From sankhesh.jhaveri at kitware.com  Mon Sep  1 12:31:10 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Mon, 01 Sep 2014 09:31:10 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Sunday, August 31 2014
Message-ID: <20140901163103.B8F65808A1@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]fedora-x64-icc-nightlymaster-debug 1
   [2]Mac10.7.5-gcc-nightlymaster-release 1
   -------------------- REPORT --------------------
   2 FAILURES
   [3]pvcs.ExtractLevel , 1 , ['fedora-x64-icc-nightlymaster-debug']
   [4]pvcs.ExportX3dPOVVRML , 1 , ['Mac10.7.5-gcc-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [5]fedora-x64-icc-nightlynext-release 1
   [6]Win32-vs9-shared-release 1
   -------------------- REPORT --------------------
   2 FAILURES
   [7]vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter , 1
   , ['Win32-vs9-shared-release']
   [8]pvcs.RenderNan , 1 , ['fedora-x64-icc-nightlynext-release']

   Issues with submitters
   The following expected submitters did not submit:
   KargadMac.kitware , Lion-gcc-pvVTK

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-01&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-01&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-gcc-nightlymaster-release
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-01&name=pvcs.ExtractLevel
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-01&name=pvcs.ExportX3dPOVVRML
   5. http://open.cdash.org/index.php?project=ParaView&date=2014-09-01&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
   6. http://open.cdash.org/index.php?project=ParaView&date=2014-09-01&filtercount=1&field1=buildname/string&compare1=61&value1=Win32-vs9-shared-release
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-01&name=vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-01&name=pvcs.RenderNan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140901/24cc827d/attachment.html>

From jlih at masonlive.gmu.edu  Mon Sep  1 22:15:02 2014
From: jlih at masonlive.gmu.edu (jlih)
Date: Tue, 2 Sep 2014 02:15:02 +0000
Subject: [Paraview-developers] Paraviewweb: possbilities of integrating
 python scripts?
In-Reply-To: <A870492F-ADE6-4694-8F0F-A2B6A64C8D8F@kitware.com>
References: <1409331056518.31974@masonlive.gmu.edu>
 <CABObKxfbRbEOmXzriZ6Z8637LaNBKx2VHcVDyoKSrm+5yQizSg@mail.gmail.com>
 <1409346343439.62598@masonlive.gmu.edu>,
 <A870492F-ADE6-4694-8F0F-A2B6A64C8D8F@kitware.com>
Message-ID: <1409624093400.59698@masonlive.gmu.edu>

?Thanks, Seb.


I will check that one.


Jing

________________________________
From: Sebastien Jourdain <sebastien.jourdain at kitware.com>
Sent: Saturday, August 30, 2014 8:55 AM
To: jlih
Cc: paraview-developers at paraview.org
Subject: Re: [Paraview-developers] Paraviewweb: possbilities of integrating python scripts?

Hi Jing,

The web visualizer is an example by itself. You can look at the vtk web too. Specially the graph one.

Seb

On Aug 29, 2014, at 15:05, jlih <jlih at masonlive.gmu.edu<mailto:jlih at masonlive.gmu.edu>> wrote:


Hi Seb,

Thanks for your quick reply! Your answer help me understand the role of Javascript and the launcher in the framework.


In terms of extending the application, could you provide me an example or point me to the related resources?


Thanks again.


Best,

Jing

________________________________
From: Sebastien Jourdain <sebastien.jourdain at kitware.com<mailto:sebastien.jourdain at kitware.com>>
Sent: Friday, August 29, 2014 1:00 PM
To: jlih
Cc: paraview-developers at paraview.org<mailto:paraview-developers at paraview.org>
Subject: Re: [Paraview-developers] Paraviewweb: possbilities of integrating python scripts?

Hi Jing,

I'm glad to hear that you setup a ParaViewWeb instance on your end. And you are right, you can use your server side python scripts to extend existing ParaViewWeb application or even create your own one.

We don't rely on JSP per say but on pure JavaScript which allow you to expose server side features on the client side.

The Python launcher is made to allow several users to start their own visualization process automatically, but if you plan to develop your own app leveraging your scripts, you can start without it.

Some useful resources can be found here: http://www.paraview.org/ParaView3/Doc/Nightly/www/js-doc/index.html

Seb


On Fri, Aug 29, 2014 at 10:50 AM, jlih <jlih at masonlive.gmu.edu<mailto:jlih at masonlive.gmu.edu>> wrote:

Dear all,

I have developed  a few python scripts that calls pvserver to perform visualization functions for my datasets. I've noticed Paraviewweb has provided quite a few good functions to perform interactive manipulation within a browser. I have installed Paraviewweb on a server.


I am wondering if I can use my python scripts as a basis to develop server side functions and use JSP functions provided by Paraviewweb as front-end to customize a website for my application. The purpose is to provide the visualization functions to users who do not have access to datasets and computing resources. Should I use "Python Launcher" to start?


Thanks in advance.


Jing




_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org>
http://public.kitware.com/mailman/listinfo/paraview-developers


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/14b176c2/attachment.html>

From frank.tt at einseinself.org  Tue Sep  2 03:28:19 2014
From: frank.tt at einseinself.org (Frank tt)
Date: Tue, 02 Sep 2014 09:28:19 +0200
Subject: [Paraview-developers] Read Image Sequence (not as time series)
Message-ID: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>

Hi,

I'm Frank and new in developing ParaView. 
My Setting is a client and a server. The server is image processing and
data server. 
I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
volume (similar to DICOM).
When I try to load a group of NIfTI files, every file will be loaded as
one data set.

Well, I want to write a file reader plugin, where I all selected files
(in alpha-num order) will be loaded as one volume. 
I did a look at the example plugins and the paraview file reader
plugins. 

I've read in the paraview mailing list archive [1] and I think the
"FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
I also know about the ITK solution [2] to convert the data first. I
think this won't work in my setup, because I have limited compiling
possibilities on the server (where the data is.)

I stuck to get started. Maybe someone can help me.

Greetings

Frank


[1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
[2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices




From akyrieleis at gmail.com  Tue Sep  2 07:03:08 2014
From: akyrieleis at gmail.com (Albrecht Kyrieleis)
Date: Tue, 2 Sep 2014 12:03:08 +0100
Subject: [Paraview-developers] IE 9
In-Reply-To: <CAOkZLxpEMo==NiPoCw7nnGp0TqVDcemRTtwx4aYNwy_8FJB3Pg@mail.gmail.com>
References: <CAOkZLxpEMo==NiPoCw7nnGp0TqVDcemRTtwx4aYNwy_8FJB3Pg@mail.gmail.com>
Message-ID: <CAOkZLxruse40fBDSFG6DHAJNoCqco42ycZiJNuBEmUf4dsppng@mail.gmail.com>

Hi,
on the paraviewweb site it is mentioned that Internet Explorer is supported
from v.10. I would like to use IE9 with my pvweb application if possible.
Which features of paraview can one expect not to work with IE9 ? Is there a
way to reliably use IE9 by not using certain features of paraview ? Is
there any plan to make paraviewweb work with IE9 at some time ? Many thanks.
Kyr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/ff8c806c/attachment.html>

From utkarsh.ayachit at kitware.com  Tue Sep  2 08:47:16 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 2 Sep 2014 08:47:16 -0400
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
Message-ID: <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>

Frank,

Are you using the AnalyzeNIfTIIO plugin or are you using the builtin
version (which is coming from the VisItBridge) for the NifTI files?

Utkarsh

On Tue, Sep 2, 2014 at 3:28 AM, Frank tt <frank.tt at einseinself.org> wrote:
> Hi,
>
> I'm Frank and new in developing ParaView.
> My Setting is a client and a server. The server is image processing and
> data server.
> I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
> volume (similar to DICOM).
> When I try to load a group of NIfTI files, every file will be loaded as
> one data set.
>
> Well, I want to write a file reader plugin, where I all selected files
> (in alpha-num order) will be loaded as one volume.
> I did a look at the example plugins and the paraview file reader
> plugins.
>
> I've read in the paraview mailing list archive [1] and I think the
> "FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
> I also know about the ITK solution [2] to convert the data first. I
> think this won't work in my setup, because I have limited compiling
> possibilities on the server (where the data is.)
>
> I stuck to get started. Maybe someone can help me.
>
> Greetings
>
> Frank
>
>
> [1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
> [2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices
>
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From frank.tt at einseinself.org  Tue Sep  2 10:08:59 2014
From: frank.tt at einseinself.org (Frank tt)
Date: Tue, 02 Sep 2014 16:08:59 +0200
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
Message-ID: <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>

Hi Utkarsh,

I use the AnalizeNIfTI plugin.

Actually I'm trying to write an own plugin based on the AnalizeNIfTI
plugin. I'm a little bit confused with this and other file readers. I
know some file reades as they get something like a
'std::vector<std::string>' including a list of the files (choosen from
the open file dialog). 
But here the readers have 'this->GetFileName()'. When I'm following
that, im comming over the Analize header to
VTK/IO/Image/vtkImageReader2.h:

062  // Description:
063  // Specify a list of file names.  Each file must be a single slice,
064  // and each slice must be of the same size. The files must be in
the
065  // correct order.
066  // Use SetFileName when reading a volume (multiple slice), since
067  // DataExtent will be modified after a SetFileNames call.
068  virtual void SetFileNames(vtkStringArray *);
069  vtkGetObjectMacro(FileNames, vtkStringArray);
[?]
246  vtkStringArray *FileNames;
247
248  char *InternalFileName;
249  char *FileName;


I it correct that '*FileNames' include the list from the open file
dialog? And can I work with it in my plugin? E.g.:

    vtkStringArray * inputArray;
    inputArray = this->GetFileNames();

I'm actually trying to get my mind in the code, so I'm sorry for some
stupid questions.


Greetings
Frank




On Tue, 2014-09-02 at 08:47 -0400, Utkarsh Ayachit wrote:
> Frank,
> 
> Are you using the AnalyzeNIfTIIO plugin or are you using the builtin
> version (which is coming from the VisItBridge) for the NifTI files?
> 
> Utkarsh
> 
> On Tue, Sep 2, 2014 at 3:28 AM, Frank tt <frank.tt at einseinself.org> wrote:
> > Hi,
> >
> > I'm Frank and new in developing ParaView.
> > My Setting is a client and a server. The server is image processing and
> > data server.
> > I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
> > volume (similar to DICOM).
> > When I try to load a group of NIfTI files, every file will be loaded as
> > one data set.
> >
> > Well, I want to write a file reader plugin, where I all selected files
> > (in alpha-num order) will be loaded as one volume.
> > I did a look at the example plugins and the paraview file reader
> > plugins.
> >
> > I've read in the paraview mailing list archive [1] and I think the
> > "FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
> > I also know about the ITK solution [2] to convert the data first. I
> > think this won't work in my setup, because I have limited compiling
> > possibilities on the server (where the data is.)
> >
> > I stuck to get started. Maybe someone can help me.
> >
> > Greetings
> >
> > Frank
> >
> >
> > [1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
> > [2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices
> >
> >
> >
> > _______________________________________________
> > Paraview-developers mailing list
> > Paraview-developers at paraview.org
> > http://public.kitware.com/mailman/listinfo/paraview-developers



From felipe.bordeu at ec-nantes.fr  Tue Sep  2 10:17:47 2014
From: felipe.bordeu at ec-nantes.fr (Felipe Bordeu)
Date: Tue, 02 Sep 2014 16:17:47 +0200
Subject: [Paraview-developers] apply a filter to only a selected block
In-Reply-To: <CADHSJjAd2AyLSFWuvL=0_P-fMnjkuGZc=FHhqhms5Pz3N-fWxA@mail.gmail.com>
References: <53FDE544.4070107@ec-nantes.fr>	<CADHSJjCBUzC=mi2weeCtt=0BxL=Um-nHogjPvonsGxt_-zpwqQ@mail.gmail.com>	<53FF5311.9090202@ec-nantes.fr>
 <CADHSJjAd2AyLSFWuvL=0_P-fMnjkuGZc=FHhqhms5Pz3N-fWxA@mail.gmail.com>
Message-ID: <5405D18B.2080705@ec-nantes.fr>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Le 28/08/2014 18:49, Utkarsh Ayachit a ?crit :
> Maybe go the custom filter route instead. You cannot have a choice of
> filters to chose from a comb-box, in that case. But you can  create a
> ExtractBlock+Filter version of every filter you're interested in.
> Something similar to how we deal with file-series readers, for
> example.
> 

It works like a charm!!!

I created a class (called vtkApplyOnBlocksFilter) to apply a filter to
only selected blocks. Then a vtkSIMetaFilterProxy (like the
vtkSIMetaReaderProxy).And now just generate one "SourceProxy" entry in
the xml description for every filter that I want to use.

Life is so much easier now. The added bonus. I put all theses new
filters in a submenu "On Block Filters" with theirs original icons.

Now any new ApplyOnBlocksFilters filters can be added to ParaView using
only a xml description.

Thanks Utkarsh and all the ParaView team.

Felipe

> 
> 
> On Thu, Aug 28, 2014 at 12:04 PM, Felipe Bordeu
> <felipe.bordeu at ec-nantes.fr> wrote:
> Thanks,
> 
> It works, but (I know there is always a but)...
> 
> 1) Paraview (at the creation of my mainfilter) calls RequestInformation
> on all the internal filters available, and because they don't have
> connections, I get plenty of :
> 
> ERROR: In
> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkExecutive.cxx,
> line 783
> vtkPVCompositeDataPipeline (0x686f5b0): Algorithm
> vtkPVArrayCalculator(0x6875210) returned failure for request:
> vtkInformation (0x686dde0)
> 
> ERROR: In
> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
> line 710
> vtkPVCompositeDataPipeline (0x6940a30): Input port 0 of algorithm
> vtkPVMetaClipDataSet(0x6921660) has 0 connections but is not optional.
> 
> 
> ERROR: In
> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
> line 710
> vtkPVCompositeDataPipeline (0x6969450): Input port 0 of algorithm
> vtkThreshold(0x696f170) has 0 connections but is not optional.
> 
> 
> ERROR: In
> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
> line 710
> vtkPVCompositeDataPipeline (0x6977740): Input port 0 of algorithm
> vtkDataSetSurfaceFilter(0x697f540) has 0 connections but is not optional.
> 
> 2) The panels of filters with custom widget are not generated correctly
> (the case of the calculator filter).
> 
> 3) Because of the filters don't have information about the input, the 3D
> widget are not plot in the renderview ( the plane of the clip filter)
> 
> I know I'm trying to do something tricky, but we do this kind of
> operation often (extract a block apply a filter, extract another block
> apply a filter, extract all the remaining block, and then do  "append
> datasets" to recover a multi-block structure). And even with this we
> lost the original structure of the multi-block.
> 
> the point 2 and tree are not that important (we can set the data from
> the panel, and for the calculator we still have the function field)
> 
> Felipe
> 
> ps: in the attachement there is an image of the filter
> 
> Le 28/08/2014 16:30, Utkarsh Ayachit a ?crit :
>>>> I am not sure I'd recommend anyone to do this :), but since you're
>>>> experimenting .. here's something to try:
>>>>
>>>> Don't do the SetInputDataObject() in RequestInformation, try passing
>>>> the output of ExtractBlock to the input in RequestData() instead.
>>>>
>>>> Utkarsh
>>>>
>>>> On Wed, Aug 27, 2014 at 10:03 AM, Felipe Bordeu
>>>> <felipe.bordeu at ec-nantes.fr> wrote:
>>>> Hi,
>>>>
>>>> I'm writing a filter to apply a filter(for example a cut, or clip ) to
>>>> only a selected block of a vtkMultiBlockData.
>>>>
>>>> So I start reading the implementation of the vtkExtractBlock and
>>>> vtkPVGlyphFilter (because I what to select interactively the filter to
>>>> apply in the same way we can choose the type of glyph).
>>>>
>>>> Now the problematic point. This is the  part of the xml to select the
>>>> filter to apply (in the same way as the glyph filter) :
>>>>
>>>>
>>>> <InputProperty command="SetalgoToApply"
>>>>                label="Filter To Apply"
>>>>                 name="Source">
>>>>
>>>>     <ProxyGroupDomain name="groups">
>>>>         <Group name="filters" />
>>>>     </ProxyGroupDomain>
>>>>
>>>>     <DataTypeDomain name="input_type" composite_data_supported="1">
>>>>         <DataType value="vtkDataObject" />
>>>>     </DataTypeDomain>
>>>>
>>>>     <ProxyListDomain name="proxy_list">
>>>>         <Proxy group="filters" name="Cut" />
>>>>         <!-- More filter here -->
>>>>     </ProxyListDomain>
>>>> </InputProperty>
>>>>
>>>> The interface is generated correctly in paraview.
>>>> In the code a receive the algorithm with this function
>>>>
>>>> vtkfiltertest::vtkSetMacro(algoToApply, vtkAlgorithmOutput*);
>>>>
>>>>
>>>> But I don't know how to set the input to the algorithm.
>>>>
>>>> I tried this (vtkDataSet* tmpinput  is a pointer to some data, )
>>>>
>>>> void vtkfiltertest::SetalgoToApply (vtkAlgorithmOutput* _arg) {
>>>>    if (this->algoToApply != _arg) {
>>>>      this->algoToApply = _arg;
>>>>      this->algoToApply->GetProducer()->SetInputDataObject(0,tmpinput);
>>>>      this->Modified();
>>>>    }
>>>> }
>>>>
>>>> and also in
>>>> vtkfiltertest::RequestInformation(){
>>>> ...
>>>> this->algoToApply->GetProducer()->SetInputDataObject(0,tmpinput);
>>>> ..
>>>> }
>>>>
>>>> but in paraview when I create the vtkfiltertest (click in the menu) I
>>>> get this:
>>>>
>>>> ERROR: In
>>>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>> line 710
>>>> vtkPVCompositeDataPipeline (0x7845e80): Input port 0 of algorithm
>>>> vtkPVMetaSliceDataSet(0x783d880) has 0 connections but is not optional.
>>>>
>>>> Is this possible of I'm asking to much???
>>>>
>>>> Thanks.
>>>>> _______________________________________________
>>>>> Paraview-developers mailing list
>>>>> Paraview-developers at paraview.org
>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
> 

- -- 
Felipe Bordeu Weldt
Ing?nieur de Recherche
- -------------------------------------
T?l. : 33 (0)2 40 37 16 57
Fax. : 33 (0)2 40 74 74 06
Felipe.Bordeu at ec-nantes.fr
Institut GeM - UMR CNRS 6183
?cole Centrale Nantes
1 Rue de La No?, 44321 Nantes, FRANCE
- -------------------------------------
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQEcBAEBAgAGBQJUBdGGAAoJEE/fMfNgU9/DcCYH/ic1RytF5WNehxqYeyerNeBQ
0IymWxeiWm+Zi5cOKwx4ebdSkl6Mmks1httD7+ojGqppSTJ6feNaQQhx/29haQv6
z99Yc5tefz5qIRsKQnfI36fgDhl+o/Fp7025L/6gmE2y+lTdgbIaiHRsp11ZVgk0
ys2wzdUbb69UFU++9pWhO6E7ebPG8jVfqMAL3QCDmQfqY1tDdGj8o5GRf8Gm2a2s
lZeSrzT0MibiyMjkKzWF3ygDiJa1BN0fJcO0+cYTxn9yUG/xaI16iImBcvpW97Rg
GD7ctUQm2ooUzMugpMekFtIJWzIIj61kBqBsmSYYQecjcqGheBgfT2whTO4enOc=
=btJx
-----END PGP SIGNATURE-----

From sebastien.jourdain at kitware.com  Tue Sep  2 10:45:51 2014
From: sebastien.jourdain at kitware.com (Sebastien Jourdain)
Date: Tue, 2 Sep 2014 08:45:51 -0600
Subject: [Paraview-developers] IE 9
In-Reply-To: <CAOkZLxruse40fBDSFG6DHAJNoCqco42ycZiJNuBEmUf4dsppng@mail.gmail.com>
References: <CAOkZLxpEMo==NiPoCw7nnGp0TqVDcemRTtwx4aYNwy_8FJB3Pg@mail.gmail.com>
 <CAOkZLxruse40fBDSFG6DHAJNoCqco42ycZiJNuBEmUf4dsppng@mail.gmail.com>
Message-ID: <CABObKxe4iwU8LsoCZssiAjjqOduW5hMuB_oZr9eV=0b1RGsHLQ@mail.gmail.com>

Currently you can not use IE9 with ParaViewWeb as IE9 does not support
WebSockets (http://caniuse.com/#feat=websockets)

But we are working on upgrading the library that we are using for the
communication between the client and the server (Autobahn http://autobahn.ws)
which will allow us to handle non-websocket communication for old user that
can't keep up with today's standard.

Seb




On Tue, Sep 2, 2014 at 5:03 AM, Albrecht Kyrieleis <akyrieleis at gmail.com>
wrote:

> Hi,
> on the paraviewweb site it is mentioned that Internet Explorer is
> supported from v.10. I would like to use IE9 with my pvweb application if
> possible. Which features of paraview can one expect not to work with IE9 ?
> Is there a way to reliably use IE9 by not using certain features of
> paraview ? Is there any plan to make paraviewweb work with IE9 at some time
> ? Many thanks.
> Kyr
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/38ae1123/attachment.html>

From utkarsh.ayachit at kitware.com  Tue Sep  2 10:47:22 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 2 Sep 2014 10:47:22 -0400
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
 <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>
Message-ID: <CADHSJjABbEeTrfxSfs8=6UU35Yv1yvaHBOaQmOWgh36-S2HmZA@mail.gmail.com>

Frank,

If you're writing your own reader, I'd suggesting adding
AddFileName(), RemoveAllFileNames() methods to your class and then
using the following property for your XML.

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

The "repeat_command" is the thing that'll tell ParaView that this
property can take an entire file list. Then you're free in you C++
code to arrange the files in whichever order you deem fit and process
them.

Utkarsh


On Tue, Sep 2, 2014 at 10:08 AM, Frank tt <frank.tt at einseinself.org> wrote:
> Hi Utkarsh,
>
> I use the AnalizeNIfTI plugin.
>
> Actually I'm trying to write an own plugin based on the AnalizeNIfTI
> plugin. I'm a little bit confused with this and other file readers. I
> know some file reades as they get something like a
> 'std::vector<std::string>' including a list of the files (choosen from
> the open file dialog).
> But here the readers have 'this->GetFileName()'. When I'm following
> that, im comming over the Analize header to
> VTK/IO/Image/vtkImageReader2.h:
>
> 062  // Description:
> 063  // Specify a list of file names.  Each file must be a single slice,
> 064  // and each slice must be of the same size. The files must be in
> the
> 065  // correct order.
> 066  // Use SetFileName when reading a volume (multiple slice), since
> 067  // DataExtent will be modified after a SetFileNames call.
> 068  virtual void SetFileNames(vtkStringArray *);
> 069  vtkGetObjectMacro(FileNames, vtkStringArray);
> [?]
> 246  vtkStringArray *FileNames;
> 247
> 248  char *InternalFileName;
> 249  char *FileName;
>
>
> I it correct that '*FileNames' include the list from the open file
> dialog? And can I work with it in my plugin? E.g.:
>
>     vtkStringArray * inputArray;
>     inputArray = this->GetFileNames();
>
> I'm actually trying to get my mind in the code, so I'm sorry for some
> stupid questions.
>
>
> Greetings
> Frank
>
>
>
>
> On Tue, 2014-09-02 at 08:47 -0400, Utkarsh Ayachit wrote:
>> Frank,
>>
>> Are you using the AnalyzeNIfTIIO plugin or are you using the builtin
>> version (which is coming from the VisItBridge) for the NifTI files?
>>
>> Utkarsh
>>
>> On Tue, Sep 2, 2014 at 3:28 AM, Frank tt <frank.tt at einseinself.org> wrote:
>> > Hi,
>> >
>> > I'm Frank and new in developing ParaView.
>> > My Setting is a client and a server. The server is image processing and
>> > data server.
>> > I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
>> > volume (similar to DICOM).
>> > When I try to load a group of NIfTI files, every file will be loaded as
>> > one data set.
>> >
>> > Well, I want to write a file reader plugin, where I all selected files
>> > (in alpha-num order) will be loaded as one volume.
>> > I did a look at the example plugins and the paraview file reader
>> > plugins.
>> >
>> > I've read in the paraview mailing list archive [1] and I think the
>> > "FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
>> > I also know about the ITK solution [2] to convert the data first. I
>> > think this won't work in my setup, because I have limited compiling
>> > possibilities on the server (where the data is.)
>> >
>> > I stuck to get started. Maybe someone can help me.
>> >
>> > Greetings
>> >
>> > Frank
>> >
>> >
>> > [1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
>> > [2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices
>> >
>> >
>> >
>> > _______________________________________________
>> > Paraview-developers mailing list
>> > Paraview-developers at paraview.org
>> > http://public.kitware.com/mailman/listinfo/paraview-developers
>
>

From utkarsh.ayachit at kitware.com  Tue Sep  2 10:42:48 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 2 Sep 2014 10:42:48 -0400
Subject: [Paraview-developers] apply a filter to only a selected block
In-Reply-To: <5405D18B.2080705@ec-nantes.fr>
References: <53FDE544.4070107@ec-nantes.fr>
 <CADHSJjCBUzC=mi2weeCtt=0BxL=Um-nHogjPvonsGxt_-zpwqQ@mail.gmail.com>
 <53FF5311.9090202@ec-nantes.fr>
 <CADHSJjAd2AyLSFWuvL=0_P-fMnjkuGZc=FHhqhms5Pz3N-fWxA@mail.gmail.com>
 <5405D18B.2080705@ec-nantes.fr>
Message-ID: <CADHSJjAPpPxVQmKDTkB2wNDr4zz3Ei_gftAtL1x3BJxQqsx+-g@mail.gmail.com>

Great! Glad it's working.

On Tue, Sep 2, 2014 at 10:17 AM, Felipe Bordeu
<felipe.bordeu at ec-nantes.fr> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Le 28/08/2014 18:49, Utkarsh Ayachit a ?crit :
>> Maybe go the custom filter route instead. You cannot have a choice of
>> filters to chose from a comb-box, in that case. But you can  create a
>> ExtractBlock+Filter version of every filter you're interested in.
>> Something similar to how we deal with file-series readers, for
>> example.
>>
>
> It works like a charm!!!
>
> I created a class (called vtkApplyOnBlocksFilter) to apply a filter to
> only selected blocks. Then a vtkSIMetaFilterProxy (like the
> vtkSIMetaReaderProxy).And now just generate one "SourceProxy" entry in
> the xml description for every filter that I want to use.
>
> Life is so much easier now. The added bonus. I put all theses new
> filters in a submenu "On Block Filters" with theirs original icons.
>
> Now any new ApplyOnBlocksFilters filters can be added to ParaView using
> only a xml description.
>
> Thanks Utkarsh and all the ParaView team.
>
> Felipe
>
>>
>>
>> On Thu, Aug 28, 2014 at 12:04 PM, Felipe Bordeu
>> <felipe.bordeu at ec-nantes.fr> wrote:
>> Thanks,
>>
>> It works, but (I know there is always a but)...
>>
>> 1) Paraview (at the creation of my mainfilter) calls RequestInformation
>> on all the internal filters available, and because they don't have
>> connections, I get plenty of :
>>
>> ERROR: In
>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkExecutive.cxx,
>> line 783
>> vtkPVCompositeDataPipeline (0x686f5b0): Algorithm
>> vtkPVArrayCalculator(0x6875210) returned failure for request:
>> vtkInformation (0x686dde0)
>>
>> ERROR: In
>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>> line 710
>> vtkPVCompositeDataPipeline (0x6940a30): Input port 0 of algorithm
>> vtkPVMetaClipDataSet(0x6921660) has 0 connections but is not optional.
>>
>>
>> ERROR: In
>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>> line 710
>> vtkPVCompositeDataPipeline (0x6969450): Input port 0 of algorithm
>> vtkThreshold(0x696f170) has 0 connections but is not optional.
>>
>>
>> ERROR: In
>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>> line 710
>> vtkPVCompositeDataPipeline (0x6977740): Input port 0 of algorithm
>> vtkDataSetSurfaceFilter(0x697f540) has 0 connections but is not optional.
>>
>> 2) The panels of filters with custom widget are not generated correctly
>> (the case of the calculator filter).
>>
>> 3) Because of the filters don't have information about the input, the 3D
>> widget are not plot in the renderview ( the plane of the clip filter)
>>
>> I know I'm trying to do something tricky, but we do this kind of
>> operation often (extract a block apply a filter, extract another block
>> apply a filter, extract all the remaining block, and then do  "append
>> datasets" to recover a multi-block structure). And even with this we
>> lost the original structure of the multi-block.
>>
>> the point 2 and tree are not that important (we can set the data from
>> the panel, and for the calculator we still have the function field)
>>
>> Felipe
>>
>> ps: in the attachement there is an image of the filter
>>
>> Le 28/08/2014 16:30, Utkarsh Ayachit a ?crit :
>>>>> I am not sure I'd recommend anyone to do this :), but since you're
>>>>> experimenting .. here's something to try:
>>>>>
>>>>> Don't do the SetInputDataObject() in RequestInformation, try passing
>>>>> the output of ExtractBlock to the input in RequestData() instead.
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Wed, Aug 27, 2014 at 10:03 AM, Felipe Bordeu
>>>>> <felipe.bordeu at ec-nantes.fr> wrote:
>>>>> Hi,
>>>>>
>>>>> I'm writing a filter to apply a filter(for example a cut, or clip ) to
>>>>> only a selected block of a vtkMultiBlockData.
>>>>>
>>>>> So I start reading the implementation of the vtkExtractBlock and
>>>>> vtkPVGlyphFilter (because I what to select interactively the filter to
>>>>> apply in the same way we can choose the type of glyph).
>>>>>
>>>>> Now the problematic point. This is the  part of the xml to select the
>>>>> filter to apply (in the same way as the glyph filter) :
>>>>>
>>>>>
>>>>> <InputProperty command="SetalgoToApply"
>>>>>                label="Filter To Apply"
>>>>>                 name="Source">
>>>>>
>>>>>     <ProxyGroupDomain name="groups">
>>>>>         <Group name="filters" />
>>>>>     </ProxyGroupDomain>
>>>>>
>>>>>     <DataTypeDomain name="input_type" composite_data_supported="1">
>>>>>         <DataType value="vtkDataObject" />
>>>>>     </DataTypeDomain>
>>>>>
>>>>>     <ProxyListDomain name="proxy_list">
>>>>>         <Proxy group="filters" name="Cut" />
>>>>>         <!-- More filter here -->
>>>>>     </ProxyListDomain>
>>>>> </InputProperty>
>>>>>
>>>>> The interface is generated correctly in paraview.
>>>>> In the code a receive the algorithm with this function
>>>>>
>>>>> vtkfiltertest::vtkSetMacro(algoToApply, vtkAlgorithmOutput*);
>>>>>
>>>>>
>>>>> But I don't know how to set the input to the algorithm.
>>>>>
>>>>> I tried this (vtkDataSet* tmpinput  is a pointer to some data, )
>>>>>
>>>>> void vtkfiltertest::SetalgoToApply (vtkAlgorithmOutput* _arg) {
>>>>>    if (this->algoToApply != _arg) {
>>>>>      this->algoToApply = _arg;
>>>>>      this->algoToApply->GetProducer()->SetInputDataObject(0,tmpinput);
>>>>>      this->Modified();
>>>>>    }
>>>>> }
>>>>>
>>>>> and also in
>>>>> vtkfiltertest::RequestInformation(){
>>>>> ...
>>>>> this->algoToApply->GetProducer()->SetInputDataObject(0,tmpinput);
>>>>> ..
>>>>> }
>>>>>
>>>>> but in paraview when I create the vtkfiltertest (click in the menu) I
>>>>> get this:
>>>>>
>>>>> ERROR: In
>>>>> /Big/fbordeu/sources/ParaView4/VTK/Common/ExecutionModel/vtkDemandDrivenPipeline.cxx,
>>>>> line 710
>>>>> vtkPVCompositeDataPipeline (0x7845e80): Input port 0 of algorithm
>>>>> vtkPVMetaSliceDataSet(0x783d880) has 0 connections but is not optional.
>>>>>
>>>>> Is this possible of I'm asking to much???
>>>>>
>>>>> Thanks.
>>>>>> _______________________________________________
>>>>>> Paraview-developers mailing list
>>>>>> Paraview-developers at paraview.org
>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>
>
> - --
> Felipe Bordeu Weldt
> Ing?nieur de Recherche
> - -------------------------------------
> T?l. : 33 (0)2 40 37 16 57
> Fax. : 33 (0)2 40 74 74 06
> Felipe.Bordeu at ec-nantes.fr
> Institut GeM - UMR CNRS 6183
> ?cole Centrale Nantes
> 1 Rue de La No?, 44321 Nantes, FRANCE
> - -------------------------------------
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.11 (GNU/Linux)
>
> iQEcBAEBAgAGBQJUBdGGAAoJEE/fMfNgU9/DcCYH/ic1RytF5WNehxqYeyerNeBQ
> 0IymWxeiWm+Zi5cOKwx4ebdSkl6Mmks1httD7+ojGqppSTJ6feNaQQhx/29haQv6
> z99Yc5tefz5qIRsKQnfI36fgDhl+o/Fp7025L/6gmE2y+lTdgbIaiHRsp11ZVgk0
> ys2wzdUbb69UFU++9pWhO6E7ebPG8jVfqMAL3QCDmQfqY1tDdGj8o5GRf8Gm2a2s
> lZeSrzT0MibiyMjkKzWF3ygDiJa1BN0fJcO0+cYTxn9yUG/xaI16iImBcvpW97Rg
> GD7ctUQm2ooUzMugpMekFtIJWzIIj61kBqBsmSYYQecjcqGheBgfT2whTO4enOc=
> =btJx
> -----END PGP SIGNATURE-----
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From sankhesh.jhaveri at kitware.com  Tue Sep  2 12:31:12 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Tue, 02 Sep 2014 09:31:12 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Monday, September 01 2014
Message-ID: <20140902163103.879FA7FECE@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]Win7x64-vs10-static-release 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pv.OpenHelp , 1 , ['Win7x64-vs10-static-release']

   Issues with submitters
   The following expected submitters did not submit:
   KargadMac.kitware , Lion-gcc-pvVTK

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-02&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-static-release
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-02&name=pv.OpenHelp

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/37dc9692/attachment.html>

From frank.tt at einseinself.org  Tue Sep  2 13:04:59 2014
From: frank.tt at einseinself.org (Frank TT)
Date: Tue, 02 Sep 2014 19:04:59 +0200
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <CADHSJjABbEeTrfxSfs8=6UU35Yv1yvaHBOaQmOWgh36-S2HmZA@mail.gmail.com>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
 <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjABbEeTrfxSfs8=6UU35Yv1yvaHBOaQmOWgh36-S2HmZA@mail.gmail.com>
Message-ID: <1409677499.1841.18.camel@Judika>

Thx. I'll try it.

I don't need my own plugin. Your suggestion looks like it's possible to
try this with the existing AnalizeNIfTI plugin? 
(I just want to make a 3D volume out of a stack of 2D foo%03d.nii
files.)

Frank


On Tue, 2014-09-02 at 10:47 -0400, Utkarsh Ayachit wrote: 
> Frank,
> 
> If you're writing your own reader, I'd suggesting adding
> AddFileName(), RemoveAllFileNames() methods to your class and then
> using the following property for your XML.
> 
>  <StringVectorProperty
>                             clean_command="RemoveAllFileNames"
>                             command="AddFileName"
>                             name="FileName"
>                             number_of_elements="1"
>                             panel_visibility="never"
>                             repeat_command="1">
>         <FileListDomain name="files" />
>         <Documentation>The list of files to be read by the
>         reader.</Documentation>
>  </StringVectorProperty>
> 
> The "repeat_command" is the thing that'll tell ParaView that this
> property can take an entire file list. Then you're free in you C++
> code to arrange the files in whichever order you deem fit and process
> them.
> 
> Utkarsh
> 
> 
> On Tue, Sep 2, 2014 at 10:08 AM, Frank tt <frank.tt at einseinself.org> wrote:
> > Hi Utkarsh,
> >
> > I use the AnalizeNIfTI plugin.
> >
> > Actually I'm trying to write an own plugin based on the AnalizeNIfTI
> > plugin. I'm a little bit confused with this and other file readers. I
> > know some file reades as they get something like a
> > 'std::vector<std::string>' including a list of the files (choosen from
> > the open file dialog).
> > But here the readers have 'this->GetFileName()'. When I'm following
> > that, im comming over the Analize header to
> > VTK/IO/Image/vtkImageReader2.h:
> >
> > 062  // Description:
> > 063  // Specify a list of file names.  Each file must be a single slice,
> > 064  // and each slice must be of the same size. The files must be in
> > the
> > 065  // correct order.
> > 066  // Use SetFileName when reading a volume (multiple slice), since
> > 067  // DataExtent will be modified after a SetFileNames call.
> > 068  virtual void SetFileNames(vtkStringArray *);
> > 069  vtkGetObjectMacro(FileNames, vtkStringArray);
> > [?]
> > 246  vtkStringArray *FileNames;
> > 247
> > 248  char *InternalFileName;
> > 249  char *FileName;
> >
> >
> > I it correct that '*FileNames' include the list from the open file
> > dialog? And can I work with it in my plugin? E.g.:
> >
> >     vtkStringArray * inputArray;
> >     inputArray = this->GetFileNames();
> >
> > I'm actually trying to get my mind in the code, so I'm sorry for some
> > stupid questions.
> >
> >
> > Greetings
> > Frank
> >
> >
> >
> >
> > On Tue, 2014-09-02 at 08:47 -0400, Utkarsh Ayachit wrote:
> >> Frank,
> >>
> >> Are you using the AnalyzeNIfTIIO plugin or are you using the builtin
> >> version (which is coming from the VisItBridge) for the NifTI files?
> >>
> >> Utkarsh
> >>
> >> On Tue, Sep 2, 2014 at 3:28 AM, Frank tt <frank.tt at einseinself.org> wrote:
> >> > Hi,
> >> >
> >> > I'm Frank and new in developing ParaView.
> >> > My Setting is a client and a server. The server is image processing and
> >> > data server.
> >> > I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
> >> > volume (similar to DICOM).
> >> > When I try to load a group of NIfTI files, every file will be loaded as
> >> > one data set.
> >> >
> >> > Well, I want to write a file reader plugin, where I all selected files
> >> > (in alpha-num order) will be loaded as one volume.
> >> > I did a look at the example plugins and the paraview file reader
> >> > plugins.
> >> >
> >> > I've read in the paraview mailing list archive [1] and I think the
> >> > "FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
> >> > I also know about the ITK solution [2] to convert the data first. I
> >> > think this won't work in my setup, because I have limited compiling
> >> > possibilities on the server (where the data is.)
> >> >
> >> > I stuck to get started. Maybe someone can help me.
> >> >
> >> > Greetings
> >> >
> >> > Frank
> >> >
> >> >
> >> > [1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
> >> > [2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices
> >> >
> >> >
> >> >
> >> > _______________________________________________
> >> > Paraview-developers mailing list
> >> > Paraview-developers at paraview.org
> >> > http://public.kitware.com/mailman/listinfo/paraview-developers
> >
> >



From utkarsh.ayachit at kitware.com  Tue Sep  2 13:11:28 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 2 Sep 2014 13:11:28 -0400
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <1409677499.1841.18.camel@Judika>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
 <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjABbEeTrfxSfs8=6UU35Yv1yvaHBOaQmOWgh36-S2HmZA@mail.gmail.com>
 <1409677499.1841.18.camel@Judika>
Message-ID: <CADHSJjAPu_uL7JqUUWGDhdF9mOBawGaEdP7dy21_b7ej_xXC+Q@mail.gmail.com>

No, I am not sure if the current AnalizeNIfTI plugin supports reading
image stacks.

On Tue, Sep 2, 2014 at 1:04 PM, Frank TT <frank.tt at einseinself.org> wrote:
> Thx. I'll try it.
>
> I don't need my own plugin. Your suggestion looks like it's possible to
> try this with the existing AnalizeNIfTI plugin?
> (I just want to make a 3D volume out of a stack of 2D foo%03d.nii
> files.)
>
> Frank
>
>
> On Tue, 2014-09-02 at 10:47 -0400, Utkarsh Ayachit wrote:
>> Frank,
>>
>> If you're writing your own reader, I'd suggesting adding
>> AddFileName(), RemoveAllFileNames() methods to your class and then
>> using the following property for your XML.
>>
>>  <StringVectorProperty
>>                             clean_command="RemoveAllFileNames"
>>                             command="AddFileName"
>>                             name="FileName"
>>                             number_of_elements="1"
>>                             panel_visibility="never"
>>                             repeat_command="1">
>>         <FileListDomain name="files" />
>>         <Documentation>The list of files to be read by the
>>         reader.</Documentation>
>>  </StringVectorProperty>
>>
>> The "repeat_command" is the thing that'll tell ParaView that this
>> property can take an entire file list. Then you're free in you C++
>> code to arrange the files in whichever order you deem fit and process
>> them.
>>
>> Utkarsh
>>
>>
>> On Tue, Sep 2, 2014 at 10:08 AM, Frank tt <frank.tt at einseinself.org> wrote:
>> > Hi Utkarsh,
>> >
>> > I use the AnalizeNIfTI plugin.
>> >
>> > Actually I'm trying to write an own plugin based on the AnalizeNIfTI
>> > plugin. I'm a little bit confused with this and other file readers. I
>> > know some file reades as they get something like a
>> > 'std::vector<std::string>' including a list of the files (choosen from
>> > the open file dialog).
>> > But here the readers have 'this->GetFileName()'. When I'm following
>> > that, im comming over the Analize header to
>> > VTK/IO/Image/vtkImageReader2.h:
>> >
>> > 062  // Description:
>> > 063  // Specify a list of file names.  Each file must be a single slice,
>> > 064  // and each slice must be of the same size. The files must be in
>> > the
>> > 065  // correct order.
>> > 066  // Use SetFileName when reading a volume (multiple slice), since
>> > 067  // DataExtent will be modified after a SetFileNames call.
>> > 068  virtual void SetFileNames(vtkStringArray *);
>> > 069  vtkGetObjectMacro(FileNames, vtkStringArray);
>> > [?]
>> > 246  vtkStringArray *FileNames;
>> > 247
>> > 248  char *InternalFileName;
>> > 249  char *FileName;
>> >
>> >
>> > I it correct that '*FileNames' include the list from the open file
>> > dialog? And can I work with it in my plugin? E.g.:
>> >
>> >     vtkStringArray * inputArray;
>> >     inputArray = this->GetFileNames();
>> >
>> > I'm actually trying to get my mind in the code, so I'm sorry for some
>> > stupid questions.
>> >
>> >
>> > Greetings
>> > Frank
>> >
>> >
>> >
>> >
>> > On Tue, 2014-09-02 at 08:47 -0400, Utkarsh Ayachit wrote:
>> >> Frank,
>> >>
>> >> Are you using the AnalyzeNIfTIIO plugin or are you using the builtin
>> >> version (which is coming from the VisItBridge) for the NifTI files?
>> >>
>> >> Utkarsh
>> >>
>> >> On Tue, Sep 2, 2014 at 3:28 AM, Frank tt <frank.tt at einseinself.org> wrote:
>> >> > Hi,
>> >> >
>> >> > I'm Frank and new in developing ParaView.
>> >> > My Setting is a client and a server. The server is image processing and
>> >> > data server.
>> >> > I have a huge stack of NIfTI files (*.nii). Every file is a slice of a
>> >> > volume (similar to DICOM).
>> >> > When I try to load a group of NIfTI files, every file will be loaded as
>> >> > one data set.
>> >> >
>> >> > Well, I want to write a file reader plugin, where I all selected files
>> >> > (in alpha-num order) will be loaded as one volume.
>> >> > I did a look at the example plugins and the paraview file reader
>> >> > plugins.
>> >> >
>> >> > I've read in the paraview mailing list archive [1] and I think the
>> >> > "FileSeriesReaderProxy" tag in the ServerManager xml won't work for me.
>> >> > I also know about the ITK solution [2] to convert the data first. I
>> >> > think this won't work in my setup, because I have limited compiling
>> >> > possibilities on the server (where the data is.)
>> >> >
>> >> > I stuck to get started. Maybe someone can help me.
>> >> >
>> >> > Greetings
>> >> >
>> >> > Frank
>> >> >
>> >> >
>> >> > [1] http://www.paraview.org/pipermail/paraview/2009-January/010621.html
>> >> > [2] http://www.itk.org/Wiki/ITK/Examples/IO/VolumeFromSlices
>> >> >
>> >> >
>> >> >
>> >> > _______________________________________________
>> >> > Paraview-developers mailing list
>> >> > Paraview-developers at paraview.org
>> >> > http://public.kitware.com/mailman/listinfo/paraview-developers
>> >
>> >
>
>

From akyrieleis at gmail.com  Tue Sep  2 14:06:03 2014
From: akyrieleis at gmail.com (Albrecht Kyrieleis)
Date: Tue, 2 Sep 2014 19:06:03 +0100
Subject: [Paraview-developers] prevent multiple pvpython processes
Message-ID: <CAOkZLxrVgTbrAPBpNvkUn8yZBdxjjv=X4aEe7ig4qo-Nqq6Uug@mail.gmail.com>

Hi,
I have been successfully running a custom application based on ParavieWeb
4.1. Everything works fine using the Chrome browser, v.35.0

When I start a new browser instance displaying my application then a new
process pvpython is started as expected. Closing the browser again does not
end the process. However, after a while the process terminates
automatically. Is this the way closed visualisation sessions are
automatically cleaned up or should one rather enforce termination of the
pvpython process as soon as the session is ended ?

Thank you

kyr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/f39a6753/attachment.html>

From sebastien.jourdain at kitware.com  Tue Sep  2 14:10:50 2014
From: sebastien.jourdain at kitware.com (Sebastien Jourdain)
Date: Tue, 2 Sep 2014 12:10:50 -0600
Subject: [Paraview-developers] prevent multiple pvpython processes
In-Reply-To: <CAOkZLxrVgTbrAPBpNvkUn8yZBdxjjv=X4aEe7ig4qo-Nqq6Uug@mail.gmail.com>
References: <CAOkZLxrVgTbrAPBpNvkUn8yZBdxjjv=X4aEe7ig4qo-Nqq6Uug@mail.gmail.com>
Message-ID: <CABObKxdwB8XQ1YQg-tFdq1EUBZnP3FVik2Sjn=j+FsgGKrGOgw@mail.gmail.com>

The client should let the server knows that its services are not needed
anymore.
We do that using the following callback:

<body onbeforeunload="stop()" onunload="stop()">

Where stop() is just:

var stop = function() {
     connection.session.call('application.exit'); // For ParaView 4.2
}

Just be aware the string in the call() method would be different on
ParaView 4.1. I think, it should be "vtk:exit". But if you look at the
sample applications of your code base, you should see what is the correct
string to use.

Seb


On Tue, Sep 2, 2014 at 12:06 PM, Albrecht Kyrieleis <akyrieleis at gmail.com>
wrote:

> Hi,
> I have been successfully running a custom application based on ParavieWeb
> 4.1. Everything works fine using the Chrome browser, v.35.0
>
> When I start a new browser instance displaying my application then a new
> process pvpython is started as expected. Closing the browser again does not
> end the process. However, after a while the process terminates
> automatically. Is this the way closed visualisation sessions are
> automatically cleaned up or should one rather enforce termination of the
> pvpython process as soon as the session is ended ?
>
> Thank you
>
> kyr
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140902/cebb5563/attachment-0001.html>

From sankhesh.jhaveri at kitware.com  Wed Sep  3 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Wed, 03 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Tuesday, September 02 2014
Message-ID: <20140903163103.87F6480D23@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]ubuntu-x64-nightlynext 1
   [2]Lion-gcc-pvVTK 3
   [3]ubuntu-x64-nightlynext 2
   [4]fedora-x64-icc-nightlynext-release 1
   -------------------- REPORT --------------------
   7 FAILURES
   [5]pv.LoadSaveStateAnimation , 1 , ['ubuntu-x64-nightlynext']
   [6]pvweb-chrome.TestApp-all , 1 , ['ubuntu-x64-nightlynext']
   [7]pvcs.CatalystLiveSetBreakpoint , 1 , ['ubuntu-x64-nightlynext']
   [8]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [9]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [10]pv.LogColorMap , 1 , ['fedora-x64-icc-nightlynext-release']
   [11]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']

   Issues with submitters
   The following expected submitters did not submit:
   kamino.kitware , Mac10.7.5-gcc-nightlymaster-release
   kamino.kitware , Mac10.7.5-clang-nightlymaster-release

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-03&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-03&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-03&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   4. http://open.cdash.org/index.php?project=ParaView&date=2014-09-03&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pv.LoadSaveStateAnimation
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pvweb-chrome.TestApp-all
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pvcs.CatalystLiveSetBreakpoint
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pvcs.HistogramSelection
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pvcrs.HistogramSelection
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pv.LogColorMap
  11. http://open.cdash.org/testSummary.php?project=9&date=2014-09-03&name=pv.HistogramSelection

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140903/600afd5c/attachment.html>

From jlih at masonlive.gmu.edu  Wed Sep  3 15:33:03 2014
From: jlih at masonlive.gmu.edu (jlih)
Date: Wed, 3 Sep 2014 19:33:03 +0000
Subject: [Paraview-developers] OpenGL support of running volume rendering on
	EC2
Message-ID: <1409772782763.67158@masonlive.gmu.edu>

Dear all,

I am running a volume rendering script on Amazon EC2 and got the following error message.

Commend:

pvbatch --use-offscreen-rendering test_parallel.py


Error message:

ERROR: In /root/ParaView-v4.1.0/VTK/Rendering/VolumeOpenGL/vtkOpenGLProjectedTetrahedraMapper.cxx, line 183
vtkOpenGLProjectedTetrahedraMapper (0x438c880): The required extensions are not supported.

Warning: In /root/ParaView-v4.1.0/VTK/Rendering/OpenGL/vtkOpenGLExtensionManager.cxx, line 750
vtkOpenGLExtensionManager (0x46ff4a0): Attempting to load GL_VERSION_1_3, which is not supported.


I guess there is something missing for the OpenGL. I have installed and set up MESA when compiling the Paraview and ran the simple example successfully.


The OS is Red Hat Enterprise Linux Server release 6.5 (Santiago). The version of Mesa is 9.2.


Thanks,

Jing


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140903/a661f692/attachment.html>

From utkarsh.ayachit at kitware.com  Thu Sep  4 09:41:06 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 4 Sep 2014 09:41:06 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjAxi-RHb2wWg6x7EV3z0KYSWJOonatNVXkp2uVhE70ykA@mail.gmail.com>

SUMMARY
---------------------------------------------
Topics merged into master:
(VTK)   add-streamline-seedids
(VTK)   bring-vktweb-fix-for-python26
        pv-cinema-add-test
        pvweb-fix-ctest-concurrency-issue
        pvweb-fix-scalarbar-positions-under-webgl
        pvweb-tests-more-portnumber-mgmt
(VTK)   split-mode-request
        sqtk-updates

---------------------------------------------
Topics reverted from next:
  drag_n_drop_file_open

From Gerald.Lodron at joanneum.at  Thu Sep  4 09:23:45 2014
From: Gerald.Lodron at joanneum.at (Lodron, Gerald)
Date: Thu, 4 Sep 2014 13:23:45 +0000
Subject: [Paraview-developers] Paraview and VS2013 compile errors
Message-ID: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>

Hi

I am compiling PV with qt 5.2.1, internal VTK and boost 1.55 (for ReebGraph), under 64bit (without python) and get following errors:

Min max unknown in protobuf, following patch works:
https://code.google.com/p/protobuf/issues/detail?id=531

I also get many
2>C:\Program Files (x86)\Microsoft Visual Studio 12.0\VC\include\type_traits(1220): error C2065: 'max_align_t' : undeclared identifier

Adding #include <stddef.h> to common.h of protobuf solved that in porotobuf but I get it also in CommandLineArguments.cxx of vtksys where that solution does not worked...

Additionally I get many linker errors:
6>QtTesting_automoc.obj : error LNK2005: "public: static struct QMetaObject const pqComboBoxEventTranslator::staticMetaObject" (?staticMetaObject at pqComboBoxEventTranslator@@2UQMetaObject@@B) already defined in moc_pqComboBoxEventTranslator.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual struct QMetaObject const * __cdecl QVTKPlugin::metaObject(void)const " (?metaObject at QVTKPlugin@@UEBAPEBUQMetaObject@@XZ) already defined in moc_Q4VTKWidgetPlugin.obj
6>QtTesting_automoc.obj : error LNK2005: "private: static void __cdecl pqComboBoxEventTranslator::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at pqComboBoxEventTranslator@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_pqComboBoxEventTranslator.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void * __cdecl QVTKPlugin::qt_metacast(char const *)" (?qt_metacast at QVTKPlugin@@UEAAPEAXPEBD at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual int __cdecl QVTKPlugin::qt_metacall(enum QMetaObject::Call,int,void * *)" (?qt_metacall at QVTKPlugin@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void __cdecl QVTKPlugin::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at QVTKPlugin@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual struct QMetaObject const * __cdecl QVTKWidget::metaObject(void)const " (?metaObject at QVTKWidget@@UEBAPEBUQMetaObject@@XZ) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void * __cdecl QVTKWidget::qt_metacast(char const *)" (?qt_metacast at QVTKWidget@@UEAAPEAXPEBD at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual int __cdecl QVTKWidget::qt_metacall(enum QMetaObject::Call,int,void * *)" (?qt_metacall at QVTKWidget@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void __cdecl QVTKWidget::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at QVTKWidget@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: qt_plugin_query_metadata already defined in moc_Q4VTKWidgetPlugin.obj
9>QVTKWidgetPlugin_automoc.obj : error LNK2005: qt_plugin_instance already defined in moc_Q4VTKWidgetPlugin.obj

Any hints?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140904/3132dd91/attachment.html>

From sankhesh.jhaveri at kitware.com  Thu Sep  4 12:31:08 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Thu, 04 Sep 2014 09:31:08 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Wednesday, September 03 2014
Message-ID: <20140904163104.510EF7FECA@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Win7x64-vs10-static-release 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pv.OpenHelp , 1 , ['Win7x64-vs10-static-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [3]ubuntu-x64-coverage 1
   [4]Lion-gcc-pvVTK 3
   [5]Win64-vs9-static-release 1
   -------------------- REPORT --------------------
   5 FAILURES
   [6]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [7]pvcs.NormalGlyphs , 1 , ['ubuntu-x64-coverage']
   [8]pvcs.Contour , 1 , ['Win64-vs9-static-release']
   [9]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [10]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-04&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-static-release
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pv.OpenHelp
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-04&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
   4. http://open.cdash.org/index.php?project=ParaView&date=2014-09-04&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   5. http://open.cdash.org/index.php?project=ParaView&date=2014-09-04&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-static-release
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pvcrs.HistogramSelection
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pvcs.NormalGlyphs
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pvcs.Contour
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pv.HistogramSelection
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-04&name=pvcs.HistogramSelection

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140904/8e2e0c11/attachment-0001.html>

From frank.tt at einseinself.org  Thu Sep  4 13:34:01 2014
From: frank.tt at einseinself.org (Frank TT)
Date: Thu, 04 Sep 2014 19:34:01 +0200
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
Message-ID: <1409852041.1875.17.camel@Judika>

On Thu, 2014-09-04 at 13:23 +0000, Lodron, Gerald wrote: 
> Hi
> 
> I am compiling PV with qt 5.2.1, internal VTK and boost 1.55 (for ReebGraph), under 64bit (without python) and get following errors:
> 
> Min max unknown in protobuf, following patch works:
> https://code.google.com/p/protobuf/issues/detail?id=531
> 
> I also get many
> 2>C:\Program Files (x86)\Microsoft Visual Studio 12.0\VC\include\type_traits(1220): error C2065: 'max_align_t' : undeclared identifier
> 
> Adding #include <stddef.h> to common.h of protobuf solved that in porotobuf but I get it also in CommandLineArguments.cxx of vtksys where that solution does not worked...
> 
> Additionally I get many linker errors:
> 6>QtTesting_automoc.obj : error LNK2005: "public: static struct QMetaObject const pqComboBoxEventTranslator::staticMetaObject" (?staticMetaObject at pqComboBoxEventTranslator@@2UQMetaObject@@B) already defined in moc_pqComboBoxEventTranslator.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual struct QMetaObject const * __cdecl QVTKPlugin::metaObject(void)const " (?metaObject at QVTKPlugin@@UEBAPEBUQMetaObject@@XZ) already defined in moc_Q4VTKWidgetPlugin.obj
> 6>QtTesting_automoc.obj : error LNK2005: "private: static void __cdecl pqComboBoxEventTranslator::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at pqComboBoxEventTranslator@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_pqComboBoxEventTranslator.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void * __cdecl QVTKPlugin::qt_metacast(char const *)" (?qt_metacast at QVTKPlugin@@UEAAPEAXPEBD at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual int __cdecl QVTKPlugin::qt_metacall(enum QMetaObject::Call,int,void * *)" (?qt_metacall at QVTKPlugin@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void __cdecl QVTKPlugin::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at QVTKPlugin@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual struct QMetaObject const * __cdecl QVTKWidget::metaObject(void)const " (?metaObject at QVTKWidget@@UEBAPEBUQMetaObject@@XZ) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void * __cdecl QVTKWidget::qt_metacast(char const *)" (?qt_metacast at QVTKWidget@@UEAAPEAXPEBD at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual int __cdecl QVTKWidget::qt_metacall(enum QMetaObject::Call,int,void * *)" (?qt_metacall at QVTKWidget@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void __cdecl QVTKWidget::qt_static_metacall(class QObject *,enum QMetaObject::Call,int,void * *)" (?qt_static_metacall at QVTKWidget@@CAXPEAVQObject@@W4Call at QMetaObject@@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: qt_plugin_query_metadata already defined in moc_Q4VTKWidgetPlugin.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: qt_plugin_instance already defined in moc_Q4VTKWidgetPlugin.obj
> 

On my first build I've got similar errors.

You see in the last pasted line "? moc_Q4VTKWidgetPlugin.obj" a hint. I
used Qt 4.8.6 and it builds. If you have a look in the CMakeLists.txt
there will it be more clear. For PV 4.1.0 you will find this in line 409
ff.
Maybe you try Qt 4.8.* to build PV.

Greetings
Frank


From Gerald.Lodron at joanneum.at  Thu Sep  4 14:04:48 2014
From: Gerald.Lodron at joanneum.at (Lodron, Gerald)
Date: Thu, 4 Sep 2014 18:04:48 +0000
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <1409852041.1875.17.camel@Judika>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
Message-ID: <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>

Hi

Omg, there is a Qt4 you are right. Should be fixed in root cmake! Any ideas of the max_alight_t issue?

No, our "institute" toolkit is also switched to qt5 so it would be adviceable that  I also switch my paraview/vtk stuff to same version. And I am also migrating from VS 2008 to 2013 at the same time, so many things to do. I also found out that I need to build qt with the no-angle and opengl switch to get the vtk opengl stuff working...., omg, everything back to beginning....

Does the knob of "ExternalVTK" work? In PV 4.1 this didn't work... Perhaps I will see it in 2-3 days when I rebuild qt, this would be really great ...

Thanks for hint and best regards,
Gerald



-----Urspr?ngliche Nachricht-----
Von: Paraview-developers [mailto:paraview-developers-bounces at paraview.org] Im Auftrag von Frank TT
Gesendet: Donnerstag, 04. September 2014 19:34
An: paraview-developers at paraview.org
Betreff: Re: [Paraview-developers] Paraview and VS2013 compile errors

On Thu, 2014-09-04 at 13:23 +0000, Lodron, Gerald wrote: 
> Hi
> 
> I am compiling PV with qt 5.2.1, internal VTK and boost 1.55 (for ReebGraph), under 64bit (without python) and get following errors:
> 
> Min max unknown in protobuf, following patch works:
> https://code.google.com/p/protobuf/issues/detail?id=531
> 
> I also get many
> 2>C:\Program Files (x86)\Microsoft Visual Studio 
> 2>12.0\VC\include\type_traits(1220): error C2065: 'max_align_t' : 
> 2>undeclared identifier
> 
> Adding #include <stddef.h> to common.h of protobuf solved that in porotobuf but I get it also in CommandLineArguments.cxx of vtksys where that solution does not worked...
> 
> Additionally I get many linker errors:
> 6>QtTesting_automoc.obj : error LNK2005: "public: static struct 
> 6>QMetaObject const pqComboBoxEventTranslator::staticMetaObject" 
> 6>(?staticMetaObject at pqComboBoxEventTranslator@@2UQMetaObject@@B) 
> 6>already defined in moc_pqComboBoxEventTranslator.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual 
> 9>struct QMetaObject const * __cdecl QVTKPlugin::metaObject(void)const 
> 9>" (?metaObject at QVTKPlugin@@UEBAPEBUQMetaObject@@XZ) already defined 
> 9>in moc_Q4VTKWidgetPlugin.obj
> 6>QtTesting_automoc.obj : error LNK2005: "private: static void __cdecl 
> 6>pqComboBoxEventTranslator::qt_static_metacall(class QObject *,enum 
> 6>QMetaObject::Call,int,void * *)" 
> 6>(?qt_static_metacall at pqComboBoxEventTranslator@@CAXPEAVQObject@@W4Ca
> 6>ll at QMetaObject@@HPEAPEAX at Z) already defined in 
> 6>moc_pqComboBoxEventTranslator.obj
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void 
> 9>* __cdecl QVTKPlugin::qt_metacast(char const *)" 
> 9>(?qt_metacast at QVTKPlugin@@UEAAPEAXPEBD at Z) already defined in 
> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error 
> 9>LNK2005: "public: virtual int __cdecl QVTKPlugin::qt_metacall(enum 
> 9>QMetaObject::Call,int,void * *)" 
> 9>(?qt_metacall at QVTKPlugin@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) 
> 9>already defined in moc_Q4VTKWidgetPlugin.obj 
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void 
> 9>__cdecl QVTKPlugin::qt_static_metacall(class QObject *,enum 
> 9>QMetaObject::Call,int,void * *)" 
> 9>(?qt_static_metacall at QVTKPlugin@@CAXPEAVQObject@@W4Call at QMetaObject@
> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj 
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual 
> 9>struct QMetaObject const * __cdecl QVTKWidget::metaObject(void)const 
> 9>" (?metaObject at QVTKWidget@@UEBAPEBUQMetaObject@@XZ) already defined 
> 9>in moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error 
> 9>LNK2005: "public: virtual void * __cdecl 
> 9>QVTKWidget::qt_metacast(char const *)" 
> 9>(?qt_metacast at QVTKWidget@@UEAAPEAXPEBD at Z) already defined in 
> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error 
> 9>LNK2005: "public: virtual int __cdecl QVTKWidget::qt_metacall(enum 
> 9>QMetaObject::Call,int,void * *)" 
> 9>(?qt_metacall at QVTKWidget@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z) 
> 9>already defined in moc_Q4VTKWidgetPlugin.obj 
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void 
> 9>__cdecl QVTKWidget::qt_static_metacall(class QObject *,enum 
> 9>QMetaObject::Call,int,void * *)" 
> 9>(?qt_static_metacall at QVTKWidget@@CAXPEAVQObject@@W4Call at QMetaObject@
> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj 
> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: 
> 9>qt_plugin_query_metadata already defined in 
> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error 
> 9>LNK2005: qt_plugin_instance already defined in 
> 9>moc_Q4VTKWidgetPlugin.obj
> 

On my first build I've got similar errors.

You see in the last pasted line "? moc_Q4VTKWidgetPlugin.obj" a hint. I used Qt 4.8.6 and it builds. If you have a look in the CMakeLists.txt there will it be more clear. For PV 4.1.0 you will find this in line 409 ff.
Maybe you try Qt 4.8.* to build PV.

Greetings
Frank

_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org
http://public.kitware.com/mailman/listinfo/paraview-developers

From utkarsh.ayachit at kitware.com  Thu Sep  4 14:56:18 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 4 Sep 2014 14:56:18 -0400
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
 <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
Message-ID: <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>

Gerald,

You should brace yourself :):
+ Qt 5 support in ParaView currently is experimental. We still aren't
officially supporting it.
+ VS2013 suppport for Superbuilds is experimental too. Should work,
but not if you're using internal Python yet.
+ External VTK is definitely not supported yet, I am afraid.

Utkarsh

On Thu, Sep 4, 2014 at 2:04 PM, Lodron, Gerald
<Gerald.Lodron at joanneum.at> wrote:
> Hi
>
> Omg, there is a Qt4 you are right. Should be fixed in root cmake! Any ideas of the max_alight_t issue?
>
> No, our "institute" toolkit is also switched to qt5 so it would be adviceable that  I also switch my paraview/vtk stuff to same version. And I am also migrating from VS 2008 to 2013 at the same time, so many things to do. I also found out that I need to build qt with the no-angle and opengl switch to get the vtk opengl stuff working...., omg, everything back to beginning....
>
> Does the knob of "ExternalVTK" work? In PV 4.1 this didn't work... Perhaps I will see it in 2-3 days when I rebuild qt, this would be really great ...
>
> Thanks for hint and best regards,
> Gerald
>
>
>
> -----Urspr?ngliche Nachricht-----
> Von: Paraview-developers [mailto:paraview-developers-bounces at paraview.org] Im Auftrag von Frank TT
> Gesendet: Donnerstag, 04. September 2014 19:34
> An: paraview-developers at paraview.org
> Betreff: Re: [Paraview-developers] Paraview and VS2013 compile errors
>
> On Thu, 2014-09-04 at 13:23 +0000, Lodron, Gerald wrote:
>> Hi
>>
>> I am compiling PV with qt 5.2.1, internal VTK and boost 1.55 (for ReebGraph), under 64bit (without python) and get following errors:
>>
>> Min max unknown in protobuf, following patch works:
>> https://code.google.com/p/protobuf/issues/detail?id=531
>>
>> I also get many
>> 2>C:\Program Files (x86)\Microsoft Visual Studio
>> 2>12.0\VC\include\type_traits(1220): error C2065: 'max_align_t' :
>> 2>undeclared identifier
>>
>> Adding #include <stddef.h> to common.h of protobuf solved that in porotobuf but I get it also in CommandLineArguments.cxx of vtksys where that solution does not worked...
>>
>> Additionally I get many linker errors:
>> 6>QtTesting_automoc.obj : error LNK2005: "public: static struct
>> 6>QMetaObject const pqComboBoxEventTranslator::staticMetaObject"
>> 6>(?staticMetaObject at pqComboBoxEventTranslator@@2UQMetaObject@@B)
>> 6>already defined in moc_pqComboBoxEventTranslator.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual
>> 9>struct QMetaObject const * __cdecl QVTKPlugin::metaObject(void)const
>> 9>" (?metaObject at QVTKPlugin@@UEBAPEBUQMetaObject@@XZ) already defined
>> 9>in moc_Q4VTKWidgetPlugin.obj
>> 6>QtTesting_automoc.obj : error LNK2005: "private: static void __cdecl
>> 6>pqComboBoxEventTranslator::qt_static_metacall(class QObject *,enum
>> 6>QMetaObject::Call,int,void * *)"
>> 6>(?qt_static_metacall at pqComboBoxEventTranslator@@CAXPEAVQObject@@W4Ca
>> 6>ll at QMetaObject@@HPEAPEAX at Z) already defined in
>> 6>moc_pqComboBoxEventTranslator.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void
>> 9>* __cdecl QVTKPlugin::qt_metacast(char const *)"
>> 9>(?qt_metacast at QVTKPlugin@@UEAAPEAXPEBD at Z) already defined in
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual int __cdecl QVTKPlugin::qt_metacall(enum
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_metacall at QVTKPlugin@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z)
>> 9>already defined in moc_Q4VTKWidgetPlugin.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void
>> 9>__cdecl QVTKPlugin::qt_static_metacall(class QObject *,enum
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_static_metacall at QVTKPlugin@@CAXPEAVQObject@@W4Call at QMetaObject@
>> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual
>> 9>struct QMetaObject const * __cdecl QVTKWidget::metaObject(void)const
>> 9>" (?metaObject at QVTKWidget@@UEBAPEBUQMetaObject@@XZ) already defined
>> 9>in moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual void * __cdecl
>> 9>QVTKWidget::qt_metacast(char const *)"
>> 9>(?qt_metacast at QVTKWidget@@UEAAPEAXPEBD at Z) already defined in
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual int __cdecl QVTKWidget::qt_metacall(enum
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_metacall at QVTKWidget@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z)
>> 9>already defined in moc_Q4VTKWidgetPlugin.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void
>> 9>__cdecl QVTKWidget::qt_static_metacall(class QObject *,enum
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_static_metacall at QVTKWidget@@CAXPEAVQObject@@W4Call at QMetaObject@
>> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005:
>> 9>qt_plugin_query_metadata already defined in
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: qt_plugin_instance already defined in
>> 9>moc_Q4VTKWidgetPlugin.obj
>>
>
> On my first build I've got similar errors.
>
> You see in the last pasted line "? moc_Q4VTKWidgetPlugin.obj" a hint. I used Qt 4.8.6 and it builds. If you have a look in the CMakeLists.txt there will it be more clear. For PV 4.1.0 you will find this in line 409 ff.
> Maybe you try Qt 4.8.* to build PV.
>
> Greetings
> Frank
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From Gerald.Lodron at joanneum.at  Thu Sep  4 16:13:18 2014
From: Gerald.Lodron at joanneum.at (Lodron, Gerald)
Date: Thu, 4 Sep 2014 20:13:18 +0000
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
 <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
 <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>
Message-ID: <5951482beedf45b68f60b8e80f760ff0@RZJMBX3.jr1.local>

Hi,

Thanks for clearification (could not find details about that stuff). Since we have our own build system I need to migrate Paraview and VTK  to it, so I am afraid that I cannot use the superbuild, I need to link to our own compiled python/qt/boost...Since I/my small group are the only ones which use vtk I could also use the internal VTK of PV  but it would be really great if paraview would get working with qt5 (since our internal's are not compatible with qt4 anymore and I/we work on an extremely old qt4 branch). 

Is there any timeline to support qt5 or the external VTK? Is PV generally not supporting VS2013 or only superbuild (which wouldn't be a problem)? The problem is that we cannot work with VS2008 any more since many other toolkits (especially PCL) use C++11 features (and our company IT only has VS2008 and 2013). 

Oh sometimes live is really hard....

Best regards,
Gerald





-----Urspr?ngliche Nachricht-----
Von: Utkarsh Ayachit [mailto:utkarsh.ayachit at kitware.com] 
Gesendet: Donnerstag, 04. September 2014 20:56
An: Lodron, Gerald
Cc: Frank TT; Paraview Developer (Paraview-developers at paraview.org)
Betreff: Re: [Paraview-developers] Paraview and VS2013 compile errors

Gerald,

You should brace yourself :):
+ Qt 5 support in ParaView currently is experimental. We still aren't
officially supporting it.
+ VS2013 suppport for Superbuilds is experimental too. Should work,
but not if you're using internal Python yet.
+ External VTK is definitely not supported yet, I am afraid.

Utkarsh

On Thu, Sep 4, 2014 at 2:04 PM, Lodron, Gerald <Gerald.Lodron at joanneum.at> wrote:
> Hi
>
> Omg, there is a Qt4 you are right. Should be fixed in root cmake! Any ideas of the max_alight_t issue?
>
> No, our "institute" toolkit is also switched to qt5 so it would be adviceable that  I also switch my paraview/vtk stuff to same version. And I am also migrating from VS 2008 to 2013 at the same time, so many things to do. I also found out that I need to build qt with the no-angle and opengl switch to get the vtk opengl stuff working...., omg, everything back to beginning....
>
> Does the knob of "ExternalVTK" work? In PV 4.1 this didn't work... Perhaps I will see it in 2-3 days when I rebuild qt, this would be really great ...
>
> Thanks for hint and best regards,
> Gerald
>
>
>
> -----Urspr?ngliche Nachricht-----
> Von: Paraview-developers 
> [mailto:paraview-developers-bounces at paraview.org] Im Auftrag von Frank 
> TT
> Gesendet: Donnerstag, 04. September 2014 19:34
> An: paraview-developers at paraview.org
> Betreff: Re: [Paraview-developers] Paraview and VS2013 compile errors
>
> On Thu, 2014-09-04 at 13:23 +0000, Lodron, Gerald wrote:
>> Hi
>>
>> I am compiling PV with qt 5.2.1, internal VTK and boost 1.55 (for ReebGraph), under 64bit (without python) and get following errors:
>>
>> Min max unknown in protobuf, following patch works:
>> https://code.google.com/p/protobuf/issues/detail?id=531
>>
>> I also get many
>> 2>C:\Program Files (x86)\Microsoft Visual Studio
>> 2>12.0\VC\include\type_traits(1220): error C2065: 'max_align_t' :
>> 2>undeclared identifier
>>
>> Adding #include <stddef.h> to common.h of protobuf solved that in porotobuf but I get it also in CommandLineArguments.cxx of vtksys where that solution does not worked...
>>
>> Additionally I get many linker errors:
>> 6>QtTesting_automoc.obj : error LNK2005: "public: static struct 
>> 6>QMetaObject const pqComboBoxEventTranslator::staticMetaObject"
>> 6>(?staticMetaObject at pqComboBoxEventTranslator@@2UQMetaObject@@B)
>> 6>already defined in moc_pqComboBoxEventTranslator.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual 
>> 9>struct QMetaObject const * __cdecl 
>> 9>QVTKPlugin::metaObject(void)const " 
>> 9>(?metaObject at QVTKPlugin@@UEBAPEBUQMetaObject@@XZ) already defined 
>> 9>in moc_Q4VTKWidgetPlugin.obj
>> 6>QtTesting_automoc.obj : error LNK2005: "private: static void 
>> 6>__cdecl pqComboBoxEventTranslator::qt_static_metacall(class QObject 
>> 6>*,enum QMetaObject::Call,int,void * *)"
>> 6>(?qt_static_metacall at pqComboBoxEventTranslator@@CAXPEAVQObject@@W4C
>> 6>a
>> 6>ll at QMetaObject@@HPEAPEAX at Z) already defined in 
>> 6>moc_pqComboBoxEventTranslator.obj
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual void
>> 9>* __cdecl QVTKPlugin::qt_metacast(char const *)"
>> 9>(?qt_metacast at QVTKPlugin@@UEAAPEAXPEBD at Z) already defined in 
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual int __cdecl QVTKPlugin::qt_metacall(enum 
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_metacall at QVTKPlugin@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z)
>> 9>already defined in moc_Q4VTKWidgetPlugin.obj 
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void 
>> 9>__cdecl QVTKPlugin::qt_static_metacall(class QObject *,enum 
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_static_metacall at QVTKPlugin@@CAXPEAVQObject@@W4Call at QMetaObject
>> 9>@
>> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj 
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "public: virtual 
>> 9>struct QMetaObject const * __cdecl 
>> 9>QVTKWidget::metaObject(void)const " 
>> 9>(?metaObject at QVTKWidget@@UEBAPEBUQMetaObject@@XZ) already defined 
>> 9>in moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual void * __cdecl 
>> 9>QVTKWidget::qt_metacast(char const *)"
>> 9>(?qt_metacast at QVTKWidget@@UEAAPEAXPEBD at Z) already defined in 
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: "public: virtual int __cdecl QVTKWidget::qt_metacall(enum 
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_metacall at QVTKWidget@@UEAAHW4Call at QMetaObject@@HPEAPEAX at Z)
>> 9>already defined in moc_Q4VTKWidgetPlugin.obj 
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005: "private: static void 
>> 9>__cdecl QVTKWidget::qt_static_metacall(class QObject *,enum 
>> 9>QMetaObject::Call,int,void * *)"
>> 9>(?qt_static_metacall at QVTKWidget@@CAXPEAVQObject@@W4Call at QMetaObject
>> 9>@
>> 9>@HPEAPEAX at Z) already defined in moc_Q4VTKWidgetPlugin.obj 
>> 9>QVTKWidgetPlugin_automoc.obj : error LNK2005:
>> 9>qt_plugin_query_metadata already defined in 
>> 9>moc_Q4VTKWidgetPlugin.obj QVTKWidgetPlugin_automoc.obj : error
>> 9>LNK2005: qt_plugin_instance already defined in 
>> 9>moc_Q4VTKWidgetPlugin.obj
>>
>
> On my first build I've got similar errors.
>
> You see in the last pasted line "? moc_Q4VTKWidgetPlugin.obj" a hint. I used Qt 4.8.6 and it builds. If you have a look in the CMakeLists.txt there will it be more clear. For PV 4.1.0 you will find this in line 409 ff.
> Maybe you try Qt 4.8.* to build PV.
>
> Greetings
> Frank
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From utkarsh.ayachit at kitware.com  Thu Sep  4 17:17:15 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 4 Sep 2014 17:17:15 -0400
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <5951482beedf45b68f60b8e80f760ff0@RZJMBX3.jr1.local>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
 <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
 <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>
 <5951482beedf45b68f60b8e80f760ff0@RZJMBX3.jr1.local>
Message-ID: <CADHSJjBF_MsC0cntSbQPUqAL439BFyTt_uivaOj70wh0M9AGfQ@mail.gmail.com>

> Is there any timeline to support qt5

The plan is to support Qt 5 for 4.3.

> or the external VTK?

There's no plan for this right now, although I personally want to
support this. It just hard to find time and energy for big tasks that
are not customer driven, alas!

> Is PV generally not supporting VS2013 or only superbuild (which wouldn't be a problem)?

PV  should build fine with VS2013 (although we don't have a dashboard
yet, I plan to put one soon). I was referring more to the superbuild.

Utkarsh

From dtaveras at andrew.cmu.edu  Thu Sep  4 17:14:27 2014
From: dtaveras at andrew.cmu.edu (dtaveras)
Date: Thu, 4 Sep 2014 14:14:27 -0700 (PDT)
Subject: [Paraview-developers] Paraview 3DWidget Plugin
Message-ID: <1409865267734-3225.post@n6.nabble.com>

I'm writing a plugin similar to the transform widget however instead on
having a box set by the databounds I wanted to use the actual datapoints
from the object. I'm wondering if I have a filter or a source how can I
communicate the data points and cells to the widget representation.




--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Paraview-3DWidget-Plugin-tp3225.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From utkarsh.ayachit at kitware.com  Fri Sep  5 00:34:44 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 5 Sep 2014 00:34:44 -0400
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <CADHSJjBF_MsC0cntSbQPUqAL439BFyTt_uivaOj70wh0M9AGfQ@mail.gmail.com>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
 <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
 <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>
 <5951482beedf45b68f60b8e80f760ff0@RZJMBX3.jr1.local>
 <CADHSJjBF_MsC0cntSbQPUqAL439BFyTt_uivaOj70wh0M9AGfQ@mail.gmail.com>
Message-ID: <CADHSJjBzDUkRsbUKt7-8vQfapbk_jDE+-aPrZY1dyy9ht-A1Pw@mail.gmail.com>

> PV  should build fine with VS2013 (although we don't have a dashboard
> yet, I plan to put one soon). I was referring more to the superbuild.


I was doing a test build with VS2013 and I am running into a lot of
missing #include<algorithm> related errors. I'd suggest holding off
till this fixes make it in. I am going to set up VS2013 dashboard
soon.

Utkarsh

From Gerald.Lodron at joanneum.at  Fri Sep  5 03:18:23 2014
From: Gerald.Lodron at joanneum.at (Lodron, Gerald)
Date: Fri, 5 Sep 2014 07:18:23 +0000
Subject: [Paraview-developers] Paraview and VS2013 compile errors
In-Reply-To: <CADHSJjBzDUkRsbUKt7-8vQfapbk_jDE+-aPrZY1dyy9ht-A1Pw@mail.gmail.com>
References: <58fef09f89374bc3aa2e8db159886be3@RZJMBX2.jr1.local>
 <1409852041.1875.17.camel@Judika>
 <b760b84f36704283b6c53c54dc11ca6b@RZJMBX3.jr1.local>
 <CADHSJjDv53_VTxit6ACr9F6hk-U7uuwEUSHi6pJoU_HbwVM9bA@mail.gmail.com>
 <5951482beedf45b68f60b8e80f760ff0@RZJMBX3.jr1.local>
 <CADHSJjBF_MsC0cntSbQPUqAL439BFyTt_uivaOj70wh0M9AGfQ@mail.gmail.com>
 <CADHSJjBzDUkRsbUKt7-8vQfapbk_jDE+-aPrZY1dyy9ht-A1Pw@mail.gmail.com>
Message-ID: <662fa6c95e1047eb85196a7c1fe0ebf0@RZJMBX2.jr1.local>

Oh, thanks. Good to hear that it was not my fault :-)

-----Urspr?ngliche Nachricht-----
Von: Utkarsh Ayachit [mailto:utkarsh.ayachit at kitware.com] 
Gesendet: Freitag, 05. September 2014 06:35
An: Lodron, Gerald
Cc: Frank TT; Paraview Developer (Paraview-developers at paraview.org)
Betreff: Re: [Paraview-developers] Paraview and VS2013 compile errors

> PV  should build fine with VS2013 (although we don't have a dashboard 
> yet, I plan to put one soon). I was referring more to the superbuild.


I was doing a test build with VS2013 and I am running into a lot of missing #include<algorithm> related errors. I'd suggest holding off till this fixes make it in. I am going to set up VS2013 dashboard soon.

Utkarsh

From felipe.bordeu at ec-nantes.fr  Fri Sep  5 07:17:41 2014
From: felipe.bordeu at ec-nantes.fr (Felipe Bordeu)
Date: Fri, 05 Sep 2014 13:17:41 +0200
Subject: [Paraview-developers] update ParaView Binaries Build Information
Message-ID: <54099BD5.5030304@ec-nantes.fr>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

Can some one update the page about the binaries build information in the
wiki.

http://www.paraview.org/Wiki/ParaView_Binaries

Thanks,



- -- 
Felipe Bordeu Weldt
Ing?nieur de Recherche
- -------------------------------------
T?l. : 33 (0)2 40 37 16 57
Fax. : 33 (0)2 40 74 74 06
Felipe.Bordeu at ec-nantes.fr
Institut GeM - UMR CNRS 6183
?cole Centrale Nantes
1 Rue de La No?, 44321 Nantes, FRANCE
- -------------------------------------
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQEcBAEBAgAGBQJUCZvKAAoJEE/fMfNgU9/DYXMH/07lDjo6KE6jjk4zK1saITt9
Hnz7mQUHI/uJ1/tsJ/yCfdtmhoGNnq4IyqIVc5gUw8kH/PJeetwvihDN/IQQW6Ez
+uLkmK9iK2hl6zCZds+LnxJRW5UChTNAYlMuJalcy+W3GW9ocZB9Xi/ekY79i/5H
+To1oKNh+JKaqYSEdV0WS9RzYkq60anbbHJ/GgV2A3HLrVoT26UWLU01WyqiDzab
0iGV0Mm80XNsPLJhzeNF+z+R1L3q3zb0jRU+oKRuDQ11Ca+lH25R9ClRS08Vq6VL
AdY9tWDvgac8TG99HBrAkCnNyvRnJzReDVyctn3U72FmP20mIpmi4NEAcrNz/KA=
=vlJk
-----END PGP SIGNATURE-----

From utkarsh.ayachit at kitware.com  Fri Sep  5 08:46:42 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 5 Sep 2014 08:46:42 -0400
Subject: [Paraview-developers] update ParaView Binaries Build Information
In-Reply-To: <54099BD5.5030304@ec-nantes.fr>
References: <54099BD5.5030304@ec-nantes.fr>
Message-ID: <CADHSJjC_4hWFRMw=_-me4X+ppJZbaPStC2wUhft=a-LrCEv61A@mail.gmail.com>

It'll get updated before the final release. I am still debating if to
change the VS compiler for this release to be VS 2013 itself.

On Fri, Sep 5, 2014 at 7:17 AM, Felipe Bordeu
<felipe.bordeu at ec-nantes.fr> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello,
>
> Can some one update the page about the binaries build information in the
> wiki.
>
> http://www.paraview.org/Wiki/ParaView_Binaries
>
> Thanks,
>
>
>
> - --
> Felipe Bordeu Weldt
> Ing?nieur de Recherche
> - -------------------------------------
> T?l. : 33 (0)2 40 37 16 57
> Fax. : 33 (0)2 40 74 74 06
> Felipe.Bordeu at ec-nantes.fr
> Institut GeM - UMR CNRS 6183
> ?cole Centrale Nantes
> 1 Rue de La No?, 44321 Nantes, FRANCE
> - -------------------------------------
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.11 (GNU/Linux)
>
> iQEcBAEBAgAGBQJUCZvKAAoJEE/fMfNgU9/DYXMH/07lDjo6KE6jjk4zK1saITt9
> Hnz7mQUHI/uJ1/tsJ/yCfdtmhoGNnq4IyqIVc5gUw8kH/PJeetwvihDN/IQQW6Ez
> +uLkmK9iK2hl6zCZds+LnxJRW5UChTNAYlMuJalcy+W3GW9ocZB9Xi/ekY79i/5H
> +To1oKNh+JKaqYSEdV0WS9RzYkq60anbbHJ/GgV2A3HLrVoT26UWLU01WyqiDzab
> 0iGV0Mm80XNsPLJhzeNF+z+R1L3q3zb0jRU+oKRuDQ11Ca+lH25R9ClRS08Vq6VL
> AdY9tWDvgac8TG99HBrAkCnNyvRnJzReDVyctn3U72FmP20mIpmi4NEAcrNz/KA=
> =vlJk
> -----END PGP SIGNATURE-----
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From sankhesh.jhaveri at kitware.com  Fri Sep  5 12:31:06 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Fri, 05 Sep 2014 09:31:06 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Thursday, September 04 2014
Message-ID: <20140905163104.3CBD480027@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Mac10.7.5-gcc-nightlymaster-release 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pvcrs.ExportX3dPOVVRML , 1 , ['Mac10.7.5-gcc-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [3]Win64-vs9-static-release 2
   [4]Lion-gcc-pvVTK 3
   -------------------- REPORT --------------------
   5 FAILURES
   [5]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [6]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [7]pvcs.Contour , 1 , ['Win64-vs9-static-release']
   [8]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [9]pvcrs.SimpleInteraction , 1 , ['Win64-vs9-static-release']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-05&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-gcc-nightlymaster-release
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pvcrs.ExportX3dPOVVRML
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-05&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-static-release
   4. http://open.cdash.org/index.php?project=ParaView&date=2014-09-05&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pvcs.HistogramSelection
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pvcrs.HistogramSelection
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pvcs.Contour
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pv.HistogramSelection
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-05&name=pvcrs.SimpleInteraction

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140905/bd4a9850/attachment.html>

From utkarsh.ayachit at kitware.com  Sun Sep  7 14:02:07 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Sun, 7 Sep 2014 14:02:07 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjA_x9ottNb-ribeY8C0SWTgTiJUNXkOW9_4Qx=6mTKFvw@mail.gmail.com>

Topics merged into master:
        add-vtk-adios-reader
        drag_n_drop_file_open
        update_quad_views_panel
(VTK)   vs2013_fixes

From joachim.pouderoux at kitware.com  Mon Sep  8 08:42:34 2014
From: joachim.pouderoux at kitware.com (Joachim Pouderoux)
Date: Mon, 8 Sep 2014 14:42:34 +0200
Subject: [Paraview-developers] ANN: ParaView Training Course in Paris, France,
	October 14-15, 2014
Message-ID: <CACJ=gThe52uEvQF47K0nwhKVO+wqjVmZ3hzcJt77bRFO8Lp0ow@mail.gmail.com>

Hello,

Kitware will be holding a 2-day ParaView course on October 14 and 15th,
2014 at Universit? Pierre et Marie Curie, Paris, France.

The specificity of this training is that it will take place in a virtual
reality room fitted with a powerwall.

Please visit our web site for more information and registration details at
either:

http://training.kitware.fr/browse/71 (in English)
or
http://formations.kitware.fr/browse/71 (in French)

Note that the course might be taught in English. If you have any question,
please contact us at formations at http://www.kitware.fr

Thank you,

*Joachim Pouderoux*

*PhD, Technical Expert*
*Kitware SAS <http://www.kitware.fr>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140908/7393ad59/attachment.html>

From jfavre at cscs.ch  Mon Sep  8 11:42:19 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Mon, 8 Sep 2014 15:42:19 +0000
Subject: [Paraview-developers] Read Image Sequence (not as time series)
In-Reply-To: <CADHSJjAPu_uL7JqUUWGDhdF9mOBawGaEdP7dy21_b7ej_xXC+Q@mail.gmail.com>
References: <1409642899.3285.14.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjCYywF2goVDhcKdmOFp=mv2_j1Bp4J5bJ6cGwtwedcLTQ@mail.gmail.com>
 <1409666939.3285.46.camel@imedv21.ime.kfa-juelich.de>
 <CADHSJjABbEeTrfxSfs8=6UU35Yv1yvaHBOaQmOWgh36-S2HmZA@mail.gmail.com>
 <1409677499.1841.18.camel@Judika>,
 <CADHSJjAPu_uL7JqUUWGDhdF9mOBawGaEdP7dy21_b7ej_xXC+Q@mail.gmail.com>
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F6828128A11@MBX21.d.ethz.ch>

hello folks

I am trying to replicate the results from http://www.paraview.org/Wiki/ParaView/ParaView_And_Mesa_3D on two different clusters (Cray + other) . The main difference with that page is that I have installed Mesa 10.2.4 compiled with gallium.

Things basically work ( I can render LIC textures for example), but I am not getting any differences in run-time trying different values of LP_NUM_THREADS.
Has someone successfully done the switch to a higher version of Mesa than the one used for the wiki (9.2.2)?

-----------------
Jean/CSCS

From jfavre at cscs.ch  Mon Sep  8 11:44:39 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Mon, 8 Sep 2014 15:44:39 +0000
Subject: [Paraview-developers] Mesa rendering with Gallium
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F6828128A2A@MBX21.d.ethz.ch>


hello folks

I am trying to replicate the results from http://www.paraview.org/Wiki/ParaView/ParaView_And_Mesa_3D on two different clusters (Cray + other) . The main difference with that page is that I have installed Mesa 10.2.4 compiled with gallium.

Things basically work ( I can render LIC textures for example), but I am not getting any differences in run-time trying different values of LP_NUM_THREADS.
Has someone successfully done the switch to a higher version of Mesa than the one used for the wiki (9.2.2)?

-----------------
Jean/CSCS

From sankhesh.jhaveri at kitware.com  Mon Sep  8 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Mon, 08 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Sunday, September 07 2014
Message-ID: <20140908163104.27AA17FF1F@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]ubuntu-x64-nightlynext 1
   [2]Win7x64-ninja-nightlynext-shared-release 3
   [3]Lion-gcc-pvVTK 3
   -------------------- REPORT --------------------
   7 FAILURES
   [4]CoProcessingFullWorkflow , 1 ,
   ['Win7x64-ninja-nightlynext-shared-release']
   [5]pvcs.CatalystLiveSetBreakpoint , 1 , ['ubuntu-x64-nightlynext']
   [6]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [7]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [8]CoProcessingTestPythonScript , 1 ,
   ['Win7x64-ninja-nightlynext-shared-release']
   [9]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [10]SavePythonState , 1 , ['Win7x64-ninja-nightlynext-shared-release']

   Issues with submitters
   The following expected submitters did not submit:
   tarvalon.kitware , Win7x64-vs10-static-release
   tarvalon.kitware , Win7x64-vs10-static-release

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-08&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-08&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-ninja-nightlynext-shared-release
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-08&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=CoProcessingFullWorkflow
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=pvcs.CatalystLiveSetBreakpoint
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=pvcs.HistogramSelection
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=pvcrs.HistogramSelection
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=CoProcessingTestPythonScript
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=pv.HistogramSelection
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-08&name=SavePythonState

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140908/ff5789e3/attachment.html>

From burlen.loring at gmail.com  Tue Sep  9 02:15:16 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Mon, 08 Sep 2014 23:15:16 -0700
Subject: [Paraview-developers] Mesa rendering with Gallium
In-Reply-To: <0EB9B6375711A04B820E6B6F5CCA9F6828128A2A@MBX21.d.ethz.ch>
References: <0EB9B6375711A04B820E6B6F5CCA9F6828128A2A@MBX21.d.ethz.ch>
Message-ID: <540E9AF4.2010107@gmail.com>

Jean,

What are you using for a test case?

The speed up is highly data and algorithm dependent. Mesa's fragment 
shaders are threaded but vertex shaders are not so results depend on how 
the work is distributed between the vertex ops and fragment ops. The LIC 
is a rare algorithm where fragment shaders do a lot of work and hence 
threading helps it.

For typical ParaView use case where you expect some massive dataset with 
a lot of vertices rendered through the fixed function pipeline the 
threading isn't a huge win. Gallium is a big win over the old mesa in 
general though.

You may insert a printf in the mesa sources to verify that it's picking 
up the enviroment variable.

Burlen

On 9/8/2014 8:44 AM, Favre Jean wrote:
> hello folks
>
> I am trying to replicate the results from http://www.paraview.org/Wiki/ParaView/ParaView_And_Mesa_3D on two different clusters (Cray + other) . The main difference with that page is that I have installed Mesa 10.2.4 compiled with gallium.
>
> Things basically work ( I can render LIC textures for example), but I am not getting any differences in run-time trying different values of LP_NUM_THREADS.
> Has someone successfully done the switch to a higher version of Mesa than the one used for the wiki (9.2.2)?
>
> -----------------
> Jean/CSCS
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers


From utkarsh.ayachit at kitware.com  Tue Sep  9 07:30:49 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 9 Sep 2014 07:30:49 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjAfKmCB20Zg+JiAxRpvDKtbH7TiBYM-Hy+Qs6-YKvug_g@mail.gmail.com>

Topics merged into master:
        fix_warnings
        pyhton_highlighting_for_python_cue
        qt5.3_fixes
        reorganize-pvweb-guides
        vr_plugin_fix

From sankhesh.jhaveri at kitware.com  Tue Sep  9 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Tue, 09 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Monday, September 08 2014
Message-ID: <20140909163103.D770E7FD6E@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]Lion-gcc-pvVTK 3
   -------------------- REPORT --------------------
   3 FAILURES
   [2]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [3]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [4]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-09&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-09&name=pvcrs.HistogramSelection
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-09&name=pv.HistogramSelection
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-09&name=pvcs.HistogramSelection

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140909/3e8b5570/attachment-0001.html>

From utkarsh.ayachit at kitware.com  Wed Sep 10 09:06:34 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 10 Sep 2014 09:06:34 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjA==WgcFjV2HWrsWocoDvNDZ2=UH3C2kas64pJq3=vLPg@mail.gmail.com>

Topics merged into master:
        fix-mpi-finalize
        pvw-fix-webgl-vgl-issues
(VTK)   vtkweb-catalyst-delayed-exit-and-alternate-launch

From sankhesh.jhaveri at kitware.com  Wed Sep 10 12:31:07 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Wed, 10 Sep 2014 09:31:07 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Tuesday, September 09 2014
Message-ID: <20140910163103.7BDDB7FD77@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]ubuntu-x64-coverage 1
   [2]Lion-gcc-pvVTK 3
   [3]ubuntu-x64-nightlynext 1
   -------------------- REPORT --------------------
   4 FAILURES
   [4]pvcrs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [5]pv.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [6]pvcs.HistogramSelection , 1 , ['Lion-gcc-pvVTK']
   [7]pvcs.CatalystLiveSetBreakpoint , 2 , ['ubuntu-x64-coverage',
   'ubuntu-x64-nightlynext']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-10&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-10&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-10&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-10&name=pvcrs.HistogramSelection
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-10&name=pv.HistogramSelection
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-10&name=pvcs.HistogramSelection
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-10&name=pvcs.CatalystLiveSetBreakpoint

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140910/8d38c599/attachment.html>

From eric.mellier at bobst.com  Thu Sep 11 06:53:26 2014
From: eric.mellier at bobst.com (Mellier)
Date: Thu, 11 Sep 2014 03:53:26 -0700 (PDT)
Subject: [Paraview-developers] Default setting for surface with edges
	representation
Message-ID: <1410432806274-3241.post@n6.nabble.com>

Hi,

Is it a way to define representation surface with edges as default setting
instead of surface ?

Thank you

Eric



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Default-setting-for-surface-with-edges-representation-tp3241.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From cory.quammen at kitware.com  Thu Sep 11 10:00:24 2014
From: cory.quammen at kitware.com (Cory Quammen)
Date: Thu, 11 Sep 2014 10:00:24 -0400
Subject: [Paraview-developers] Default setting for surface with edges
	representation
In-Reply-To: <1410432806274-3241.post@n6.nabble.com>
References: <1410432806274-3241.post@n6.nabble.com>
Message-ID: <CAB5Fpx7rnvEey98-Odfsk_=Lc_ByR4s2O-tr6rN8C_2hqds0ug@mail.gmail.com>

Hi Eric,

In principle, this should be possible using the new settings mechanism
in ParaView. However, representations are handled specially and the
custom default setting for representation type is overridden by
internal logic, so it is not possible in ParaView 4.2 RC1 to set the
representation default to surface with edges.

Cory

On Thu, Sep 11, 2014 at 6:53 AM, Mellier <eric.mellier at bobst.com> wrote:
> Hi,
>
> Is it a way to define representation surface with edges as default setting
> instead of surface ?
>
> Thank you
>
> Eric
>
>
>
> --
> View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Default-setting-for-surface-with-edges-representation-tp3241.html
> Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From sankhesh.jhaveri at kitware.com  Thu Sep 11 12:31:12 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Thu, 11 Sep 2014 09:31:12 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Wednesday, September 10 2014
Message-ID: <20140911163105.62E477FFF6@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]ubuntu-x64-nightlymaster 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pv.ServerConnectDialog , 1 , ['ubuntu-x64-nightlymaster']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [3]ubuntu-x64-coverage 1
   [4]ubuntu-x64-nightlynext 1
   -------------------- REPORT --------------------
   1 FAILURES
   [5]pvcs.CatalystLiveSetBreakpoint , 2 , ['ubuntu-x64-coverage',
   'ubuntu-x64-nightlynext']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-11&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlymaster
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-11&name=pv.ServerConnectDialog
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-11&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
   4. http://open.cdash.org/index.php?project=ParaView&date=2014-09-11&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-11&name=pvcs.CatalystLiveSetBreakpoint

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140911/2dc66870/attachment.html>

From jfavre at cscs.ch  Thu Sep 11 15:51:39 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Thu, 11 Sep 2014 19:51:39 +0000
Subject: [Paraview-developers] ghost cells in parallel reader
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F682812989B@MBX21.d.ethz.ch>

Hello

I have developed a parallel reader for a custom HDF5 format providing a single structured mesh. I am using version 4.2 RC1, i.e. the syntax of the new pipeline. It distributes my sub-extents very well and I can see my full data.

Things get out of hands (crash) when I try to do a contour with normals, i.e. requesting ghosts. My reader sees the request for ghost-level and then crash. I have not been able to find the source code of a parallel reader where the part handling ghosts is not commented out. I mean they don't check outInfo->Has(vtkStreamingDemandDrivenPipeline::UPDATE_NUMBER_OF_GHOST_LEVELS()). Can you point me to a working reader example?

My understanding is that I am not supposed to create and add the vtkGhostLevels array. The pipeline does that on my behalf, right? So when the pipeline asks me for a level of ghost cells, I get extents that are augmented by +/- 1 in each direction. Is it assuming that I can always fulfill the request?

Thanks for any tip on the subject.

Jean/CSCS




From eric.mellier at bobst.com  Fri Sep 12 01:40:27 2014
From: eric.mellier at bobst.com (Mellier)
Date: Thu, 11 Sep 2014 22:40:27 -0700 (PDT)
Subject: [Paraview-developers] Default setting for surface with edges
	representation
In-Reply-To: <CAB5Fpx7rnvEey98-Odfsk_=Lc_ByR4s2O-tr6rN8C_2hqds0ug@mail.gmail.com>
References: <1410432806274-3241.post@n6.nabble.com>
 <CAB5Fpx7rnvEey98-Odfsk_=Lc_ByR4s2O-tr6rN8C_2hqds0ug@mail.gmail.com>
Message-ID: <1410500427239-3245.post@n6.nabble.com>

Hi Cory,

Thank's for your answer.

I hope what this will be modified in the final release of Paraview 4.2
Is it a way to put the correction in the pipe ? 

Eric



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Default-setting-for-surface-with-edges-representation-tp3241p3245.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From cory.quammen at kitware.com  Fri Sep 12 08:43:39 2014
From: cory.quammen at kitware.com (Cory Quammen)
Date: Fri, 12 Sep 2014 08:43:39 -0400
Subject: [Paraview-developers] Default setting for surface with edges
	representation
In-Reply-To: <1410500427239-3245.post@n6.nabble.com>
References: <1410432806274-3241.post@n6.nabble.com>
 <CAB5Fpx7rnvEey98-Odfsk_=Lc_ByR4s2O-tr6rN8C_2hqds0ug@mail.gmail.com>
 <1410500427239-3245.post@n6.nabble.com>
Message-ID: <CAB5Fpx5Gv4Zg_Fsw8Prg4ztMGGgAosGCuE=pKtmdiR-HogMTPA@mail.gmail.com>

Eric,

I have filed a bug report: http://www.paraview.org/Bug/view.php?id=14992

Unfortunately, I can't guarantee at this point that it will make it
into the final release of 4.2.

Best,
Cory

On Fri, Sep 12, 2014 at 1:40 AM, Mellier <eric.mellier at bobst.com> wrote:
> Hi Cory,
>
> Thank's for your answer.
>
> I hope what this will be modified in the final release of Paraview 4.2
> Is it a way to put the correction in the pipe ?
>
> Eric
>
>
>
> --
> View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Default-setting-for-surface-with-edges-representation-tp3241p3245.html
> Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From eric.mellier at bobst.com  Fri Sep 12 08:56:18 2014
From: eric.mellier at bobst.com (Mellier)
Date: Fri, 12 Sep 2014 05:56:18 -0700 (PDT)
Subject: [Paraview-developers] Default setting for surface with edges
	representation
In-Reply-To: <CAB5Fpx5Gv4Zg_Fsw8Prg4ztMGGgAosGCuE=pKtmdiR-HogMTPA@mail.gmail.com>
References: <1410432806274-3241.post@n6.nabble.com>
 <CAB5Fpx7rnvEey98-Odfsk_=Lc_ByR4s2O-tr6rN8C_2hqds0ug@mail.gmail.com>
 <1410500427239-3245.post@n6.nabble.com>
 <CAB5Fpx5Gv4Zg_Fsw8Prg4ztMGGgAosGCuE=pKtmdiR-HogMTPA@mail.gmail.com>
Message-ID: <1410526578332-3247.post@n6.nabble.com>

Cory,

Thank's a lot for your help.

Eric



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/Default-setting-for-surface-with-edges-representation-tp3241p3247.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.

From berk.geveci at kitware.com  Fri Sep 12 09:53:01 2014
From: berk.geveci at kitware.com (Berk Geveci)
Date: Fri, 12 Sep 2014 09:53:01 -0400
Subject: [Paraview-developers] ghost cells in parallel reader
In-Reply-To: <0EB9B6375711A04B820E6B6F5CCA9F682812989B@MBX21.d.ethz.ch>
References: <0EB9B6375711A04B820E6B6F5CCA9F682812989B@MBX21.d.ethz.ch>
Message-ID: <CAE32kpVMLBYUuny_yYM4kbO1UoxDAm82S6m1AS56ZsfavAjcng@mail.gmail.com>

Hi Jean,

You probably won't find a lot of structured data readers that handle pieces
and ghost cells/points directly. Although it is possible to develop them if
you are not happy with the default domain decomposition the pipeline gives
you. To find an example of a reader that supports ghost request, you need
to look at unstructured readers, such as vtkXMLUnstructuredDataReader.

You have 3 choices in dealing with partitioning and ghosts in a structured
reader:

1. Don't deal with piece or extent requests.

2. Set CAN_PRODUCE_SUB_EXTENT and let the pipeline ask the reader extent
using UPDATE_EXTENT. Ghost levels will be handled automatically. The reader
has to only support UPDATE_EXTENT.

3. Set CAN_HANDLE_PIECE_REQUEST. Then deal with a combination of
UPDATE_EXTENT and UPDATE_PIECE_NUMBER / NUMBER_OF_PIECES /
NUMBER_OF_GHOST_LEVELS.

In 3, you still have to support UPDATE_EXTENT because downstream can ask
for a sub-region of the data (such as vtkExtractVOI). However, this
UPDATE_EXTENT will be the same for all MPI ranks - it is up to the reader
to decide how to split it.

On the other hand in 2, each rank with get a subset of the whole
UPDATE_EXTENT with ghost levels factored in.

I hope that this helps.

PS: Please do me a favor and check if what I described here comes across
clearly in here http://www.vtk.org/Wiki/VTK/Parallel_Pipeline. If not, I
need to update that page.

Best,
Best,
-berk

On Thu, Sep 11, 2014 at 3:51 PM, Favre Jean <jfavre at cscs.ch> wrote:

> Hello
>
> I have developed a parallel reader for a custom HDF5 format providing a
> single structured mesh. I am using version 4.2 RC1, i.e. the syntax of the
> new pipeline. It distributes my sub-extents very well and I can see my full
> data.
>
> Things get out of hands (crash) when I try to do a contour with normals,
> i.e. requesting ghosts. My reader sees the request for ghost-level and then
> crash. I have not been able to find the source code of a parallel reader
> where the part handling ghosts is not commented out. I mean they don't
> check
> outInfo->Has(vtkStreamingDemandDrivenPipeline::UPDATE_NUMBER_OF_GHOST_LEVELS()).
> Can you point me to a working reader example?
>
> My understanding is that I am not supposed to create and add the
> vtkGhostLevels array. The pipeline does that on my behalf, right? So when
> the pipeline asks me for a level of ghost cells, I get extents that are
> augmented by +/- 1 in each direction. Is it assuming that I can always
> fulfill the request?
>
> Thanks for any tip on the subject.
>
> Jean/CSCS
>
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140912/182f8bd2/attachment.html>

From sankhesh.jhaveri at kitware.com  Fri Sep 12 12:31:14 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Fri, 12 Sep 2014 09:31:14 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Thursday, September 11 2014
Message-ID: <20140912163104.13EA783BCC@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Win7x64-ninja-nightlymaster-static-release 1
   [2]fedora-x64-icc-nightlymaster-debug 1
   [3]ubuntu-x64-nightlymaster 1
   -------------------- REPORT --------------------
   3 FAILURES
   [4]pv.Contour , 1 , ['ubuntu-x64-nightlymaster']
   [5]pvcrs.TestIsoVolume , 1 ,
   ['Win7x64-ninja-nightlymaster-static-release']
   [6]pvcs-collab.CreateDelete , 1 ,
   ['fedora-x64-icc-nightlymaster-debug']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [7]ubuntu-x64-nightlynext 3
   -------------------- REPORT --------------------
   3 FAILURES
   [8]pvcs.CatalystLiveSetBreakpoint , 1 , ['ubuntu-x64-nightlynext']
   [9]pv.CameraLink , 1 , ['ubuntu-x64-nightlynext']
   [10]pv.ChartAxisRangeAndLabels , 1 , ['ubuntu-x64-nightlynext']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-12&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-ninja-nightlymaster-static-release
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-12&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-12&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlymaster
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pv.Contour
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pvcrs.TestIsoVolume
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pvcs-collab.CreateDelete
   7. http://open.cdash.org/index.php?project=ParaView&date=2014-09-12&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pvcs.CatalystLiveSetBreakpoint
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pv.CameraLink
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-12&name=pv.ChartAxisRangeAndLabels

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140912/b3a6836f/attachment.html>

From jfavre at cscs.ch  Fri Sep 12 15:38:03 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Fri, 12 Sep 2014 19:38:03 +0000
Subject: [Paraview-developers] ghost cells in parallel reader
In-Reply-To: <CAE32kpVMLBYUuny_yYM4kbO1UoxDAm82S6m1AS56ZsfavAjcng@mail.gmail.com>
References: <0EB9B6375711A04B820E6B6F5CCA9F682812989B@MBX21.d.ethz.ch>,
 <CAE32kpVMLBYUuny_yYM4kbO1UoxDAm82S6m1AS56ZsfavAjcng@mail.gmail.com>
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F6828137D98@MBX11.d.ethz.ch>

Thanks Berk

the explanation was clear. I got the reader working, and generating proper ghost cells when needed. I used method 2.

-----------------
Jean


From utkarsh.ayachit at kitware.com  Sun Sep 14 11:30:58 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Sun, 14 Sep 2014 11:30:58 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjBQjm5D_=a1Ntv20agQpqoyT_5V+a6vBVb6bmwv7srRBQ@mail.gmail.com>

Topics merged into master:
        10643_chart_legend_properties
(VTK)   13139-plot-small-geometry
        14931-cluttered-histogram
        14983_catalyst_programmablefilter
        14990_finalize_catalyst_pipelines
        add-flexibility-to-catalyst-exporter
        catalyst_fix_plugins
        fix_text_source
        pvweb-docs-further-reorganization
        pvweb-move-firefox-test-to-port-8080

From utkarsh.ayachit at kitware.com  Mon Sep 15 11:27:40 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 15 Sep 2014 11:27:40 -0400
Subject: [Paraview-developers] Tagging 4.2.0 soon
Message-ID: <CADHSJjCYFZeN192tV+zqLm1bOQxZbBx9WMcYUYU5YX8wSbKTNA@mail.gmail.com>

Folks,

I'll be tagging the final 4.2 on Wed/Thu (depending on dashboard
state). Please let know if there's a strong reason to delay this.

Thanks,
Utkarsh

From sankhesh.jhaveri at kitware.com  Mon Sep 15 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Mon, 15 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Sunday, September 14 2014
Message-ID: <20140915163104.0FE9B7FE37@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Mac10.7.5-gcc-nightlymaster-release 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter , 1
   , ['Mac10.7.5-gcc-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [3]Win32-vs9-shared-release 1
   -------------------- REPORT --------------------
   1 FAILURES
   [4]pv.UndoRedo2 , 1 , ['Win32-vs9-shared-release']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-15&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-gcc-nightlymaster-release
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-15&name=vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-15&filtercount=1&field1=buildname/string&compare1=61&value1=Win32-vs9-shared-release
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-15&name=pv.UndoRedo2

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140915/500119d8/attachment-0001.html>

From sankhesh.jhaveri at kitware.com  Tue Sep 16 12:31:06 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Tue, 16 Sep 2014 09:31:06 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Monday, September 15 2014
Message-ID: <20140916163104.501A98070A@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Mac10.7.5-gcc-nightlymaster-release 6
   -------------------- REPORT --------------------
   6 FAILURES
   [2]pvcrs.TraceMultiViews , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [3]pvcrs.XYHistogram , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [4]pvcrs.FindDataDialog , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [5]vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter , 1
   , ['Mac10.7.5-gcc-nightlymaster-release']
   [6]pvcs-collab.XdmfRead , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [7]pvcrs.TensorGlyph , 1 , ['Mac10.7.5-gcc-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [8]ubuntu-x64-nightlynext 3
   [9]Win7x64-vs10-static-release 3
   [10]Lion-gcc-pvVTK 3
   [11]Win64-vs9-shared-debug-nocollab 3
   [12]Win64-vs9-static-release 3
   [13]ubuntu-x64-nightlynext-static 3
   [14]fedora-x64-icc-nightlynext-release 4
   [15]Win7x64-ninja-nightlynext-shared-release 3
   [16]ubuntu-x64-nightlynext 4
   [17]Win7x64-vs9-nightlynext-static-release 3
   [18]Win7x64-vs10-shared-release 3
   [19]ubuntu-x64-coverage 3
   [20]Win32-vs9-shared-release 3
   -------------------- REPORT --------------------
   5 FAILURES
   [21]pv.ExtractLevel , 1 , ['fedora-x64-icc-nightlynext-release']
   [22]pvcs.CatalystLiveSetBreakpoint , 1 , ['ubuntu-x64-nightlynext']
   [23]pv.UpdateTransferFunctionRanges , 13 , ['ubuntu-x64-nightlynext',
   'Win7x64-vs10-static-release', 'Lion-gcc-pvVTK',
   'Win64-vs9-shared-debug-nocollab', 'Win64-vs9-static-release',
   'ubuntu-x64-nightlynext-static', 'fedora-x64-icc-nightlynext-release',
   'Win7x64-ninja-nightlynext-shared-release', 'ubuntu-x64-nightlynext',
   'Win7x64-vs9-nightlynext-static-release',
   'Win7x64-vs10-shared-release', 'ubuntu-x64-coverage',
   'Win32-vs9-shared-release']
   [24]pvcrs.UpdateTransferFunctionRanges , 13 ,
   ['ubuntu-x64-nightlynext', 'Win7x64-vs10-static-release',
   'Lion-gcc-pvVTK', 'Win64-vs9-shared-debug-nocollab',
   'Win64-vs9-static-release', 'ubuntu-x64-nightlynext-static',
   'fedora-x64-icc-nightlynext-release',
   'Win7x64-ninja-nightlynext-shared-release', 'ubuntu-x64-nightlynext',
   'Win7x64-vs9-nightlynext-static-release',
   'Win7x64-vs10-shared-release', 'ubuntu-x64-coverage',
   'Win32-vs9-shared-release']
   [25]pvcs.UpdateTransferFunctionRanges , 13 , ['ubuntu-x64-nightlynext',
   'Win7x64-vs10-static-release', 'Lion-gcc-pvVTK',
   'Win64-vs9-shared-debug-nocollab', 'Win64-vs9-static-release',
   'ubuntu-x64-nightlynext-static', 'fedora-x64-icc-nightlynext-release',
   'Win7x64-ninja-nightlynext-shared-release', 'ubuntu-x64-nightlynext',
   'Win7x64-vs9-nightlynext-static-release',
   'Win7x64-vs10-shared-release', 'ubuntu-x64-coverage',
   'Win32-vs9-shared-release']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-gcc-nightlymaster-release
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcrs.TraceMultiViews
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcrs.XYHistogram
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcrs.FindDataDialog
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcs-collab.XdmfRead
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcrs.TensorGlyph
   8. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   9. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-static-release
  10. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
  11. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-shared-debug-nocollab
  12. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-static-release
  13. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext-static
  14. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
  15. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-ninja-nightlynext-shared-release
  16. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
  17. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs9-nightlynext-static-release
  18. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-shared-release
  19. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
  20. http://open.cdash.org/index.php?project=ParaView&date=2014-09-16&filtercount=1&field1=buildname/string&compare1=61&value1=Win32-vs9-shared-release
  21. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pv.ExtractLevel
  22. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcs.CatalystLiveSetBreakpoint
  23. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pv.UpdateTransferFunctionRanges
  24. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcrs.UpdateTransferFunctionRanges
  25. http://open.cdash.org/testSummary.php?project=9&date=2014-09-16&name=pvcs.UpdateTransferFunctionRanges

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140916/854ec15e/attachment.html>

From Orlando.RIVERA at mtu.de  Wed Sep 17 08:11:21 2014
From: Orlando.RIVERA at mtu.de (RIVERA, Orlando)
Date: Wed, 17 Sep 2014 12:11:21 +0000
Subject: [Paraview-developers] vtkExtractBlock  Search option
Message-ID: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>

Hi

I hope I am posting in the right place ?

I would like to add a search textBox  in the panel  Properties of a Extract Block filte. In the attached picture  I  drew , more or less, what I would like to achieve.   Once I typed a text  I would like to be able to have checked the tree leaves that match . (I have done  something similar for a reader,   but  that time I have to write the GUI code my self . so I knew where was everything.)

I could understand how  the vtkExtrackBlocks filter works .  However , I couldn?t  find  more that the filter.xml definition ( Although I could add a StringVectorProperty ,  I don?t know how to act upon the tree view)

 Where can I find the GUI implementation for this filter or the Tree view in order to adapt it or extend it ?

Any Help will be gladly appreciated

Regards



Orlando Rivera

--
MTU Aero Engines AG
Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer Martens, Michael Schrey?gg, Dr. Stefan Weingartner
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/20140917/161926cf/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ExtractBlock.jpeg
Type: image/jpeg
Size: 65060 bytes
Desc: ExtractBlock.jpeg
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140917/161926cf/attachment-0001.jpeg>

From utkarsh.ayachit at kitware.com  Wed Sep 17 08:31:32 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 17 Sep 2014 08:31:32 -0400
Subject: [Paraview-developers] vtkExtractBlock Search option
In-Reply-To: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>
References: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>
Message-ID: <CADHSJjBp4pX+wL2-L7w9R9=H5gjxb5RHxJBcmMiq+12w=kEPOQ@mail.gmail.com>

Something like that is already supported. See [1].

Utkarsh

[1] http://www.itk.org/Wiki/ParaView/Users_Guide/Introduction#Modality



On Wed, Sep 17, 2014 at 8:11 AM, RIVERA, Orlando <Orlando.RIVERA at mtu.de> wrote:
> Hi
>
> I hope I am posting in the right place ?
>
> I would like to add a search textBox  in the panel  Properties of a Extract
> Block filte. In the attached picture  I  drew , more or less, what I would
> like to achieve.   Once I typed a text  I would like to be able to have
> checked the tree leaves that match . (I have done  something similar for a
> reader,   but  that time I have to write the GUI code my self . so I knew
> where was everything.)
>
> I could understand how  the vtkExtrackBlocks filter works .  However , I
> couldn?t  find  more that the filter.xml definition ( Although I could add a
> StringVectorProperty ,  I don?t know how to act upon the tree view)
>
>  Where can I find the GUI implementation for this filter or the Tree view in
> order to adapt it or extend it ?
>
> Any Help will be gladly appreciated
>
> Regards
>
>
>
> Orlando Rivera
>
>
>
> --
> MTU Aero Engines AG
> Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer
> Martens, Michael Schrey?gg, Dr. Stefan Weingartner
> 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.
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>

From jfavre at cscs.ch  Wed Sep 17 09:58:56 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Wed, 17 Sep 2014 13:58:56 +0000
Subject: [Paraview-developers] vtkExtractBlock Search option
In-Reply-To: <CADHSJjBp4pX+wL2-L7w9R9=H5gjxb5RHxJBcmMiq+12w=kEPOQ@mail.gmail.com>
References: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>,
 <CADHSJjBp4pX+wL2-L7w9R9=H5gjxb5RHxJBcmMiq+12w=kEPOQ@mail.gmail.com>
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F6828139E4B@MBX11.d.ethz.ch>

Utkarsh

I have the exact same case. The Ctr-F and search does work, but I have not figured out how to apply the selection. I wish to type my search string and have all matching blocks pre-selected. The only thing I manage to do with the up/down arrows is to go to the next/previous block whose name matches the search. I can see that it skips over the other blocks. Am I missing a step? I wish to select all blocks in Extract Block based on a search string and one other keystroke. (I could have hundreds of block matching)

Jean

From Orlando.RIVERA at mtu.de  Wed Sep 17 10:03:26 2014
From: Orlando.RIVERA at mtu.de (RIVERA, Orlando)
Date: Wed, 17 Sep 2014 14:03:26 +0000
Subject: [Paraview-developers] vtkExtractBlock Search option
In-Reply-To: <0EB9B6375711A04B820E6B6F5CCA9F6828139E4B@MBX11.d.ethz.ch>
References: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>,
 <CADHSJjBp4pX+wL2-L7w9R9=H5gjxb5RHxJBcmMiq+12w=kEPOQ@mail.gmail.com>
 <0EB9B6375711A04B820E6B6F5CCA9F6828139E4B@MBX11.d.ethz.ch>
Message-ID: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B177@memc022.de.mtu.corp>

I was just writing  the  same answer



-----Urspr?ngliche Nachricht-----
Von: Favre Jean [mailto:jfavre at cscs.ch]
Gesendet: 17 September 2014 15:59
An: Utkarsh Ayachit; RIVERA, Orlando
Cc: paraview-developers at paraview.org
Betreff: RE: [Paraview-developers] vtkExtractBlock Search option

Utkarsh

I have the exact same case. The Ctr-F and search does work, but I have not figured out how to apply the selection. I wish to type my search string and have all matching blocks pre-selected. The only thing I manage to do with the up/down arrows is to go to the next/previous block whose name matches the search. I can see that it skips over the other blocks. Am I missing a step? I wish to select all blocks in Extract Block based on a search string and one other keystroke. (I could have hundreds of block matching)

Jean
--
MTU Aero Engines AG
Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer Martens, Michael Schrey?gg, Dr. Stefan Weingartner
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.

From utkarsh.ayachit at kitware.com  Wed Sep 17 10:08:18 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 17 Sep 2014 10:08:18 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjA9DM5vq2DuicUE-HnWHNAxyj3G2m0c4_OHYitVCKDzEA@mail.gmail.com>

SUMMARY
---------------------------------------------
Topics merged into master:
        14975-add-dicom-reader
        14998_catalyst_fix_PYTHONPATH
(VTK)   fix_vrml_importer_segfault
        live-breakpoint
        macro-warning-14970
(VTK)   misc-adios-fixes
        use-system-libraries

From utkarsh.ayachit at kitware.com  Wed Sep 17 10:12:33 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 17 Sep 2014 10:12:33 -0400
Subject: [Paraview-developers] vtkExtractBlock Search option
In-Reply-To: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B177@memc022.de.mtu.corp>
References: <ACE279CCDDBEFA43BD2F0EE07671EA98A933B13B@memc022.de.mtu.corp>
 <CADHSJjBp4pX+wL2-L7w9R9=H5gjxb5RHxJBcmMiq+12w=kEPOQ@mail.gmail.com>
 <0EB9B6375711A04B820E6B6F5CCA9F6828139E4B@MBX11.d.ethz.ch>
 <ACE279CCDDBEFA43BD2F0EE07671EA98A933B177@memc022.de.mtu.corp>
Message-ID: <CADHSJjDxFodkWMBj_eexAmn1g3bdfrDOZQyiKn03G-kd-8LE4Q@mail.gmail.com>

Ah! Okay, I should read emails fully :). Yea, that's not supported
currently, but a good feature indeed!

Orlando, I'd start by looking at this searching class itself [1].
Looking at where it gets instantiated will provide you the locations
where these are created. Maybe extending this widget to add support to
"check/uncheck" matching items, would be the way to go.

Utkarsh

[1] http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classpqItemViewSearchWidget.html

On Wed, Sep 17, 2014 at 10:03 AM, RIVERA, Orlando <Orlando.RIVERA at mtu.de> wrote:
> I was just writing  the  same answer
>
>
>
> -----Urspr?ngliche Nachricht-----
> Von: Favre Jean [mailto:jfavre at cscs.ch]
> Gesendet: 17 September 2014 15:59
> An: Utkarsh Ayachit; RIVERA, Orlando
> Cc: paraview-developers at paraview.org
> Betreff: RE: [Paraview-developers] vtkExtractBlock Search option
>
> Utkarsh
>
> I have the exact same case. The Ctr-F and search does work, but I have not figured out how to apply the selection. I wish to type my search string and have all matching blocks pre-selected. The only thing I manage to do with the up/down arrows is to go to the next/previous block whose name matches the search. I can see that it skips over the other blocks. Am I missing a step? I wish to select all blocks in Extract Block based on a search string and one other keystroke. (I could have hundreds of block matching)
>
> Jean
> --
> MTU Aero Engines AG
> Vorstand/Board of Management: Reiner Winkler, Vorsitzender/CEO; Dr. Rainer Martens, Michael Schrey?gg, Dr. Stefan Weingartner
> 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.

From sankhesh.jhaveri at kitware.com  Wed Sep 17 12:31:06 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Wed, 17 Sep 2014 09:31:06 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Tuesday, September 16 2014
Message-ID: <20140917163104.A354280B4A@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]fedora-x64-icc-nightlymaster-debug 1
   [2]Mac10.7.5-gcc-nightlymaster-release 13
   -------------------- REPORT --------------------
   14 FAILURES
   [3]pvcrs.RenderNan , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [4]pvcs.Tessellate , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [5]pvcs.TensorGlyph , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [6]pvcrs.SaveLargeScreenshot , 1 ,
   ['Mac10.7.5-gcc-nightlymaster-release']
   [7]pvcrs.TemporalInterpolator , 1 ,
   ['Mac10.7.5-gcc-nightlymaster-release']
   [8]pvcrs.MemoryInspectorPanel , 1 ,
   ['Mac10.7.5-gcc-nightlymaster-release']
   [9]vtkSciberQuestCxx-MPI-TestFieldTracer , 1 ,
   ['Mac10.7.5-gcc-nightlymaster-release']
   [10]pvcrs.TensorGlyph , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [11]vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter ,
   1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [12]pvcs.TemporalShiftScale , 1 ,
   ['Mac10.7.5-gcc-nightlymaster-release']
   [13]pv.FindWidget , 1 , ['fedora-x64-icc-nightlymaster-debug']
   [14]pvcs.Threshold , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [15]pvcrs.TabbedViews , 1 , ['Mac10.7.5-gcc-nightlymaster-release']
   [16]pvcrs.TestIsoVolume , 1 , ['Mac10.7.5-gcc-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [17]Win64-vs9-shared-debug-nocollab 2
   [18]ubuntu-x64-nightlynext-static 2
   [19]Win7x64-vs10-static-release 2
   [20]ubuntu-x64-coverage 2
   [21]fedora-x64-icc-nightlynext-release 2
   [22]Win32-vs9-shared-release 2
   [23]Lion-gcc-pvVTK 2
   [24]Win7x64-vs10-shared-release 2
   [25]ubuntu-x64-nightlynext 2
   [26]ubuntu-x64-nightlynext 2
   [27]Win64-vs9-static-release 2
   -------------------- REPORT --------------------
   2 FAILURES
   [28]pvcs.RandomAttributes , 11 , ['Win64-vs9-shared-debug-nocollab',
   'ubuntu-x64-nightlynext-static', 'Win7x64-vs10-static-release',
   'ubuntu-x64-coverage', 'fedora-x64-icc-nightlynext-release',
   'Win32-vs9-shared-release', 'Lion-gcc-pvVTK',
   'Win7x64-vs10-shared-release', 'ubuntu-x64-nightlynext',
   'ubuntu-x64-nightlynext', 'Win64-vs9-static-release']
   [29]pvcrs.RandomAttributes , 11 , ['Win64-vs9-shared-debug-nocollab',
   'ubuntu-x64-nightlynext-static', 'Win7x64-vs10-static-release',
   'ubuntu-x64-coverage', 'fedora-x64-icc-nightlynext-release',
   'Win32-vs9-shared-release', 'Lion-gcc-pvVTK',
   'Win7x64-vs10-shared-release', 'ubuntu-x64-nightlynext',
   'ubuntu-x64-nightlynext', 'Win64-vs9-static-release']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-gcc-nightlymaster-release
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.RenderNan
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcs.Tessellate
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcs.TensorGlyph
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.SaveLargeScreenshot
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.TemporalInterpolator
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.MemoryInspectorPanel
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=vtkSciberQuestCxx-MPI-TestFieldTracer
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.TensorGlyph
  11. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=vtkPVServerManagerDefaultPython-MPI-SymmetricParallelImageWriter
  12. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcs.TemporalShiftScale
  13. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pv.FindWidget
  14. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcs.Threshold
  15. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.TabbedViews
  16. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.TestIsoVolume
  17. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-shared-debug-nocollab
  18. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext-static
  19. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-static-release
  20. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
  21. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
  22. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Win32-vs9-shared-release
  23. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
  24. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Win7x64-vs10-shared-release
  25. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
  26. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
  27. http://open.cdash.org/index.php?project=ParaView&date=2014-09-17&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-static-release
  28. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcs.RandomAttributes
  29. http://open.cdash.org/testSummary.php?project=9&date=2014-09-17&name=pvcrs.RandomAttributes

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140917/4fbc4065/attachment-0001.html>

From utkarsh.ayachit at kitware.com  Wed Sep 17 14:17:52 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Wed, 17 Sep 2014 14:17:52 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjALKcyxtqNKHF_Y8jjdYK1ijzEBZ6fV2UOiomELTUg3og@mail.gmail.com>

Topics merged into master:
        miscellaneous_volume_rendering_fixes
(VTK)   vtkPolyData_removecells

From burlen.loring at gmail.com  Wed Sep 17 19:35:49 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Wed, 17 Sep 2014 16:35:49 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in 4.2.0 rc1
Message-ID: <541A1AD5.1070508@gmail.com>

Hi All,

I'm having trouble with our NERSC install, plugins don't seem to be 
working at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?

Burlen

From utkarsh.ayachit at kitware.com  Thu Sep 18 08:37:03 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 18 Sep 2014 08:37:03 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <541A1AD5.1070508@gmail.com>
References: <541A1AD5.1070508@gmail.com>
Message-ID: <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>

Yes, you need to load them manually, however. They are no longer auto-loaded.

http://www.paraview.org/Bug/view.php?id=14542

Utkarsh

On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
> Hi All,
>
> I'm having trouble with our NERSC install, plugins don't seem to be working
> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>
> Burlen
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From reuterma at ornl.gov  Thu Sep 18 09:57:42 2014
From: reuterma at ornl.gov (Reuter, Michael A.)
Date: Thu, 18 Sep 2014 13:57:42 +0000
Subject: [Paraview-developers] Git-master and Windows
Message-ID: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>

Hi,

I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.

Thanks,
M

Dr. Michael Reuter
Scientific Data Analysis Group
Neutron Data Analysis and Visualization Division
Oak Ridge National Laboratory

Office: 1-865-241-7216
Fax: 1-865-574-6080
Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov>







From utkarsh.ayachit at kitware.com  Thu Sep 18 10:17:46 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 18 Sep 2014 10:17:46 -0400
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
Message-ID: <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>

Since you said you build with your own installation of Python, is
Python in PATH? Try using the dependency walker[1] to see if it can
find all the dependencies correctly.

Utkarsh

[1] http://www.dependencywalker.com/

On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov> wrote:
> Hi,
>
> I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.
>
> Thanks,
> M
>
> Dr. Michael Reuter
> Scientific Data Analysis Group
> Neutron Data Analysis and Visualization Division
> Oak Ridge National Laboratory
>
> Office: 1-865-241-7216
> Fax: 1-865-574-6080
> Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov>
>
>
>
>
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From reuterma at ornl.gov  Thu Sep 18 11:41:24 2014
From: reuterma at ornl.gov (Reuter, Michael A.)
Date: Thu, 18 Sep 2014 15:41:24 +0000
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
 <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
Message-ID: <4A355AC2-A898-4888-963C-80EA93A2629B@ornl.gov>

The only dependency it can?t find is IESHIMS.DLL. I?m not sure if this is an issue but the VTKSYS-PV4.2.DLL file is showing a number of red function points like vtksys::SystemTools::SplitString, Split, CollapseFullPath, FileIsDirectory, RemoveADirectory, RemoveFile, SplitPathRootComponent, Load.
Also, VTKCOMMONEXECUTIONMODEL-PV4.2.DLL has vtkCompositeDataPipeline::COMPOSITE_INDICIES(void) as red.

M

On Sep 18, 2014, at 10:17 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:

Since you said you build with your own installation of Python, is
Python in PATH? Try using the dependency walker[1] to see if it can
find all the dependencies correctly.

Utkarsh

[1] http://www.dependencywalker.com/

On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov>> wrote:
Hi,

I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.

Thanks,
M

Dr. Michael Reuter
Scientific Data Analysis Group
Neutron Data Analysis and Visualization Division
Oak Ridge National Laboratory

Office: 1-865-241-7216
Fax: 1-865-574-6080
Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>






_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org>
http://public.kitware.com/mailman/listinfo/paraview-developers


From reuterma at ornl.gov  Thu Sep 18 11:49:03 2014
From: reuterma at ornl.gov (Reuter, Michael A.)
Date: Thu, 18 Sep 2014 15:49:03 +0000
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
 <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
Message-ID: <FDF9BB35-15EF-4B42-B70A-98D06579EC2C@ornl.gov>

Ignore the red function points thing. Had other PV libs in the path. Cleaned that up, but the dependency thing still stands. The executable can find all the necessary libraries except IESHIMS.DLL.

M

On Sep 18, 2014, at 10:17 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:

Since you said you build with your own installation of Python, is
Python in PATH? Try using the dependency walker[1] to see if it can
find all the dependencies correctly.

Utkarsh

[1] http://www.dependencywalker.com/

On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov>> wrote:
Hi,

I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.

Thanks,
M

Dr. Michael Reuter
Scientific Data Analysis Group
Neutron Data Analysis and Visualization Division
Oak Ridge National Laboratory

Office: 1-865-241-7216
Fax: 1-865-574-6080
Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>






_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org>
http://public.kitware.com/mailman/listinfo/paraview-developers


From sankhesh.jhaveri at kitware.com  Thu Sep 18 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Thu, 18 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Wednesday, September 17 2014
Message-ID: <20140918163104.3E7688131F@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   Cdash returned nothing useful as of this report.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140918/862e3426/attachment-0001.html>

From burlen.loring at gmail.com  Thu Sep 18 17:05:07 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Thu, 18 Sep 2014 14:05:07 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
Message-ID: <541B4903.6010002@gmail.com>

Hi Utkarsh,

Good, glad it's still supported! There seem to be some issues though.

On the Cray the remote plugin list is empty except for 
vtkPVLPuginInitializer. RGBZView is automatically loaded on the client 
but not the server and thus as soon as the server connects the plugin 
manager opens to report the issue. The remote plugin list is empty. 
Browsing to the install and loading  the plugin doesn't work. it's a .a 
file so I'm not surprised. Judging from what I see on my workstation, 
the remote plugin list should be populated right?

I've been trying to reproduce it locally on my workstation. I have 
different problems. The remote plugin list is fully populated and 
RGBZView loads on both sides. But when I try to load a plugin  via "load 
selected" ParaView crashes(stack below). That's running from the build. 
When I try the same on an install the server segv's as soon as I 
connect. The stack is the same as the in-build crash.

By the way this is in client-server mode the server is built with sahred 
libs off, with mpi and python and without qt.

Burlen

on local linux workstation, client server, server built without qt, with 
python and mpi
Program received signal SIGSEGV, Segmentation fault.
0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0, 
session=0xb4880b0) at 
/work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
473       if (this->Internals->ActiveSessionStack.back() != session)

(gdb) where
#0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0, 
session=0xb4880b0) at 
/work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
#1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at 
/work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
#2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream 
(this=0xb4880b0, location=21, stream=..., ignore_errors=false) at 
/work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
#3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI 
(this=0xb4880b0, message=0xbc82030, message_length=16) at 
/work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
#4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback 
(localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at 
/work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
#5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI 
(this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16, 
rmiTag=55625) at 
/work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
#6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs 
(this=0xbc79460, reportErrors=0, dont_loop=1) at 
/work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
#7  0x00000000037f8ae7 in 
vtkTCPNetworkAccessManager::ProcessEventsInternal (this=0xb461a70, 
timeout_msecs=0, do_processing=true) at 
/work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
#8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents 
(this=0xb461a70, timeout_msecs=0) at 
/work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
#9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8, 
type=vtkProcessModule::PROCESS_SERVER) at 
/work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
#10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at 
/work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27



On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
> Yes, you need to load them manually, however. They are no longer auto-loaded.
>
> http://www.paraview.org/Bug/view.php?id=14542
>
> Utkarsh
>
> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Hi All,
>>
>> I'm having trouble with our NERSC install, plugins don't seem to be working
>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>
>> Burlen
>> _______________________________________________
>> Paraview-developers mailing list
>> Paraview-developers at paraview.org
>> http://public.kitware.com/mailman/listinfo/paraview-developers


From utkarsh.ayachit at kitware.com  Thu Sep 18 17:36:33 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Thu, 18 Sep 2014 17:36:33 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <541B4903.6010002@gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
Message-ID: <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>

Burlen,

A few things:
+ RGBZView should not be autoloaded since 6575a800a . That happened
after RC1, I'm afraid.
+ The server would still need a .plugins file (which should be
autogenerated). Can you verify such a file exists on the Cray? Also,
try running with PV_PLUGIN_DEBUG set. It should print more info. I get
the following with my static build:

>  PV_PLUGIN_DEBUG=1 ./bin/pvserver
Locate and load distributed plugin list.
Looking for static plugin '.plugins'
Loading plugin configuration xml:
/home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
Trying to locate plugin with name: SLACTools
Looking for static plugin 'SLACTools'
Found static plugin 'SLACTools'
--- Found SLACTools
....
Static build. Skipping PLUGIN_PATHS.Waiting for client...
Connection URL: cs://blight:11111
Accepting connection(s): blight:11111

I'm not sure what the RGBZView issue on your workstation is. Mind
updating to git/master on your workstation to see if the problem
persists? Actually, if it's only failing for RGBZView, it's less of a
concern than if fails for other plugins too.

Utkarsh

On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
> Hi Utkarsh,
>
> Good, glad it's still supported! There seem to be some issues though.
>
> On the Cray the remote plugin list is empty except for
> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client but
> not the server and thus as soon as the server connects the plugin manager
> opens to report the issue. The remote plugin list is empty. Browsing to the
> install and loading  the plugin doesn't work. it's a .a file so I'm not
> surprised. Judging from what I see on my workstation, the remote plugin list
> should be populated right?
>
> I've been trying to reproduce it locally on my workstation. I have different
> problems. The remote plugin list is fully populated and RGBZView loads on
> both sides. But when I try to load a plugin  via "load selected" ParaView
> crashes(stack below). That's running from the build. When I try the same on
> an install the server segv's as soon as I connect. The stack is the same as
> the in-build crash.
>
> By the way this is in client-server mode the server is built with sahred
> libs off, with mpi and python and without qt.
>
> Burlen
>
> on local linux workstation, client server, server built without qt, with
> python and mpi
> Program received signal SIGSEGV, Segmentation fault.
> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
> session=0xb4880b0) at
> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
> 473       if (this->Internals->ActiveSessionStack.back() != session)
>
> (gdb) where
> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
> session=0xb4880b0) at
> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream (this=0xb4880b0,
> location=21, stream=..., ignore_errors=false) at
> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
> (this=0xb4880b0, message=0xbc82030, message_length=16) at
> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
> rmiTag=55625) at
> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
> (this=0xbc79460, reportErrors=0, dont_loop=1) at
> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
> #7  0x00000000037f8ae7 in vtkTCPNetworkAccessManager::ProcessEventsInternal
> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
> (this=0xb461a70, timeout_msecs=0) at
> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
> type=vtkProcessModule::PROCESS_SERVER) at
> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>
>
>
>
> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>
>> Yes, you need to load them manually, however. They are no longer
>> auto-loaded.
>>
>> http://www.paraview.org/Bug/view.php?id=14542
>>
>> Utkarsh
>>
>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>> wrote:
>>>
>>> Hi All,
>>>
>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>> working
>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>
>>> Burlen
>>> _______________________________________________
>>> Paraview-developers mailing list
>>> Paraview-developers at paraview.org
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>

From burlen.loring at gmail.com  Thu Sep 18 19:48:53 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Thu, 18 Sep 2014 16:48:53 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>	<CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>	<541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
Message-ID: <541B6F65.1030005@gmail.com>

Hi Utkarsh,

I updated to master on both Cray and workstation.

On the Cray the .plugins file is installed in lib/paraview-4.2. The 
PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib. 
Cp'ing the file into the bin dir fixed that issue, I was able to load 
and use a plugin!!

The issues on the Cray and my workstation seem to be unrelated. On the 
workstation I'm seeing the same segv after updating to master. Now it 
happens in both build and install during connection. From the stack it 
looks unrelated to plugins. Happily I don't need the static build on my 
workstation, so it's not a huge issue.

Burlen

On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
> Burlen,
>
> A few things:
> + RGBZView should not be autoloaded since 6575a800a . That happened
> after RC1, I'm afraid.
> + The server would still need a .plugins file (which should be
> autogenerated). Can you verify such a file exists on the Cray? Also,
> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
> the following with my static build:
>
>>   PV_PLUGIN_DEBUG=1 ./bin/pvserver
> Locate and load distributed plugin list.
> Looking for static plugin '.plugins'
> Loading plugin configuration xml:
> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
> Trying to locate plugin with name: SLACTools
> Looking for static plugin 'SLACTools'
> Found static plugin 'SLACTools'
> --- Found SLACTools
> ....
> Static build. Skipping PLUGIN_PATHS.Waiting for client...
> Connection URL: cs://blight:11111
> Accepting connection(s): blight:11111
>
> I'm not sure what the RGBZView issue on your workstation is. Mind
> updating to git/master on your workstation to see if the problem
> persists? Actually, if it's only failing for RGBZView, it's less of a
> concern than if fails for other plugins too.
>
> Utkarsh
>
> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Hi Utkarsh,
>>
>> Good, glad it's still supported! There seem to be some issues though.
>>
>> On the Cray the remote plugin list is empty except for
>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client but
>> not the server and thus as soon as the server connects the plugin manager
>> opens to report the issue. The remote plugin list is empty. Browsing to the
>> install and loading  the plugin doesn't work. it's a .a file so I'm not
>> surprised. Judging from what I see on my workstation, the remote plugin list
>> should be populated right?
>>
>> I've been trying to reproduce it locally on my workstation. I have different
>> problems. The remote plugin list is fully populated and RGBZView loads on
>> both sides. But when I try to load a plugin  via "load selected" ParaView
>> crashes(stack below). That's running from the build. When I try the same on
>> an install the server segv's as soon as I connect. The stack is the same as
>> the in-build crash.
>>
>> By the way this is in client-server mode the server is built with sahred
>> libs off, with mpi and python and without qt.
>>
>> Burlen
>>
>> on local linux workstation, client server, server built without qt, with
>> python and mpi
>> Program received signal SIGSEGV, Segmentation fault.
>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>> session=0xb4880b0) at
>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>
>> (gdb) where
>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>> session=0xb4880b0) at
>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream (this=0xb4880b0,
>> location=21, stream=..., ignore_errors=false) at
>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>> rmiTag=55625) at
>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>> #7  0x00000000037f8ae7 in vtkTCPNetworkAccessManager::ProcessEventsInternal
>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>> (this=0xb461a70, timeout_msecs=0) at
>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>> type=vtkProcessModule::PROCESS_SERVER) at
>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>
>>
>>
>>
>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>> Yes, you need to load them manually, however. They are no longer
>>> auto-loaded.
>>>
>>> http://www.paraview.org/Bug/view.php?id=14542
>>>
>>> Utkarsh
>>>
>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>>> wrote:
>>>> Hi All,
>>>>
>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>> working
>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>
>>>> Burlen
>>>> _______________________________________________
>>>> Paraview-developers mailing list
>>>> Paraview-developers at paraview.org
>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>


From utkarsh.ayachit at kitware.com  Fri Sep 19 00:40:42 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 00:40:42 -0400
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <FDF9BB35-15EF-4B42-B70A-98D06579EC2C@ornl.gov>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
 <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
 <FDF9BB35-15EF-4B42-B70A-98D06579EC2C@ornl.gov>
Message-ID: <CADHSJjA+zXf9m30ZHG6NunvcXRpxAJesf9z=KoAph2qQE7nG2A@mail.gmail.com>

The IESHIM thing can be ignored. It often shows up missing. Are the
right Python dlls being loaded? Is it accidentally picking them up
from some other path?

On Thu, Sep 18, 2014 at 11:49 AM, Reuter, Michael A. <reuterma at ornl.gov> wrote:
> Ignore the red function points thing. Had other PV libs in the path. Cleaned that up, but the dependency thing still stands. The executable can find all the necessary libraries except IESHIMS.DLL.
>
> M
>
> On Sep 18, 2014, at 10:17 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:
>
> Since you said you build with your own installation of Python, is
> Python in PATH? Try using the dependency walker[1] to see if it can
> find all the dependencies correctly.
>
> Utkarsh
>
> [1] http://www.dependencywalker.com/
>
> On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov>> wrote:
> Hi,
>
> I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.
>
> Thanks,
> M
>
> Dr. Michael Reuter
> Scientific Data Analysis Group
> Neutron Data Analysis and Visualization Division
> Oak Ridge National Laboratory
>
> Office: 1-865-241-7216
> Fax: 1-865-574-6080
> Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>
>
>
>
>
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org>
> http://public.kitware.com/mailman/listinfo/paraview-developers
>

From jfavre at cscs.ch  Fri Sep 19 05:33:07 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Fri, 19 Sep 2014 09:33:07 +0000
Subject: [Paraview-developers] detecting X display
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F682813B8CD@MBX11.d.ethz.ch>

After checking out the latest changes, my server-side rendering on the Cray does not work anymore because the DISPLAY is not correctly identified.

Error message is:

"Display is not accessible on the server side.
Remote rendering will be disabled."

I reverted to a tag just before the fix to detect X

 ad33f7548abf4b0b643e2ed15dd9ba1f8690c04f

and got everything working again.

P.S. I use hardware-rendering and specifically start X servers and set DISPLAYs to :0 for every job.
-----------------
Jean 

From utkarsh.ayachit at kitware.com  Fri Sep 19 08:34:11 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 08:34:11 -0400
Subject: [Paraview-developers] detecting X display
In-Reply-To: <0EB9B6375711A04B820E6B6F5CCA9F682813B8CD@MBX11.d.ethz.ch>
References: <0EB9B6375711A04B820E6B6F5CCA9F682813B8CD@MBX11.d.ethz.ch>
Message-ID: <CADHSJjCcK97+B--_QMzz4OCVxYsiN61tPa9VO1QHwEnhWnSYjw@mail.gmail.com>

Jean,

Can you try the attached patch and see if the output on the
server-side has any useful information that could help diagnose this?

Thanks,
Utkarsh

On Fri, Sep 19, 2014 at 5:33 AM, Favre  Jean <jfavre at cscs.ch> wrote:
> After checking out the latest changes, my server-side rendering on the Cray does not work anymore because the DISPLAY is not correctly identified.
>
> Error message is:
>
> "Display is not accessible on the server side.
> Remote rendering will be disabled."
>
> I reverted to a tag just before the fix to detect X
>
>  ad33f7548abf4b0b643e2ed15dd9ba1f8690c04f
>
> and got everything working again.
>
> P.S. I use hardware-rendering and specifically start X servers and set DISPLAYs to :0 for every job.
> -----------------
> Jean
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
-------------- next part --------------
diff --git a/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx b/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx
index 3575fcf..cbf0c75 100644
--- a/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx
+++ b/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx
@@ -178,6 +178,7 @@ bool vtkProcessModule::Initialize(ProcessTypes type, int &argc, char** &argv)
       {
       char *displayenv = new char[strlen(argv[i+1]) + 10];
       sprintf(displayenv, "DISPLAY=%s", argv[i+1]);
+      vtkGenericWarningMacro("Setting display from command line args as: " << displayenv);
       vtksys::SystemTools::PutEnv(displayenv);
       delete [] displayenv;
       // safe to delete since PutEnv keeps a copy of the string.
diff --git a/ParaViewCore/ClientServerCore/Rendering/vtkPVDisplayInformation.cxx b/ParaViewCore/ClientServerCore/Rendering/vtkPVDisplayInformation.cxx
index a9a2a00..1011034 100644
--- a/ParaViewCore/ClientServerCore/Rendering/vtkPVDisplayInformation.cxx
+++ b/ParaViewCore/ClientServerCore/Rendering/vtkPVDisplayInformation.cxx
@@ -19,7 +19,7 @@
 #include "vtkPVOptions.h"
 #include "vtkProcessModule.h"
 #include "vtkToolkits.h"
-
+#include "vtksys/SystemTools.hxx"
 #include "vtkRenderingOpenGLConfigure.h" // needed for VTK_USE_X
 #if defined(VTK_USE_X)
 # include <X11/Xlib.h>
@@ -50,15 +50,18 @@ void vtkPVDisplayInformation::PrintSelf(ostream& os, vtkIndent indent)
 bool vtkPVDisplayInformation::CanOpenDisplayLocally()
 {
 #if defined(VTK_USE_X)
+  vtkWarningMacro("Testing  DISPLAY=" << vtksys::SystemTools::GetEnv("DISPLAY"));
   Display* dId = XOpenDisplay((char *)NULL);
   if (dId)
     {
+    vtkWarningMacro("Success");
     XCloseDisplay(dId);
     return true;
     }
+  vtkWarningMacro("Failed");
   return false;
 #endif
-
+  vtkWarningMacro("Success by default");
   return true;
 }
 

From utkarsh.ayachit at kitware.com  Fri Sep 19 09:49:17 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 09:49:17 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <541B6F65.1030005@gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
Message-ID: <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>

Ah! I know why it's looking for .plugins in the wrong place. I'll push
a fix, Thanks!

Utkarsh

On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
> Hi Utkarsh,
>
> I updated to master on both Cray and workstation.
>
> On the Cray the .plugins file is installed in lib/paraview-4.2. The
> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib. Cp'ing
> the file into the bin dir fixed that issue, I was able to load and use a
> plugin!!
>
> The issues on the Cray and my workstation seem to be unrelated. On the
> workstation I'm seeing the same segv after updating to master. Now it
> happens in both build and install during connection. From the stack it looks
> unrelated to plugins. Happily I don't need the static build on my
> workstation, so it's not a huge issue.
>
> Burlen
>
>
> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>
>> Burlen,
>>
>> A few things:
>> + RGBZView should not be autoloaded since 6575a800a . That happened
>> after RC1, I'm afraid.
>> + The server would still need a .plugins file (which should be
>> autogenerated). Can you verify such a file exists on the Cray? Also,
>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>> the following with my static build:
>>
>>>   PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>
>> Locate and load distributed plugin list.
>> Looking for static plugin '.plugins'
>> Loading plugin configuration xml:
>>
>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>> Trying to locate plugin with name: SLACTools
>> Looking for static plugin 'SLACTools'
>> Found static plugin 'SLACTools'
>> --- Found SLACTools
>> ....
>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>> Connection URL: cs://blight:11111
>> Accepting connection(s): blight:11111
>>
>> I'm not sure what the RGBZView issue on your workstation is. Mind
>> updating to git/master on your workstation to see if the problem
>> persists? Actually, if it's only failing for RGBZView, it's less of a
>> concern than if fails for other plugins too.
>>
>> Utkarsh
>>
>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com>
>> wrote:
>>>
>>> Hi Utkarsh,
>>>
>>> Good, glad it's still supported! There seem to be some issues though.
>>>
>>> On the Cray the remote plugin list is empty except for
>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client
>>> but
>>> not the server and thus as soon as the server connects the plugin manager
>>> opens to report the issue. The remote plugin list is empty. Browsing to
>>> the
>>> install and loading  the plugin doesn't work. it's a .a file so I'm not
>>> surprised. Judging from what I see on my workstation, the remote plugin
>>> list
>>> should be populated right?
>>>
>>> I've been trying to reproduce it locally on my workstation. I have
>>> different
>>> problems. The remote plugin list is fully populated and RGBZView loads on
>>> both sides. But when I try to load a plugin  via "load selected" ParaView
>>> crashes(stack below). That's running from the build. When I try the same
>>> on
>>> an install the server segv's as soon as I connect. The stack is the same
>>> as
>>> the in-build crash.
>>>
>>> By the way this is in client-server mode the server is built with sahred
>>> libs off, with mpi and python and without qt.
>>>
>>> Burlen
>>>
>>> on local linux workstation, client server, server built without qt, with
>>> python and mpi
>>> Program received signal SIGSEGV, Segmentation fault.
>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>> session=0xb4880b0) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>
>>> (gdb) where
>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>> session=0xb4880b0) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>> (this=0xb4880b0,
>>> location=21, stream=..., ignore_errors=false) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>> rmiTag=55625) at
>>>
>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>
>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>> #7  0x00000000037f8ae7 in
>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>> (this=0xb461a70, timeout_msecs=0) at
>>>
>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>> type=vtkProcessModule::PROCESS_SERVER) at
>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>
>>>
>>>
>>>
>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>
>>>> Yes, you need to load them manually, however. They are no longer
>>>> auto-loaded.
>>>>
>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>
>>>> Utkarsh
>>>>
>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>>>> wrote:
>>>>>
>>>>> Hi All,
>>>>>
>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>> working
>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>
>>>>> Burlen
>>>>> _______________________________________________
>>>>> Paraview-developers mailing list
>>>>> Paraview-developers at paraview.org
>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>
>>>
>

From reuterma at ornl.gov  Fri Sep 19 10:55:27 2014
From: reuterma at ornl.gov (Reuter, Michael A.)
Date: Fri, 19 Sep 2014 14:55:27 +0000
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <CADHSJjA+zXf9m30ZHG6NunvcXRpxAJesf9z=KoAph2qQE7nG2A@mail.gmail.com>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
 <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
 <FDF9BB35-15EF-4B42-B70A-98D06579EC2C@ornl.gov>
 <CADHSJjA+zXf9m30ZHG6NunvcXRpxAJesf9z=KoAph2qQE7nG2A@mail.gmail.com>
Message-ID: <6435380C-D8D9-4DEE-88CB-CA5EAEE29BD1@ornl.gov>

The PYTHON27.DLL is coming from the expected location. That?s the only system Python library being reported by dependency walker.


On Sep 19, 2014, at 12:40 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:

The IESHIM thing can be ignored. It often shows up missing. Are the
right Python dlls being loaded? Is it accidentally picking them up
from some other path?

On Thu, Sep 18, 2014 at 11:49 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov>> wrote:
Ignore the red function points thing. Had other PV libs in the path. Cleaned that up, but the dependency thing still stands. The executable can find all the necessary libraries except IESHIMS.DLL.

M

On Sep 18, 2014, at 10:17 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com><mailto:utkarsh.ayachit at kitware.com>> wrote:

Since you said you build with your own installation of Python, is
Python in PATH? Try using the dependency walker[1] to see if it can
find all the dependencies correctly.

Utkarsh

[1] http://www.dependencywalker.com/

On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>> wrote:
Hi,

I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.

Thanks,
M

Dr. Michael Reuter
Scientific Data Analysis Group
Neutron Data Analysis and Visualization Division
Oak Ridge National Laboratory

Office: 1-865-241-7216
Fax: 1-865-574-6080
Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>






_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org><mailto:Paraview-developers at paraview.org>
http://public.kitware.com/mailman/listinfo/paraview-developers



From utkarsh.ayachit at kitware.com  Fri Sep 19 10:57:10 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 10:57:10 -0400
Subject: [Paraview-developers] Git-master and Windows
In-Reply-To: <6435380C-D8D9-4DEE-88CB-CA5EAEE29BD1@ornl.gov>
References: <40513B88-F05F-4247-8422-B4CF33B90F11@ornl.gov>
 <CADHSJjBvOr4PVN4tnyDWZobpfrzvUKUfO6u0vHvxVFVcjsRQ9g@mail.gmail.com>
 <FDF9BB35-15EF-4B42-B70A-98D06579EC2C@ornl.gov>
 <CADHSJjA+zXf9m30ZHG6NunvcXRpxAJesf9z=KoAph2qQE7nG2A@mail.gmail.com>
 <6435380C-D8D9-4DEE-88CB-CA5EAEE29BD1@ornl.gov>
Message-ID: <CADHSJjB9QWMA-g8_M15bmK2hR=MWLh5b0dUBxWes3wiEHrG4ww@mail.gmail.com>

I am out of ideas at this point. Any Windows gurus out there?

On Fri, Sep 19, 2014 at 10:55 AM, Reuter, Michael A. <reuterma at ornl.gov> wrote:
> The PYTHON27.DLL is coming from the expected location. That?s the only system Python library being reported by dependency walker.
>
>
> On Sep 19, 2014, at 12:40 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>> wrote:
>
> The IESHIM thing can be ignored. It often shows up missing. Are the
> right Python dlls being loaded? Is it accidentally picking them up
> from some other path?
>
> On Thu, Sep 18, 2014 at 11:49 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov>> wrote:
> Ignore the red function points thing. Had other PV libs in the path. Cleaned that up, but the dependency thing still stands. The executable can find all the necessary libraries except IESHIMS.DLL.
>
> M
>
> On Sep 18, 2014, at 10:17 AM, Utkarsh Ayachit <utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com><mailto:utkarsh.ayachit at kitware.com>> wrote:
>
> Since you said you build with your own installation of Python, is
> Python in PATH? Try using the dependency walker[1] to see if it can
> find all the dependencies correctly.
>
> Utkarsh
>
> [1] http://www.dependencywalker.com/
>
> On Thu, Sep 18, 2014 at 9:57 AM, Reuter, Michael A. <reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>> wrote:
> Hi,
>
> I?m having a very strange issue with trying to use a git-master version of ParaView on Windows using ninja/VS2012. I can get the build to complete just fine, but running the paraview.exe does nothing. It thinks for a bit and then exits without any hit of an error. A debug build does the same thing. I can?t attach any debugging process since subsequent attempts at execution exit much faster. I think the issue has to do with Python because I?ve made a build using PARAVIEW_ENABLE_PYTHON=OFF and paraview.exe works fine in that case. I?ve been left scratching my head, so I?m willing to entertain any and all ideas about what further things I can look at to determine the problem.
>
> Thanks,
> M
>
> Dr. Michael Reuter
> Scientific Data Analysis Group
> Neutron Data Analysis and Visualization Division
> Oak Ridge National Laboratory
>
> Office: 1-865-241-7216
> Fax: 1-865-574-6080
> Email: reuterma at ornl.gov<mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov><mailto:reuterma at ornl.gov>
>
>
>
>
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org<mailto:Paraview-developers at paraview.org><mailto:Paraview-developers at paraview.org>
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
>

From utkarsh.ayachit at kitware.com  Fri Sep 19 11:27:42 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 11:27:42 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
Message-ID: <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>

Burlen,

I was also able to reproduce the same segfault you're getting on your
workstation. It seems to be a problem with incremental builds for the
static build. A clean build (starting with an empty directory)
addressed the problem for me.

Utkarsh

On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Ah! I know why it's looking for .plugins in the wrong place. I'll push
> a fix, Thanks!
>
> Utkarsh
>
> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Hi Utkarsh,
>>
>> I updated to master on both Cray and workstation.
>>
>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib. Cp'ing
>> the file into the bin dir fixed that issue, I was able to load and use a
>> plugin!!
>>
>> The issues on the Cray and my workstation seem to be unrelated. On the
>> workstation I'm seeing the same segv after updating to master. Now it
>> happens in both build and install during connection. From the stack it looks
>> unrelated to plugins. Happily I don't need the static build on my
>> workstation, so it's not a huge issue.
>>
>> Burlen
>>
>>
>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>
>>> Burlen,
>>>
>>> A few things:
>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>> after RC1, I'm afraid.
>>> + The server would still need a .plugins file (which should be
>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>>> the following with my static build:
>>>
>>>>   PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>
>>> Locate and load distributed plugin list.
>>> Looking for static plugin '.plugins'
>>> Loading plugin configuration xml:
>>>
>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>> Trying to locate plugin with name: SLACTools
>>> Looking for static plugin 'SLACTools'
>>> Found static plugin 'SLACTools'
>>> --- Found SLACTools
>>> ....
>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>> Connection URL: cs://blight:11111
>>> Accepting connection(s): blight:11111
>>>
>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>> updating to git/master on your workstation to see if the problem
>>> persists? Actually, if it's only failing for RGBZView, it's less of a
>>> concern than if fails for other plugins too.
>>>
>>> Utkarsh
>>>
>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com>
>>> wrote:
>>>>
>>>> Hi Utkarsh,
>>>>
>>>> Good, glad it's still supported! There seem to be some issues though.
>>>>
>>>> On the Cray the remote plugin list is empty except for
>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client
>>>> but
>>>> not the server and thus as soon as the server connects the plugin manager
>>>> opens to report the issue. The remote plugin list is empty. Browsing to
>>>> the
>>>> install and loading  the plugin doesn't work. it's a .a file so I'm not
>>>> surprised. Judging from what I see on my workstation, the remote plugin
>>>> list
>>>> should be populated right?
>>>>
>>>> I've been trying to reproduce it locally on my workstation. I have
>>>> different
>>>> problems. The remote plugin list is fully populated and RGBZView loads on
>>>> both sides. But when I try to load a plugin  via "load selected" ParaView
>>>> crashes(stack below). That's running from the build. When I try the same
>>>> on
>>>> an install the server segv's as soon as I connect. The stack is the same
>>>> as
>>>> the in-build crash.
>>>>
>>>> By the way this is in client-server mode the server is built with sahred
>>>> libs off, with mpi and python and without qt.
>>>>
>>>> Burlen
>>>>
>>>> on local linux workstation, client server, server built without qt, with
>>>> python and mpi
>>>> Program received signal SIGSEGV, Segmentation fault.
>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>> session=0xb4880b0) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>>
>>>> (gdb) where
>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>> session=0xb4880b0) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>> (this=0xb4880b0,
>>>> location=21, stream=..., ignore_errors=false) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>> rmiTag=55625) at
>>>>
>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>
>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>> #7  0x00000000037f8ae7 in
>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>
>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>
>>>>
>>>>
>>>>
>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>
>>>>> Yes, you need to load them manually, however. They are no longer
>>>>> auto-loaded.
>>>>>
>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>>>>> wrote:
>>>>>>
>>>>>> Hi All,
>>>>>>
>>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>>> working
>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>
>>>>>> Burlen
>>>>>> _______________________________________________
>>>>>> Paraview-developers mailing list
>>>>>> Paraview-developers at paraview.org
>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>
>>>>
>>

From utkarsh.ayachit at kitware.com  Fri Sep 19 11:31:42 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 11:31:42 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
Message-ID: <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>

I spoke too soon, grrrr :~/.

Utkarsh

On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Burlen,
>
> I was also able to reproduce the same segfault you're getting on your
> workstation. It seems to be a problem with incremental builds for the
> static build. A clean build (starting with an empty directory)
> addressed the problem for me.
>
> Utkarsh
>
> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
> <utkarsh.ayachit at kitware.com> wrote:
>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>> a fix, Thanks!
>>
>> Utkarsh
>>
>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>>> Hi Utkarsh,
>>>
>>> I updated to master on both Cray and workstation.
>>>
>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib. Cp'ing
>>> the file into the bin dir fixed that issue, I was able to load and use a
>>> plugin!!
>>>
>>> The issues on the Cray and my workstation seem to be unrelated. On the
>>> workstation I'm seeing the same segv after updating to master. Now it
>>> happens in both build and install during connection. From the stack it looks
>>> unrelated to plugins. Happily I don't need the static build on my
>>> workstation, so it's not a huge issue.
>>>
>>> Burlen
>>>
>>>
>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>
>>>> Burlen,
>>>>
>>>> A few things:
>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>> after RC1, I'm afraid.
>>>> + The server would still need a .plugins file (which should be
>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>>>> the following with my static build:
>>>>
>>>>>   PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>
>>>> Locate and load distributed plugin list.
>>>> Looking for static plugin '.plugins'
>>>> Loading plugin configuration xml:
>>>>
>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>> Trying to locate plugin with name: SLACTools
>>>> Looking for static plugin 'SLACTools'
>>>> Found static plugin 'SLACTools'
>>>> --- Found SLACTools
>>>> ....
>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>> Connection URL: cs://blight:11111
>>>> Accepting connection(s): blight:11111
>>>>
>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>> updating to git/master on your workstation to see if the problem
>>>> persists? Actually, if it's only failing for RGBZView, it's less of a
>>>> concern than if fails for other plugins too.
>>>>
>>>> Utkarsh
>>>>
>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com>
>>>> wrote:
>>>>>
>>>>> Hi Utkarsh,
>>>>>
>>>>> Good, glad it's still supported! There seem to be some issues though.
>>>>>
>>>>> On the Cray the remote plugin list is empty except for
>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client
>>>>> but
>>>>> not the server and thus as soon as the server connects the plugin manager
>>>>> opens to report the issue. The remote plugin list is empty. Browsing to
>>>>> the
>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm not
>>>>> surprised. Judging from what I see on my workstation, the remote plugin
>>>>> list
>>>>> should be populated right?
>>>>>
>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>> different
>>>>> problems. The remote plugin list is fully populated and RGBZView loads on
>>>>> both sides. But when I try to load a plugin  via "load selected" ParaView
>>>>> crashes(stack below). That's running from the build. When I try the same
>>>>> on
>>>>> an install the server segv's as soon as I connect. The stack is the same
>>>>> as
>>>>> the in-build crash.
>>>>>
>>>>> By the way this is in client-server mode the server is built with sahred
>>>>> libs off, with mpi and python and without qt.
>>>>>
>>>>> Burlen
>>>>>
>>>>> on local linux workstation, client server, server built without qt, with
>>>>> python and mpi
>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>> session=0xb4880b0) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>>>
>>>>> (gdb) where
>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>> session=0xb4880b0) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>> (this=0xb4880b0,
>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>> rmiTag=55625) at
>>>>>
>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>
>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>> #7  0x00000000037f8ae7 in
>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>
>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>
>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>> auto-loaded.
>>>>>>
>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>
>>>>>> Utkarsh
>>>>>>
>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>>>>>> wrote:
>>>>>>>
>>>>>>> Hi All,
>>>>>>>
>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>>>> working
>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>
>>>>>>> Burlen
>>>>>>> _______________________________________________
>>>>>>> Paraview-developers mailing list
>>>>>>> Paraview-developers at paraview.org
>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>
>>>>>
>>>

From burlen.loring at gmail.com  Fri Sep 19 12:05:09 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Fri, 19 Sep 2014 09:05:09 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>	<CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>	<541B4903.6010002@gmail.com>	<CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>	<541B6F65.1030005@gmail.com>	<CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>	<CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
Message-ID: <541C5435.4090008@gmail.com>

Thanks for looking into that, and thanks for fixing the .plugins issue! 
The updated static plugin support in 4.2 is great!

On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
> I spoke too soon, grrrr :~/.
>
> Utkarsh
>
> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
> <utkarsh.ayachit at kitware.com> wrote:
>> Burlen,
>>
>> I was also able to reproduce the same segfault you're getting on your
>> workstation. It seems to be a problem with incremental builds for the
>> static build. A clean build (starting with an empty directory)
>> addressed the problem for me.
>>
>> Utkarsh
>>
>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>> <utkarsh.ayachit at kitware.com> wrote:
>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>> a fix, Thanks!
>>>
>>> Utkarsh
>>>
>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>>>> Hi Utkarsh,
>>>>
>>>> I updated to master on both Cray and workstation.
>>>>
>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib. Cp'ing
>>>> the file into the bin dir fixed that issue, I was able to load and use a
>>>> plugin!!
>>>>
>>>> The issues on the Cray and my workstation seem to be unrelated. On the
>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>> happens in both build and install during connection. From the stack it looks
>>>> unrelated to plugins. Happily I don't need the static build on my
>>>> workstation, so it's not a huge issue.
>>>>
>>>> Burlen
>>>>
>>>>
>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>> Burlen,
>>>>>
>>>>> A few things:
>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>> after RC1, I'm afraid.
>>>>> + The server would still need a .plugins file (which should be
>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>>>>> the following with my static build:
>>>>>
>>>>>>    PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>> Locate and load distributed plugin list.
>>>>> Looking for static plugin '.plugins'
>>>>> Loading plugin configuration xml:
>>>>>
>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>> Trying to locate plugin with name: SLACTools
>>>>> Looking for static plugin 'SLACTools'
>>>>> Found static plugin 'SLACTools'
>>>>> --- Found SLACTools
>>>>> ....
>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>> Connection URL: cs://blight:11111
>>>>> Accepting connection(s): blight:11111
>>>>>
>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>> updating to git/master on your workstation to see if the problem
>>>>> persists? Actually, if it's only failing for RGBZView, it's less of a
>>>>> concern than if fails for other plugins too.
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring <burlen.loring at gmail.com>
>>>>> wrote:
>>>>>> Hi Utkarsh,
>>>>>>
>>>>>> Good, glad it's still supported! There seem to be some issues though.
>>>>>>
>>>>>> On the Cray the remote plugin list is empty except for
>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the client
>>>>>> but
>>>>>> not the server and thus as soon as the server connects the plugin manager
>>>>>> opens to report the issue. The remote plugin list is empty. Browsing to
>>>>>> the
>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm not
>>>>>> surprised. Judging from what I see on my workstation, the remote plugin
>>>>>> list
>>>>>> should be populated right?
>>>>>>
>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>> different
>>>>>> problems. The remote plugin list is fully populated and RGBZView loads on
>>>>>> both sides. But when I try to load a plugin  via "load selected" ParaView
>>>>>> crashes(stack below). That's running from the build. When I try the same
>>>>>> on
>>>>>> an install the server segv's as soon as I connect. The stack is the same
>>>>>> as
>>>>>> the in-build crash.
>>>>>>
>>>>>> By the way this is in client-server mode the server is built with sahred
>>>>>> libs off, with mpi and python and without qt.
>>>>>>
>>>>>> Burlen
>>>>>>
>>>>>> on local linux workstation, client server, server built without qt, with
>>>>>> python and mpi
>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>> session=0xb4880b0) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>>>>
>>>>>> (gdb) where
>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>> session=0xb4880b0) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>> (this=0xb4880b0,
>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>> #3  0x00000000034ac735 in vtkPVSessionServer::OnClientServerMessageRMI
>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>> rmiTag=55625) at
>>>>>>
>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>
>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>> #7  0x00000000037f8ae7 in
>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>
>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>> auto-loaded.
>>>>>>>
>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>
>>>>>>> Utkarsh
>>>>>>>
>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring <burlen.loring at gmail.com>
>>>>>>> wrote:
>>>>>>>> Hi All,
>>>>>>>>
>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>>>>> working
>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>
>>>>>>>> Burlen
>>>>>>>> _______________________________________________
>>>>>>>> Paraview-developers mailing list
>>>>>>>> Paraview-developers at paraview.org
>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>


From sankhesh.jhaveri at kitware.com  Fri Sep 19 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Fri, 19 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Thursday, September 18 2014
Message-ID: <20140919163103.5D5D87FFD7@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]Win32-vs9-shared-release 2
   [2]Win64-vs9-static-release 1
   [3]Win64-vs9-shared-debug-nocollab 2
   [4]Lion-gcc-pvVTK 1
   [5]ubuntu-x64-nightlynext 1
   [6]fedora-x64-icc-nightlynext-release 1
   -------------------- REPORT --------------------
   5 FAILURES
   [7]pvcs.ExtractLevel , 1 , ['fedora-x64-icc-nightlynext-release']
   [8]pvcs-collab.VariableSelector , 1 , ['ubuntu-x64-nightlynext']
   [9]pvcs-collab.SpreadSheet1 , 1 , ['Lion-gcc-pvVTK']
   [10]vtkPVServerManagerDefaultPython-MPI-SymmetricLinePlotInScripts , 2
   , ['Win32-vs9-shared-release', 'Win64-vs9-shared-debug-nocollab']
   [11]pvcs-tile-display-2x1.TileDisplay3DTesting-2x1 , 3 ,
   ['Win32-vs9-shared-release', 'Win64-vs9-static-release',
   'Win64-vs9-shared-debug-nocollab']

   Issues with submitters
   The following expected submitters did not submit:
   amber12.kitware , Master-Win64-vs9-static-release

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=Win32-vs9-shared-release
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-static-release
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=Win64-vs9-shared-debug-nocollab
   4. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=Lion-gcc-pvVTK
   5. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   6. http://open.cdash.org/index.php?project=ParaView&date=2014-09-19&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
   7. http://open.cdash.org/testSummary.php?project=9&date=2014-09-19&name=pvcs.ExtractLevel
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-19&name=pvcs-collab.VariableSelector
   9. http://open.cdash.org/testSummary.php?project=9&date=2014-09-19&name=pvcs-collab.SpreadSheet1
  10. http://open.cdash.org/testSummary.php?project=9&date=2014-09-19&name=vtkPVServerManagerDefaultPython-MPI-SymmetricLinePlotInScripts
  11. http://open.cdash.org/testSummary.php?project=9&date=2014-09-19&name=pvcs-tile-display-2x1.TileDisplay3DTesting-2x1

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140919/dd0f55bd/attachment.html>

From utkarsh.ayachit at kitware.com  Fri Sep 19 14:30:41 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 14:30:41 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <541C5435.4090008@gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
Message-ID: <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>

Burlen,

There are the fixes that should address both the static pvserver not
finding ".plugins" as well as the issue on your local workstation.

http://review.source.kitware.com/#/t/4692

For the latter, this is more of an hack, but I'll have that addressed
properly for 4.3.

Utkarsh


On Fri, Sep 19, 2014 at 12:05 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
> Thanks for looking into that, and thanks for fixing the .plugins issue! The
> updated static plugin support in 4.2 is great!
>
>
> On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
>>
>> I spoke too soon, grrrr :~/.
>>
>> Utkarsh
>>
>> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
>> <utkarsh.ayachit at kitware.com> wrote:
>>>
>>> Burlen,
>>>
>>> I was also able to reproduce the same segfault you're getting on your
>>> workstation. It seems to be a problem with incremental builds for the
>>> static build. A clean build (starting with an empty directory)
>>> addressed the problem for me.
>>>
>>> Utkarsh
>>>
>>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>
>>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>>> a fix, Thanks!
>>>>
>>>> Utkarsh
>>>>
>>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com>
>>>> wrote:
>>>>>
>>>>> Hi Utkarsh,
>>>>>
>>>>> I updated to master on both Cray and workstation.
>>>>>
>>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib.
>>>>> Cp'ing
>>>>> the file into the bin dir fixed that issue, I was able to load and use
>>>>> a
>>>>> plugin!!
>>>>>
>>>>> The issues on the Cray and my workstation seem to be unrelated. On the
>>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>>> happens in both build and install during connection. From the stack it
>>>>> looks
>>>>> unrelated to plugins. Happily I don't need the static build on my
>>>>> workstation, so it's not a huge issue.
>>>>>
>>>>> Burlen
>>>>>
>>>>>
>>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>>>
>>>>>> Burlen,
>>>>>>
>>>>>> A few things:
>>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>>> after RC1, I'm afraid.
>>>>>> + The server would still need a .plugins file (which should be
>>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>>>>>> the following with my static build:
>>>>>>
>>>>>>>    PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>>>
>>>>>> Locate and load distributed plugin list.
>>>>>> Looking for static plugin '.plugins'
>>>>>> Loading plugin configuration xml:
>>>>>>
>>>>>>
>>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>>> Trying to locate plugin with name: SLACTools
>>>>>> Looking for static plugin 'SLACTools'
>>>>>> Found static plugin 'SLACTools'
>>>>>> --- Found SLACTools
>>>>>> ....
>>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>>> Connection URL: cs://blight:11111
>>>>>> Accepting connection(s): blight:11111
>>>>>>
>>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>>> updating to git/master on your workstation to see if the problem
>>>>>> persists? Actually, if it's only failing for RGBZView, it's less of a
>>>>>> concern than if fails for other plugins too.
>>>>>>
>>>>>> Utkarsh
>>>>>>
>>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring
>>>>>> <burlen.loring at gmail.com>
>>>>>> wrote:
>>>>>>>
>>>>>>> Hi Utkarsh,
>>>>>>>
>>>>>>> Good, glad it's still supported! There seem to be some issues though.
>>>>>>>
>>>>>>> On the Cray the remote plugin list is empty except for
>>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the
>>>>>>> client
>>>>>>> but
>>>>>>> not the server and thus as soon as the server connects the plugin
>>>>>>> manager
>>>>>>> opens to report the issue. The remote plugin list is empty. Browsing
>>>>>>> to
>>>>>>> the
>>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm
>>>>>>> not
>>>>>>> surprised. Judging from what I see on my workstation, the remote
>>>>>>> plugin
>>>>>>> list
>>>>>>> should be populated right?
>>>>>>>
>>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>>> different
>>>>>>> problems. The remote plugin list is fully populated and RGBZView
>>>>>>> loads on
>>>>>>> both sides. But when I try to load a plugin  via "load selected"
>>>>>>> ParaView
>>>>>>> crashes(stack below). That's running from the build. When I try the
>>>>>>> same
>>>>>>> on
>>>>>>> an install the server segv's as soon as I connect. The stack is the
>>>>>>> same
>>>>>>> as
>>>>>>> the in-build crash.
>>>>>>>
>>>>>>> By the way this is in client-server mode the server is built with
>>>>>>> sahred
>>>>>>> libs off, with mpi and python and without qt.
>>>>>>>
>>>>>>> Burlen
>>>>>>>
>>>>>>> on local linux workstation, client server, server built without qt,
>>>>>>> with
>>>>>>> python and mpi
>>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>>> session=0xb4880b0) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>>>>>
>>>>>>> (gdb) where
>>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession
>>>>>>> (this=0x0,
>>>>>>> session=0xb4880b0) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>>> (this=0xb4880b0,
>>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>>> #3  0x00000000034ac735 in
>>>>>>> vtkPVSessionServer::OnClientServerMessageRMI
>>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>>> rmiTag=55625) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>>> #7  0x00000000037f8ae7 in
>>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>>
>>>>>>>
>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>>>
>>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>>> auto-loaded.
>>>>>>>>
>>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>>
>>>>>>>> Utkarsh
>>>>>>>>
>>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring
>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> Hi All,
>>>>>>>>>
>>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>>>>>> working
>>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>>
>>>>>>>>> Burlen
>>>>>>>>> _______________________________________________
>>>>>>>>> Paraview-developers mailing list
>>>>>>>>> Paraview-developers at paraview.org
>>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>>
>>>>>>>
>

From burlen.loring at gmail.com  Fri Sep 19 19:53:03 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Fri, 19 Sep 2014 16:53:03 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>	<CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>	<541B4903.6010002@gmail.com>	<CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>	<541B6F65.1030005@gmail.com>	<CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>	<CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>	<CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>	<541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
Message-ID: <541CC1DF.6060307@gmail.com>

Thanks Utkarsh,

I tested the patch but still had a segv.

Re: plugins in static linked server, how are they loaded from pvbatch? 
Does LoadPlugin call (which takes a file name) work?

On 09/19/2014 11:30 AM, Utkarsh Ayachit wrote:
> Burlen,
>
> There are the fixes that should address both the static pvserver not
> finding ".plugins" as well as the issue on your local workstation.
>
> http://review.source.kitware.com/#/t/4692
>
> For the latter, this is more of an hack, but I'll have that addressed
> properly for 4.3.
>
> Utkarsh
>
>
> On Fri, Sep 19, 2014 at 12:05 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Thanks for looking into that, and thanks for fixing the .plugins issue! The
>> updated static plugin support in 4.2 is great!
>>
>>
>> On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
>>> I spoke too soon, grrrr :~/.
>>>
>>> Utkarsh
>>>
>>> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
>>> <utkarsh.ayachit at kitware.com> wrote:
>>>> Burlen,
>>>>
>>>> I was also able to reproduce the same segfault you're getting on your
>>>> workstation. It seems to be a problem with incremental builds for the
>>>> static build. A clean build (starting with an empty directory)
>>>> addressed the problem for me.
>>>>
>>>> Utkarsh
>>>>
>>>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>>>> a fix, Thanks!
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring <burlen.loring at gmail.com>
>>>>> wrote:
>>>>>> Hi Utkarsh,
>>>>>>
>>>>>> I updated to master on both Cray and workstation.
>>>>>>
>>>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib.
>>>>>> Cp'ing
>>>>>> the file into the bin dir fixed that issue, I was able to load and use
>>>>>> a
>>>>>> plugin!!
>>>>>>
>>>>>> The issues on the Cray and my workstation seem to be unrelated. On the
>>>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>>>> happens in both build and install during connection. From the stack it
>>>>>> looks
>>>>>> unrelated to plugins. Happily I don't need the static build on my
>>>>>> workstation, so it's not a huge issue.
>>>>>>
>>>>>> Burlen
>>>>>>
>>>>>>
>>>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>>>> Burlen,
>>>>>>>
>>>>>>> A few things:
>>>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>>>> after RC1, I'm afraid.
>>>>>>> + The server would still need a .plugins file (which should be
>>>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I get
>>>>>>> the following with my static build:
>>>>>>>
>>>>>>>>     PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>>>> Locate and load distributed plugin list.
>>>>>>> Looking for static plugin '.plugins'
>>>>>>> Loading plugin configuration xml:
>>>>>>>
>>>>>>>
>>>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>>>> Trying to locate plugin with name: SLACTools
>>>>>>> Looking for static plugin 'SLACTools'
>>>>>>> Found static plugin 'SLACTools'
>>>>>>> --- Found SLACTools
>>>>>>> ....
>>>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>>>> Connection URL: cs://blight:11111
>>>>>>> Accepting connection(s): blight:11111
>>>>>>>
>>>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>>>> updating to git/master on your workstation to see if the problem
>>>>>>> persists? Actually, if it's only failing for RGBZView, it's less of a
>>>>>>> concern than if fails for other plugins too.
>>>>>>>
>>>>>>> Utkarsh
>>>>>>>
>>>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring
>>>>>>> <burlen.loring at gmail.com>
>>>>>>> wrote:
>>>>>>>> Hi Utkarsh,
>>>>>>>>
>>>>>>>> Good, glad it's still supported! There seem to be some issues though.
>>>>>>>>
>>>>>>>> On the Cray the remote plugin list is empty except for
>>>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the
>>>>>>>> client
>>>>>>>> but
>>>>>>>> not the server and thus as soon as the server connects the plugin
>>>>>>>> manager
>>>>>>>> opens to report the issue. The remote plugin list is empty. Browsing
>>>>>>>> to
>>>>>>>> the
>>>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm
>>>>>>>> not
>>>>>>>> surprised. Judging from what I see on my workstation, the remote
>>>>>>>> plugin
>>>>>>>> list
>>>>>>>> should be populated right?
>>>>>>>>
>>>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>>>> different
>>>>>>>> problems. The remote plugin list is fully populated and RGBZView
>>>>>>>> loads on
>>>>>>>> both sides. But when I try to load a plugin  via "load selected"
>>>>>>>> ParaView
>>>>>>>> crashes(stack below). That's running from the build. When I try the
>>>>>>>> same
>>>>>>>> on
>>>>>>>> an install the server segv's as soon as I connect. The stack is the
>>>>>>>> same
>>>>>>>> as
>>>>>>>> the in-build crash.
>>>>>>>>
>>>>>>>> By the way this is in client-server mode the server is built with
>>>>>>>> sahred
>>>>>>>> libs off, with mpi and python and without qt.
>>>>>>>>
>>>>>>>> Burlen
>>>>>>>>
>>>>>>>> on local linux workstation, client server, server built without qt,
>>>>>>>> with
>>>>>>>> python and mpi
>>>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>>>> session=0xb4880b0) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>> 473       if (this->Internals->ActiveSessionStack.back() != session)
>>>>>>>>
>>>>>>>> (gdb) where
>>>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession
>>>>>>>> (this=0x0,
>>>>>>>> session=0xb4880b0) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>>>> (this=0xb4880b0,
>>>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>>>> #3  0x00000000034ac735 in
>>>>>>>> vtkPVSessionServer::OnClientServerMessageRMI
>>>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>>>> rmiTag=55625) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>>>> #7  0x00000000037f8ae7 in
>>>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>>>
>>>>>>>>
>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>>>> auto-loaded.
>>>>>>>>>
>>>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>>>
>>>>>>>>> Utkarsh
>>>>>>>>>
>>>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring
>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>> Hi All,
>>>>>>>>>>
>>>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to be
>>>>>>>>>> working
>>>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>>>
>>>>>>>>>> Burlen
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Paraview-developers mailing list
>>>>>>>>>> Paraview-developers at paraview.org
>>>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>>>


From utkarsh.ayachit at kitware.com  Fri Sep 19 19:58:17 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 19:58:17 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <541CC1DF.6060307@gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
 <541CC1DF.6060307@gmail.com>
Message-ID: <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>

Burlen,

You did rebuild both the client and the server right? You could not
get the same segfault anymore due to this change [1]. The code short
circuits when loading a shared lib. The segfault was a happening
because the plugin.so from the shared build was being loaded.

LoadPlugin() takes the filename indeed. For static builds, however,
the filename reported for the distributed plugins is just their name
itself. I need to clean that up, but didn't want to make those changes
so close to the release. I'll refactor that piece of code for 4.3.
Plugin cleanups are long overdue :).

[1] http://review.source.kitware.com/#/c/17131/1/ParaViewCore/ClientServerCore/Core/vtkPVPluginLoader.cxx

On Fri, Sep 19, 2014 at 7:53 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
> Thanks Utkarsh,
>
> I tested the patch but still had a segv.
>
> Re: plugins in static linked server, how are they loaded from pvbatch? Does
> LoadPlugin call (which takes a file name) work?
>
>
> On 09/19/2014 11:30 AM, Utkarsh Ayachit wrote:
>>
>> Burlen,
>>
>> There are the fixes that should address both the static pvserver not
>> finding ".plugins" as well as the issue on your local workstation.
>>
>> http://review.source.kitware.com/#/t/4692
>>
>> For the latter, this is more of an hack, but I'll have that addressed
>> properly for 4.3.
>>
>> Utkarsh
>>
>>
>> On Fri, Sep 19, 2014 at 12:05 PM, Burlen Loring <burlen.loring at gmail.com>
>> wrote:
>>>
>>> Thanks for looking into that, and thanks for fixing the .plugins issue!
>>> The
>>> updated static plugin support in 4.2 is great!
>>>
>>>
>>> On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
>>>>
>>>> I spoke too soon, grrrr :~/.
>>>>
>>>> Utkarsh
>>>>
>>>> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>
>>>>> Burlen,
>>>>>
>>>>> I was also able to reproduce the same segfault you're getting on your
>>>>> workstation. It seems to be a problem with incremental builds for the
>>>>> static build. A clean build (starting with an empty directory)
>>>>> addressed the problem for me.
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>
>>>>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>>>>> a fix, Thanks!
>>>>>>
>>>>>> Utkarsh
>>>>>>
>>>>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring
>>>>>> <burlen.loring at gmail.com>
>>>>>> wrote:
>>>>>>>
>>>>>>> Hi Utkarsh,
>>>>>>>
>>>>>>> I updated to master on both Cray and workstation.
>>>>>>>
>>>>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib.
>>>>>>> Cp'ing
>>>>>>> the file into the bin dir fixed that issue, I was able to load and
>>>>>>> use
>>>>>>> a
>>>>>>> plugin!!
>>>>>>>
>>>>>>> The issues on the Cray and my workstation seem to be unrelated. On
>>>>>>> the
>>>>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>>>>> happens in both build and install during connection. From the stack
>>>>>>> it
>>>>>>> looks
>>>>>>> unrelated to plugins. Happily I don't need the static build on my
>>>>>>> workstation, so it's not a huge issue.
>>>>>>>
>>>>>>> Burlen
>>>>>>>
>>>>>>>
>>>>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>>>>>
>>>>>>>> Burlen,
>>>>>>>>
>>>>>>>> A few things:
>>>>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>>>>> after RC1, I'm afraid.
>>>>>>>> + The server would still need a .plugins file (which should be
>>>>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I
>>>>>>>> get
>>>>>>>> the following with my static build:
>>>>>>>>
>>>>>>>>>     PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>>>>>
>>>>>>>> Locate and load distributed plugin list.
>>>>>>>> Looking for static plugin '.plugins'
>>>>>>>> Loading plugin configuration xml:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>>>>> Trying to locate plugin with name: SLACTools
>>>>>>>> Looking for static plugin 'SLACTools'
>>>>>>>> Found static plugin 'SLACTools'
>>>>>>>> --- Found SLACTools
>>>>>>>> ....
>>>>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>>>>> Connection URL: cs://blight:11111
>>>>>>>> Accepting connection(s): blight:11111
>>>>>>>>
>>>>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>>>>> updating to git/master on your workstation to see if the problem
>>>>>>>> persists? Actually, if it's only failing for RGBZView, it's less of
>>>>>>>> a
>>>>>>>> concern than if fails for other plugins too.
>>>>>>>>
>>>>>>>> Utkarsh
>>>>>>>>
>>>>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring
>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> Hi Utkarsh,
>>>>>>>>>
>>>>>>>>> Good, glad it's still supported! There seem to be some issues
>>>>>>>>> though.
>>>>>>>>>
>>>>>>>>> On the Cray the remote plugin list is empty except for
>>>>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the
>>>>>>>>> client
>>>>>>>>> but
>>>>>>>>> not the server and thus as soon as the server connects the plugin
>>>>>>>>> manager
>>>>>>>>> opens to report the issue. The remote plugin list is empty.
>>>>>>>>> Browsing
>>>>>>>>> to
>>>>>>>>> the
>>>>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm
>>>>>>>>> not
>>>>>>>>> surprised. Judging from what I see on my workstation, the remote
>>>>>>>>> plugin
>>>>>>>>> list
>>>>>>>>> should be populated right?
>>>>>>>>>
>>>>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>>>>> different
>>>>>>>>> problems. The remote plugin list is fully populated and RGBZView
>>>>>>>>> loads on
>>>>>>>>> both sides. But when I try to load a plugin  via "load selected"
>>>>>>>>> ParaView
>>>>>>>>> crashes(stack below). That's running from the build. When I try the
>>>>>>>>> same
>>>>>>>>> on
>>>>>>>>> an install the server segv's as soon as I connect. The stack is the
>>>>>>>>> same
>>>>>>>>> as
>>>>>>>>> the in-build crash.
>>>>>>>>>
>>>>>>>>> By the way this is in client-server mode the server is built with
>>>>>>>>> sahred
>>>>>>>>> libs off, with mpi and python and without qt.
>>>>>>>>>
>>>>>>>>> Burlen
>>>>>>>>>
>>>>>>>>> on local linux workstation, client server, server built without qt,
>>>>>>>>> with
>>>>>>>>> python and mpi
>>>>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>> 473       if (this->Internals->ActiveSessionStack.back() !=
>>>>>>>>> session)
>>>>>>>>>
>>>>>>>>> (gdb) where
>>>>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession
>>>>>>>>> (this=0x0,
>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0)
>>>>>>>>> at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>>>>> (this=0xb4880b0,
>>>>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>>>>> #3  0x00000000034ac735 in
>>>>>>>>> vtkPVSessionServer::OnClientServerMessageRMI
>>>>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>>>>> rmiTag=55625) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>>>>> #7  0x00000000037f8ae7 in
>>>>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>>>>>
>>>>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>>>>> auto-loaded.
>>>>>>>>>>
>>>>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>>>>
>>>>>>>>>> Utkarsh
>>>>>>>>>>
>>>>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring
>>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>> Hi All,
>>>>>>>>>>>
>>>>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to
>>>>>>>>>>> be
>>>>>>>>>>> working
>>>>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>>>>
>>>>>>>>>>> Burlen
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> Paraview-developers mailing list
>>>>>>>>>>> Paraview-developers at paraview.org
>>>>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>>>>
>>>>>>>>>
>

From utkarsh.ayachit at kitware.com  Fri Sep 19 20:08:00 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 19 Sep 2014 20:08:00 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
 <541CC1DF.6060307@gmail.com>
 <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>
Message-ID: <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>

Anycayse, the problem is indeed stemming from a bug that Jean and Alan
reported earlier about plugins
(http://www.paraview.org/Bug/view.php?id=14732).

I am going to target this for 4.2.1. Cleaning this up properly may
take a little more love and care.

Utkarsh

On Fri, Sep 19, 2014 at 7:58 PM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Burlen,
>
> You did rebuild both the client and the server right? You could not
> get the same segfault anymore due to this change [1]. The code short
> circuits when loading a shared lib. The segfault was a happening
> because the plugin.so from the shared build was being loaded.
>
> LoadPlugin() takes the filename indeed. For static builds, however,
> the filename reported for the distributed plugins is just their name
> itself. I need to clean that up, but didn't want to make those changes
> so close to the release. I'll refactor that piece of code for 4.3.
> Plugin cleanups are long overdue :).
>
> [1] http://review.source.kitware.com/#/c/17131/1/ParaViewCore/ClientServerCore/Core/vtkPVPluginLoader.cxx
>
> On Fri, Sep 19, 2014 at 7:53 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Thanks Utkarsh,
>>
>> I tested the patch but still had a segv.
>>
>> Re: plugins in static linked server, how are they loaded from pvbatch? Does
>> LoadPlugin call (which takes a file name) work?
>>
>>
>> On 09/19/2014 11:30 AM, Utkarsh Ayachit wrote:
>>>
>>> Burlen,
>>>
>>> There are the fixes that should address both the static pvserver not
>>> finding ".plugins" as well as the issue on your local workstation.
>>>
>>> http://review.source.kitware.com/#/t/4692
>>>
>>> For the latter, this is more of an hack, but I'll have that addressed
>>> properly for 4.3.
>>>
>>> Utkarsh
>>>
>>>
>>> On Fri, Sep 19, 2014 at 12:05 PM, Burlen Loring <burlen.loring at gmail.com>
>>> wrote:
>>>>
>>>> Thanks for looking into that, and thanks for fixing the .plugins issue!
>>>> The
>>>> updated static plugin support in 4.2 is great!
>>>>
>>>>
>>>> On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
>>>>>
>>>>> I spoke too soon, grrrr :~/.
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>
>>>>>> Burlen,
>>>>>>
>>>>>> I was also able to reproduce the same segfault you're getting on your
>>>>>> workstation. It seems to be a problem with incremental builds for the
>>>>>> static build. A clean build (starting with an empty directory)
>>>>>> addressed the problem for me.
>>>>>>
>>>>>> Utkarsh
>>>>>>
>>>>>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>>
>>>>>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>>>>>> a fix, Thanks!
>>>>>>>
>>>>>>> Utkarsh
>>>>>>>
>>>>>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring
>>>>>>> <burlen.loring at gmail.com>
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> Hi Utkarsh,
>>>>>>>>
>>>>>>>> I updated to master on both Cray and workstation.
>>>>>>>>
>>>>>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>>>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib.
>>>>>>>> Cp'ing
>>>>>>>> the file into the bin dir fixed that issue, I was able to load and
>>>>>>>> use
>>>>>>>> a
>>>>>>>> plugin!!
>>>>>>>>
>>>>>>>> The issues on the Cray and my workstation seem to be unrelated. On
>>>>>>>> the
>>>>>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>>>>>> happens in both build and install during connection. From the stack
>>>>>>>> it
>>>>>>>> looks
>>>>>>>> unrelated to plugins. Happily I don't need the static build on my
>>>>>>>> workstation, so it's not a huge issue.
>>>>>>>>
>>>>>>>> Burlen
>>>>>>>>
>>>>>>>>
>>>>>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>>>>>>
>>>>>>>>> Burlen,
>>>>>>>>>
>>>>>>>>> A few things:
>>>>>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>>>>>> after RC1, I'm afraid.
>>>>>>>>> + The server would still need a .plugins file (which should be
>>>>>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I
>>>>>>>>> get
>>>>>>>>> the following with my static build:
>>>>>>>>>
>>>>>>>>>>     PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>>>>>>
>>>>>>>>> Locate and load distributed plugin list.
>>>>>>>>> Looking for static plugin '.plugins'
>>>>>>>>> Loading plugin configuration xml:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>>>>>> Trying to locate plugin with name: SLACTools
>>>>>>>>> Looking for static plugin 'SLACTools'
>>>>>>>>> Found static plugin 'SLACTools'
>>>>>>>>> --- Found SLACTools
>>>>>>>>> ....
>>>>>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>>>>>> Connection URL: cs://blight:11111
>>>>>>>>> Accepting connection(s): blight:11111
>>>>>>>>>
>>>>>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>>>>>> updating to git/master on your workstation to see if the problem
>>>>>>>>> persists? Actually, if it's only failing for RGBZView, it's less of
>>>>>>>>> a
>>>>>>>>> concern than if fails for other plugins too.
>>>>>>>>>
>>>>>>>>> Utkarsh
>>>>>>>>>
>>>>>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring
>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>> Hi Utkarsh,
>>>>>>>>>>
>>>>>>>>>> Good, glad it's still supported! There seem to be some issues
>>>>>>>>>> though.
>>>>>>>>>>
>>>>>>>>>> On the Cray the remote plugin list is empty except for
>>>>>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the
>>>>>>>>>> client
>>>>>>>>>> but
>>>>>>>>>> not the server and thus as soon as the server connects the plugin
>>>>>>>>>> manager
>>>>>>>>>> opens to report the issue. The remote plugin list is empty.
>>>>>>>>>> Browsing
>>>>>>>>>> to
>>>>>>>>>> the
>>>>>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm
>>>>>>>>>> not
>>>>>>>>>> surprised. Judging from what I see on my workstation, the remote
>>>>>>>>>> plugin
>>>>>>>>>> list
>>>>>>>>>> should be populated right?
>>>>>>>>>>
>>>>>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>>>>>> different
>>>>>>>>>> problems. The remote plugin list is fully populated and RGBZView
>>>>>>>>>> loads on
>>>>>>>>>> both sides. But when I try to load a plugin  via "load selected"
>>>>>>>>>> ParaView
>>>>>>>>>> crashes(stack below). That's running from the build. When I try the
>>>>>>>>>> same
>>>>>>>>>> on
>>>>>>>>>> an install the server segv's as soon as I connect. The stack is the
>>>>>>>>>> same
>>>>>>>>>> as
>>>>>>>>>> the in-build crash.
>>>>>>>>>>
>>>>>>>>>> By the way this is in client-server mode the server is built with
>>>>>>>>>> sahred
>>>>>>>>>> libs off, with mpi and python and without qt.
>>>>>>>>>>
>>>>>>>>>> Burlen
>>>>>>>>>>
>>>>>>>>>> on local linux workstation, client server, server built without qt,
>>>>>>>>>> with
>>>>>>>>>> python and mpi
>>>>>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>>> 473       if (this->Internals->ActiveSessionStack.back() !=
>>>>>>>>>> session)
>>>>>>>>>>
>>>>>>>>>> (gdb) where
>>>>>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession
>>>>>>>>>> (this=0x0,
>>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0)
>>>>>>>>>> at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>>>>>> (this=0xb4880b0,
>>>>>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>>>>>> #3  0x00000000034ac735 in
>>>>>>>>>> vtkPVSessionServer::OnClientServerMessageRMI
>>>>>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>>>>>> rmiTag=55625) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>>>>>> #7  0x00000000037f8ae7 in
>>>>>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>>>>>>
>>>>>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>>>>>> auto-loaded.
>>>>>>>>>>>
>>>>>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>>>>>
>>>>>>>>>>> Utkarsh
>>>>>>>>>>>
>>>>>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring
>>>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>>>> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> Hi All,
>>>>>>>>>>>>
>>>>>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to
>>>>>>>>>>>> be
>>>>>>>>>>>> working
>>>>>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>>>>>
>>>>>>>>>>>> Burlen
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> Paraview-developers mailing list
>>>>>>>>>>>> Paraview-developers at paraview.org
>>>>>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>>>>>
>>>>>>>>>>
>>

From burlen.loring at gmail.com  Fri Sep 19 20:22:16 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Fri, 19 Sep 2014 17:22:16 -0700
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>	<CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>	<541B4903.6010002@gmail.com>	<CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>	<541B6F65.1030005@gmail.com>	<CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>	<CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>	<CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>	<541C5435.4090008@gmail.com>	<CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>	<541CC1DF.6060307@gmail.com>	<CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>
 <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
Message-ID: <541CC8B8.3030308@gmail.com>

Thanks for the explanations!

I did the incremental patch and ninja on both my shared client and 
static server builds. I can look closer into it, may be a new issue.

On 09/19/2014 05:08 PM, Utkarsh Ayachit wrote:
> Anycayse, the problem is indeed stemming from a bug that Jean and Alan
> reported earlier about plugins
> (http://www.paraview.org/Bug/view.php?id=14732).
>
> I am going to target this for 4.2.1. Cleaning this up properly may
> take a little more love and care.
>
> Utkarsh
>
> On Fri, Sep 19, 2014 at 7:58 PM, Utkarsh Ayachit
> <utkarsh.ayachit at kitware.com> wrote:
>> Burlen,
>>
>> You did rebuild both the client and the server right? You could not
>> get the same segfault anymore due to this change [1]. The code short
>> circuits when loading a shared lib. The segfault was a happening
>> because the plugin.so from the shared build was being loaded.
>>
>> LoadPlugin() takes the filename indeed. For static builds, however,
>> the filename reported for the distributed plugins is just their name
>> itself. I need to clean that up, but didn't want to make those changes
>> so close to the release. I'll refactor that piece of code for 4.3.
>> Plugin cleanups are long overdue :).
>>
>> [1] http://review.source.kitware.com/#/c/17131/1/ParaViewCore/ClientServerCore/Core/vtkPVPluginLoader.cxx
>>
>> On Fri, Sep 19, 2014 at 7:53 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>>> Thanks Utkarsh,
>>>
>>> I tested the patch but still had a segv.
>>>
>>> Re: plugins in static linked server, how are they loaded from pvbatch? Does
>>> LoadPlugin call (which takes a file name) work?
>>>
>>>
>>> On 09/19/2014 11:30 AM, Utkarsh Ayachit wrote:
>>>> Burlen,
>>>>
>>>> There are the fixes that should address both the static pvserver not
>>>> finding ".plugins" as well as the issue on your local workstation.
>>>>
>>>> http://review.source.kitware.com/#/t/4692
>>>>
>>>> For the latter, this is more of an hack, but I'll have that addressed
>>>> properly for 4.3.
>>>>
>>>> Utkarsh
>>>>
>>>>
>>>> On Fri, Sep 19, 2014 at 12:05 PM, Burlen Loring <burlen.loring at gmail.com>
>>>> wrote:
>>>>> Thanks for looking into that, and thanks for fixing the .plugins issue!
>>>>> The
>>>>> updated static plugin support in 4.2 is great!
>>>>>
>>>>>
>>>>> On 09/19/2014 08:31 AM, Utkarsh Ayachit wrote:
>>>>>> I spoke too soon, grrrr :~/.
>>>>>>
>>>>>> Utkarsh
>>>>>>
>>>>>> On Fri, Sep 19, 2014 at 11:27 AM, Utkarsh Ayachit
>>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>> Burlen,
>>>>>>>
>>>>>>> I was also able to reproduce the same segfault you're getting on your
>>>>>>> workstation. It seems to be a problem with incremental builds for the
>>>>>>> static build. A clean build (starting with an empty directory)
>>>>>>> addressed the problem for me.
>>>>>>>
>>>>>>> Utkarsh
>>>>>>>
>>>>>>> On Fri, Sep 19, 2014 at 9:49 AM, Utkarsh Ayachit
>>>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>>> Ah! I know why it's looking for .plugins in the wrong place. I'll push
>>>>>>>> a fix, Thanks!
>>>>>>>>
>>>>>>>> Utkarsh
>>>>>>>>
>>>>>>>> On Thu, Sep 18, 2014 at 7:48 PM, Burlen Loring
>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>> wrote:
>>>>>>>>> Hi Utkarsh,
>>>>>>>>>
>>>>>>>>> I updated to master on both Cray and workstation.
>>>>>>>>>
>>>>>>>>> On the Cray the .plugins file is installed in lib/paraview-4.2. The
>>>>>>>>> PV_PLUGIN_DEBUG output shows that it's expected to be in bin or lib.
>>>>>>>>> Cp'ing
>>>>>>>>> the file into the bin dir fixed that issue, I was able to load and
>>>>>>>>> use
>>>>>>>>> a
>>>>>>>>> plugin!!
>>>>>>>>>
>>>>>>>>> The issues on the Cray and my workstation seem to be unrelated. On
>>>>>>>>> the
>>>>>>>>> workstation I'm seeing the same segv after updating to master. Now it
>>>>>>>>> happens in both build and install during connection. From the stack
>>>>>>>>> it
>>>>>>>>> looks
>>>>>>>>> unrelated to plugins. Happily I don't need the static build on my
>>>>>>>>> workstation, so it's not a huge issue.
>>>>>>>>>
>>>>>>>>> Burlen
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 09/18/2014 02:36 PM, Utkarsh Ayachit wrote:
>>>>>>>>>> Burlen,
>>>>>>>>>>
>>>>>>>>>> A few things:
>>>>>>>>>> + RGBZView should not be autoloaded since 6575a800a . That happened
>>>>>>>>>> after RC1, I'm afraid.
>>>>>>>>>> + The server would still need a .plugins file (which should be
>>>>>>>>>> autogenerated). Can you verify such a file exists on the Cray? Also,
>>>>>>>>>> try running with PV_PLUGIN_DEBUG set. It should print more info. I
>>>>>>>>>> get
>>>>>>>>>> the following with my static build:
>>>>>>>>>>
>>>>>>>>>>>      PV_PLUGIN_DEBUG=1 ./bin/pvserver
>>>>>>>>>> Locate and load distributed plugin list.
>>>>>>>>>> Looking for static plugin '.plugins'
>>>>>>>>>> Loading plugin configuration xml:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> /home/kitware/Dashboards/MyTests/NightlyNext/ParaView-Debug-Static/bin/.plugins
>>>>>>>>>> Trying to locate plugin with name: SLACTools
>>>>>>>>>> Looking for static plugin 'SLACTools'
>>>>>>>>>> Found static plugin 'SLACTools'
>>>>>>>>>> --- Found SLACTools
>>>>>>>>>> ....
>>>>>>>>>> Static build. Skipping PLUGIN_PATHS.Waiting for client...
>>>>>>>>>> Connection URL: cs://blight:11111
>>>>>>>>>> Accepting connection(s): blight:11111
>>>>>>>>>>
>>>>>>>>>> I'm not sure what the RGBZView issue on your workstation is. Mind
>>>>>>>>>> updating to git/master on your workstation to see if the problem
>>>>>>>>>> persists? Actually, if it's only failing for RGBZView, it's less of
>>>>>>>>>> a
>>>>>>>>>> concern than if fails for other plugins too.
>>>>>>>>>>
>>>>>>>>>> Utkarsh
>>>>>>>>>>
>>>>>>>>>> On Thu, Sep 18, 2014 at 5:05 PM, Burlen Loring
>>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>>> wrote:
>>>>>>>>>>> Hi Utkarsh,
>>>>>>>>>>>
>>>>>>>>>>> Good, glad it's still supported! There seem to be some issues
>>>>>>>>>>> though.
>>>>>>>>>>>
>>>>>>>>>>> On the Cray the remote plugin list is empty except for
>>>>>>>>>>> vtkPVLPuginInitializer. RGBZView is automatically loaded on the
>>>>>>>>>>> client
>>>>>>>>>>> but
>>>>>>>>>>> not the server and thus as soon as the server connects the plugin
>>>>>>>>>>> manager
>>>>>>>>>>> opens to report the issue. The remote plugin list is empty.
>>>>>>>>>>> Browsing
>>>>>>>>>>> to
>>>>>>>>>>> the
>>>>>>>>>>> install and loading  the plugin doesn't work. it's a .a file so I'm
>>>>>>>>>>> not
>>>>>>>>>>> surprised. Judging from what I see on my workstation, the remote
>>>>>>>>>>> plugin
>>>>>>>>>>> list
>>>>>>>>>>> should be populated right?
>>>>>>>>>>>
>>>>>>>>>>> I've been trying to reproduce it locally on my workstation. I have
>>>>>>>>>>> different
>>>>>>>>>>> problems. The remote plugin list is fully populated and RGBZView
>>>>>>>>>>> loads on
>>>>>>>>>>> both sides. But when I try to load a plugin  via "load selected"
>>>>>>>>>>> ParaView
>>>>>>>>>>> crashes(stack below). That's running from the build. When I try the
>>>>>>>>>>> same
>>>>>>>>>>> on
>>>>>>>>>>> an install the server segv's as soon as I connect. The stack is the
>>>>>>>>>>> same
>>>>>>>>>>> as
>>>>>>>>>>> the in-build crash.
>>>>>>>>>>>
>>>>>>>>>>> By the way this is in client-server mode the server is built with
>>>>>>>>>>> sahred
>>>>>>>>>>> libs off, with mpi and python and without qt.
>>>>>>>>>>>
>>>>>>>>>>> Burlen
>>>>>>>>>>>
>>>>>>>>>>> on local linux workstation, client server, server built without qt,
>>>>>>>>>>> with
>>>>>>>>>>> python and mpi
>>>>>>>>>>> Program received signal SIGSEGV, Segmentation fault.
>>>>>>>>>>> 0x00000000037a44ff in vtkProcessModule::PopActiveSession (this=0x0,
>>>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>>>> 473       if (this->Internals->ActiveSessionStack.back() !=
>>>>>>>>>>> session)
>>>>>>>>>>>
>>>>>>>>>>> (gdb) where
>>>>>>>>>>> #0  0x00000000037a44ff in vtkProcessModule::PopActiveSession
>>>>>>>>>>> (this=0x0,
>>>>>>>>>>> session=0xb4880b0) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkProcessModule.cxx:473
>>>>>>>>>>> #1  0x00000000037f6d94 in vtkSession::DeActivate (this=0xb4880b0)
>>>>>>>>>>> at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkSession.cxx:39
>>>>>>>>>>> #2  0x00000000034a01c5 in vtkPVSessionBase::ExecuteStream
>>>>>>>>>>> (this=0xb4880b0,
>>>>>>>>>>> location=21, stream=..., ignore_errors=false) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionBase.cxx:170
>>>>>>>>>>> #3  0x00000000034ac735 in
>>>>>>>>>>> vtkPVSessionServer::OnClientServerMessageRMI
>>>>>>>>>>> (this=0xb4880b0, message=0xbc82030, message_length=16) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:552
>>>>>>>>>>> #4  0x00000000034aa8f8 in (anonymous namespace)::RMICallback
>>>>>>>>>>> (localArg=0xb4880b0, remoteArg=0xbc82030, remoteArgLength=16) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ServerImplementation/Core/vtkPVSessionServer.cxx:56
>>>>>>>>>>> #5  0x000000000489d3c4 in vtkMultiProcessController::ProcessRMI
>>>>>>>>>>> (this=0xbc79460, remoteProcessId=1, arg=0xbc82030, argLength=16,
>>>>>>>>>>> rmiTag=55625) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:774
>>>>>>>>>>> #6  0x000000000489cf98 in vtkMultiProcessController::ProcessRMIs
>>>>>>>>>>> (this=0xbc79460, reportErrors=0, dont_loop=1) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/VTK/Parallel/Core/vtkMultiProcessController.cxx:720
>>>>>>>>>>> #7  0x00000000037f8ae7 in
>>>>>>>>>>> vtkTCPNetworkAccessManager::ProcessEventsInternal
>>>>>>>>>>> (this=0xb461a70, timeout_msecs=0, do_processing=true) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:268
>>>>>>>>>>> #8  0x00000000037f868a in vtkTCPNetworkAccessManager::ProcessEvents
>>>>>>>>>>> (this=0xb461a70, timeout_msecs=0) at
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> /work/ParaView/ParaView/ParaViewCore/ClientServerCore/Core/vtkTCPNetworkAccessManager.cxx:165
>>>>>>>>>>> #9  0x0000000001c6d896 in RealMain (argc=1, argv=0x7fff33cd48f8,
>>>>>>>>>>> type=vtkProcessModule::PROCESS_SERVER) at
>>>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver_common.h:91
>>>>>>>>>>> #10 0x0000000001c6d94f in main (argc=1, argv=0x7fff33cd48f8) at
>>>>>>>>>>> /work/ParaView/ParaView/CommandLineExecutables/pvserver.cxx:27
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On 09/18/2014 05:37 AM, Utkarsh Ayachit wrote:
>>>>>>>>>>>> Yes, you need to load them manually, however. They are no longer
>>>>>>>>>>>> auto-loaded.
>>>>>>>>>>>>
>>>>>>>>>>>> http://www.paraview.org/Bug/view.php?id=14542
>>>>>>>>>>>>
>>>>>>>>>>>> Utkarsh
>>>>>>>>>>>>
>>>>>>>>>>>> On Wed, Sep 17, 2014 at 7:35 PM, Burlen Loring
>>>>>>>>>>>> <burlen.loring at gmail.com>
>>>>>>>>>>>> wrote:
>>>>>>>>>>>>> Hi All,
>>>>>>>>>>>>>
>>>>>>>>>>>>> I'm having trouble with our NERSC install, plugins don't seem to
>>>>>>>>>>>>> be
>>>>>>>>>>>>> working
>>>>>>>>>>>>> at all.  Should plugins work with BUILD_SHARED_LIBS=OFF in 4.2?
>>>>>>>>>>>>>
>>>>>>>>>>>>> Burlen
>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>> Paraview-developers mailing list
>>>>>>>>>>>>> Paraview-developers at paraview.org
>>>>>>>>>>>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>>>>>>>>>>


From utkarsh.ayachit at kitware.com  Sat Sep 20 09:15:44 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Sat, 20 Sep 2014 09:15:44 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjCgZsW9qUOF5HB2ypenwbkXp=uHKB2Zu3y-3zDReWippw@mail.gmail.com>

---------------------------------------------
Topics merged into master:
        chart_catalyst_fixes
        fix_static_plugins
        xdisplay_test_workaround

From biddisco at cscs.ch  Mon Sep 22 12:32:07 2014
From: biddisco at cscs.ch (Biddiscombe, John A.)
Date: Mon, 22 Sep 2014 16:32:07 +0000
Subject: [Paraview-developers] new error after wiping build and git pull
Message-ID: <50320452A334BD42A5EC72BAD214509918ED35B7@MBX110.d.ethz.ch>

For some reason, none of the paraview exes can be built - if I manually create the dirs. first then it's ok, but doing it for every exe on this mac is taking me time.
Did anyone change something that might cause this?

did my bin permissions get munged when I wiped my build dir? Is there an easy way to tell

(Xcode did upgrade itself a few days ago - I'm not using it, but google tells me that this might be a cause of trouble)

thanks

JB


Linking CXX executable ../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests
ld: can't open output file for writing: ../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests, errno=2 for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests] Error 1
make[2]: Target `ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerManagerRenderingCxxTests.dir/build' not remade because of errors.
make[1]: *** [ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerManagerRenderingCxxTests.dir/all] Error 2
Linking CXX executable ../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator
ld: can't open output file for writing: ../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator, errno=2 for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator] Error 1
make[2]: Target `VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommunicator.dir/build' not remade because of errors.
make[1]: *** [VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommunicator.dir/all] Error 2
Linking CXX executable ../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests
ld: can't open output file for writing: ../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests, errno=2 for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests] Error 1
make[2]: Target `VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/build' not remade because of errors.
make[1]: *** [VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/all] Error 2
Linking CXX executable ../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI
ld: can't open output file for writing: ../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI, errno=2 for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI] Error 1
make[2]: Target `VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/build' not remade because of errors.
make[1]: *** [VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/all] Error 2
Linking CXX executable ../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkRenderingAnnotationCxxTests
ld: can't open output file for writing: ../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkRenderingAnnotationCxxTests, errno=2 for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140922/fa51ca49/attachment.html>

From utkarsh.ayachit at kitware.com  Mon Sep 22 12:35:52 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 22 Sep 2014 12:35:52 -0400
Subject: [Paraview-developers] new error after wiping build and git pull
In-Reply-To: <50320452A334BD42A5EC72BAD214509918ED35B7@MBX110.d.ethz.ch>
References: <50320452A334BD42A5EC72BAD214509918ED35B7@MBX110.d.ethz.ch>
Message-ID: <CADHSJjBdea9zQ=Du=3=tjZnD8gV91PO0QaJre559eVZfFCnbCQ@mail.gmail.com>

That's weird. vtkPVServerManagerRenderingCxxTests etc. are not meant
to be apps, but command line tools. Ben, can you follow up with John
to figure out what could be going on please? Thanks.

Utkarsh

On Mon, Sep 22, 2014 at 12:32 PM, Biddiscombe, John A. <biddisco at cscs.ch> wrote:
> For some reason, none of the paraview exes can be built - if I manually
> create the dirs. first then it?s ok, but doing it for every exe on this mac
> is taking me time.
>
> Did anyone change something that might cause this?
>
>
>
> did my bin permissions get munged when I wiped my build dir? Is there an
> easy way to tell
>
>
>
> (Xcode did upgrade itself a few days ago - I?m not using it, but google
> tells me that this might be a cause of trouble)
>
>
>
> thanks
>
>
>
> JB
>
>
>
>
>
> Linking CXX executable
> ../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests
>
> ld: can't open output file for writing:
> ../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests,
> errno=2 for architecture x86_64
>
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
>
> make[2]: ***
> [bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerManagerRenderingCxxTests]
> Error 1
>
> make[2]: Target
> `ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerManagerRenderingCxxTests.dir/build'
> not remade because of errors.
>
> make[1]: ***
> [ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerManagerRenderingCxxTests.dir/all]
> Error 2
>
> Linking CXX executable
> ../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator
>
> ld: can't open output file for writing:
> ../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator,
> errno=2 for architecture x86_64
>
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
>
> make[2]: ***
> [bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParallelCore-TestSocketCommunicator]
> Error 1
>
> make[2]: Target
> `VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommunicator.dir/build'
> not remade because of errors.
>
> make[1]: ***
> [VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommunicator.dir/all]
> Error 2
>
> Linking CXX executable
> ../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests
>
> ld: can't open output file for writing:
> ../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests,
> errno=2 for architecture x86_64
>
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
>
> make[2]: ***
> [bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests]
> Error 1
>
> make[2]: Target
> `VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/build'
> not remade because of errors.
>
> make[1]: ***
> [VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/all]
> Error 2
>
> Linking CXX executable
> ../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI
>
> ld: can't open output file for writing:
> ../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI,
> errno=2 for architecture x86_64
>
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
>
> make[2]: ***
> [bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests-MPI]
> Error 1
>
> make[2]: Target
> `VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/build'
> not remade because of errors.
>
> make[1]: ***
> [VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/all]
> Error 2
>
> Linking CXX executable
> ../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkRenderingAnnotationCxxTests
>
> ld: can't open output file for writing:
> ../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkRenderingAnnotationCxxTests,
> errno=2 for architecture x86_64
>
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>

From biddisco at cscs.ch  Mon Sep 22 18:39:30 2014
From: biddisco at cscs.ch (Biddiscombe, John A.)
Date: Mon, 22 Sep 2014 22:39:30 +0000
Subject: [Paraview-developers] new error after wiping build and git pull
In-Reply-To: <CADHSJjBdea9zQ=Du=3=tjZnD8gV91PO0QaJre559eVZfFCnbCQ@mail.gmail.com>
Message-ID: <D0467168.68D44%biddisco@cscs.ch>

The problem went away on its own when I wiped the build tree again. No
idea what was going on other than cmake can be very temperamental when you
have old cache values - The first wipe I was all the lib/bin stuff but I
left the cache file.Seoncd I wiped everything.

JB

On 22/09/14 18:35, "Utkarsh Ayachit" <utkarsh.ayachit at kitware.com> wrote:

>That's weird. vtkPVServerManagerRenderingCxxTests etc. are not meant
>to be apps, but command line tools. Ben, can you follow up with John
>to figure out what could be going on please? Thanks.
>
>Utkarsh
>
>On Mon, Sep 22, 2014 at 12:32 PM, Biddiscombe, John A. <biddisco at cscs.ch>
>wrote:
>> For some reason, none of the paraview exes can be built - if I manually
>> create the dirs. first then it?s ok, but doing it for every exe on this
>>mac
>> is taking me time.
>>
>> Did anyone change something that might cause this?
>>
>>
>>
>> did my bin permissions get munged when I wiped my build dir? Is there an
>> easy way to tell
>>
>>
>>
>> (Xcode did upgrade itself a few days ago - I?m not using it, but google
>> tells me that this might be a cause of trouble)
>>
>>
>>
>> thanks
>>
>>
>>
>> JB
>>
>>
>>
>>
>>
>> Linking CXX executable
>> 
>>../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS
>>/vtkPVServerManagerRenderingCxxTests
>>
>> ld: can't open output file for writing:
>> 
>>../../../../../bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS
>>/vtkPVServerManagerRenderingCxxTests,
>> errno=2 for architecture x86_64
>>
>> clang: error: linker command failed with exit code 1 (use -v to see
>> invocation)
>>
>> make[2]: ***
>> 
>>[bin/vtkPVServerManagerRenderingCxxTests.app/Contents/MacOS/vtkPVServerMa
>>nagerRenderingCxxTests]
>> Error 1
>>
>> make[2]: Target
>> 
>>`ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerM
>>anagerRenderingCxxTests.dir/build'
>> not remade because of errors.
>>
>> make[1]: ***
>> 
>>[ParaViewCore/ServerManager/Rendering/Testing/Cxx/CMakeFiles/vtkPVServerM
>>anagerRenderingCxxTests.dir/all]
>> Error 2
>>
>> Linking CXX executable
>> 
>>../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/Ma
>>cOS/vtkParallelCore-TestSocketCommunicator
>>
>> ld: can't open output file for writing:
>> 
>>../../../../../bin/vtkParallelCore-TestSocketCommunicator.app/Contents/Ma
>>cOS/vtkParallelCore-TestSocketCommunicator,
>> errno=2 for architecture x86_64
>>
>> clang: error: linker command failed with exit code 1 (use -v to see
>> invocation)
>>
>> make[2]: ***
>> 
>>[bin/vtkParallelCore-TestSocketCommunicator.app/Contents/MacOS/vtkParalle
>>lCore-TestSocketCommunicator]
>> Error 1
>>
>> make[2]: Target
>> 
>>`VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommu
>>nicator.dir/build'
>> not remade because of errors.
>>
>> make[1]: ***
>> 
>>[VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCore-TestSocketCommu
>>nicator.dir/all]
>> Error 2
>>
>> Linking CXX executable
>> 
>>../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallel
>>CoreCxxTests
>>
>> ld: can't open output file for writing:
>> 
>>../../../../../bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallel
>>CoreCxxTests,
>> errno=2 for architecture x86_64
>>
>> clang: error: linker command failed with exit code 1 (use -v to see
>> invocation)
>>
>> make[2]: ***
>> [bin/vtkParallelCoreCxxTests.app/Contents/MacOS/vtkParallelCoreCxxTests]
>> Error 1
>>
>> make[2]: Target
>> 
>>`VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/bui
>>ld'
>> not remade because of errors.
>>
>> make[1]: ***
>> 
>>[VTK/Parallel/Core/Testing/Cxx/CMakeFiles/vtkParallelCoreCxxTests.dir/all
>>]
>> Error 2
>>
>> Linking CXX executable
>> 
>>../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParal
>>lelMPICxxTests-MPI
>>
>> ld: can't open output file for writing:
>> 
>>../../../../../bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParal
>>lelMPICxxTests-MPI,
>> errno=2 for architecture x86_64
>>
>> clang: error: linker command failed with exit code 1 (use -v to see
>> invocation)
>>
>> make[2]: ***
>> 
>>[bin/vtkParallelMPICxxTests-MPI.app/Contents/MacOS/vtkParallelMPICxxTests
>>-MPI]
>> Error 1
>>
>> make[2]: Target
>> 
>>`VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/b
>>uild'
>> not remade because of errors.
>>
>> make[1]: ***
>> 
>>[VTK/Parallel/MPI/Testing/Cxx/CMakeFiles/vtkParallelMPICxxTests-MPI.dir/a
>>ll]
>> Error 2
>>
>> Linking CXX executable
>> 
>>../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkR
>>enderingAnnotationCxxTests
>>
>> ld: can't open output file for writing:
>> 
>>../../../../../bin/vtkRenderingAnnotationCxxTests.app/Contents/MacOS/vtkR
>>enderingAnnotationCxxTests,
>> errno=2 for architecture x86_64
>>
>> clang: error: linker command failed with exit code 1 (use -v to see
>> invocation)
>>
>>
>> _______________________________________________
>> Paraview-developers mailing list
>> Paraview-developers at paraview.org
>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>


From utkarsh.ayachit at kitware.com  Tue Sep 23 09:17:46 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 23 Sep 2014 09:17:46 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjDPjWxHuMsNZcTL_zjqNR13jKL8u3fejnBRmyeTZohbcg@mail.gmail.com>

Topics merged into master:
        fix-mpi-tag-ub
        python_api_cleanup
        python_syntax_highlighting_keep_scroll_position
        xdisplay_test_workaround

From utkarsh.ayachit at kitware.com  Tue Sep 23 10:34:51 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 23 Sep 2014 10:34:51 -0400
Subject: [Paraview-developers] ANN: ParaView 4.2.0 Tagged!
Message-ID: <CADHSJjA4Mb9LMbreaFxBvrjF0x1Krkk41dNsQP+whqMxf19Cqg@mail.gmail.com>

Good morning folks!

ParaView 4.2.0 has been tagged!

git tag name: v4.2.0

Binaries will be available within the next few days.

Utkarsh

From sankhesh.jhaveri at kitware.com  Tue Sep 23 12:31:08 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Tue, 23 Sep 2014 09:31:08 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Monday, September 22 2014
Message-ID: <20140923163104.2AA217FFE3@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   -------------------- REPORT --------------------
   0 FAILURES

   Dashboard status for Nightly (next)

   Cdash returned nothing useful as of this report.

   Issues with submitters
   The following expected submitters did not submit:
   kamino.kitware , Mac10.7.5-gcc-nightlymaster-release
   kamino.kitware , Mac10.7.5-clang-nightlymaster-release

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140923/153d2655/attachment.html>

From burlen.loring at gmail.com  Tue Sep 23 12:35:58 2014
From: burlen.loring at gmail.com (Burlen Loring)
Date: Tue, 23 Sep 2014 09:35:58 -0700
Subject: [Paraview-developers] ANN: ParaView 4.2.0 Tagged!
In-Reply-To: <CADHSJjCao9icSPDw2-PNJxGcmXwje6q36a_nBT8C0OrM4DBGVA@mail.gmail.com>
References: <CADHSJjA4Mb9LMbreaFxBvrjF0x1Krkk41dNsQP+whqMxf19Cqg@mail.gmail.com>	<54219FC2.6050706@gmail.com>
 <CADHSJjCao9icSPDw2-PNJxGcmXwje6q36a_nBT8C0OrM4DBGVA@mail.gmail.com>
Message-ID: <5421A16E.5080302@gmail.com>

Thanks for clarifying! Forwarding on to to list in case others have 
similar question.


On 09/23/2014 09:33 AM, Utkarsh Ayachit wrote:
> Burlen,
>
> "Delete All" has become "Reset Session". It does more than what
> "Delete All" did. It also deletes LUTs and other helper proxies.
>
> The image compression settings are available under "Render VIew" tab.
> They are "advanced" by default, so either search for "compression" or
> use the "gear" icon to see them.
>
> Utkarsh
>
>
>
>
>
> On Tue, Sep 23, 2014 at 12:28 PM, Burlen Loring <burlen.loring at gmail.com> wrote:
>> Congrats!
>>
>> Delete all function is missing from the edit menu. Has it moved?
>> Is image compression still available for remote rendering? Where are those
>> settings found?
>>
>>
>> On 09/23/2014 07:34 AM, Utkarsh Ayachit wrote:
>>> Good morning folks!
>>>
>>> ParaView 4.2.0 has been tagged!
>>>
>>> git tag name: v4.2.0
>>>
>>> Binaries will be available within the next few days.
>>>
>>> Utkarsh
>>> _______________________________________________
>>> Paraview-developers mailing list
>>> Paraview-developers at paraview.org
>>> http://public.kitware.com/mailman/listinfo/paraview-developers
>>


From sankhesh.jhaveri at kitware.com  Thu Sep 25 12:31:05 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Thu, 25 Sep 2014 09:31:05 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Wednesday, September 24 2014
Message-ID: <20140925163104.1BD227FF03@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]Mac10.7.5-clang-nightlymaster-release 3
   [2]fedora-x64-icc-nightlymaster-debug 1
   -------------------- REPORT --------------------
   4 FAILURES
   [3]pvcrs.CheckableHeader , 1 ,
   ['Mac10.7.5-clang-nightlymaster-release']
   [4]pv.AnalyzeReaderWriterPlugin , 1 ,
   ['Mac10.7.5-clang-nightlymaster-release']
   [5]pvcrs.TensorGlyph , 1 , ['fedora-x64-icc-nightlymaster-debug']
   [6]pv.SimpleInteraction , 1 , ['Mac10.7.5-clang-nightlymaster-release']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [7]ubuntu-x64-coverage 1
   -------------------- REPORT --------------------
   1 FAILURES
   [8]pvcs.ImageVolumeRendering , 1 , ['ubuntu-x64-coverage']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-25&filtercount=1&field1=buildname/string&compare1=61&value1=Mac10.7.5-clang-nightlymaster-release
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-25&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-25&name=pvcrs.CheckableHeader
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-25&name=pv.AnalyzeReaderWriterPlugin
   5. http://open.cdash.org/testSummary.php?project=9&date=2014-09-25&name=pvcrs.TensorGlyph
   6. http://open.cdash.org/testSummary.php?project=9&date=2014-09-25&name=pv.SimpleInteraction
   7. http://open.cdash.org/index.php?project=ParaView&date=2014-09-25&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
   8. http://open.cdash.org/testSummary.php?project=9&date=2014-09-25&name=pvcs.ImageVolumeRendering

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140925/29c72988/attachment.html>

From jfavre at cscs.ch  Fri Sep 26 09:20:29 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Fri, 26 Sep 2014 13:20:29 +0000
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins
	in	4.2.0 rc1
In-Reply-To: <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
 <541CC1DF.6060307@gmail.com>
 <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>,
 <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F682814708A@MBX21.d.ethz.ch>


I am trying to test a private plugin with static builds.

For ease of use (I thought), I placed my plugin in the same place at the other PV plugins, i.e. in /path-to-source-code/Plugins, and I followed the same syntax as H5PartReader or GMVReader

I have both a shared_libs and a static builds in seperate places. Both of them build and place my plugin in the $CMAKE_INSTALL_PREFIX/lib/paraview-4.2 directories/libprivate.so and
$CMAKE_INSTALL_PREFIX/lib/paraview-4.2 directories/libprivate.a in the other install place for the static build.

I can load and use my plugin with the shared_libs build. I see it in the GUI and can load it. I can also load it in python via LoadPlugin("...",  ns=globals()) for pvbatch use.

The questions related to the static build.

1) I cannot see my plugin in the GUI, although it is in the same place as H5PartReader and GMVReader for example.
2) I can't figure out how to load any plugins (I also need libEyeDomeLightingView.a ) via LoadPlugin()

Any tip?


From utkarsh.ayachit at kitware.com  Fri Sep 26 10:11:53 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Fri, 26 Sep 2014 10:11:53 -0400
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
	4.2.0 rc1
In-Reply-To: <0EB9B6375711A04B820E6B6F5CCA9F682814708A@MBX21.d.ethz.ch>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
 <541CC1DF.6060307@gmail.com>
 <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>
 <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
 <0EB9B6375711A04B820E6B6F5CCA9F682814708A@MBX21.d.ethz.ch>
Message-ID: <CADHSJjCNkysiKCHTiaU2ZQZfiusAhM7kJnE1VKP0nrXXq3kW5A@mail.gmail.com>

1. You cannot dynamically load static plugins. Static plugins need to
be compiled into the ParaView executable. When building ParaView you
can set the PARAVIEW_EXTERNAL_PLUGIN_DIRS cmake variable to point to
external plugin source directories Note, each plugins must include a
plugin.cmake, similar to the Plugins under ParaView source.

2. To load a "compiled in" plugin, use LoadDistributedPlugin() [1]. It
also works with shared builds.

Utkarsh

[1] http://www.paraview.org/ParaView3/Doc/Nightly/www/py-doc/paraview.simple.html#paraview.simple.LoadDistributedPlugin

On Fri, Sep 26, 2014 at 9:20 AM, Favre  Jean <jfavre at cscs.ch> wrote:
>
> I am trying to test a private plugin with static builds.
>
> For ease of use (I thought), I placed my plugin in the same place at the other PV plugins, i.e. in /path-to-source-code/Plugins, and I followed the same syntax as H5PartReader or GMVReader
>
> I have both a shared_libs and a static builds in seperate places. Both of them build and place my plugin in the $CMAKE_INSTALL_PREFIX/lib/paraview-4.2 directories/libprivate.so and
> $CMAKE_INSTALL_PREFIX/lib/paraview-4.2 directories/libprivate.a in the other install place for the static build.
>
> I can load and use my plugin with the shared_libs build. I see it in the GUI and can load it. I can also load it in python via LoadPlugin("...",  ns=globals()) for pvbatch use.
>
> The questions related to the static build.
>
> 1) I cannot see my plugin in the GUI, although it is in the same place as H5PartReader and GMVReader for example.
> 2) I can't figure out how to load any plugins (I also need libEyeDomeLightingView.a ) via LoadPlugin()
>
> Any tip?
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From jfavre at cscs.ch  Fri Sep 26 12:19:49 2014
From: jfavre at cscs.ch (Favre  Jean)
Date: Fri, 26 Sep 2014 16:19:49 +0000
Subject: [Paraview-developers] BUILD_SHARED_LIBS=OFF and plugins in
 4.2.0 rc1
In-Reply-To: <CADHSJjCNkysiKCHTiaU2ZQZfiusAhM7kJnE1VKP0nrXXq3kW5A@mail.gmail.com>
References: <541A1AD5.1070508@gmail.com>
 <CADHSJjBc1S1n01Qpge1QSZAB3+Lx_dNuyw9o6L6ZAf0BxZLbsA@mail.gmail.com>
 <541B4903.6010002@gmail.com>
 <CADHSJjCdyz3vvcG8-E6L3NO+c0wCtQDTO5+-VReU2amxs4cL+A@mail.gmail.com>
 <541B6F65.1030005@gmail.com>
 <CADHSJjDZ4vZqyTaj3AVi1s1xvqVp7P=x8GLgjXtgb_W1SXvXrw@mail.gmail.com>
 <CADHSJjCaGuZ+s_Bf-ZxRaLBHFzwtBX7vATcxgHoZDtkk4_OtGQ@mail.gmail.com>
 <CADHSJjCDe+404uUB+DxHZnCzr8_Xhp5p_ehFTG-vfnxpV8rhTA@mail.gmail.com>
 <541C5435.4090008@gmail.com>
 <CADHSJjDj51UyJ=V2=O+PKcO0UFoNBHUGgWYPKjHNF1gh7u7fhQ@mail.gmail.com>
 <541CC1DF.6060307@gmail.com>
 <CADHSJjDXEzGaM78Gmexifz5K7kvH70UKzk4Nh1XmiQgd+pSoDQ@mail.gmail.com>
 <CADHSJjAgxzQO9vheudabZbag=QfY0fxGHpLiK0hE7+_LFx1tiQ@mail.gmail.com>
 <0EB9B6375711A04B820E6B6F5CCA9F682814708A@MBX21.d.ethz.ch>,
 <CADHSJjCNkysiKCHTiaU2ZQZfiusAhM7kJnE1VKP0nrXXq3kW5A@mail.gmail.com>
Message-ID: <0EB9B6375711A04B820E6B6F5CCA9F682814A1A6@MBX21.d.ethz.ch>

Utkarsh

of course I understand that I cannot dynamically load static plugins. 

I have now switched to using PARAVIEW_EXTERNAL_PLUGIN_DIRS and LoadDistributedPlugin(), and I can confirm that I can use a static build and my private plugins with python (batch-mode) or connecting from a client in interactive mode. My earlier problem was most likely due to a name mismatch.

If it matters, this is on the Cray XC30, and it is the just-released version 4.2

thanks again to Utkarsh for putting me in the right direction.

Jean/CSCS


From sankhesh.jhaveri at kitware.com  Fri Sep 26 12:31:06 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Fri, 26 Sep 2014 09:31:06 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Thursday, September 25 2014
Message-ID: <20140926163104.5B8E67FF26@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]fedora-x64-icc-nightlymaster-debug 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pvcs.RenderNan , 1 , ['fedora-x64-icc-nightlymaster-debug']

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [3]ubuntu-x64-nightlynext 1
   -------------------- REPORT --------------------
   1 FAILURES
   [4]pvcs-collab.VariableSelector , 1 , ['ubuntu-x64-nightlynext']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-26&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-26&name=pvcs.RenderNan
   3. http://open.cdash.org/index.php?project=ParaView&date=2014-09-26&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-nightlynext
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-26&name=pvcs-collab.VariableSelector

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140926/bbdc69f2/attachment.html>

From bwvb at xs4all.nl  Fri Sep 26 13:05:03 2014
From: bwvb at xs4all.nl (B.W.H. van Beest)
Date: Fri, 26 Sep 2014 19:05:03 +0200
Subject: [Paraview-developers] How to find the selected object?
Message-ID: <54259CBF.4050806@xs4all.nl>

Hello,

I'm trying to build my own plugin (C++) in Paraview.
I wonder how one can programattically find which object (point, box,
sphere, ... etc) is selected/clicked by a user.

Any hint is much appreciated.

Regards,
Bertwim



From redkite at virgin.net  Fri Sep 26 13:32:34 2014
From: redkite at virgin.net (redkite)
Date: Fri, 26 Sep 2014 18:32:34 +0100
Subject: [Paraview-developers] Extracted selection moves with choice of
 extracted block (Paraview 4.1)
Message-ID: <5425A332.8050806@virgin.net>


Hi all,

I'm writing a plugin which attaches boundary conditions to voxel data 
for FE analysis. We wish to represent our data as a multi-block in order 
to allow points/conditions to be defined on the surface of each block.

Let me know if there's another way -- but using the 'Select Points On' 
option with 'Extract Selection' button to define a set of points seems 
to have some strange behaviour with respect to multi-blocks. In the 
Paraview interface, under the Information tab, selecting and extracting 
points in this way generates a multi-block dataset but my plugin filter 
(which operates on the extracted object) actually receives 
vtkUnstructuredGrid data.

Stranger still (it seems to me), if the point selection is made across 2 
blocks, instead of receiving a multi-block with 2 blocks, my 
'add-boundary-condition' filter is called twice with different chunks of 
of vtkUnstructuredGrid data. Since these arrive in no particular order, 
it is difficult to know how best to reassemble them. A static member 
might do it, but would also get mixed with other conditions. Clumsy, but 
it might be made to work.

I then thought using the extract block filter might make things easier, 
but the extracted selection of points seems to move around with the 
choice of block. Here's an example to illustrate:-

1) Load the can.ex2 example and apply the 'Extract Block' filter.
2) Navigate to the 'Block indices' menu and switch on 'Unnamed block ID 
1' so you can see the semi-circular can object.
3) Choose 'vtkCompositeIndex' in the 'Solid Color' drop-down.
4) Click the 'Select Points On' selection button and draw a rectangle of 
points on the can.
5) Click the 'Extract Selection' button.
6) Go back to the 'Block indices' menu, switch off 'Unnamed block ID 1' 
and turn on 'Unnamed block ID 2'
7) Adjust the opacity if the selection is hidden, but you should the 
selection move between blocks as you move from ID 1 to 2 and back (if 
both are on the selection is hidden). Making another selection & 
extraction on the rectangular block object will move to the can, and so on.

Maybe I need to write another extraction procedure which acts as a 
source ie. cannot then be altered? Or has this perhaps been fixed in the 
next release?

regards,
Richard

ps I see this behaviour both on Ubuntu 12.10 and Windows 7.



From frank.tt at einseinself.org  Sun Sep 28 22:28:34 2014
From: frank.tt at einseinself.org (Frank TT)
Date: Mon, 29 Sep 2014 04:28:34 +0200
Subject: [Paraview-developers] Filter Plugin
Message-ID: <1411957714.20122.15.camel@Judika>

Hi,


I want to write my own filter plugin.

I'm i little confused, how to do. I read already the 'how to'[1], had a
look at the example plugins and the PV plugins and searched the ML
archive.

I'm trying to implement a algorithm that works on loaded volume data and
creates a suface data. What I wrote below [2].

Now I'm not sure how to access the marked data set from the pipeline
browser, manipulate the data and then write the new output (with load
and display).

Can anyone give me some hints? Or is there a plugin I have to see?


Greetings
Frank


### ### ### ### ### ### ###
//marchingCubes.h

#ifndef __marchingCubes_h
#define __marchingCubes_h

//#include <vtkSmartPointer.h>
#include <vtkImageData.h>


#include "vtkElevationFilter.h"
//#include "vtkImageAlgorithm.h"
//#include "vtkAlgorithm.h"

class VTK_EXPORT marchingCubes : public vtkElevationFilter
{
public:
  static marchingCubes* New();
  vtkTypeMacro(marchingCubes, vtkElevationFilter);
  void PrintSelf(ostream& os, vtkIndent indent);

protected:
  marchingCubes();
  ~marchingCubes();

  void setIsoValue(double isoValue);
  virtual void Execute();

private:
  marchingCubes(const marchingCubes&);  // Not implemented.
  void operator=(const marchingCubes&);  // Not implemented.
  double isoValue;
};

#endif

[1] www.paraview.org/Wiki/ParaView/Plugin_HowTo

[2]
### ### ### ### ### ### ###
//marchingCubes.h

#ifndef __marchingCubes_h
#define __marchingCubes_h

#include "vtkElevationFilter.h"
//#include "vtkImageAlgorithm.h"
//#include "vtkAlgorithm.h"

class VTK_EXPORT marchingCubes : public vtkElevationFilter
{
public:
  static marchingCubes* New();
  vtkTypeMacro(marchingCubes, vtkElevationFilter);
  void PrintSelf(ostream& os, vtkIndent indent);

  void setIsoValue(double pFromGUI);

protected:
  marchingCubes();
  ~marchingCubes();

private:
  marchingCubes(const marchingCubes&);  // Not implemented.
  void operator=(const marchingCubes&);  // Not implemented.
};

#endif

### ### ### ### ### ### ###
//marchingCubes.cxx

#include "marchingCubes.h"

#include "vtkObjectFactory.h"
#include <iostream>

vtkStandardNewMacro(marchingCubes);

//----------------------------------------------------------------------------
marchingCubes::marchingCubes()
{
}

//----------------------------------------------------------------------------
marchingCubes::~marchingCubes()
{
}

//----------------------------------------------------------------------------
void marchingCubes::PrintSelf(ostream& os, vtkIndent indent)
{
  this->Superclass::PrintSelf(os,indent);
}

//----------------------------------------------------------------------------
// From server manager xml file. Does no output to console.
void marchingCubes::setIsoValue(double isoValue)
{
    std:cout<<"iso value:" << (char)isoValue;
}



From utkarsh.ayachit at kitware.com  Mon Sep 29 08:37:22 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 29 Sep 2014 08:37:22 -0400
Subject: [Paraview-developers] Filter Plugin
In-Reply-To: <1411957714.20122.15.camel@Judika>
References: <1411957714.20122.15.camel@Judika>
Message-ID: <CADHSJjApAYveFUA9Jx5yC1+W-=_YJ3PUrbVzbJbWfJcmfiswnw@mail.gmail.com>

Frank,

ParaView is based on VTK, and relies on VTK to provide the execution
model and data model. Thus, when writing you a new data processing
filter, you're basically going to wrote a VTK filter. The ParaView
plugin infrastructure comes into play to make such a VTK filter
available/accessible in ParaView.

I'd suggesting starting with a existing VTK filter that closely
matches what you're doing e.g. vtkMarchingCubes [1].

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

On Sun, Sep 28, 2014 at 10:28 PM, Frank TT <frank.tt at einseinself.org> wrote:
> Hi,
>
>
> I want to write my own filter plugin.
>
> I'm i little confused, how to do. I read already the 'how to'[1], had a
> look at the example plugins and the PV plugins and searched the ML
> archive.
>
> I'm trying to implement a algorithm that works on loaded volume data and
> creates a suface data. What I wrote below [2].
>
> Now I'm not sure how to access the marked data set from the pipeline
> browser, manipulate the data and then write the new output (with load
> and display).
>
> Can anyone give me some hints? Or is there a plugin I have to see?
>
>
> Greetings
> Frank
>
>
> ### ### ### ### ### ### ###
> //marchingCubes.h
>
> #ifndef __marchingCubes_h
> #define __marchingCubes_h
>
> //#include <vtkSmartPointer.h>
> #include <vtkImageData.h>
>
>
> #include "vtkElevationFilter.h"
> //#include "vtkImageAlgorithm.h"
> //#include "vtkAlgorithm.h"
>
> class VTK_EXPORT marchingCubes : public vtkElevationFilter
> {
> public:
>   static marchingCubes* New();
>   vtkTypeMacro(marchingCubes, vtkElevationFilter);
>   void PrintSelf(ostream& os, vtkIndent indent);
>
> protected:
>   marchingCubes();
>   ~marchingCubes();
>
>   void setIsoValue(double isoValue);
>   virtual void Execute();
>
> private:
>   marchingCubes(const marchingCubes&);  // Not implemented.
>   void operator=(const marchingCubes&);  // Not implemented.
>   double isoValue;
> };
>
> #endif
>
> [1] www.paraview.org/Wiki/ParaView/Plugin_HowTo
>
> [2]
> ### ### ### ### ### ### ###
> //marchingCubes.h
>
> #ifndef __marchingCubes_h
> #define __marchingCubes_h
>
> #include "vtkElevationFilter.h"
> //#include "vtkImageAlgorithm.h"
> //#include "vtkAlgorithm.h"
>
> class VTK_EXPORT marchingCubes : public vtkElevationFilter
> {
> public:
>   static marchingCubes* New();
>   vtkTypeMacro(marchingCubes, vtkElevationFilter);
>   void PrintSelf(ostream& os, vtkIndent indent);
>
>   void setIsoValue(double pFromGUI);
>
> protected:
>   marchingCubes();
>   ~marchingCubes();
>
> private:
>   marchingCubes(const marchingCubes&);  // Not implemented.
>   void operator=(const marchingCubes&);  // Not implemented.
> };
>
> #endif
>
> ### ### ### ### ### ### ###
> //marchingCubes.cxx
>
> #include "marchingCubes.h"
>
> #include "vtkObjectFactory.h"
> #include <iostream>
>
> vtkStandardNewMacro(marchingCubes);
>
> //----------------------------------------------------------------------------
> marchingCubes::marchingCubes()
> {
> }
>
> //----------------------------------------------------------------------------
> marchingCubes::~marchingCubes()
> {
> }
>
> //----------------------------------------------------------------------------
> void marchingCubes::PrintSelf(ostream& os, vtkIndent indent)
> {
>   this->Superclass::PrintSelf(os,indent);
> }
>
> //----------------------------------------------------------------------------
> // From server manager xml file. Does no output to console.
> void marchingCubes::setIsoValue(double isoValue)
> {
>     std:cout<<"iso value:" << (char)isoValue;
> }
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From utkarsh.ayachit at kitware.com  Mon Sep 29 09:00:42 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 29 Sep 2014 09:00:42 -0400
Subject: [Paraview-developers] How to find the selected object?
In-Reply-To: <54259CBF.4050806@xs4all.nl>
References: <54259CBF.4050806@xs4all.nl>
Message-ID: <CADHSJjAVS2MhvhShpU-PAO--aE8DaqET=AB3Aw1O=yVz2=ChVg@mail.gmail.com>

Depends on what you mean by "selection". If you're referring to the
selection in the Pipeline browser i.e. the "active" source, then use
pqActiveObjects [1]. If you're looking for selection by
clicking/selecting in the render view, pqSelectionManager [2]. [1]
Would often suffice for [2] too depending on what you're looking for
to do once you've determined what was selected since selecting in
render view, also makes the source active.

Utkarsh

[1] http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classpqActiveObjects.html
[2] http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classpqSelectionManager.html

On Fri, Sep 26, 2014 at 1:05 PM, B.W.H. van Beest <bwvb at xs4all.nl> wrote:
> Hello,
>
> I'm trying to build my own plugin (C++) in Paraview.
> I wonder how one can programattically find which object (point, box,
> sphere, ... etc) is selected/clicked by a user.
>
> Any hint is much appreciated.
>
> Regards,
> Bertwim
>
>
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers

From utkarsh.ayachit at kitware.com  Mon Sep 29 11:52:00 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Mon, 29 Sep 2014 11:52:00 -0400
Subject: [Paraview-developers] ANN: ParaView 4.2.0 is available for download
Message-ID: <CADHSJjD7kuWDN3BWzcrquWVmxeAgMkZcJ+ufatSfpjEwh8U=1Q@mail.gmail.com>

Folks,

ParaView 4.2.0 is now available for download. Checkout to the release
notes on the Kitware blog [1].

As always, we look forward to your feedback [2].

Also stay tuned to the Kitware Blog [3] for upcoming features and
enhancements to ParaView, ParaView Catalyst, ParaViewWeb and much
more!

- The ParaView team

[1] http://www.kitware.com/blog/home/post/744
[2] http://paraview.uservoice.com
[3] http://www.kitware.com/blog/

From sankhesh.jhaveri at kitware.com  Mon Sep 29 12:31:06 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Mon, 29 Sep 2014 09:31:06 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Sunday, September 28 2014
Message-ID: <20140929163104.BD9307FDF3@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   Cdash returned nothing useful as of this report.

   Dashboard status for Nightly (next)

   -------------------- ANALYZING --------------------
   [1]fedora-x64-icc-nightlynext-release 1
   [2]ubuntu-x64-coverage 1
   -------------------- REPORT --------------------
   2 FAILURES
   [3]pv.LogColorMap , 1 , ['fedora-x64-icc-nightlynext-release']
   [4]pvcs.LoadPartiallyUpdatedStateFiles , 1 , ['ubuntu-x64-coverage']

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-29&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlynext-release
   2. http://open.cdash.org/index.php?project=ParaView&date=2014-09-29&filtercount=1&field1=buildname/string&compare1=61&value1=ubuntu-x64-coverage
   3. http://open.cdash.org/testSummary.php?project=9&date=2014-09-29&name=pv.LogColorMap
   4. http://open.cdash.org/testSummary.php?project=9&date=2014-09-29&name=pvcs.LoadPartiallyUpdatedStateFiles

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140929/e99730b6/attachment.html>

From rroemer at gmail.com  Mon Sep 29 13:46:02 2014
From: rroemer at gmail.com (=?UTF-8?Q?Ronald_R=C3=B6mer?=)
Date: Mon, 29 Sep 2014 19:46:02 +0200
Subject: [Paraview-developers] How to find the selected object?
In-Reply-To: <CADHSJjAVS2MhvhShpU-PAO--aE8DaqET=AB3Aw1O=yVz2=ChVg@mail.gmail.com>
References: <54259CBF.4050806@xs4all.nl>
 <CADHSJjAVS2MhvhShpU-PAO--aE8DaqET=AB3Aw1O=yVz2=ChVg@mail.gmail.com>
Message-ID: <CAPK9r69gQE9OBetTT_fFAL2FG4BCwWTz96sig5hbRrsvW1Vy3A@mail.gmail.com>

This sounds interesting! Could you write a snippet of how to get the
selected sources from the pipeline browser? Let us assume that a user
selects two sources. What's the simplest way to get both output-ports? I
could use this information for my plugin, that consumes two polydata
inputs. Currently a dialog is opened, where the user must select the second
input. This takes too much time ...

Thanks.
Am 29.09.2014 15:00 schrieb "Utkarsh Ayachit" <utkarsh.ayachit at kitware.com>:

> Depends on what you mean by "selection". If you're referring to the
> selection in the Pipeline browser i.e. the "active" source, then use
> pqActiveObjects [1]. If you're looking for selection by
> clicking/selecting in the render view, pqSelectionManager [2]. [1]
> Would often suffice for [2] too depending on what you're looking for
> to do once you've determined what was selected since selecting in
> render view, also makes the source active.
>
> Utkarsh
>
> [1]
> http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classpqActiveObjects.html
> [2]
> http://www.paraview.org/ParaView3/Doc/Nightly/www/cxx-doc/classpqSelectionManager.html
>
> On Fri, Sep 26, 2014 at 1:05 PM, B.W.H. van Beest <bwvb at xs4all.nl> wrote:
> > Hello,
> >
> > I'm trying to build my own plugin (C++) in Paraview.
> > I wonder how one can programattically find which object (point, box,
> > sphere, ... etc) is selected/clicked by a user.
> >
> > Any hint is much appreciated.
> >
> > Regards,
> > Bertwim
> >
> >
> > _______________________________________________
> > Paraview-developers mailing list
> > Paraview-developers at paraview.org
> > http://public.kitware.com/mailman/listinfo/paraview-developers
> _______________________________________________
> Paraview-developers mailing list
> Paraview-developers at paraview.org
> http://public.kitware.com/mailman/listinfo/paraview-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140929/0503b851/attachment.html>

From utkarsh.ayachit at kitware.com  Tue Sep 30 08:28:57 2014
From: utkarsh.ayachit at kitware.com (Utkarsh Ayachit)
Date: Tue, 30 Sep 2014 08:28:57 -0400
Subject: [Paraview-developers] Gatekeeper Review Summary
Message-ID: <CADHSJjCooZbpdUPqq-cUL-Kczi3q65C0NAZ6VpbpVrgqcdAgcA@mail.gmail.com>

Topics merged into master:
        14308_fix_nonconvex_polygon_vis
(VTK)   bar-chart-colors

From sankhesh.jhaveri at kitware.com  Tue Sep 30 12:31:10 2014
From: sankhesh.jhaveri at kitware.com (Sankhesh Jhaveri)
Date: Tue, 30 Sep 2014 09:31:10 -0700 (PDT)
Subject: [Paraview-developers] Dashboard Status: Monday, September 29 2014
Message-ID: <20140930163103.E888E7FF21@sanganak-ubuntu>

   Dashboard status for Nightly (master)

   -------------------- ANALYZING --------------------
   [1]fedora-x64-icc-nightlymaster-debug 1
   -------------------- REPORT --------------------
   1 FAILURES
   [2]pvcs-collab.CreateDelete , 1 ,
   ['fedora-x64-icc-nightlymaster-debug']

   Dashboard status for Nightly (next)

   Cdash returned nothing useful as of this report.

References

   1. http://open.cdash.org/index.php?project=ParaView&date=2014-09-30&filtercount=1&field1=buildname/string&compare1=61&value1=fedora-x64-icc-nightlymaster-debug
   2. http://open.cdash.org/testSummary.php?project=9&date=2014-09-30&name=pvcs-collab.CreateDelete

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview-developers/attachments/20140930/0097f3a8/attachment.html>

From nthl.collard at gmail.com  Tue Sep 30 19:59:31 2014
From: nthl.collard at gmail.com (tacana)
Date: Tue, 30 Sep 2014 16:59:31 -0700 (PDT)
Subject: [Paraview-developers] building a 'stair' shaped grid
Message-ID: <1412121571646-3312.post@n6.nabble.com>

Hi everyone,

I am a new Paraview user and the following question might seem quite simple
for advanced users but here it goes:

I would like to build a grid which has a 'stair' shape. For example, the
grid would have
6 blocks symbolized by the following figure:

X
XX
XXX

Initially I was thinking to use the following vtk file format:

# vtk DataFile Version 2.0
scalar data
ASCII

DATASET STRUCTURED_GRID
DIMENSIONS    1  3  3
POINTS    6 float
0.5000E+02 0.5000E+02 0.5000E+02
0.5000E+02 0.5000E+02 0.1500E+03
0.5000E+02 0.5000E+02 0.2500E+03
0.5000E+02 0.1500E+03 0.5000E+02
0.5000E+02 0.1500E+03 0.1500E+03
0.5000E+02 0.1500E+03 0.5000E+02

POINT_DATA      6
SCALARS   scalars  float
LOOKUP_TABLE default
1
2
3
4
5
6

The X-direction is just for giving a volume to my blocks, but it is not
really considered.

What would be the right values to use for the 'DIMENSIONS' of the grid since
it is not a regular shape?
Am I using the correct type of grid: STRUCTURED_GRID ?

Many thanks in advance!

Best,

Tacana



--
View this message in context: http://the-unofficial-paraview-developers-forum.34153.x6.nabble.com/building-a-stair-shaped-grid-tp3312.html
Sent from the The Unofficial ParaView Developers Forum mailing list archive at Nabble.com.