtkPVPostFilterExecutive : Input for connection index 0 on input port 
index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type 
vtkDataObject, but a vtkPolyData is required.

Some logic is still missing here : could somebody explain me what and 
why ?

Thanks,

Franck


From mathieu.westphal at kitware.com  Wed Jul 15 01:21:44 2015
From: mathieu.westphal at kitware.com (Mathieu Westphal)
Date: Wed, 15 Jul 2015 07:21:44 +0200
Subject: [Paraview-developers] How to implement an identity filter ?
In-Reply-To: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
Message-ID: <CAC2t4UQHrkiQnHWszdBGDH=mf80UseS4Hkm4S42N-Apaf6pfLQ@mail.gmail.com>

Du you inherit from vtkPolydataAlgorithm ? You should inherit from
vtkAlgorithm .
 On 14 Jul 2015 23:40, "houssen" <houssen at ipgp.fr> wrote:

> How to write an identity filter ? (1 input port, 1 output port, output =
> input whatever input may be)
>
> I need to write a custom filter. I am used to VTK "from outside"
> (creating/handling vtk pipelines/scene/object from a main program) that is
> basically without the need to understand all internal VTK mechanics (except
> the SetInputConnection concept). I am not yet used to VTK internal
> mechanics that is necessary for writing a ParaView plugin.
>
> I need my filter to do "complex" stuffs. As I couldn't make it work, I
> simplified as much as possible to obtain finally an identity filter (!) :
> this is still not working. I feel I still miss something from VTK internal
> mechanics.
>
> As far as I understand, in the constructor, I need :
> myFilter::myFilter () { SetNumberOfInputPorts  ( 1 );
> SetNumberOfOutputPorts ( 1 ); }
>
> So then I need to implement the FillXXXPortInformation :
> int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ),
> vtkInformation * ipInfo )
> {
>   if ( ipInfo ) ipInfo -> Set ( vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (),
> "vtkDataObject" ); // Filter allowed data types
>   return 1;
> }
> int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort ),
> vtkInformation * ipInfo )
> {
>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
> "vtkDataObject" ); // Output data type to be created
>   return 1;
> }
>
> So, now I only need to implement the ProcessRequest gateway:
> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
> vtkInformationVector ** ipInputVector, vtkInformationVector *
> opOutputVector )
> {
>   if ( ipRequest && ipRequest -> Has (
> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>     if ( ipInputVector && opOutputVector ) opOutputVector =
> ipInputVector[0];
>    return Superclass::ProcessRequest ( ipRequest, ipInputVector,
> opOutputVector );
> }
>
> I believed that I would be the only steps to proceed to get an identity
> filter. When I use my identity filter in Paraview, I get errors like :
> ERROR: In
> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
> line 810
> vtkPVPostFilterExecutive : Input for connection index 0 on input port
> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type vtkDataObject,
> but a vtkPolyData is required.
>
> Some logic is still missing here : could somebody explain me what and why ?
>
> Thanks,
>
> Franck
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150715/b34f278f/attachment.html>

From houssen at ipgp.fr  Wed Jul 15 04:13:25 2015
From: houssen at ipgp.fr (houssen)
Date: Wed, 15 Jul 2015 10:13:25 +0200
Subject: [Paraview-developers]
 =?utf-8?q?How_to_implement_an_identity_filt?= =?utf-8?q?er_=3F?=
In-Reply-To: <CAC2t4UQHrkiQnHWszdBGDH=mf80UseS4Hkm4S42N-Apaf6pfLQ@mail.gmail.com>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <CAC2t4UQHrkiQnHWszdBGDH=mf80UseS4Hkm4S42N-Apaf6pfLQ@mail.gmail.com>
Message-ID: <c95e10a69315f9a2cfd322089db69642@imap.ipgp.fr>

Yes, myFilter inherits from vtkAlgorithm and re-implements only 
FillInputPortInformation, FillOutputPortInformation, ProcessRequest as 
described below.

Le 2015-07-15 7:21, Mathieu Westphal a ?crit?:
> Du you inherit from vtkPolydataAlgorithm ? You should inherit from
> vtkAlgorithm .
>
> On 14 Jul 2015 23:40, "houssen" <houssen at ipgp.fr [5]> wrote:
>
>> How to write an identity filter ? (1 input port, 1 output port,
>> output = input whatever input may be)
>>
>> I need to write a custom filter. I am used to VTK "from outside"
>> (creating/handling vtk pipelines/scene/object from a main program)
>> that is basically without the need to understand all internal VTK
>> mechanics (except the SetInputConnection concept). I am not yet
> used
>> to VTK internal mechanics that is necessary for writing a ParaView
>> plugin.
>>
>> I need my filter to do "complex" stuffs. As I couldn't make it
>> work, I simplified as much as possible to obtain finally an
> identity
>> filter (!) : this is still not working. I feel I still miss
>> something from VTK internal mechanics.
>>
>> As far as I understand, in the constructor, I need :
>> myFilter::myFilter () { SetNumberOfInputPorts? ( 1 );
>> SetNumberOfOutputPorts ( 1 ); }
>>
>> So then I need to implement the FillXXXPortInformation :
>> int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ),
>> vtkInformation * ipInfo )
>> {
>> ? if ( ipInfo ) ipInfo -> Set (
>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" ); //
>> Filter allowed data types
>> ? return 1;
>> }
>> int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort
>> ), vtkInformation * ipInfo )
>> {
>> ? if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
>> "vtkDataObject" ); // Output data type to be created
>> ? return 1;
>> }
>>
>> So, now I only need to implement the ProcessRequest gateway:
>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>> opOutputVector )
>> {
>> ? if ( ipRequest && ipRequest -> Has (
>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>> ? ? if ( ipInputVector && opOutputVector ) opOutputVector =
>> ipInputVector[0];
>> ? ?return Superclass::ProcessRequest ( ipRequest, ipInputVector,
>> opOutputVector );
>> }
>>
>> I believed that I would be the only steps to proceed to get an
>> identity filter. When I use my identity filter in Paraview, I get
>> errors like :
>> ERROR: In
>>
> 
> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>> line 810
>> vtkPVPostFilterExecutive : Input for connection index 0 on input
>> port index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>> vtkDataObject, but a vtkPolyData is required.
>>
>> Some logic is still missing here : could somebody explain me what
>> and why ?
>>
>> Thanks,
>>
>> Franck
>>
>> _______________________________________________
>> Powered by www.kitware.com [1]
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html [2]
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers [3]
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers [4]
>
>
> Links:
> ------
> [1] http://www.kitware.com
> [2] http://www.kitware.com/opensource/opensource.html
> [3] http://markmail.org/search/?q=Paraview-developers
> [4] http://public.kitware.com/mailman/listinfo/paraview-developers
> [5] mailto:houssen at ipgp.fr


From houssen at ipgp.fr  Wed Jul 15 05:57:50 2015
From: houssen at ipgp.fr (houssen)
Date: Wed, 15 Jul 2015 11:57:50 +0200
Subject: [Paraview-developers]
 =?utf-8?q?How_to_implement_an_identity_filt?= =?utf-8?q?er_=3F?=
In-Reply-To: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
Message-ID: <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>

Seems that it is not possible to filter on vtkDataObject type.
But with less generic types (like vtkUnstrucutredGrid for instance), 
the myFilter works : no error occur, but, viewpoint ends up to be empty. 
I get "zero" instead of "identity" : why ?!

Franck

Le 2015-07-14 23:38, houssen a ?crit?:
> How to write an identity filter ? (1 input port, 1 output port,
> output = input whatever input may be)
>
> I need to write a custom filter. I am used to VTK "from outside"
> (creating/handling vtk pipelines/scene/object from a main program)
> that is basically without the need to understand all internal VTK
> mechanics (except the SetInputConnection concept). I am not yet used
> to VTK internal mechanics that is necessary for writing a ParaView
> plugin.
>
> I need my filter to do "complex" stuffs. As I couldn't make it work,
> I simplified as much as possible to obtain finally an identity filter
> (!) : this is still not working. I feel I still miss something from
> VTK internal mechanics.
>
> As far as I understand, in the constructor, I need :
> myFilter::myFilter () { SetNumberOfInputPorts  ( 1 );
> SetNumberOfOutputPorts ( 1 ); }
>
> So then I need to implement the FillXXXPortInformation :
> int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ),
> vtkInformation * ipInfo )
> {
>   if ( ipInfo ) ipInfo -> Set (
> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" ); //
> Filter allowed data types
>   return 1;
> }
> int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort ),
> vtkInformation * ipInfo )
> {
>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
> "vtkDataObject" ); // Output data type to be created
>   return 1;
> }
>
> So, now I only need to implement the ProcessRequest gateway:
> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
> vtkInformationVector ** ipInputVector, vtkInformationVector *
> opOutputVector )
> {
>   if ( ipRequest && ipRequest -> Has (
> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>     if ( ipInputVector && opOutputVector ) opOutputVector = 
> ipInputVector[0];
>    return Superclass::ProcessRequest ( ipRequest, ipInputVector,
> opOutputVector );
> }
>
> I believed that I would be the only steps to proceed to get an
> identity filter. When I use my identity filter in Paraview, I get
> errors like :
> ERROR: In
> 
> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
> line 810
> vtkPVPostFilterExecutive : Input for connection index 0 on input port
> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
> vtkDataObject, but a vtkPolyData is required.
>
> Some logic is still missing here : could somebody explain me what and 
> why ?
>
> Thanks,
>
> Franck
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers


From joachim.pouderoux at kitware.com  Wed Jul 15 07:27:23 2015
From: joachim.pouderoux at kitware.com (Joachim Pouderoux)
Date: Wed, 15 Jul 2015 13:27:23 +0200
Subject: [Paraview-developers] How to implement an identity filter ?
In-Reply-To: <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
Message-ID: <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>

Franck,

You will have to override the RequestDataObject() method in order to make
your filter generic. This method is the one called to generate the output
data object.
Take a look at vtkGenerateIndexArray.cxx for instance, I think you can just
copy it in your class (which should be derivated from
vtkDataObjectAlgorithm).

Best,


*Joachim Pouderoux*

*PhD, Technical Expert*
*Kitware SAS <http://www.kitware.fr>*


2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr>:

> Seems that it is not possible to filter on vtkDataObject type.
> But with less generic types (like vtkUnstrucutredGrid for instance), the
> myFilter works : no error occur, but, viewpoint ends up to be empty. I get
> "zero" instead of "identity" : why ?!
>
> Franck
>
>
> Le 2015-07-14 23:38, houssen a ?crit :
>
>> How to write an identity filter ? (1 input port, 1 output port,
>> output = input whatever input may be)
>>
>> I need to write a custom filter. I am used to VTK "from outside"
>> (creating/handling vtk pipelines/scene/object from a main program)
>> that is basically without the need to understand all internal VTK
>> mechanics (except the SetInputConnection concept). I am not yet used
>> to VTK internal mechanics that is necessary for writing a ParaView
>> plugin.
>>
>> I need my filter to do "complex" stuffs. As I couldn't make it work,
>> I simplified as much as possible to obtain finally an identity filter
>> (!) : this is still not working. I feel I still miss something from
>> VTK internal mechanics.
>>
>> As far as I understand, in the constructor, I need :
>> myFilter::myFilter () { SetNumberOfInputPorts  ( 1 );
>> SetNumberOfOutputPorts ( 1 ); }
>>
>> So then I need to implement the FillXXXPortInformation :
>> int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ),
>> vtkInformation * ipInfo )
>> {
>>   if ( ipInfo ) ipInfo -> Set (
>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" ); //
>> Filter allowed data types
>>   return 1;
>> }
>> int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort ),
>> vtkInformation * ipInfo )
>> {
>>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
>> "vtkDataObject" ); // Output data type to be created
>>   return 1;
>> }
>>
>> So, now I only need to implement the ProcessRequest gateway:
>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>> opOutputVector )
>> {
>>   if ( ipRequest && ipRequest -> Has (
>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>     if ( ipInputVector && opOutputVector ) opOutputVector =
>> ipInputVector[0];
>>    return Superclass::ProcessRequest ( ipRequest, ipInputVector,
>> opOutputVector );
>> }
>>
>> I believed that I would be the only steps to proceed to get an
>> identity filter. When I use my identity filter in Paraview, I get
>> errors like :
>> ERROR: In
>>
>>
>> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>> line 810
>> vtkPVPostFilterExecutive : Input for connection index 0 on input port
>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>> vtkDataObject, but a vtkPolyData is required.
>>
>> Some logic is still missing here : could somebody explain me what and why
>> ?
>>
>> Thanks,
>>
>> Franck
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150715/d50d736d/attachment-0001.html>

From shawn.waldon at kitware.com  Wed Jul 15 09:47:36 2015
From: shawn.waldon at kitware.com (Shawn Waldon)
Date: Wed, 15 Jul 2015 09:47:36 -0400
Subject: [Paraview-developers] How to implement an identity filter ?
In-Reply-To: <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
 <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
Message-ID: <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>

You might want to take a look at inheriting from
vtkPassInputTypeAlgorithm[1] if you want the output type to be the same as
the input type.  It already implements RequestDataObject the way that I
think Joachim is suggesting.

HTH,
Shawn

[1]: http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html

On Wed, Jul 15, 2015 at 7:27 AM, Joachim Pouderoux <
joachim.pouderoux at kitware.com> wrote:

> Franck,
>
> You will have to override the RequestDataObject() method in order to make
> your filter generic. This method is the one called to generate the output
> data object.
> Take a look at vtkGenerateIndexArray.cxx for instance, I think you can
> just copy it in your class (which should be derivated from
> vtkDataObjectAlgorithm).
>
> Best,
>
>
> *Joachim Pouderoux*
>
> *PhD, Technical Expert*
> *Kitware SAS <http://www.kitware.fr>*
>
>
> 2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr>:
>
>> Seems that it is not possible to filter on vtkDataObject type.
>> But with less generic types (like vtkUnstrucutredGrid for instance), the
>> myFilter works : no error occur, but, viewpoint ends up to be empty. I get
>> "zero" instead of "identity" : why ?!
>>
>> Franck
>>
>>
>> Le 2015-07-14 23:38, houssen a ?crit :
>>
>>> How to write an identity filter ? (1 input port, 1 output port,
>>> output = input whatever input may be)
>>>
>>> I need to write a custom filter. I am used to VTK "from outside"
>>> (creating/handling vtk pipelines/scene/object from a main program)
>>> that is basically without the need to understand all internal VTK
>>> mechanics (except the SetInputConnection concept). I am not yet used
>>> to VTK internal mechanics that is necessary for writing a ParaView
>>> plugin.
>>>
>>> I need my filter to do "complex" stuffs. As I couldn't make it work,
>>> I simplified as much as possible to obtain finally an identity filter
>>> (!) : this is still not working. I feel I still miss something from
>>> VTK internal mechanics.
>>>
>>> As far as I understand, in the constructor, I need :
>>> myFilter::myFilter () { SetNumberOfInputPorts  ( 1 );
>>> SetNumberOfOutputPorts ( 1 ); }
>>>
>>> So then I need to implement the FillXXXPortInformation :
>>> int myFilter::FillInputPortInformation ( int vtkNotUsed ( iPort ),
>>> vtkInformation * ipInfo )
>>> {
>>>   if ( ipInfo ) ipInfo -> Set (
>>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" ); //
>>> Filter allowed data types
>>>   return 1;
>>> }
>>> int myFilter::FillOutputPortInformation ( int vtkNotUsed ( iPort ),
>>> vtkInformation * ipInfo )
>>> {
>>>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
>>> "vtkDataObject" ); // Output data type to be created
>>>   return 1;
>>> }
>>>
>>> So, now I only need to implement the ProcessRequest gateway:
>>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>>> opOutputVector )
>>> {
>>>   if ( ipRequest && ipRequest -> Has (
>>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>>     if ( ipInputVector && opOutputVector ) opOutputVector =
>>> ipInputVector[0];
>>>    return Superclass::ProcessRequest ( ipRequest, ipInputVector,
>>> opOutputVector );
>>> }
>>>
>>> I believed that I would be the only steps to proceed to get an
>>> identity filter. When I use my identity filter in Paraview, I get
>>> errors like :
>>> ERROR: In
>>>
>>>
>>> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>> line 810
>>> vtkPVPostFilterExecutive : Input for connection index 0 on input port
>>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>>> vtkDataObject, but a vtkPolyData is required.
>>>
>>> Some logic is still missing here : could somebody explain me what and
>>> why ?
>>>
>>> Thanks,
>>>
>>> Franck
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>
>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150715/0c53a7fa/attachment.html>

From houssen at ipgp.fr  Wed Jul 15 17:49:56 2015
From: houssen at ipgp.fr (houssen)
Date: Wed, 15 Jul 2015 23:49:56 +0200
Subject: [Paraview-developers]
 =?utf-8?q?How_to_implement_an_identity_filt?= =?utf-8?q?er_=3F?=
In-Reply-To: <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
 <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
 <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>
Message-ID: <535c216abf12af284f43241d3465516d@imap.ipgp.fr>

Shawn, Joachim, thanks.
After a quick look at the code, if I got you right, on a 
REQUEST_DATA_OBJECT request, I still need to create a new instance of 
the input and to set it as the DATA_OBJECT of the vtkInfo of the output. 
I still have to test this, but, I got the idea !

Franck

Le 2015-07-15 15:47, Shawn Waldon a ?crit?:
> You might want to take a look at inheriting from
> vtkPassInputTypeAlgorithm[1] if you want the output type to be the
> same as the input type.? It already implements RequestDataObject the
> way that I think Joachim is suggesting.
>
> HTH,
> Shawn
>
> [1]:
> 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
> [14]
>
> On Wed, Jul 15, 2015 at 7:27 AM, Joachim Pouderoux
> <joachim.pouderoux at kitware.com [15]> wrote:
>
>> Franck,
>>
>> You will have to override the RequestDataObject() method in order
>> to make your filter generic. This method is the one called to
>> generate the output data object.
>> Take a look at vtkGenerateIndexArray.cxx for instance, I think you
>> can just copy it in your class (which should be derivated from
>> vtkDataObjectAlgorithm).
>>
>> Best,
>>
>> JOACHIM POUDEROUX
>>
>>> _PhD, Technical Expert_
>>
>> 2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr [9]>:
>>
>>> Seems that it is not possible to filter on vtkDataObject type.
>>> But with less generic types (like vtkUnstrucutredGrid for
>>> instance), the myFilter works : no error occur, but, viewpoint
>>> ends up to be empty. I get "zero" instead of "identity" : why ?!
>>>
>>> Franck
>>>
>>> Le 2015-07-14 23:38, houssen a ?crit?:
>>>
>>>> How to write an identity filter ? (1 input port, 1 output port,
>>>> output = input whatever input may be)
>>>>
>>>> I need to write a custom filter. I am used to VTK "from
>>>> outside"
>>>> (creating/handling vtk pipelines/scene/object from a main
>>>> program)
>>>> that is basically without the need to understand all internal
>>>> VTK
>>>> mechanics (except the SetInputConnection concept). I am not
>>>> yet used
>>>> to VTK internal mechanics that is necessary for writing a
>>>> ParaView
>>>> plugin.
>>>>
>>>> I need my filter to do "complex" stuffs. As I couldn't make it
>>>> work,
>>>> I simplified as much as possible to obtain finally an identity
>>>> filter
>>>> (!) : this is still not working. I feel I still miss something
>>>> from
>>>> VTK internal mechanics.
>>>>
>>>> As far as I understand, in the constructor, I need :
>>>> myFilter::myFilter () { SetNumberOfInputPorts? ( 1 );
>>>> SetNumberOfOutputPorts ( 1 ); }
>>>>
>>>> So then I need to implement the FillXXXPortInformation :
>>>> int myFilter::FillInputPortInformation ( int vtkNotUsed (
>>>> iPort ),
>>>> vtkInformation * ipInfo )
>>>> {
>>>> ? if ( ipInfo ) ipInfo -> Set (
>>>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" );
>>>> //
>>>> Filter allowed data types
>>>> ? return 1;
>>>> }
>>>> int myFilter::FillOutputPortInformation ( int vtkNotUsed (
>>>> iPort ),
>>>> vtkInformation * ipInfo )
>>>> {
>>>> ? if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME
>>>> (),
>>>> "vtkDataObject" ); // Output data type to be created
>>>> ? return 1;
>>>> }
>>>>
>>>> So, now I only need to implement the ProcessRequest gateway:
>>>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>>>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>>>> opOutputVector )
>>>> {
>>>> ? if ( ipRequest && ipRequest -> Has (
>>>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>>> ? ? if ( ipInputVector && opOutputVector ) opOutputVector =
>>>> ipInputVector[0];
>>>> ? ?return Superclass::ProcessRequest ( ipRequest,
>>>> ipInputVector,
>>>> opOutputVector );
>>>> }
>>>>
>>>> I believed that I would be the only steps to proceed to get an
>>>> identity filter. When I use my identity filter in Paraview, I
>>>> get
>>>> errors like :
>>>> ERROR: In
>>>>
>>>>
>>>>
>>>
>>
> 
> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>> line 810
>>>> vtkPVPostFilterExecutive : Input for connection index 0 on
>>>> input port
>>>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>>>> vtkDataObject, but a vtkPolyData is required.
>>>>
>>>> Some logic is still missing here : could somebody explain me
>>>> what and why ?
>>>>
>>>> Thanks,
>>>>
>>>> Franck
>>>>
>>>> _______________________________________________
>>>> Powered by www.kitware.com [1]
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html [2]
>>>>
>>>> Search the list archives at:
>>>> http://markmail.org/search/?q=Paraview-developers [3]
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [4]
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com [5]
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html [6]
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers [7]
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>> [8]
>>
>> _______________________________________________
>> Powered by www.kitware.com [10]
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html [11]
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers [12]
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>> [13]
>
>
>
> Links:
> ------
> [1] http://www.kitware.com
> [2] http://www.kitware.com/opensource/opensource.html
> [3] http://markmail.org/search/?q=Paraview-developers
> [4] http://public.kitware.com/mailman/listinfo/paraview-developers
> [5] http://www.kitware.com
> [6] http://www.kitware.com/opensource/opensource.html
> [7] http://markmail.org/search/?q=Paraview-developers
> [8] http://public.kitware.com/mailman/listinfo/paraview-developers
> [9] mailto:houssen at ipgp.fr
> [10] http://www.kitware.com
> [11] http://www.kitware.com/opensource/opensource.html
> [12] http://markmail.org/search/?q=Paraview-developers
> [13] http://public.kitware.com/mailman/listinfo/paraview-developers
> [14] 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
> [15] mailto:joachim.pouderoux at kitware.com


From chuck.atkins at kitware.com  Thu Jul 16 11:51:40 2015
From: chuck.atkins at kitware.com (Chuck Atkins)
Date: Thu, 16 Jul 2015 11:51:40 -0400
Subject: [Paraview-developers] Paraview Runtime Segfault when linking
 with external c++ library
In-Reply-To: <E493A24A2D2E32408F6F053DB5082EDD319167B3@INBLCORMX01.quest-global.com>
References: <E493A24A2D2E32408F6F053DB5082EDD319167B3@INBLCORMX01.quest-global.com>
Message-ID: <CAOXZzG4tiUyU+p5nJ+_S9tig8TsZM26DXCCuNd_4HBkSoeYVxA@mail.gmail.com>

Hi Prashant,

 Is it a linker problem because my external library is compiled through
> ICPC(intel c++) but Paraview is compiled with G++?
>

Does this happen for both a debug build and release build of your external
library? If it only happens in the release build, try making sure that
you're using the Intel likers since they handle optimizations, specifically
IPO and LTO, in a very specific and coordinating way that the GNU linkers
are unaware of.  You can do this by setting the environment variables
LD=xild and AR=xiar prior to running CMake.  Or if you're not using CMake
then be sure to use Intel linkers xild and xiar in place of the GNU linkers
ld and ar respectively for whatever build tool you are using.

- Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150716/08ef1e3c/attachment.html>

From cgrow4 at gmail.com  Thu Jul 16 17:01:01 2015
From: cgrow4 at gmail.com (Charles Grow)
Date: Thu, 16 Jul 2015 14:01:01 -0700 (PDT)
Subject: [Paraview-developers] Paraview Adios
Message-ID: <1437080461110-3893.post@n6.nabble.com>

Hello,

        I am having trouble with the adios plugin and Paraview, I have
compiled the plugin and it loads properly, but when I load in some data,
Paraview crashes with error:

ERROR: In
/Users/cgrow/Projects/paraview-new/ParaView/ParaViewCore/ServerImplementation/Core/vtkSIProxy.cxx,
line 307
vtkSISourceProxy (0x7fde81c8f5f0): Failed to create vtkAdiosPixieReader.
Aborting for debugging purposes.


I have tried both the example data in adios, as well as some example data I
found while searching for info on adios and Paraview, both result in the
same crash.

Using Paraview 4.3.1 

Any ideas?

Thanks,
Charles



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-Adios-tp3893.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From sebastien.jourdain at kitware.com  Thu Jul 16 17:51:35 2015
From: sebastien.jourdain at kitware.com (Sebastien Jourdain)
Date: Thu, 16 Jul 2015 15:51:35 -0600
Subject: [Paraview-developers] Paraview Adios
In-Reply-To: <1437080461110-3893.post@n6.nabble.com>
References: <1437080461110-3893.post@n6.nabble.com>
Message-ID: <CABObKxc_GUcXcLC6Gk9NuXtgH2aEMceAx-L-1yOrDreQ0T+-EQ@mail.gmail.com>

Hi Charles,

do you know what kind of Adios data are you trying to read?
If I'm not mistaking that plugin is pretty old and was meant at beginning
of Adios to try out in-transit setup for Supercomputing.
So it might be possible that the pixie format has changed a little bit as
well as the version of adios which probably has a slightly different API
now or expectation.

None the less, I think that you must run ParaView with MPI in order for
Adios to be properly initialized.
Try to run ParaView with --mpi as argument or in client server mode using
MPI.

Let us know if you managed to read your data.

Seb

On Thu, Jul 16, 2015 at 3:01 PM, Charles Grow <cgrow4 at gmail.com> wrote:

> Hello,
>
>         I am having trouble with the adios plugin and Paraview, I have
> compiled the plugin and it loads properly, but when I load in some data,
> Paraview crashes with error:
>
> ERROR: In
>
> /Users/cgrow/Projects/paraview-new/ParaView/ParaViewCore/ServerImplementation/Core/vtkSIProxy.cxx,
> line 307
> vtkSISourceProxy (0x7fde81c8f5f0): Failed to create vtkAdiosPixieReader.
> Aborting for debugging purposes.
>
>
> I have tried both the example data in adios, as well as some example data I
> found while searching for info on adios and Paraview, both result in the
> same crash.
>
> Using Paraview 4.3.1
>
> Any ideas?
>
> Thanks,
> Charles
>
>
>
> --
> View this message in context:
> http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-Adios-tp3893.html
> Sent from the The Unofficial ParaView Developers Forum mailing list
> archive at Nabble.com.
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150716/021bbae7/attachment.html>

From cgrow4 at gmail.com  Thu Jul 16 17:56:35 2015
From: cgrow4 at gmail.com (Charles Grow)
Date: Thu, 16 Jul 2015 14:56:35 -0700 (PDT)
Subject: [Paraview-developers] Paraview Adios
In-Reply-To: <CABObKxc_GUcXcLC6Gk9NuXtgH2aEMceAx-L-1yOrDreQ0T+-EQ@mail.gmail.com>
References: <1437080461110-3893.post@n6.nabble.com>
 <CABObKxc_GUcXcLC6Gk9NuXtgH2aEMceAx-L-1yOrDreQ0T+-EQ@mail.gmail.com>
Message-ID: <1437083795925-3895.post@n6.nabble.com>

I tried launching Paraview with mpi, and it still crashes.

I am using the example data from this thread:

        http://www.paraview.org/pipermail/paraview/2013-January/027330.html

Do you know if there are any other adios related plugins for Paraview?

Thanks,
Charles



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-Adios-tp3893p3895.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From sebastien.jourdain at kitware.com  Thu Jul 16 18:47:43 2015
From: sebastien.jourdain at kitware.com (Sebastien Jourdain)
Date: Thu, 16 Jul 2015 16:47:43 -0600
Subject: [Paraview-developers] Paraview Adios
In-Reply-To: <1437083795925-3895.post@n6.nabble.com>
References: <1437080461110-3893.post@n6.nabble.com>
 <CABObKxc_GUcXcLC6Gk9NuXtgH2aEMceAx-L-1yOrDreQ0T+-EQ@mail.gmail.com>
 <1437083795925-3895.post@n6.nabble.com>
Message-ID: <CABObKxdCSvjLH56dPR-h=MSS7K3YnukmSKEhn8KRonDsrPm47Q@mail.gmail.com>

Hum,

I guess that should have worked. Reading the error again, it seems that the
CS wrapping is not properly done and the server can not instantiate that
class.
I'm wondering if Utkarsh could give a quick look and see if he doesn't see
anything obvious in the CMake code of the plugin that could explain it.

For reading pixie data with Adios, I don't think we have anything else.
I know that some other works have been done using Adios but for other data
format.
I'm not aware of any progress regarding pixie.

Seb

On Thu, Jul 16, 2015 at 3:56 PM, Charles Grow <cgrow4 at gmail.com> wrote:

> I tried launching Paraview with mpi, and it still crashes.
>
> I am using the example data from this thread:
>
>
> http://www.paraview.org/pipermail/paraview/2013-January/027330.html
>
> Do you know if there are any other adios related plugins for Paraview?
>
> Thanks,
> Charles
>
>
>
> --
> View this message in context:
> http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-Adios-tp3893p3895.html
> Sent from the The Unofficial ParaView Developers Forum mailing list
> archive at Nabble.com.
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150716/f2b95afb/attachment.html>

From joachim.pouderoux at kitware.com  Fri Jul 17 04:42:24 2015
From: joachim.pouderoux at kitware.com (Joachim Pouderoux)
Date: Fri, 17 Jul 2015 10:42:24 +0200
Subject: [Paraview-developers] Paraview Adios
In-Reply-To: <1437080461110-3893.post@n6.nabble.com>
References: <1437080461110-3893.post@n6.nabble.com>
Message-ID: <CACJ=gThe85y77eRy+9j=DoJxzmRb1N5JohXg4ckzOa5N73bcKQ@mail.gmail.com>

Charles,

Beware, there are two Adios file readers in ParaView.
The ADIOS Pixie reader (that seems to be used here but is it what you want)
to load Pixie Image data: vtkAdiosPixieReader.
This reader is enabled when you enable the
PARAVIEW_BUILD_PLUGIN_AdiosReader in ParaView's CMake.

Then there are recent ADIOS writers/readers for VTK data structures
(developed last year by Chuck Atkins at Kitware): vtkADIOSReader.
Those one will be available in ParaView if you enable the Module_vtkIOADIOS.
But note that this reader is not a universal reader, it will read ADIOS
files generated by the corresponding ADIOS writer (vtkADIOSWriter).

There is no other ADIOS reader in ParaView.

Hope it helps,

Joachim

*Joachim Pouderoux*

*PhD, Technical Expert*
*Kitware SAS <http://www.kitware.fr>*


2015-07-16 23:01 GMT+02:00 Charles Grow <cgrow4 at gmail.com>:

> Hello,
>
>         I am having trouble with the adios plugin and Paraview, I have
> compiled the plugin and it loads properly, but when I load in some data,
> Paraview crashes with error:
>
> ERROR: In
>
> /Users/cgrow/Projects/paraview-new/ParaView/ParaViewCore/ServerImplementation/Core/vtkSIProxy.cxx,
> line 307
> vtkSISourceProxy (0x7fde81c8f5f0): Failed to create vtkAdiosPixieReader.
> Aborting for debugging purposes.
>
>
> I have tried both the example data in adios, as well as some example data I
> found while searching for info on adios and Paraview, both result in the
> same crash.
>
> Using Paraview 4.3.1
>
> Any ideas?
>
> Thanks,
> Charles
>
>
>
> --
> View this message in context:
> http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-Adios-tp3893.html
> Sent from the The Unofficial ParaView Developers Forum mailing list
> archive at Nabble.com.
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/f2bcd28f/attachment.html>

From dave.demarle at kitware.com  Fri Jul 17 08:52:02 2015
From: dave.demarle at kitware.com (David E DeMarle)
Date: Fri, 17 Jul 2015 08:52:02 -0400
Subject: [Paraview-developers] Intel Developer Forum,
 software-defined visualization, San Francisco August 18-20
Message-ID: <CANjZAi9OmrUXio85uL15Eff2NWJ1BdtdJX3caNaf=pvCK5i1EA@mail.gmail.com>

Greetings lists,

The vis and rendering folks at Intel will be showing off the things they've
been working on and are bringing to VTK, ParaView and the rest.

See the attached flyer for registration promo-codes if you plan to attend
the conference.

Hope to see you there.

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: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/6796d574/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IDF_wPromoCodes.pdf
Type: application/pdf
Size: 71463 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/6796d574/attachment-0001.pdf>

From utkarsh.ayachit at kitware.com  Fri Jul 17 14:29:43 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 17 Jul 2015 18:29:43 +0000
Subject: [Paraview-developers] Update: Upcoming ParaView Releases (4.4 and
	5.0)
Message-ID: <CADHSJjAJS1YR9gHfAte4f5Ga+ij0vGghSQ21Hto+uYX+JbhWiw@mail.gmail.com>

Folks,

Since we're getting close to the next release, I just wanted to send out an
email to let everyone know how we're planning to do the next release.

+ We'll do a ParaView 4.4 release will all features/fixes planned for the
next release. This will use the older rendering code (same as previous
release) that doesn't using OpenGL 2+ features.

+  Following the same pattern as VTK, soon after (at the same time, if
possible), we'll release ParaView 5.0. This will be ParaView 4.4 but with
OpenGL2 backend.

The main reason for this separation would be to have a fallback in case
showstopper issues are noticed with the major rendering changes and for
legacy systems that don't have a newer OpenGL.

Any thoughts? Comments? Suggestions? If things go according to plan,
release candidates should start popping up mid/late August.

Utkarsh

p.s. Originally, we also intended to upgrade ParaView 5.0 to use Qt 5, by
default. However, there are a few outstanding issues with Qt 5 (some in Qt
5 itself, and some in our testing code) and hence the upgrade to Qt 5 has
been tabled for now.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/2da9dc62/attachment.html>

From berk.geveci at kitware.com  Fri Jul 17 14:38:58 2015
From: berk.geveci at kitware.com (Berk Geveci)
Date: Fri, 17 Jul 2015 14:38:58 -0400
Subject: [Paraview-developers] Open Position Research Programmer
Message-ID: <CAE32kpXu004+DuR4nUnvtBj8PtWYEY=y6RFr3Z-q_3mqcVEN5w@mail.gmail.com>

Hi folks,

This is from our friends at the UIUC Parallel Programming Laboratory:

Now accepting applications for Visiting Research Programmer

Close date: Monday, August 3, 2015 (extended)

The primary functions of the position are to extend the OpenAtom software
and to develop the Argo exascale operating system prototype. Working with
the senior research programmer and research staff, the visiting research
programmer will help improve Charm++ functionalities as necessary for these
and other projects as well as write software for parallel supercomputers
and clusters.

See full description at:

http://charm.cs.illinois.edu/career

Best,
-berk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/1614f496/attachment.html>

From berk.geveci at kitware.com  Fri Jul 17 19:59:18 2015
From: berk.geveci at kitware.com (Berk Geveci)
Date: Fri, 17 Jul 2015 19:59:18 -0400
Subject: [Paraview-developers] CFP: SC '15 Scientific Visualization Showcase
Message-ID: <CAE32kpVcWLAKqNTH87H+RKe40EYKGy8VT6eL7w4axK3Lk5jw5g@mail.gmail.com>

Hi folks,

Just a reminder that the deadline for the Scientific Visualization Showcase
is near.

Best,
-berk

On Tue, May 26, 2015 at 8:19 AM, Berk Geveci <berk.geveci at kitware.com>
wrote:

> Hi folks,
>
> The Call for Participation for the SC '15 Scientific Visualization
> Showcase is out. This is a great opportunity for the community to highlight
> the wonderful visualization work going on. Please consider submitting an
> entry (or more).
>
> http://sc15.supercomputing.org/program/scientific-visualization-showcase
>
> SC15?s Visualization and Data Analytics Showcase Program provides a forum
> for the year's most instrumental movies in HPC.  Six finalists will compete
> for the Best Visualization Award, and each finalist will present his or her
> movie during a dedicated session at SC15 in a 15-minute presentation.
> Movies are judged based on how their movie illuminates science, by the
> quality of the movie, and for innovations in the process used for creating
> the movie.
>
> Best,
> -berk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150717/78f3e751/attachment.html>

From m.schlottke-lakemper at aia.rwth-aachen.de  Fri Jul 17 23:21:24 2015
From: m.schlottke-lakemper at aia.rwth-aachen.de (Schlottke-Lakemper, Michael)
Date: Sat, 18 Jul 2015 03:21:24 +0000
Subject: [Paraview-developers] Update: Upcoming ParaView Releases (4.4
 and	5.0)
In-Reply-To: <CADHSJjAJS1YR9gHfAte4f5Ga+ij0vGghSQ21Hto+uYX+JbhWiw@mail.gmail.com>
References: <CADHSJjAJS1YR9gHfAte4f5Ga+ij0vGghSQ21Hto+uYX+JbhWiw@mail.gmail.com>
Message-ID: <DBFD69E2-F2C2-4917-AB1A-4D967F00F1A4@aia.rwth-aachen.de>

Hi Utkarsh,

For those of us maintaining our own ParaView builds (and also our own set of custom-built plugins), it would be great if you can provide a document specifically outlining breaking changes in v5.0 for the Plugin API, differences in required library versions or different flags for compiling ParaView etc. Comments on changes for compiling with OSMesa (if any) are also welcome.

Other than that, kudos for following the dual approach with releasing 4.4/5.0 simultaneously. This is certainly going to make things much easier for us in the transition process, as it allows us to verify if things breaking on our side are due to new features or due to the new backend.

Regards,

Michael
--
Michael Schlottke-Lakemper

SimLab Highly Scalable Fluids & Solids Engineering
J?lich Aachen Research Alliance (JARA-HPC)
RWTH Aachen University
W?llnerstra?e 5a
52062 Aachen
Germany

Phone: +49 (241) 80 95188
Fax: +49 (241) 80 92257
Mail: m.schlottke-lakemper at aia.rwth-aachen.de<mailto:m.schlottke-lakemper at aia.rwth-aachen.de>
Web: http://www.jara.org/jara-hpc

On 17 Jul 2015, at 20:29 , Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:

Folks,

Since we're getting close to the next release, I just wanted to send out an email to let everyone know how we're planning to do the next release.

+ We'll do a ParaView 4.4 release will all features/fixes planned for the next release. This will use the older rendering code (same as previous release) that doesn't using OpenGL 2+ features.

+  Following the same pattern as VTK, soon after (at the same time, if possible), we'll release ParaView 5.0. This will be ParaView 4.4 but with OpenGL2 backend.

The main reason for this separation would be to have a fallback in case showstopper issues are noticed with the major rendering changes and for legacy systems that don't have a newer OpenGL.

Any thoughts? Comments? Suggestions? If things go according to plan, release candidates should start popping up mid/late August.

Utkarsh

p.s. Originally, we also intended to upgrade ParaView 5.0 to use Qt 5, by default. However, there are a few outstanding issues with Qt 5 (some in Qt 5 itself, and some in our testing code) and hence the upgrade to Qt 5 has been tabled for now.
_______________________________________________
Powered by www.kitware.com<http://www.kitware.com>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Search the list archives at: http://markmail.org/search/?q=Paraview-developers

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview-developers

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150718/0ed49db2/attachment-0001.html>

From houssen at ipgp.fr  Sun Jul 19 13:53:22 2015
From: houssen at ipgp.fr (houssen)
Date: Sun, 19 Jul 2015 19:53:22 +0200
Subject: [Paraview-developers]
 =?utf-8?q?How_to_implement_an_identity_filt?= =?utf-8?q?er_=3F?=
In-Reply-To: <535c216abf12af284f43241d3465516d@imap.ipgp.fr>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
 <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
 <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>
 <535c216abf12af284f43241d3465516d@imap.ipgp.fr>
Message-ID: <97197a455a6b1beb451d1749a8801761@imap.ipgp.fr>

Still get the zero filter ?!... Don't understand !? If somebody has an 
idea ?... Something to check ? Verify ? Track ? Something missing ? Add 
tag in xml ? Missing rendering call ?

Now, I derive from vtkDataObjectAlgorithm, I re-implement only 
RequestDataObject. I tried without success to create a new instance 
(from pInputObj) to pass to pOutputInfo (with and without using 
DeepCopy) : finally I tried to pass directly pInputObj. In the GUI, in 
the information tab, I get the expected type but zero point / cell. When 
I delete the filter, the initial data are still here (tab info indicates 
non zero number of points / cells) but the 3D viewpoint is empty : can 
this be a clue of some problem ?

Franck

int myFilter::RequestDataObject ( vtkInformation * ipRequest, 
vtkInformationVector ** ipInputVector, vtkInformationVector * 
opOutputVector )
{
   if ( !ipRequest ) return 0;

   vtkInformation * pInputInfo = ( ipInputVector ) ? ipInputVector[0] -> 
GetInformationObject ( 0 )            : NULL;
   vtkDataObject  * pInputObj  = ( pInputInfo    ) ? pInputInfo       -> 
Get ( vtkDataObject::DATA_OBJECT () ) : NULL;
   if ( pInputObj )
   {
     vtkInformation * pOutputInfo = ( opOutputVector ) ? opOutputVector 
-> GetInformationObject ( 0 )            : NULL;
     vtkDataObject  * pOutputObj  = ( pOutputInfo    ) ? pOutputInfo    
-> Get ( vtkDataObject::DATA_OBJECT () ) : NULL;
     if ( !pOutputObj )
     {
       _OutputDataType = pInputObj -> GetClassName (); // For later call 
to FillOutputPortInformation
       pOutputInfo -> Set ( vtkDataObject::DATA_OBJECT (), pInputObj );
       return 1;
     }
   }
   return 0;
}

int myFilter::FillOutputPortInformation ( int iPort, vtkInformation * 
ipInfo )
{
   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (), 
_OutputDataType.c_str () ); // Output data type to create (called after 
ProcessRequest)
   return 1;
}

Le 2015-07-15 23:49, houssen a ?crit?:
> Shawn, Joachim, thanks.
> After a quick look at the code, if I got you right, on a
> REQUEST_DATA_OBJECT request, I still need to create a new instance of
> the input and to set it as the DATA_OBJECT of the vtkInfo of the
> output. I still have to test this, but, I got the idea !
>
> Franck
>
> Le 2015-07-15 15:47, Shawn Waldon a ?crit?:
>> You might want to take a look at inheriting from
>> vtkPassInputTypeAlgorithm[1] if you want the output type to be the
>> same as the input type.? It already implements RequestDataObject the
>> way that I think Joachim is suggesting.
>>
>> HTH,
>> Shawn
>>
>> [1]:
>> 
>> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>> [14]
>>
>> On Wed, Jul 15, 2015 at 7:27 AM, Joachim Pouderoux
>> <joachim.pouderoux at kitware.com [15]> wrote:
>>
>>> Franck,
>>>
>>> You will have to override the RequestDataObject() method in order
>>> to make your filter generic. This method is the one called to
>>> generate the output data object.
>>> Take a look at vtkGenerateIndexArray.cxx for instance, I think you
>>> can just copy it in your class (which should be derivated from
>>> vtkDataObjectAlgorithm).
>>>
>>> Best,
>>>
>>> JOACHIM POUDEROUX
>>>
>>>> _PhD, Technical Expert_
>>>
>>> 2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr [9]>:
>>>
>>>> Seems that it is not possible to filter on vtkDataObject type.
>>>> But with less generic types (like vtkUnstrucutredGrid for
>>>> instance), the myFilter works : no error occur, but, viewpoint
>>>> ends up to be empty. I get "zero" instead of "identity" : why ?!
>>>>
>>>> Franck
>>>>
>>>> Le 2015-07-14 23:38, houssen a ?crit?:
>>>>
>>>>> How to write an identity filter ? (1 input port, 1 output port,
>>>>> output = input whatever input may be)
>>>>>
>>>>> I need to write a custom filter. I am used to VTK "from
>>>>> outside"
>>>>> (creating/handling vtk pipelines/scene/object from a main
>>>>> program)
>>>>> that is basically without the need to understand all internal
>>>>> VTK
>>>>> mechanics (except the SetInputConnection concept). I am not
>>>>> yet used
>>>>> to VTK internal mechanics that is necessary for writing a
>>>>> ParaView
>>>>> plugin.
>>>>>
>>>>> I need my filter to do "complex" stuffs. As I couldn't make it
>>>>> work,
>>>>> I simplified as much as possible to obtain finally an identity
>>>>> filter
>>>>> (!) : this is still not working. I feel I still miss something
>>>>> from
>>>>> VTK internal mechanics.
>>>>>
>>>>> As far as I understand, in the constructor, I need :
>>>>> myFilter::myFilter () { SetNumberOfInputPorts? ( 1 );
>>>>> SetNumberOfOutputPorts ( 1 ); }
>>>>>
>>>>> So then I need to implement the FillXXXPortInformation :
>>>>> int myFilter::FillInputPortInformation ( int vtkNotUsed (
>>>>> iPort ),
>>>>> vtkInformation * ipInfo )
>>>>> {
>>>>> ? if ( ipInfo ) ipInfo -> Set (
>>>>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" );
>>>>> //
>>>>> Filter allowed data types
>>>>> ? return 1;
>>>>> }
>>>>> int myFilter::FillOutputPortInformation ( int vtkNotUsed (
>>>>> iPort ),
>>>>> vtkInformation * ipInfo )
>>>>> {
>>>>> ? if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME
>>>>> (),
>>>>> "vtkDataObject" ); // Output data type to be created
>>>>> ? return 1;
>>>>> }
>>>>>
>>>>> So, now I only need to implement the ProcessRequest gateway:
>>>>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>>>>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>>>>> opOutputVector )
>>>>> {
>>>>> ? if ( ipRequest && ipRequest -> Has (
>>>>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>>>> ? ? if ( ipInputVector && opOutputVector ) opOutputVector =
>>>>> ipInputVector[0];
>>>>> ? ?return Superclass::ProcessRequest ( ipRequest,
>>>>> ipInputVector,
>>>>> opOutputVector );
>>>>> }
>>>>>
>>>>> I believed that I would be the only steps to proceed to get an
>>>>> identity filter. When I use my identity filter in Paraview, I
>>>>> get
>>>>> errors like :
>>>>> ERROR: In
>>>>>
>>>>>
>>>>>
>>>>
>>>
>> 
>> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>>> line 810
>>>>> vtkPVPostFilterExecutive : Input for connection index 0 on
>>>>> input port
>>>>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>>>>> vtkDataObject, but a vtkPolyData is required.
>>>>>
>>>>> Some logic is still missing here : could somebody explain me
>>>>> what and why ?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Franck
>>>>>
>>>>> _______________________________________________
>>>>> Powered by www.kitware.com [1]
>>>>>
>>>>> Visit other Kitware open-source projects at
>>>>> http://www.kitware.com/opensource/opensource.html [2]
>>>>>
>>>>> Search the list archives at:
>>>>> http://markmail.org/search/?q=Paraview-developers [3]
>>>>>
>>>>> Follow this link to subscribe/unsubscribe:
>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>> [4]
>>>>
>>>> _______________________________________________
>>>> Powered by www.kitware.com [5]
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html [6]
>>>>
>>>> Search the list archives at:
>>>> http://markmail.org/search/?q=Paraview-developers [7]
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [8]
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com [10]
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html [11]
>>>
>>> Search the list archives at:
>>> http://markmail.org/search/?q=Paraview-developers [12]
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>> [13]
>>
>>
>>
>> Links:
>> ------
>> [1] http://www.kitware.com
>> [2] http://www.kitware.com/opensource/opensource.html
>> [3] http://markmail.org/search/?q=Paraview-developers
>> [4] http://public.kitware.com/mailman/listinfo/paraview-developers
>> [5] http://www.kitware.com
>> [6] http://www.kitware.com/opensource/opensource.html
>> [7] http://markmail.org/search/?q=Paraview-developers
>> [8] http://public.kitware.com/mailman/listinfo/paraview-developers
>> [9] mailto:houssen at ipgp.fr
>> [10] http://www.kitware.com
>> [11] http://www.kitware.com/opensource/opensource.html
>> [12] http://markmail.org/search/?q=Paraview-developers
>> [13] http://public.kitware.com/mailman/listinfo/paraview-developers
>> [14] 
>> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>> [15] mailto:joachim.pouderoux at kitware.com


From joachim.pouderoux at kitware.com  Sun Jul 19 15:25:51 2015
From: joachim.pouderoux at kitware.com (Joachim Pouderoux)
Date: Sun, 19 Jul 2015 21:25:51 +0200
Subject: [Paraview-developers] How to implement an identity filter ?
In-Reply-To: <97197a455a6b1beb451d1749a8801761@imap.ipgp.fr>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
 <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
 <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>
 <535c216abf12af284f43241d3465516d@imap.ipgp.fr>
 <97197a455a6b1beb451d1749a8801761@imap.ipgp.fr>
Message-ID: <CACJ=gTgkRmZH9fU5P=_9DG3wLFUR5yG3VCgBeCHD0pXNyu9B2Q@mail.gmail.com>

Franck,

What do you do in RequestData()? You are supposed to shallow copy input to
the output, right?

Joachim

*Joachim Pouderoux*

*PhD, Technical Expert*
*Kitware SAS <http://www.kitware.fr>*


2015-07-19 19:53 GMT+02:00 houssen <houssen at ipgp.fr>:

> Still get the zero filter ?!... Don't understand !? If somebody has an
> idea ?... Something to check ? Verify ? Track ? Something missing ? Add tag
> in xml ? Missing rendering call ?
>
> Now, I derive from vtkDataObjectAlgorithm, I re-implement only
> RequestDataObject. I tried without success to create a new instance (from
> pInputObj) to pass to pOutputInfo (with and without using DeepCopy) :
> finally I tried to pass directly pInputObj. In the GUI, in the information
> tab, I get the expected type but zero point / cell. When I delete the
> filter, the initial data are still here (tab info indicates non zero number
> of points / cells) but the 3D viewpoint is empty : can this be a clue of
> some problem ?
>
> Franck
>
> int myFilter::RequestDataObject ( vtkInformation * ipRequest,
> vtkInformationVector ** ipInputVector, vtkInformationVector *
> opOutputVector )
> {
>   if ( !ipRequest ) return 0;
>
>   vtkInformation * pInputInfo = ( ipInputVector ) ? ipInputVector[0] ->
> GetInformationObject ( 0 )            : NULL;
>   vtkDataObject  * pInputObj  = ( pInputInfo    ) ? pInputInfo       ->
> Get ( vtkDataObject::DATA_OBJECT () ) : NULL;
>   if ( pInputObj )
>   {
>     vtkInformation * pOutputInfo = ( opOutputVector ) ? opOutputVector ->
> GetInformationObject ( 0 )            : NULL;
>     vtkDataObject  * pOutputObj  = ( pOutputInfo    ) ? pOutputInfo    ->
> Get ( vtkDataObject::DATA_OBJECT () ) : NULL;
>     if ( !pOutputObj )
>     {
>       _OutputDataType = pInputObj -> GetClassName (); // For later call to
> FillOutputPortInformation
>       pOutputInfo -> Set ( vtkDataObject::DATA_OBJECT (), pInputObj );
>       return 1;
>     }
>   }
>   return 0;
> }
>
> int myFilter::FillOutputPortInformation ( int iPort, vtkInformation *
> ipInfo )
> {
>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
> _OutputDataType.c_str () ); // Output data type to create (called after
> ProcessRequest)
>   return 1;
> }
>
> Le 2015-07-15 23:49, houssen a ?crit :
>
>> Shawn, Joachim, thanks.
>> After a quick look at the code, if I got you right, on a
>> REQUEST_DATA_OBJECT request, I still need to create a new instance of
>> the input and to set it as the DATA_OBJECT of the vtkInfo of the
>> output. I still have to test this, but, I got the idea !
>>
>> Franck
>>
>> Le 2015-07-15 15:47, Shawn Waldon a ?crit :
>>
>>> You might want to take a look at inheriting from
>>> vtkPassInputTypeAlgorithm[1] if you want the output type to be the
>>> same as the input type.  It already implements RequestDataObject the
>>> way that I think Joachim is suggesting.
>>>
>>> HTH,
>>> Shawn
>>>
>>> [1]:
>>>
>>> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>>> [14]
>>>
>>> On Wed, Jul 15, 2015 at 7:27 AM, Joachim Pouderoux
>>> <joachim.pouderoux at kitware.com [15]> wrote:
>>>
>>>  Franck,
>>>>
>>>> You will have to override the RequestDataObject() method in order
>>>> to make your filter generic. This method is the one called to
>>>> generate the output data object.
>>>> Take a look at vtkGenerateIndexArray.cxx for instance, I think you
>>>> can just copy it in your class (which should be derivated from
>>>> vtkDataObjectAlgorithm).
>>>>
>>>> Best,
>>>>
>>>> JOACHIM POUDEROUX
>>>>
>>>>  _PhD, Technical Expert_
>>>>>
>>>>
>>>> 2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr [9]>:
>>>>
>>>>  Seems that it is not possible to filter on vtkDataObject type.
>>>>> But with less generic types (like vtkUnstrucutredGrid for
>>>>> instance), the myFilter works : no error occur, but, viewpoint
>>>>> ends up to be empty. I get "zero" instead of "identity" : why ?!
>>>>>
>>>>> Franck
>>>>>
>>>>> Le 2015-07-14 23:38, houssen a ?crit :
>>>>>
>>>>>  How to write an identity filter ? (1 input port, 1 output port,
>>>>>> output = input whatever input may be)
>>>>>>
>>>>>> I need to write a custom filter. I am used to VTK "from
>>>>>> outside"
>>>>>> (creating/handling vtk pipelines/scene/object from a main
>>>>>> program)
>>>>>> that is basically without the need to understand all internal
>>>>>> VTK
>>>>>> mechanics (except the SetInputConnection concept). I am not
>>>>>> yet used
>>>>>> to VTK internal mechanics that is necessary for writing a
>>>>>> ParaView
>>>>>> plugin.
>>>>>>
>>>>>> I need my filter to do "complex" stuffs. As I couldn't make it
>>>>>> work,
>>>>>> I simplified as much as possible to obtain finally an identity
>>>>>> filter
>>>>>> (!) : this is still not working. I feel I still miss something
>>>>>> from
>>>>>> VTK internal mechanics.
>>>>>>
>>>>>> As far as I understand, in the constructor, I need :
>>>>>> myFilter::myFilter () { SetNumberOfInputPorts  ( 1 );
>>>>>> SetNumberOfOutputPorts ( 1 ); }
>>>>>>
>>>>>> So then I need to implement the FillXXXPortInformation :
>>>>>> int myFilter::FillInputPortInformation ( int vtkNotUsed (
>>>>>> iPort ),
>>>>>> vtkInformation * ipInfo )
>>>>>> {
>>>>>>   if ( ipInfo ) ipInfo -> Set (
>>>>>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (), "vtkDataObject" );
>>>>>> //
>>>>>> Filter allowed data types
>>>>>>   return 1;
>>>>>> }
>>>>>> int myFilter::FillOutputPortInformation ( int vtkNotUsed (
>>>>>> iPort ),
>>>>>> vtkInformation * ipInfo )
>>>>>> {
>>>>>>   if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME
>>>>>> (),
>>>>>> "vtkDataObject" ); // Output data type to be created
>>>>>>   return 1;
>>>>>> }
>>>>>>
>>>>>> So, now I only need to implement the ProcessRequest gateway:
>>>>>> int myFilter::ProcessRequest ( vtkInformation * ipRequest,
>>>>>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>>>>>> opOutputVector )
>>>>>> {
>>>>>>   if ( ipRequest && ipRequest -> Has (
>>>>>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>>>>>     if ( ipInputVector && opOutputVector ) opOutputVector =
>>>>>> ipInputVector[0];
>>>>>>    return Superclass::ProcessRequest ( ipRequest,
>>>>>> ipInputVector,
>>>>>> opOutputVector );
>>>>>> }
>>>>>>
>>>>>> I believed that I would be the only steps to proceed to get an
>>>>>> identity filter. When I use my identity filter in Paraview, I
>>>>>> get
>>>>>> errors like :
>>>>>> ERROR: In
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>
>>>> line 810
>>>>>> vtkPVPostFilterExecutive : Input for connection index 0 on
>>>>>> input port
>>>>>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of type
>>>>>> vtkDataObject, but a vtkPolyData is required.
>>>>>>
>>>>>> Some logic is still missing here : could somebody explain me
>>>>>> what and why ?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Franck
>>>>>>
>>>>>> _______________________________________________
>>>>>> Powered by www.kitware.com [1]
>>>>>>
>>>>>> Visit other Kitware open-source projects at
>>>>>> http://www.kitware.com/opensource/opensource.html [2]
>>>>>>
>>>>>> Search the list archives at:
>>>>>> http://markmail.org/search/?q=Paraview-developers [3]
>>>>>>
>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>> [4]
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Powered by www.kitware.com [5]
>>>>>
>>>>> Visit other Kitware open-source projects at
>>>>> http://www.kitware.com/opensource/opensource.html [6]
>>>>>
>>>>> Search the list archives at:
>>>>> http://markmail.org/search/?q=Paraview-developers [7]
>>>>>
>>>>> Follow this link to subscribe/unsubscribe:
>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>> [8]
>>>>>
>>>>
>>>> _______________________________________________
>>>> Powered by www.kitware.com [10]
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html [11]
>>>>
>>>> Search the list archives at:
>>>> http://markmail.org/search/?q=Paraview-developers [12]
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [13]
>>>>
>>>
>>>
>>>
>>> Links:
>>> ------
>>> [1] http://www.kitware.com
>>> [2] http://www.kitware.com/opensource/opensource.html
>>> [3] http://markmail.org/search/?q=Paraview-developers
>>> [4] http://public.kitware.com/mailman/listinfo/paraview-developers
>>> [5] http://www.kitware.com
>>> [6] http://www.kitware.com/opensource/opensource.html
>>> [7] http://markmail.org/search/?q=Paraview-developers
>>> [8] http://public.kitware.com/mailman/listinfo/paraview-developers
>>> [9] mailto:houssen at ipgp.fr
>>> [10] http://www.kitware.com
>>> [11] http://www.kitware.com/opensource/opensource.html
>>> [12] http://markmail.org/search/?q=Paraview-developers
>>> [13] http://public.kitware.com/mailman/listinfo/paraview-developers
>>> [14]
>>> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>>> [15] mailto:joachim.pouderoux at kitware.com
>>>
>>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150719/94aa778d/attachment-0001.html>

From utkarsh.ayachit at kitware.com  Sun Jul 19 16:08:17 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Sun, 19 Jul 2015 20:08:17 +0000
Subject: [Paraview-developers] [Paraview] Update: Upcoming ParaView
	Releases (4.4 and 5.0)
In-Reply-To: <CAE3taFDRHUwDY=46R2SLJQLoJd0bMcbvnu9LEF1Pv3NnqyEQkg@mail.gmail.com>
References: <CADHSJjAJS1YR9gHfAte4f5Ga+ij0vGghSQ21Hto+uYX+JbhWiw@mail.gmail.com>
 <CAE3taFDRHUwDY=46R2SLJQLoJd0bMcbvnu9LEF1Pv3NnqyEQkg@mail.gmail.com>
Message-ID: <CADHSJjDB8KZNXEowrAHb+Lmem9ah4Xe=0aps93wmLt2nMkw9WA@mail.gmail.com>

Lester,

ParaView 4.4  won't have any major OpenGL related changes so it should work
on platforms ParaView 4.3 or earlier worked. ParaView 5 (with its use of
OpenGL 2.1/3.2 features) should indeed still work on Windows 7/8 with
relatively modern drivers. Note, OpenGL 3.2 is still quite old (~2009). If
you're using remote-desktop, however, you may want to switch to using VNC.
Remote desktop uses a software OpenGL implementation, which (when last I
checked) was only OpenGL 1.1 compatible.

(cc-ing Ken to correct me if I am mistaken)

Utkarsh

On Fri, Jul 17, 2015 at 6:50 PM Lester Anderson <arctica1963 at gmail.com>
wrote:

> With reference to OpenGL, will the new updates via 4,4 or 5 be compatible
> with Windows 7(or 8) or does it expect a newer version of OpenGL on the
> target system?
> thanks
>
> On 17 July 2015 at 19:29, Utkarsh Ayachit <utkarsh.ayachit at kitware.com>
> wrote:
>
>> Folks,
>>
>> Since we're getting close to the next release, I just wanted to send out
>> an email to let everyone know how we're planning to do the next release.
>>
>> + We'll do a ParaView 4.4 release will all features/fixes planned for the
>> next release. This will use the older rendering code (same as previous
>> release) that doesn't using OpenGL 2+ features.
>>
>> +  Following the same pattern as VTK, soon after (at the same time, if
>> possible), we'll release ParaView 5.0. This will be ParaView 4.4 but with
>> OpenGL2 backend.
>>
>> The main reason for this separation would be to have a fallback in case
>> showstopper issues are noticed with the major rendering changes and for
>> legacy systems that don't have a newer OpenGL.
>>
>> Any thoughts? Comments? Suggestions? If things go according to plan,
>> release candidates should start popping up mid/late August.
>>
>> Utkarsh
>>
>> p.s. Originally, we also intended to upgrade ParaView 5.0 to use Qt 5,
>> by default. However, there are a few outstanding issues with Qt 5 (some in
>> Qt 5 itself, and some in our testing code) and hence the upgrade to Qt 5
>> has been tabled for now.
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150719/22d94b57/attachment.html>

From utkarsh.ayachit at kitware.com  Sun Jul 19 16:14:44 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Sun, 19 Jul 2015 20:14:44 +0000
Subject: [Paraview-developers] Update: Upcoming ParaView Releases (4.4
	and 5.0)
In-Reply-To: <DBFD69E2-F2C2-4917-AB1A-4D967F00F1A4@aia.rwth-aachen.de>
References: <CADHSJjAJS1YR9gHfAte4f5Ga+ij0vGghSQ21Hto+uYX+JbhWiw@mail.gmail.com>
 <DBFD69E2-F2C2-4917-AB1A-4D967F00F1A4@aia.rwth-aachen.de>
Message-ID: <CADHSJjA7v-3uHJtF08UQ9LTfQF=yLsKyQMXaZm7B6ATdt83T=A@mail.gmail.com>

Michael,


it would be great if you can provide a document specifically outlining
> breaking changes in v5.0 for the Plugin API
>

We've been documenting major API changes on the doxygen pages here:
http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/index.html

Of course, if you do notice missing information, holler and we'll update
it. As a team we should document changes to make it easier for plugin and
custom app developers.

differences in required library versions or different flags for compiling
> ParaView etc. Comments on changes for compiling with OSMesa (if any) are
> also welcome.
>

These will be updated soon. We're working on simplifying the OpenGL/OSMesa
flags that needed to be set for things to work correctly. You can follow
the development here:
https://gitlab.kitware.com/vtk/vtk/merge_requests/427


Other than that, kudos for following the dual approach with releasing
> 4.4/5.0 simultaneously. This is certainly going to make things much easier
> for us in the transition process, as it allows us to verify if things
> breaking on our side are due to new features or due to the new backend.
>

Thanks! That is indeed the idea.

Utkarsh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150719/0590c651/attachment.html>

From houssen at ipgp.fr  Sun Jul 19 16:54:29 2015
From: houssen at ipgp.fr (houssen)
Date: Sun, 19 Jul 2015 22:54:29 +0200
Subject: [Paraview-developers]
 =?utf-8?q?How_to_implement_an_identity_filt?= =?utf-8?q?er_=3F?=
In-Reply-To: <CACJ=gTgkRmZH9fU5P=_9DG3wLFUR5yG3VCgBeCHD0pXNyu9B2Q@mail.gmail.com>
References: <733cf4958adeaf46abe47809b4a72290@imap.ipgp.fr>
 <15037cc2161ffabe9221687445bab41e@imap.ipgp.fr>
 <CACJ=gTgE0FMcm+E_11ocYS_ac-GmkbYM66+ihDBOThJ+t19EPg@mail.gmail.com>
 <CALSpCT0bLc=9rhxUudfXZ6+n48W_Xugxxi4X7OXgGTM0oxck5Q@mail.gmail.com>
 <535c216abf12af284f43241d3465516d@imap.ipgp.fr>
 <97197a455a6b1beb451d1749a8801761@imap.ipgp.fr>
 <CACJ=gTgkRmZH9fU5P=_9DG3wLFUR5yG3VCgBeCHD0pXNyu9B2Q@mail.gmail.com>
Message-ID: <58a7dc77dfd9d2eebfeddc4da9f15abf@imap.ipgp.fr>

Finally working ! Thanks !
I guess my mistake was to do the shallow copy in RequestDataObject, not 
in RequestData.

Franck

Le 2015-07-19 21:25, Joachim Pouderoux a ?crit?:
> Franck,
>
> What do you do in RequestData()? You are supposed to shallow copy
> input to the output, right?
>
> Joachim
>
> JOACHIM POUDEROUX
>
>> _PhD, Technical Expert_
>
> 2015-07-19 19:53 GMT+02:00 houssen <houssen at ipgp.fr [35]>:
>
>> Still get the zero filter ?!... Don't understand !? If somebody has
>> an idea ?... Something to check ? Verify ? Track ? Something
> missing
>> ? Add tag in xml ? Missing rendering call ?
>>
>> Now, I derive from vtkDataObjectAlgorithm, I re-implement only
>> RequestDataObject. I tried without success to create a new instance
>> (from pInputObj) to pass to pOutputInfo (with and without using
>> DeepCopy) : finally I tried to pass directly pInputObj. In the GUI,
>> in the information tab, I get the expected type but zero point /
>> cell. When I delete the filter, the initial data are still here
> (tab
>> info indicates non zero number of points / cells) but the 3D
>> viewpoint is empty : can this be a clue of some problem ?
>>
>> Franck
>>
>> int myFilter::RequestDataObject ( vtkInformation * ipRequest,
>> vtkInformationVector ** ipInputVector, vtkInformationVector *
>> opOutputVector )
>> {
>> ? if ( !ipRequest ) return 0;
>>
>> ? vtkInformation * pInputInfo = ( ipInputVector ) ?
>> ipInputVector[0] -> GetInformationObject ( 0 )? ? ? ? ? ? :
>> NULL;
>> ? vtkDataObject? * pInputObj? = ( pInputInfo? ? ) ?
>> pInputInfo? ? ? ?-> Get ( vtkDataObject::DATA_OBJECT () ) :
>> NULL;
>> ? if ( pInputObj )
>> ? {
>> ? ? vtkInformation * pOutputInfo = ( opOutputVector ) ?
>> opOutputVector -> GetInformationObject ( 0 )? ? ? ? ? ? :
>> NULL;
>> ? ? vtkDataObject? * pOutputObj? = ( pOutputInfo? ? ) ?
>> pOutputInfo? ? -> Get ( vtkDataObject::DATA_OBJECT () ) : NULL;
>> ? ? if ( !pOutputObj )
>> ? ? {
>> ? ? ? _OutputDataType = pInputObj -> GetClassName (); // For
>> later call to FillOutputPortInformation
>> ? ? ? pOutputInfo -> Set ( vtkDataObject::DATA_OBJECT (),
>> pInputObj );
>> ? ? ? return 1;
>> ? ? }
>> ? }
>> ? return 0;
>> }
>>
>> int myFilter::FillOutputPortInformation ( int iPort,
>> vtkInformation * ipInfo )
>> {
>> ? if ( ipInfo ) ipInfo -> Set ( vtkDataObject::DATA_TYPE_NAME (),
>> _OutputDataType.c_str () ); // Output data type to create (called
>> after ProcessRequest)
>> ? return 1;
>> }
>>
>> Le 2015-07-15 23:49, houssen a ?crit?:
>>
>>> Shawn, Joachim, thanks.
>>> After a quick look at the code, if I got you right, on a
>>> REQUEST_DATA_OBJECT request, I still need to create a new
>>> instance of
>>> the input and to set it as the DATA_OBJECT of the vtkInfo of the
>>> output. I still have to test this, but, I got the idea !
>>>
>>> Franck
>>>
>>> Le 2015-07-15 15:47, Shawn Waldon a ?crit?:
>>>
>>>> You might want to take a look at inheriting from
>>>> vtkPassInputTypeAlgorithm[1] if you want the output type to be
>>>> the
>>>> same as the input type.? It already implements
>>>> RequestDataObject the
>>>> way that I think Joachim is suggesting.
>>>>
>>>> HTH,
>>>> Shawn
>>>>
>>>> [1]:
>>>>
>>>>
>>>>
>>>
>>
> 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>>>> [14]
>>>> [14]
>>>>
>>>> On Wed, Jul 15, 2015 at 7:27 AM, Joachim Pouderoux
>>>> <joachim.pouderoux at kitware.com [15] [15]> wrote:
>>>>
>>>>> Franck,
>>>>>
>>>>> You will have to override the RequestDataObject() method in
>>>>> order
>>>>> to make your filter generic. This method is the one called
>>>>> to
>>>>> generate the output data object.
>>>>> Take a look at vtkGenerateIndexArray.cxx for instance, I
>>>>> think you
>>>>> can just copy it in your class (which should be derivated
>>>>> from
>>>>> vtkDataObjectAlgorithm).
>>>>>
>>>>> Best,
>>>>>
>>>>> JOACHIM POUDEROUX
>>>>>
>>>>>> _PhD, Technical Expert_
>>>>>
>>>>> 2015-07-15 11:57 GMT+02:00 houssen <houssen at ipgp.fr [1]
>>>>> [9]>:
>>>>>
>>>>>> Seems that it is not possible to filter on vtkDataObject
>>>>>> type.
>>>>>> But with less generic types (like vtkUnstrucutredGrid for
>>>>>> instance), the myFilter works : no error occur, but,
>>>>>> viewpoint
>>>>>> ends up to be empty. I get "zero" instead of "identity" :
>>>>>> why ?!
>>>>>>
>>>>>> Franck
>>>>>>
>>>>>> Le 2015-07-14 23:38, houssen a ?crit?:
>>>>>>
>>>>>>> How to write an identity filter ? (1 input port, 1 output
>>>>>>> port,
>>>>>>> output = input whatever input may be)
>>>>>>>
>>>>>>> I need to write a custom filter. I am used to VTK "from
>>>>>>> outside"
>>>>>>> (creating/handling vtk pipelines/scene/object from a
>>>>>>> main
>>>>>>> program)
>>>>>>> that is basically without the need to understand all
>>>>>>> internal
>>>>>>> VTK
>>>>>>> mechanics (except the SetInputConnection concept). I am
>>>>>>> not
>>>>>>> yet used
>>>>>>> to VTK internal mechanics that is necessary for writing
>>>>>>> a
>>>>>>> ParaView
>>>>>>> plugin.
>>>>>>>
>>>>>>> I need my filter to do "complex" stuffs. As I couldn't
>>>>>>> make it
>>>>>>> work,
>>>>>>> I simplified as much as possible to obtain finally an
>>>>>>> identity
>>>>>>> filter
>>>>>>> (!) : this is still not working. I feel I still miss
>>>>>>> something
>>>>>>> from
>>>>>>> VTK internal mechanics.
>>>>>>>
>>>>>>> As far as I understand, in the constructor, I need :
>>>>>>> myFilter::myFilter () { SetNumberOfInputPorts? ( 1 );
>>>>>>> SetNumberOfOutputPorts ( 1 ); }
>>>>>>>
>>>>>>> So then I need to implement the FillXXXPortInformation :
>>>>>>> int myFilter::FillInputPortInformation ( int vtkNotUsed
>>>>>>> (
>>>>>>> iPort ),
>>>>>>> vtkInformation * ipInfo )
>>>>>>> {
>>>>>>> ? if ( ipInfo ) ipInfo -> Set (
>>>>>>> vtkAlgorithm::INPUT_REQUIRED_DATA_TYPE (),
>>>>>>> "vtkDataObject" );
>>>>>>> //
>>>>>>> Filter allowed data types
>>>>>>> ? return 1;
>>>>>>> }
>>>>>>> int myFilter::FillOutputPortInformation ( int vtkNotUsed
>>>>>>> (
>>>>>>> iPort ),
>>>>>>> vtkInformation * ipInfo )
>>>>>>> {
>>>>>>> ? if ( ipInfo ) ipInfo -> Set (
>>>>>>> vtkDataObject::DATA_TYPE_NAME
>>>>>>> (),
>>>>>>> "vtkDataObject" ); // Output data type to be created
>>>>>>> ? return 1;
>>>>>>> }
>>>>>>>
>>>>>>> So, now I only need to implement the ProcessRequest
>>>>>>> gateway:
>>>>>>> int myFilter::ProcessRequest ( vtkInformation *
>>>>>>> ipRequest,
>>>>>>> vtkInformationVector ** ipInputVector,
>>>>>>> vtkInformationVector *
>>>>>>> opOutputVector )
>>>>>>> {
>>>>>>> ? if ( ipRequest && ipRequest -> Has (
>>>>>>> vtkDemandDrivenPipeline::REQUEST_DATA () ) )
>>>>>>> ? ? if ( ipInputVector && opOutputVector )
>>>>>>> opOutputVector =
>>>>>>> ipInputVector[0];
>>>>>>> ? ?return Superclass::ProcessRequest ( ipRequest,
>>>>>>> ipInputVector,
>>>>>>> opOutputVector );
>>>>>>> }
>>>>>>>
>>>>>>> I believed that I would be the only steps to proceed to
>>>>>>> get an
>>>>>>> identity filter. When I use my identity filter in
>>>>>>> Paraview, I
>>>>>>> get
>>>>>>> errors like :
>>>>>>> ERROR: In
>>>>
>>>>
>>>>
>>>
>>
> 
> /.../ParaView-v4.3.1-source/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>>
>>>>>>> line 810
>>>>>>> vtkPVPostFilterExecutive : Input for connection index 0
>>>>>>> on
>>>>>>> input port
>>>>>>> index 0 for algorithm vtkPVPostFilter(0x16d0ec0) is of
>>>>>>> type
>>>>>>> vtkDataObject, but a vtkPolyData is required.
>>>>>>>
>>>>>>> Some logic is still missing here : could somebody
>>>>>>> explain me
>>>>>>> what and why ?
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> Franck
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Powered by www.kitware.com [2] [1]
>>>>>>>
>>>>>>> Visit other Kitware open-source projects at
>>>>>>> http://www.kitware.com/opensource/opensource.html [3]
>>>>>>> [2]
>>>>>>>
>>>>>>> Search the list archives at:
>>>>>>> http://markmail.org/search/?q=Paraview-developers [4]
>>>>>>> [3]
>>>>>>>
>>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>> [5]
>>>>>>> [4]
>>>>>>
>>>>>> _______________________________________________
>>>>>> Powered by www.kitware.com [6] [5]
>>>>>>
>>>>>> Visit other Kitware open-source projects at
>>>>>> http://www.kitware.com/opensource/opensource.html [7] [6]
>>>>>>
>>>>>> Search the list archives at:
>>>>>> http://markmail.org/search/?q=Paraview-developers [8] [7]
>>>>>>
>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>>
>>>>>>
>>>>>
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>> [9]
>>>>>> [8]
>>>>>
>>>>> _______________________________________________
>>>>> Powered by www.kitware.com [10] [10]
>>>>>
>>>>> Visit other Kitware open-source projects at
>>>>> http://www.kitware.com/opensource/opensource.html [11] [11]
>>>>>
>>>>> Search the list archives at:
>>>>> http://markmail.org/search/?q=Paraview-developers [12] [12]
>>>>>
>>>>> Follow this link to subscribe/unsubscribe:
>>>>>
>>>>>
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>> [13]
>>>>> [13]
>>>>
>>>> Links:
>>>> ------
>>>> [1] http://www.kitware.com [16]
>>>> [2] http://www.kitware.com/opensource/opensource.html [17]
>>>> [3] http://markmail.org/search/?q=Paraview-developers [18]
>>>> [4]
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [19]
>>>> [5] http://www.kitware.com [20]
>>>> [6] http://www.kitware.com/opensource/opensource.html [21]
>>>> [7] http://markmail.org/search/?q=Paraview-developers [22]
>>>> [8]
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [23]
>>>> [9] mailto:houssen at ipgp.fr [24]
>>>> [10] http://www.kitware.com [25]
>>>> [11] http://www.kitware.com/opensource/opensource.html [26]
>>>> [12] http://markmail.org/search/?q=Paraview-developers [27]
>>>> [13]
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>> [28]
>>>> [14]
>>>>
>>>
>>
> 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
>>>> [29]
>>>> [15] mailto:joachim.pouderoux at kitware.com [30]
>>
>> _______________________________________________
>> Powered by www.kitware.com [31]
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html [32]
>>
>> Search the list archives at:
>> http://markmail.org/search/?q=Paraview-developers [33]
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>> [34]
>
>
>
> Links:
> ------
> [1] mailto:houssen at ipgp.fr
> [2] http://www.kitware.com
> [3] http://www.kitware.com/opensource/opensource.html
> [4] http://markmail.org/search/?q=Paraview-developers
> [5] http://public.kitware.com/mailman/listinfo/paraview-developers
> [6] http://www.kitware.com
> [7] http://www.kitware.com/opensource/opensource.html
> [8] http://markmail.org/search/?q=Paraview-developers
> [9] http://public.kitware.com/mailman/listinfo/paraview-developers
> [10] http://www.kitware.com
> [11] http://www.kitware.com/opensource/opensource.html
> [12] http://markmail.org/search/?q=Paraview-developers
> [13] http://public.kitware.com/mailman/listinfo/paraview-developers
> [14] 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
> [15] mailto:joachim.pouderoux at kitware.com
> [16] http://www.kitware.com
> [17] http://www.kitware.com/opensource/opensource.html
> [18] http://markmail.org/search/?q=Paraview-developers
> [19] http://public.kitware.com/mailman/listinfo/paraview-developers
> [20] http://www.kitware.com
> [21] http://www.kitware.com/opensource/opensource.html
> [22] http://markmail.org/search/?q=Paraview-developers
> [23] http://public.kitware.com/mailman/listinfo/paraview-developers
> [24] mailto:houssen at ipgp.fr
> [25] http://www.kitware.com
> [26] http://www.kitware.com/opensource/opensource.html
> [27] http://markmail.org/search/?q=Paraview-developers
> [28] http://public.kitware.com/mailman/listinfo/paraview-developers
> [29] 
> http://www.vtk.org/doc/nightly/html/classvtkPassInputTypeAlgorithm.html
> [30] mailto:joachim.pouderoux at kitware.com
> [31] http://www.kitware.com
> [32] http://www.kitware.com/opensource/opensource.html
> [33] http://markmail.org/search/?q=Paraview-developers
> [34] http://public.kitware.com/mailman/listinfo/paraview-developers
> [35] mailto:houssen at ipgp.fr


From jfavre at cscs.ch  Mon Jul 27 16:17:09 2015
From: jfavre at cscs.ch (Favre  Jean)
Date: Mon, 27 Jul 2015 20:17:09 +0000
Subject: [Paraview-developers] CFP: SC '15 Scientific Visualization Showcase
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F6841F212D9@MBX211.d.ethz.ch>


Folks

We have some great news. The deadline for submission has been pushed back to August 7. Take advantage of the extra time to submit the best visualizations ever.

Jean,
SC15,  Scientific Visualization Showcase Chair

________________________________
From: ParaView [paraview-bounces at paraview.org] on behalf of Berk Geveci [berk.geveci at kitware.com]
Sent: Saturday, July 18, 2015 1:59 AM
To: ParaView Developers; ParaView Mailing List
Subject: [Paraview] CFP: SC '15 Scientific Visualization Showcase

Hi folks,

Just a reminder that the deadline for the Scientific Visualization Showcase is near.

Best,
-berk

On Tue, May 26, 2015 at 8:19 AM, Berk Geveci <berk.geveci at kitware.com<mailto:berk.geveci at kitware.com>> wrote:
Hi folks,

The Call for Participation for the SC '15 Scientific Visualization Showcase is out. This is a great opportunity for the community to highlight the wonderful visualization work going on. Please consider submitting an entry (or more).

http://sc15.supercomputing.org/program/scientific-visualization-showcase

SC15?s Visualization and Data Analytics Showcase Program provides a forum for the year's most instrumental movies in HPC.  Six finalists will compete for the Best Visualization Award, and each finalist will present his or her movie during a dedicated session at SC15 in a 15-minute presentation. Movies are judged based on how their movie illuminates science, by the quality of the movie, and for innovations in the process used for creating the movie.

Best,
-berk


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150727/5c2229e6/attachment.html>

From berk.geveci at kitware.com  Tue Jul 28 08:53:38 2015
From: berk.geveci at kitware.com (Berk Geveci)
Date: Tue, 28 Jul 2015 08:53:38 -0400
Subject: [Paraview-developers] CFP: SC '15 Scientific Visualization
	Showcase
In-Reply-To: <CAE32kpVHFU4qMnTj508FT25Di9gp7c68JmOBy90LjzGPv8cE_A@mail.gmail.com>
References: <CAE32kpVHFU4qMnTj508FT25Di9gp7c68JmOBy90LjzGPv8cE_A@mail.gmail.com>
Message-ID: <CAE32kpWinJkXA=RUjyM9uqH3uvdU4Km4-ZPvbGCWGGCaor8yoQ@mail.gmail.com>

Update!

The deadline for submission of your best scientific visualizations to the
Supercomputing 2015 Showcase has been pushed back to August 7.

hxxp://sc15.supercomputing.org/program/scientific-visualization-showcase

Best,
-berk

On Tue, May 26, 2015 at 8:19 AM, Berk Geveci <berk.geveci at kitware.com>
wrote:

> Hi folks,
>
> The Call for Participation for the SC '15 Scientific Visualization
> Showcase is out. This is a great opportunity for the community to highlight
> the wonderful visualization work going on. Please consider submitting an
> entry (or more).
>
> http://sc15.supercomputing.org/program/scientific-visualization-showcase
>
> SC15?s Visualization and Data Analytics Showcase Program provides a forum
> for the year's most instrumental movies in HPC.  Six finalists will compete
> for the Best Visualization Award, and each finalist will present his or her
> movie during a dedicated session at SC15 in a 15-minute presentation.
> Movies are judged based on how their movie illuminates science, by the
> quality of the movie, and for innovations in the process used for creating
> the movie.
>
> Best,
> -berk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150728/d5744695/attachment.html>

From Orlando.RIVERA at mtu.de  Wed Jul 29 08:51:26 2015
From: Orlando.RIVERA at mtu.de (RIVERA ROLDAN, Jorge Orlando)
Date: Wed, 29 Jul 2015 12:51:26 +0000
Subject: [Paraview-developers] Reading files as  Times steps, short  question
Message-ID: <ACE279CCDDBEFA43BD2F0EE07671EA98A99DE634@memc022.de.mtu.corp>

Dear  Paraview Developers

I have a quick , and most probably already asked, question

I have a bunch of  files
File1.txt
File2.txt?.

Or
File0.5.txt
File1.0.txt
File1.5txt ?.

In the first case, is the time step number one , number two ?
In the second case , is  time 0.5s, time 1.s , time 1.5 sec ?.

My reader is actually reading each file well,  how to tell paraview to open then  as a series , and set the time steps

I was looking at something similar  , and the key would be in the xml to add  a proxy group ? something like :

<ProxyGroup name="sources">
    <SourceProxy name=?MyReader"
                           class="vtkFileSeriesReader"
                                           si_class="vtkSIFileSeriesReaderProxy"
                           file_name_method="SetFileName">?.

Still I don?t know how to  adapt that to my case ,

Do you have any pointer to me , example or documentation

Any Help would be greatly appreciated  !

Orlando


--
MTU Aero Engines AG
Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer Martens, Michael Schreyoegg
Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: Klaus Eberhardt
Sitz der Gesellschaft/Registered Office: Muenchen
Handelsregister/Commercial Register: Muenchen HRB 157206

Diese E-Mail sowie ihre Anhaenge enthalten MTU-eigene vertrauliche oder rechtlich geschuetzte Informationen.
Wenn Sie nicht der beabsichtigte Empfaenger sind, informieren Sie bitte den Absender und loeschen Sie diese
E-Mail sowie die Anhaenge. Das unbefugte Speichern, Kopieren oder Weiterleiten ist nicht gestattet.

This e-mail and any attached documents are proprietary to MTU, confidential or protected by law.
If you are not the intended recipient, please advise the sender and delete this message and its attachments.
Any unauthorised storing, copying or distribution is prohibited.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20150729/21e2520e/attachment.html>

From utkarsh.ayachit at kitware.com  Wed Jul 29 11:35:04 2015
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 29 Jul 2015 11:35:04 -0400
Subject: [Paraview-developers] Reading files as Times steps,
	short question
In-Reply-To: <ACE279CCDDBEFA43BD2F0EE07671EA98A99DE634@memc022.de.mtu.corp>
References: <ACE279CCDDBEFA43BD2F0EE07671EA98A99DE634@memc022.de.mtu.corp>
Message-ID: <CADHSJjDkfgD2iRZUBBDkNZrdNNC8oQMsODbdGXkUaM_wpE3Phw@mail.gmail.com>

The file series reader infrastructure will not handle the case where
the time value is deduced from the filename. You're better of simply
teaching your reader to load a set a files and determine what time
values each file corresponds to. ParaView can pass entire set of
selected files to the reader, using the following XML. You need to
provide RemoveAllFileNames/AddFileName methods on your reader class.

       <StringVectorProperty
                            clean_command="RemoveAllFileNames"
                            command="AddFileName"
                            name="FileNames"
                            number_of_elements="0"
                            panel_visibility="never"
                            repeat_command="1">
        <FileListDomain name="files" />
        <Documentation>The list of files to be read by the
        reader.</Documentation>
      </StringVectorProperty>

On Wed, Jul 29, 2015 at 8:51 AM, RIVERA ROLDAN, Jorge Orlando
<Orlando.RIVERA at mtu.de> wrote:
> Dear  Paraview Developers
>
>
>
> I have a quick , and most probably already asked, question
>
>
>
> I have a bunch of  files
>
> File1.txt
>
> File2.txt?.
>
>
>
> Or
>
> File0.5.txt
>
> File1.0.txt
>
> File1.5txt ?.
>
>
>
> In the first case, is the time step number one , number two ?
>
> In the second case , is  time 0.5s, time 1.s , time 1.5 sec ?.
>
>
>
> My reader is actually reading each file well,  how to tell paraview to open
> then  as a series , and set the time steps
>
>
>
> I was looking at something similar  , and the key would be in the xml to add
> a proxy group ? something like :
>
>
>
> <ProxyGroup name="sources">
>
>     <SourceProxy name=?MyReader"
>
>                            class="vtkFileSeriesReader"
>
>
> si_class="vtkSIFileSeriesReaderProxy"
>
>                            file_name_method="SetFileName">?.
>
>
>
> Still I don?t know how to  adapt that to my case ,
>
>
>
> Do you have any pointer to me , example or documentation
>
>
>
> Any Help would be greatly appreciated  !
>
>
>
> Orlando
>
>
>
>
>
> --
> MTU Aero Engines AG
> Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer
> Martens, Michael Schreyoegg
> Vorsitzender des Aufsichtsrats/Chairman of the Supervisory Board: Klaus
> Eberhardt
> Sitz der Gesellschaft/Registered Office: Muenchen
> Handelsregister/Commercial Register: Muenchen HRB 157206
>
> Diese E-Mail sowie ihre Anhaenge enthalten MTU-eigene vertrauliche oder
> rechtlich geschuetzte Informationen.
> Wenn Sie nicht der beabsichtigte Empfaenger sind, informieren Sie bitte den
> Absender und loeschen Sie diese
> E-Mail sowie die Anhaenge. Das unbefugte Speichern, Kopieren oder
> Weiterleiten ist nicht gestattet.
>
> This e-mail and any attached documents are proprietary to MTU, confidential
> or protected by law.
> If you are not the intended recipient, please advise the sender and delete
> this message and its attachments.
> Any unauthorised storing, copying or distribution is prohibited.
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Search the list archives at:
> http://markmail.org/search/?q=Paraview-developers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview-developers
>