[Paraview] Paraview Visualization Ambiguity an a 2nd order Tetrahedra

[David Lonie]

On Tue, Oct 30, 2012 at 10:57 AM, Berk Geveci berk.gev...@kitware.com wrote:
 Merhaba Umut,

 I would give it a bit more time before compiling ParaView for vector
 graphics. We are still working on resolving issues and making improvements.
 We also don't have any documentation yet. I CC'ed David. He is the main
 developer and can provide some more details.

 On Wed, Oct 24, 2012 at 10:26 AM, Umut Tabak umut.ta...@gmail.com wrote:
 Thanks for this information, I would like to compile Paraview with vector
 graphics export facilites, especially svg if possible.

The superbuild packages already have the vector graphics export built
in (I verified with last night's build). If you'd like to build your
own version, the current 'master' branch on git will have support for
this feature (see
http://www.paraview.org/Wiki/ParaView:Build_And_Install and follow the
directions for the development version from git).

To export a scene, use File - Export Scene, and select a vector
graphics format. Currently, we have basic support for exporting 3D
scenes and 2D charts as .ps, .eps, .svg, and .pdf files.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Undefined symbols when building an example reader plugin

[David Lonie]

On Tue, Nov 13, 2012 at 10:44 AM, Cory Quammen cquam...@cs.unc.edu wrote:
 Just a followup:

 My plugins build and link against ParaView 3.98-RC1 without error, so
 a regression has been introduced since then.

 Cory

 On Tue, Nov 13, 2012 at 9:50 AM, Cory Quammen cquam...@cs.unc.edu wrote:
 Takuya,

 I have a similar problem with a simple VTK filter plugin and a
 representation plugin. Unfortunately, I don't have a better solution
 than what you propose, but I just wanted to confirm that this is a new
 problem, at least on Mac OS X.

 Cory

The quick work-around is to link your plugin against
vtkRenderingMatplotlib when python is enabled. We do this for part of
the CoProcessing plugin that comes with ParaView. This is not a good
long term solution IMO, but hopefully someone with a better knowledge
of the modular build system can propose a better solution...

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Undefined symbols when building an example reader plugin

[David Lonie]

On Tue, Nov 13, 2012 at 11:05 AM, David Lonie david.lo...@kitware.com wrote:
 On Tue, Nov 13, 2012 at 10:44 AM, Cory Quammen cquam...@cs.unc.edu wrote:
 Just a followup:

 My plugins build and link against ParaView 3.98-RC1 without error, so
 a regression has been introduced since then.

 Cory

 On Tue, Nov 13, 2012 at 9:50 AM, Cory Quammen cquam...@cs.unc.edu wrote:
 Takuya,

 I have a similar problem with a simple VTK filter plugin and a
 representation plugin. Unfortunately, I don't have a better solution
 than what you propose, but I just wanted to confirm that this is a new
 problem, at least on Mac OS X.

 Cory

 The quick work-around is to link your plugin against
 vtkRenderingMatplotlib when python is enabled. We do this for part of
 the CoProcessing plugin that comes with ParaView. This is not a good
 long term solution IMO, but hopefully someone with a better knowledge
 of the modular build system can propose a better solution...

A better solution was pointed out to me off-list -- try adding

include(${PARAVIEW_USE_FILE})

and then linking against ${VTK_LIBRARIES}. This should be better in
the long-term.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Compiling error

[David Lonie]

Hi Tom,

On Mon, Dec 3, 2012 at 11:43 AM, Tom T tomt...@yahoo.com wrote:
 Hi everyone,

 I've been trying so far unsuccessfully to compile paraview and I'm not sure
 what the error message means.

 [snip]

 /scratch/fronk/software/ParaView/VTK/Rendering/Matplotlib/vtkMatplotlibMathTextUtilities.cxx:
 In member function ‘virtual bool
 vtkMatplotlibMathTextUtilities::StringToPath(const char*, vtkPath*,
 vtkTextProperty*)’:
 /scratch/fronk/software/ParaView/VTK/Rendering/Matplotlib/vtkMatplotlibMathTextUtilities.cxx:581:
 error: ‘Py_ssize_t’ was not declared in this scope

Py_ssize_t is a type defined by the python C API, but it looks like it
wasn't introduced until Python 2.5, so I'm guess that you have an
older version of python installed? What is the output of python
--version?

The matplotlib library requires python = 2.6, so if your python is
older than this, you'll need to update your python installation, or
disable matplotlib in the ParaView build by editing
CMake/VTKModules.cmake and commenting out this section (around line
350):

# Enable vtkRenderingMatplotlib for math text rendering if python is available
if (PARAVIEW_ENABLE_PYTHON)
  list (APPEND _vtk_modules vtkRenderingMatplotlib)
endif()

I'll push a patch to only enable the matplotlib module when python is
an appropriate version.

Thanks for the report!

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Compiling error

[David Lonie]

Bug created: http://paraview.org/Bug/view.php?id=13700

On Tue, Dec 4, 2012 at 8:44 AM, David Lonie david.lo...@kitware.com wrote:
 Hi Tom,

 On Mon, Dec 3, 2012 at 11:43 AM, Tom T tomt...@yahoo.com wrote:
 Hi everyone,

 I've been trying so far unsuccessfully to compile paraview and I'm not sure
 what the error message means.

 [snip]

 /scratch/fronk/software/ParaView/VTK/Rendering/Matplotlib/vtkMatplotlibMathTextUtilities.cxx:
 In member function ‘virtual bool
 vtkMatplotlibMathTextUtilities::StringToPath(const char*, vtkPath*,
 vtkTextProperty*)’:
 /scratch/fronk/software/ParaView/VTK/Rendering/Matplotlib/vtkMatplotlibMathTextUtilities.cxx:581:
 error: ‘Py_ssize_t’ was not declared in this scope

 Py_ssize_t is a type defined by the python C API, but it looks like it
 wasn't introduced until Python 2.5, so I'm guess that you have an
 older version of python installed? What is the output of python
 --version?

 The matplotlib library requires python = 2.6, so if your python is
 older than this, you'll need to update your python installation, or
 disable matplotlib in the ParaView build by editing
 CMake/VTKModules.cmake and commenting out this section (around line
 350):

 # Enable vtkRenderingMatplotlib for math text rendering if python is available
 if (PARAVIEW_ENABLE_PYTHON)
   list (APPEND _vtk_modules vtkRenderingMatplotlib)
 endif()

 I'll push a patch to only enable the matplotlib module when python is
 an appropriate version.

 Thanks for the report!

 Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Line thickness vector image

[David Lonie]

On Sat, Mar 16, 2013 at 4:25 PM, Nikolay Kostov kos...@rice.edu wrote:

 Is it possible to specify the line thickness in some way when exporting an
 image as a vector format (e.g. pdf)?

 **

 I know it used to be possible before*.*


It is not possible to change this at the moment.

In which version was this possible? The PDF export feature was just
introduced in 3.98, unless there was some earlier implementation I'm not
aware of.

What sort of issue are you seeing? We can probably expose an option to
adjust this if there's a need.

Thanks,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] VRPN Plugin problem

[David Lonie]

On Thu, Mar 21, 2013 at 12:55 PM, Aashish Chaudhary 
aashish.chaudh...@kitware.com wrote:

 On Thu, Mar 21, 2013 at 11:00 AM, Emiliano Pastorelli
 emiliano.pastore...@gmail.com wrote:
  Hi Aashish,
 
  actually it happened on both the actual operating system (Debian Weezy)
 and
  the older one that I still have on another hard disk (Ubuntu 12.10), and
  both of them with Qt 4.8.2

 I see.  Let us see if we can reproduce this bug. I am using Ubuntu
 10.10 and Qt 4.8.4


Can you run 'grep VR CMakeCache.txt' on the CMakeCache.txt in your build
directory and send me the output?

Thanks,

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] VRPN Plugin problem

[David Lonie]

On Thu, Mar 21, 2013 at 1:01 PM, Emiliano Pastorelli hoganc...@gmail.comwrote:

 Thanks Aashish,

 on the ubuntu 12.10 system i had temporary solved the problem loading
 the VR plugin from a slightly older paraview build i had around, it worked,
 but that means that i was still not using all the fixes you made in the VR
 plugin, that i would be really interested to test.

 Hi David,

 I can do it tomorrow 'cause i just came back from the office now. I'll
 send it as soon as I have it.


Thanks. The other file of interest is [build
directory]/Plugins/VR/vtkPVVRConfig.h. Please include that with the
CMakeCache info.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] VRPN Plugin problem

[David Lonie]

On Fri, Mar 22, 2013 at 7:54 AM, Emiliano Pastorelli hoganc...@gmail.comwrote:

  Hi Guys, checking in the CMakeCache was probably enough to enlighten me
 about the fact that sometimes i'm dumb ;) I had activated the VRPlugin, but
 not the PARAVIEW_USE_VRUI and PARAVIEW_USE_VRPN (i don't remember doing it
 in the 3.98 RC1, but it might be that i just forgot it).

 Now it's building, hopefully and probably that was the only needed fix to
 make it work. Should I have further problem i'll let you know, thanks for
 all the help, you saved my day ;)

 Best,
 Emiliano


 Il 21/03/13 19:03, David Lonie ha scritto:

 On Thu, Mar 21, 2013 at 1:01 PM, Emiliano Pastorelli 
 hoganc...@gmail.comwrote:

 Thanks Aashish,

 on the ubuntu 12.10 system i had temporary solved the problem loading
 the VR plugin from a slightly older paraview build i had around, it worked,
 but that means that i was still not using all the fixes you made in the VR
 plugin, that i would be really interested to test.

 Hi David,

 I can do it tomorrow 'cause i just came back from the office now. I'll
 send it as soon as I have it.


  Thanks. The other file of interest is [build
 directory]/Plugins/VR/vtkPVVRConfig.h. Please include that with the
 CMakeCache info.

  Dave



___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] VRPN Plugin problem

[David Lonie]

On Fri, Mar 22, 2013 at 7:54 AM, Emiliano Pastorelli hoganc...@gmail.comwrote:

  Hi Guys, checking in the CMakeCache was probably enough to enlighten me
 about the fact that sometimes i'm dumb ;) I had activated the VRPlugin, but
 not the PARAVIEW_USE_VRUI and PARAVIEW_USE_VRPN (i don't remember doing it
 in the 3.98 RC1, but it might be that i just forgot it).

 Now it's building, hopefully and probably that was the only needed fix to
 make it work. Should I have further problem i'll let you know, thanks for
 all the help, you saved my day ;)


Sounds like that would be the issue :-) Let us know if you run into any
other problems.

By the way, there's a new wiki page detailing configuration of the VR
plugin using the new gui that may be of interested to you:

http://www.paraview.org/Wiki/ParaView_VR_in_4.0

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Line thickness vector image

[David Lonie]

Hi Nikolay,

The new GL2PS line and point scaling factor branch has been merged into
ParaView's master branch today.

The easiest way to use the feature is to go to File  Export Scene and
select one of the vector graphics formats (ps, eps, svg, or pdf), and the
configuration dialog that pops up will let you toggle the scaling factors.
The default value is 5/7, like it was before.

If you want to continue using your script, the mechanism is a little
different. Rather than calling the global functions to set the scaling
factors, the factors are now a property of the vtkGL2PSExporter instance
with their own setters/getters. They will need to be set for each
vtkGL2PSExporter used.

Let me know if you have any problems with the feature or notice any
regressions with the new implementation.

Dave


On Fri, Mar 22, 2013 at 10:58 AM, Nikolay Kostov kos...@rice.edu wrote:

  Hi Dave,
 Please let me know if you think this is something which can be solved in
 the next few weeks or if I should look at other ways of exporting the
 images (non-vector)?

 Thanks a lot,
 Nik



 On 3/19/2013 2:14 PM, David Lonie wrote:

 On Tue, Mar 19, 2013 at 2:26 PM, Nikolay Kostov kos...@rice.edu wrote:

  The pdf image has very thick lines whereas what I see in paraview
 before I export looks fine.
 Before I was using a python script which called vtkGL2PSExporter and I
 had:
 gl2ps.SetGlobalPointSizeFactor(size)
 gl2ps.SetGlobalLineWidthFactor(size)

 This code doesn't work with paraview 3.98 anymore and I am not able to
 select line width and point size anywhere.
 Maybe the problem has to do with the newer version of vtkGL2PSExporter?


  These functions were removed for the new vtkGL2PSExporter class. Let me
 see what I can do, something similar can likely be added.

  Dave



___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] ParaView 3.98.1: Problems with Space Navigator

[David Lonie]

On Mon, Apr 1, 2013 at 3:35 PM, Nenad Vujicic nena...@gmail.com wrote:
 Hello everyone!

 I'm trying to make my Space Navigator working in ParaView v3.98.1 on my
 Windows 8 machine. I built sources without problems (using VS2012, with qt
 4.8.4, python 2.7.3, boost 1.53.0, OpenMPI 1.6.2, VRPN 07_30 - all rebuilt
 in x64 mode) with turned on PARAVIEW_USE_VRPN and
 PARAVIEW_BUILD_PLUGIN_VRPlugin (+ flags for enabling shared libraries,
 python, mpi, visit bridge, etc). But, I'm unable to make it working from Qt
 client. Here are my running steps:

 1. I start vrpn_server in background (with enabled line with my device).
 When I test it with vrpn_print_devices, everything works OK - whenever I
 touch my device, vrpn client prints some values,
 2. I start ParaView's Qt client, load VR plugin, create Cone sourse, define
 VR connection and Interactions and push Start button,
 3. I click on Cone source or built in: node in Pipeline Browser and touch
 my Space Navigator but nothing changes (except vrpn_server reports
 connection is established). Actually, for some actions like pushing button
 (on 3d mouse) I get several instances of following warning in Output Window:

 Generic Warning: In
 ..\..\..\..\ParaViewCore\ServerManager\Core\vtkSMPropertyHelper.cxx, line
 439

 Call not supported for the current property type.

 I get same behavior when I save options from my VR Panel to state file and
 load it. Here is VR content from state file:

 VRConnectionManager
 VRPNConnection name=vrconnection address=device0@localhost
   Button id=0 name=MyRightButton/
   Tracker id=0 name=MyTracker/
 /VRPNConnection
   /VRConnectionManager
   VRInteractorStyles
 Style class=vtkVRGrabWorldStyle proxy=2125 property=Visibility
   Button role=Reset world name=vrconnection.MyRightButton/
   Button role=Rotate world name=vrconnection.MyRightButton/
   Button role=Translate world name=vrconnection.MyRightButton/
   Tracker role=Tracker name=vrconnection.MyTracker/
 /Style
   /VRInteractorStyles

 I hope I made some small mistake:-). Can anyone suggest where could be the
 problem?

It's a bit odd to map all of the actions to the same button. Does your
input device only have a single button? The GrabWorldStyle requires a
button press to rotate or translate the view, but that same button is
also resetting any modifications it makes (since Reset world is
mapped to MyRightButton along with translate/rotate). These should
each be set to unique buttons.

Try using other buttons for reset/translate/rotate world. You can add
dummy buttons to the VR config if needed, just set the button's id
number higher than your device supports.

Hope this helps,

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] ParaView 3.98.1: Problems with Space Navigator

[David Lonie]

On Tue, Apr 2, 2013 at 12:39 PM, Nenad Vujicic nena...@gmail.com wrote:
 Dear Dave, Aashish,

 Thank you very much for your responses.

 I'm sorry, I tried dummy buttons with both v3.98.1 and current master
 ParaView sources, but they didn't help me. Also, I'm still able to select
 visibility with grab world style from ParaView's UI (in master sources).
 Btw, my 3d mouse has 2 buttons
 (http://www.3dconnexion.com/products/spacenavigator.html). Here is one of
 snippets from state file I tried:


 VRConnectionManager
VRPNConnection name=vrconnection address=device0@localhost
   Button id=0 name=MyRightButton/
   Button id=1 name=MyLeftButton/
   Button id=2 name=MyDummyButton/
   Analog id=0 name=MyAnalog/

   Tracker id=0 name=MyTracker/
 /VRPNConnection
   /VRConnectionManager
   VRInteractorStyles
 Style class=vtkVRGrabWorldStyle proxyName=RenderView1 proxy=269
 property=ModelTransformMatrix
   Button role=Reset world name=vrconnection.MyLeftButton/

   Button role=Rotate world name=vrconnection.MyRightButton/
   Button role=Translate world name=vrconnection.MyDummyButton/

   Tracker role=Tracker name=vrconnection.MyTracker/
/Style
 /VRInteractorStyles

 Do you have some other idea?

What happens when you press the right button and move the tracker?
Does the scene move at all, or are you still just seeing the error
messages in the log?

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] ParaView 3.98.1: Problems with Space Navigator

[David Lonie]

On Tue, Apr 2, 2013 at 12:52 PM, Nenad Vujicic nena...@gmail.com wrote:
 Dear Dave,

 It doesn't happen anything, like I didn't touch my 3d mouse. Btw, here is
 what is outputted on standard output when I execute 'vrpn_print_devices
 device0@localhost':

 on left button pressed:

 Button device0@localhost, number 1 was just pressed
 Button device0@localhost, number 1 was just released

 on right button pressed:

 Button device0@localhost, number 0 was just pressed
 Button device0@localhost, number 0 was just released

 on touching tracker:

 Analog device0@localhost:

 0.00, 0.00, 0.00, 0.00, 0.00, 0.00 (6 chans)
 (several above lines with different values)

Ok, so the vrpn server is definitely working, and it sounds like the
Call not supported for the current property type error was from
trying to control the Visibility property, rather than the
ModelTransformMatrix. Double check that the VR plugin is listening
(i.e. the Start button is disabled and Stop is enabled). Also try
restarting paraview. The configuration looks fine to me, I'm not sure
why it isn't working for you.

Are any errors/warnings emitted during startup or while running ParaView?

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] ParaView 3.98.1: Problems with Space Navigator

[David Lonie]

On Wed, Apr 3, 2013 at 7:58 AM, Nenad Vujicic nena...@gmail.com wrote:
 Dear Dave,

 Thank you for your ideas. Unfortunately, there are no errors / warnings
 emitted during startup or while running ParaView. Also, I tried several
 times with restarting ParaView and always checked if listening is started
 (i.e. Start button is disabled and Stop is enabled). I even turned off my
 firewall, but nothing helped.

 Btw, when I put some logging calls to pqVRStarter.cxx Ln 56 (beginning of
 pqVREventPlayer::playEvent() routine), I didn't get any message (during my
 trying to work with 3d mouse). Of course, pqVREventPlayer class definition
 is enabled (because PARAVIEW_USE_VRPN is turned ON). Did I maybe need to
 enable some special flag during compilation of Qt to get this event playing
 working or this is used only during automated testing?

That was only for automated testing. If you want to debug this, try
putting some debugging messages in

vtkVRGrabWorldStyle.cxx: HandleTracker() and HandleButton()
pqVRPNConnection.cxx: newButtonValue() and new TrackerValue()

The first one is the style where the events from the vrpn are
translated into camera events, and the second file will check that the
events are coming in from VRPN correctly.

Did ParaView/VRPN work on this machine in the past? AFAIK no one else
has tried this on Windows 8 yet.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] ParaView 3.98.1: Problems with Space Navigator

[David Lonie]

I just remembered this patch where Cory had rewritten the GrabWorld
style for the space navigator:

http://review.source.kitware.com/#/c/8374/

Nenad: If you get a chance to try it, please let us know (better yet,
make an account on the review board and leave feedback there, if you
have time. Otherwise an email to the list will do).

Cory: I +1'd these patch, but they seem to have been forgotten. Are
you still interested in contributing them? Can you rebase them against
current master and see if they still work?

Dave

On Wed, Apr 3, 2013 at 3:11 PM, Nenad Vujicic nena...@gmail.com wrote:
 Dear Dave, Aashish,

 Thank you very much for your help!

 I put logging calls to routines of above 2 classes (vtkVRGrabWorldStyle and
 pqVRPNConnection) and I started catching messages on both places. Now, looks
 like the problem is that current interaction styles don't like my device
 (most likely because of reason Aashish mentioned). I'll dig into these
 classes and try to fix them or create new interaction style. Then, someone
 from the team will submit the patch.

 Thanks again for your help!

 Best regards,
 Nenad.




 On Wed, Apr 3, 2013 at 3:10 PM, Aashish Chaudhary
 aashish.chaudh...@kitware.com wrote:

 One another possibility could be is that SN does not really offer tracker
 and we are expecting tracker data so the code
 somewhere is broke.

 Can you create a new interactor style and just listen for buttons and
 analogs?




 On Wed, Apr 3, 2013 at 8:38 AM, David Lonie david.lo...@kitware.com
 wrote:

 On Wed, Apr 3, 2013 at 7:58 AM, Nenad Vujicic nena...@gmail.com wrote:
  Dear Dave,
 
  Thank you for your ideas. Unfortunately, there are no errors / warnings
  emitted during startup or while running ParaView. Also, I tried several
  times with restarting ParaView and always checked if listening is
  started
  (i.e. Start button is disabled and Stop is enabled). I even turned off
  my
  firewall, but nothing helped.
 
  Btw, when I put some logging calls to pqVRStarter.cxx Ln 56 (beginning
  of
  pqVREventPlayer::playEvent() routine), I didn't get any message (during
  my
  trying to work with 3d mouse). Of course, pqVREventPlayer class
  definition
  is enabled (because PARAVIEW_USE_VRPN is turned ON). Did I maybe need
  to
  enable some special flag during compilation of Qt to get this event
  playing
  working or this is used only during automated testing?

 That was only for automated testing. If you want to debug this, try
 putting some debugging messages in

 vtkVRGrabWorldStyle.cxx: HandleTracker() and HandleButton()
 pqVRPNConnection.cxx: newButtonValue() and new TrackerValue()

 The first one is the style where the events from the vrpn are
 translated into camera events, and the second file will check that the
 events are coming in from VRPN correctly.

 Did ParaView/VRPN work on this machine in the past? AFAIK no one else
 has tried this on Windows 8 yet.

 Dave

 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

 Follow this link to subscribe/unsubscribe:
 http://www.paraview.org/mailman/listinfo/paraview




 --
 | Aashish Chaudhary
 | RD Engineer
 | Kitware Inc.
 | www.kitware.com



 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

 Follow this link to subscribe/unsubscribe:
 http://www.paraview.org/mailman/listinfo/paraview

___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Export image without background

[David Lonie]

Hi Roberto,

To remove the background, uncheck the Draw background and Rasterize 3D
geometry checkboxes in the Export Options dialog that appears after
setting the filename.

Dave


On Thu, Sep 26, 2013 at 6:53 AM, Roberto Di remigio 
roberto.diremi...@gmail.com wrote:

 Dear all,
 is there a way to export an image without the background?
 For example, in the attached svg file I would like to export only the
 contour map
 and not the map+background. Is that possible?
 With best regards.

 --
 Roberto Di Remigio


 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

 Follow this link to subscribe/unsubscribe:
 http://www.paraview.org/mailman/listinfo/paraview


___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] display bonds between atoms using paraview

[David Lonie]

On Thu, Oct 3, 2013 at 5:10 PM, Jian Luo laurenjaso...@gmail.com wrote:

 Hi everyone,

 I am in the field of atomic simulation.
 Paraview can display atoms beautifully, but I currently cannot find a way
 to draw bonds between atoms within certain distance.
 Can anyone help me with this? Thanks very much.


As far as I know, ParaView is only able to visualize molecular data, and
does not include such editing capabilities. I'd recommend using software
like Avogadro[1] to modify the molecule, which can then be saved as .cml
and displayed in ParaView if desired.

Hope this helps,
Dave

[1] http://avogadro.openmolecules.net
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

[Paraview] PDF export missing colors

[David Lonie]

Conversation moved here from:

http://www.kitware.com/blog/home/post/508

Jon,

In the Coloring section of the Properties tab, what is selected in the
drop-down box?

Also, is Interpolate Scalars Before Mapping enabled?

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] PDF export missing colors

[David Lonie]

Hi Jon,

Good, that was going to be my next suggestion -- disabling Interpolate
Scalars Before Mapping helps preserve colors that may otherwise be lost
during the export.

As far as the axis, I think that might be as good as it gets. In ParaView,
there are ways to render an object in an overlay so that it always appears
on top, but the gl2ps library (which is used to generate the pdfs from the
3d geometry) internally depth sorts all of the primitives, so any
overlay-tricks are ignored.

A workaround would be to add an Outline source that is the same size as the
axis you want, then enable it's axis and set the box's opacity to 0. This
will in effect give you a custom axis that you can resize/relocate anywhere
you'd like. By making it slightly larger than your geometry, it shouldn't
clash like the other axis.

Hope this helps,
Dave


On Tue, Feb 25, 2014 at 2:56 AM, Jon Vegard Venås zeti...@gmail.com wrote:

 A new problem arises though. How can I get the axis to be at the top
 layer:

 http://folk.ntnu.no/jonvegar/pdf/ydispNew.pdf


 2014-02-25 8:31 GMT+01:00 Jon Vegard Venås zeti...@gmail.com:

 I found a solution! I disabled Interpolate Scalars Before Mapping and I
 set Backface Representation to Cull Backface (the latter was to get
 around a bug). Thank you for your help!


 2014-02-25 8:02 GMT+01:00 Jon Vegard Venås zeti...@gmail.com:

 I found  Interpolate Scalars Before Mapping now. It is enabled.


 2014-02-24 21:46 GMT+01:00 Jon Vegard Venås zeti...@gmail.com:

 Where do I find Interpolate Scalars Before Mapping?

 Here is a screenshot of my window:

 http://folk.ntnu.no/jonvegar/pdf/screenshot.jpg


 2014-02-24 19:03 GMT+01:00 David Lonie david.lo...@kitware.com:

 Conversation moved here from:

 http://www.kitware.com/blog/home/post/508

 Jon,

 In the Coloring section of the Properties tab, what is selected in
 the drop-down box?

 Also, is Interpolate Scalars Before Mapping enabled?

  Dave




 --


 mvh
 Jon Vegard




 --


 mvh
 Jon Vegard




 --


 mvh
 Jon Vegard




 --


 mvh
 Jon Vegard

___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Export vectorial images in Paraview

[David Lonie]

On Wed, Mar 26, 2014 at 1:43 PM, Enrique Lizarraga-Garcia 
enriliz...@gmail.com wrote:

 Dear Utkarsh,

 Thank you for your reply. However, I still have a similar problem: the bar
 title and the bar numbers are vectorial, whereas the streamlines, glyph,
 etc are all in raster format (see example.pdf attached). Is there any way
 to overcome this?


This option is the one you're looking for: (it may be hidden as advanced,
click on the gear icon on the export dialog to find it)

http://www.kitware.com/blog/home/post/508
Rasterize 3D geometry: Rasterize all 3D objects in the scene into the
background of the image, overlaying 2D primitives and text. This is a
workaround for a limitation of the vector graphics formats, some of which
cannot handle texture-mapped or translucent 3D objects. This option also
speeds up exporting of large scenes, but should be disabled when possible
for the highest quality images.

Disabling this option will force the geometry to be exported in a vector
format, but may cause issues with color/transparency as described above.

Hope this helps,

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] exporting selection labels to an SVG as editable/movable text-objects

[David Lonie]

Hi Roman,

On Thu, May 22, 2014 at 6:22 AM, Dr. Roman Grothausmann
grothausmann.ro...@mh-hannover.de wrote:
 Many thanks for fixing the Label exporter issue. It does work with 4.1 to
 get the text exported into e.g. an SVG. However, the text is always rendered
 as paths (always consists of triangles). Is it also possible to have the
 text being exported as actual text-fields (like e.g. gnuplot does) that can
 be edited with the text-tool in e.g. inkscape?

There is an checkbox in the export options Render text as paths (see
http://www.kitware.com/blog/home/post/508 ). It's enabled by default
for portability, but disabling this option should do what you need.

Hope this helps,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] exporting selection labels to an SVG as editable/movable text-objects

[David Lonie]

Hi Roman,

Aha, the cube axis actors are still using vtkVectorText to render the text.
vtkVectorText is a text source that produces polydata to forms the
characters of the text string, and the actual text information is lost by
the time the SVG exporter processes the actors.

There was an effort in VTK to convert the cube axes to use vtkTextActor3D,
which would allow proper text exporting, but this is disabled in ParaView,
and I believe it is only experimental at this point. So unfortunately, this
is the best it gets for that particular text source at the moment.

Sorry for the misinformation earlier -- I thought you were referring to a
ParaView text source, which should export as editable text.

Dave


On Thu, Jun 5, 2014 at 8:57 AM, Dr. Roman Grothausmann 
grothausmann.ro...@mh-hannover.de wrote:

 Hi Dave,


 Many thanks for Your reply. I tried this option, neither disabling nor
 enabling Render text as paths creates an SVG with editable text for me
 with paraview 4.1.0. See e.g. attached SVGs which show simply a cone where
 I enabled Show Axis. Opening theses with inkscape reveals that in both
 cases the text consist of triangles and not of text objects.
 Any idea what I could be doing wrong?

 Many thanks for looking into this.
 Roman

 On 22/05/14 14:11, David Lonie wrote:

 Hi Roman,

 On Thu, May 22, 2014 at 6:22 AM, Dr. Roman Grothausmann
 grothausmann.ro...@mh-hannover.de wrote:

 Many thanks for fixing the Label exporter issue. It does work with 4.1 to
 get the text exported into e.g. an SVG. However, the text is always
 rendered
 as paths (always consists of triangles). Is it also possible to have the
 text being exported as actual text-fields (like e.g. gnuplot does) that
 can
 be edited with the text-tool in e.g. inkscape?


 There is an checkbox in the export options Render text as paths (see
 http://www.kitware.com/blog/home/post/508 ). It's enabled by default
 for portability, but disabling this option should do what you need.

 Hope this helps,
 Dave


 --
 Dr. Roman Grothausmann

 Tomographie und Digitale Bildverarbeitung
 Tomography and Digital Image Analysis

 Institut für Funktionelle und Angewandte Anatomie, OE 4120
 Medizinische Hochschule Hannover
 Carl-Neuberg-Str. 1
 D-30625 Hannover

 Tel. +49 511 532-9574

___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] exporting selection labels to an SVG as editable/movable text-objects

[David Lonie]

On Fri, Jun 6, 2014 at 3:17 AM, Dr. Roman Grothausmann 
grothausmann.ro...@mh-hannover.de wrote:

 Hi Dave,


 Many thanks for Your quick answer. I checked exporting the labels of
 selections, where the text export works as expected (SVG attached).
 For using Paraview SVG export in latex it would still be very nice to have
 the same export functionality for the cube axes. Should I reopen the former
 feature-request/fix or open a new one just to keep track in the system
 should the VTK implementation become stable?


Feel free to open a new ticket in ParaView for exposing the vtkTextActor3D
option for cube axes.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Re: [Paraview] Questions on vector images output

[David Lonie]

Hi Mikhail,

1) The triangulation artifacts you're seeing are likely due to antialiasing
in the viewer. Turning AA off (or using a different viewer) should make the
gaps between the triangles disappear and the surface should appear smooth
again.

2) The axis labels are 3D objects with rotations away from the plane of the
image, while the scalar bar labels are in the plane of the image. This
means that the scalar bar labels can be inserted as actual text (i.e. tell
the viewer to draw the string interior_bf_0), but the axis labels have to
be drawn as bezier curves (i.e. tells the viewer to draw the shape of the
text) since the output image formats can't embed arbitrarily oriented text.
To fix the issue with the gaps, it looks like viewer antialiasing is the
likely culprit again.

Hope this helps,
Dave

On Sat, Oct 18, 2014 at 4:57 PM, Mikhail Artemyev 
artemiev.mikh...@gmail.com wrote:

 Dear all,

 I have two questions related to the vector images output from ParaView for
 3D. I use ParaView 4.1.0 32-bit on Ubuntu.

 1. I've read here (http://www.kitware.com/blog/home/post/508) in the
 discussion after the post that if the colors are generated from
 point/cell attributes, the exporter should be able to handle it once
 'Rasterize 3D geometry' turned off. Otherwise you may end up with your
 exported object appearing grey/colorless. That's exactly what happens in
 my case. However my 3D picture is based on a rectangular grid with values
 of the computed field defined at the vertices of the grid. I understand
 that for visualization ParaView convert the rectangles to triangles first
 (and it's clearly seen from the grey/colorless image I have), therefore I'm
 wondering if it'll help if I replace rectangular mesh to triangular from my
 side.

 2. The second question is why the labels of the axis have some
 triangulation inside them (here is a picture
 http://drive.google.com/file/d/0B-0jjlhN_6HyWkR1bWoxc3dVazA/view?usp=sharing).
 Is it GL2PS feature? How can I make them look like labels on colour bar?

 Thank you,
 Mikhail

 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

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


___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

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

Re: [Paraview] PDF-Export

[David Lonie]

What settings are you using for the export? This wiki page explains what
the options do, and what the limitations of the exporter are:

http://www.paraview.org/Wiki/ParaView/Vector_Graphics_Export

The grey coloring sounds like there's a texture mapped surface being
rendered, which cannot be exported unless the data is rasterized.

Hope this helps,
Dave

On Thu, Nov 6, 2014 at 11:35 AM, yannic.eich...@hsu-hh.de wrote:

 Hello,

 I´m working in the area of Numerical Computational Engineering and we want
 to use Paraview to visualize our simulation results. To add a scene from
 Paraview to a documentation we need to export this scene to a PDF-file. At
 the moment if we export, a PDF-file will be created, but opening this file
 an error occurs, the scene is grey-coloured and transparency is gone. I
 hope you could help me exporting a well-working PDF-File.

 Greetings

 Yannic Eichler

 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

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

___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

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

Re: [Paraview] PDF-Export

[David Lonie]

Forgot to CC the list:

On Fri, Nov 7, 2014 at 9:16 AM, David Lonie david.lo...@kitware.com wrote:

 On Fri, Nov 7, 2014 at 8:46 AM, yannic.eich...@hsu-hh.de wrote:

 Thank you for your fast reply.
 If tried different settings, but nothing really works. If i make a tick
 at rasterize 3D geometry the exported pdf also occurs an error and the
 shown graphic is kind of a bitmap graphic and not a real vector graphic
 like I need. You are rigth it isn´t grey-colored anymore. The errorin adobe
 acrobat reader is like: The graphic could not be presented correctly,
 please contact the producer of the file.


 We use an external library to produce the exports (gl2ps), and this is a
 limitation on their side -- the textures that are used to color the data
 cannot be detected by gl2ps, so the only way to get the colors to appear is
 to rasterize the data. Indeed, this defeats the purpose of using a vector
 format :-(

 There was a similar issue on the developer's list a few months ago, see
 this thread:

 http://public.kitware.com/pipermail/paraview/2014-February/030639.html

 Is there a Interpolate Scalars Before Mapping option enabled in the
 paraview properties panel? Disabling that will change how the colors are
 stored, and seems to work around the texture issue.

 As for the client error, that's hard to track down without seeing the
 file. By chance, do you have any multi-line text being rendered in your
 scene? There was a bug (fixed in the development version now) that would
 cause similar problems when exporting multiline text as a path.


 Is it possible that the problem depends on the code I use to convert my
 simulation results into a inp-file to visualize with Paraview. In Paraview
 I don´t get any errors and problems and everything works still fine, but
 maybe I have forgotten to add some kind of special code?


 Not likely. GL2PS works by watching the OpenGL feedback buffer --
 basically, it watches what is being drawn in the window and converts it
 into a vector format. This is independent of the original data format.

 Dave

___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

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

Re: [Paraview] Labeling of axis

[David Lonie]

On Mon, May 4, 2015 at 3:20 AM, Julia Weißflog 
julia.weissf...@geophysik.tu-freiberg.de wrote:

 Thanks for your reply, i have uploaded a sample here
 https://mathmaster.mathe.tu-freiberg.de/index.php/s/UoQ506K29Db6b7E

 Notice how the axis labels are blurry while the colobar text looks fine.


This is due to the Cube Axes rendering using an older text rendering
method, vtkVectorText. These don't always export cleanly through the vector
graphics backend.

There is a replacement for the Cube Axes under development that uses newer
methods for rendering text, and this should export much better, and be
similar to the labels you see on the color legend.

Also, there is experimental support in the old Cube Axes actor for using
the newer text renderer. If you can access the cube actor in the python
shell, calling SetUseTextActor3D(True) on it should switch it on.

Hope this helps,

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] eps outputting in paraview

[David Lonie]

Hi Zuoguang,

For the first image -- what happens with Rasterize 3D Geometry left on? The
GL2PS library used to generate the vector graphics formats can't handle
geometry which is shaded using textures, which I suspect is the issue here.
Rasterizing these objects is, unfortunately, the only way to preserve the
colors that I'm aware of.

For the second image, what does the output ps look like?

Dave

On Thu, May 7, 2015 at 5:53 AM, 付作光 fzg...@126.com wrote:

 Dear all the Paraview users，

 I am the new student of Paraview and easily lost in some basic details of
 its application, one of them is eps outputting. I have tried it many times
 but failed all the while. The color maps may transform automatically into
 grayscale ones when I use

 Export Scene  choosing eps format  canceling Rasterize 3D geometry

 There are two examples for describing what I just lost in in E-mail
 attachments. I want to ask either something error in my operation or
 something I miss in the basic concepts of Paraview’s usage for my purpose.
 Seeking your help!



 Best regards
 Zuoguang Fu



 ___
 Powered by www.kitware.com

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

 Please keep messages on-topic and check the ParaView Wiki at:
 http://paraview.org/Wiki/ParaView

 Search the list archives at: http://markmail.org/search/?q=ParaView

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


___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Latex preamble (bold math ) in paraview

[David Lonie]

On Mon, Jun 1, 2015 at 12:35 PM, Raj Kumar Manna rajphysics@gmail.com
wrote:

 I have seen the answer
 http://tex.stackexchange.com/questions/595/how-can-i-get-bold-math-symbolsbefore
 But \mathbf does not give the latex bold font, it just gives bold font
 without the latex font. I would prefer font of \boldmath or \bm. Thanks for
 your help.


It is not possible to load additional packages or set the matplotlib latex
preamble at this time.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] segmentation fault with netCDF when loading state file

[David Lonie]

Hi Ryan,

I'm looking into this issue.

I download the sample data you provided, and the state file loads and
renders fine for me with the current (as of this morning) master branch of
ParaView. What version of ParaView are you using? Can you test the latest
master branch? It appears that whatever the bug was, it has been fixed.

Dave

On Wed, Oct 21, 2015 at 9:50 AM, Aashish Chaudhary <
aashish.chaudh...@kitware.com> wrote:

> Ryan,
>
> would it be possible for you to share a sample dataset? The NetCDF reader
> has fails to read data in certain corner cases but without having the data
> to reproduce the issue it is kind of little tricky to get to the problem.
>
> - Aashish
>
> On Wed, Oct 21, 2015 at 9:44 AM, Ryan Abernathey <
> ryan.abernat...@gmail.com> wrote:
>
>> Does anyone have any feedback on this issue?
>>
>> I have no idea how to debug or continue. I will have to abandon paraview
>> for my project unless I can get some help somehow.
>>
>> Thanks,
>>
>> Ryan Abernathey
>> Assistant Professor
>> Columbia University, Department of Earth & Environmental Sciences
>> Lamont-Doherty Earth Observatory, Division of Ocean & Climate Physics
>> 205 C Oceanography
>> 61 Route 9W - PO Box 1000
>> Palisades, NY 10964-8000
>> http://rabernat.github.io
>> r...@ldeo.columbia.edu
>>
>>
>> On Mon, Oct 19, 2015 at 4:59 PM, Ryan Abernathey <
>> ryan.abernat...@gmail.com> wrote:
>>
>>> Hello,
>>>
>>> I am extremely frustrated and stuck on this problem. Any advice would be
>>> appreciated.
>>>
>>> I am working with a dataset of 2400 sequentially numbered netCDF files
>>> (generic & CF conventions).
>>>
>>> In a fresh pipeline, Paraview is able to recognizes these files as a
>>> single source and load them correctly. I then create a standard pipeline
>>> involving various filters (contour, etc., nothing funny) and save a state
>>> file.
>>>
>>> When I attempt to load the state file, I get the following segmentation
>>> fault. Further down you can see a ncdump of the offending file.
>>>
>>> Would appreciate any advice on how to debug / overcome this issue. Is it
>>> a bug? It sure feels like one...
>>>
>>> Best,
>>> Ryan Abernathey
>>>
>>>
>>> ERROR: In
>>> /home/kitware/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/IO/NetCDF/vtkNetCDFReader.cxx,
>>> line 822
>>> vtkNetCDFCFReader (0x55c0620): netCDF Error: NetCDF: Index exceeds
>>> dimension bound
>>>
>>>
>>> ERROR: In
>>> /home/kitware/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/Common/ExecutionModel/vtkExecutive.cxx,
>>> line 784
>>> vtkPVCompositeDataPipeline (0x5be23d0): Algorithm
>>> vtkFileSeriesReader(0x4c0c3e0) returned failure for request: vtkInformation
>>> (0x8141700)
>>>   Debug: Off
>>>   Modified Time: 05
>>>   Reference Count: 1
>>>   Registered Events: (none)
>>>   Request: REQUEST_DATA
>>>   FORWARD_DIRECTION: 0
>>>   FROM_OUTPUT_PORT: 0
>>>   ALGORITHM_AFTER_FORWARD: 1
>>>
>>>
>>>
>>>
>>> ERROR: In
>>> /home/kitware/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/Common/ExecutionModel/vtkExecutive.cxx,
>>> line 784
>>> vtkCompositeDataPipeline (0x744fde0): Algorithm
>>> vtkPVGeometryFilter(0x7436500) returned failure for request: vtkInformation
>>> (0x5737250)
>>>   Debug: Off
>>>   Modified Time: 43
>>>   Reference Count: 1
>>>   Registered Events: (none)
>>>   Request: REQUEST_DATA_OBJECT
>>>   FORWARD_DIRECTION: 0
>>>   FROM_OUTPUT_PORT: 0
>>>   ALGORITHM_AFTER_FORWARD: 1
>>>
>>>
>>>
>>>
>>> Segmentation fault (core dumped)
>>>
>>>
>>>
>>>
>>>
>>> 
>>> $ ncdump -h rce_rrtm303x512dx3f20_512x512x64_3km_12s_128_000300.nc
>>> netcdf rce_rrtm303x512dx3f20_512x512x64_3km_12s_128_000300 {
>>> dimensions:
>>> x = 512 ;
>>> y = 512 ;
>>> z = 64 ;
>>> time = UNLIMITED ; // (1 currently)
>>> variables:
>>> float x(x) ;
>>> x:units = "m" ;
>>> float y(y) ;
>>> y:units = "m" ;
>>> float z(z) ;
>>> z:units = "m" ;
>>> z:long_name = "height" ;
>>> float time(time) ;
>>> time:units = "d" ;
>>> time:long_name = "time" ;
>>> float p(z) ;
>>> p:units = "mb" ;
>>> p:long_name = "pressure" ;
>>> float zi(z) ;
>>> zi:units = "m" ;
>>> zi:long_name = "intfc_ht" ;
>>> float U(time, z, y, x) ;
>>> U:long_name = "X Wind
>>> Component" ;
>>> U:units = "m/s   " ;
>>> float V(time, z, y, x) ;
>>> V:long_name = "Y Wind
>>> Component" ;
>>> V:units = "m/s   " ;
>>> float W(time, z, y, x) ;
>>> W:long_name = "Z Wind
>>> Component" ;
>>> W:units = "m/s   " ;
>>> float PP(time, z, y, x) ;
>>> PP:long_name = "Pressure
>>> Perturbation 

Re: [Paraview] segmentation fault with netCDF when loading state file

[David Lonie]

On Wed, Oct 28, 2015 at 9:47 PM, Ryan Abernathey 
wrote:

> Yes, I can confirm that this state loads without crashing in 4.4.
> The fields are not actually accessible at the current timestep (appear
> with a "?"), but can be rewound to the beginning and restarted without
> crashing.
>

By fields, I assume you are referring to the data arrays in the information
panel? These seem to load correctly for me here, but then the sample data I
have appears to only be a single timestep. Do you have a small dataset that
reproduces this issue?

Thanks,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Labels using mathtext notation

[David Lonie]

The bugfix is in master, see
https://gitlab.kitware.com/vtk/vtk/merge_requests/955.

Thanks again for the report!

On Tue, Dec 1, 2015 at 1:56 PM, David Lonie <david.lo...@kitware.com> wrote:

> Hi Steven,
>
> Thanks for the report. I've confirmed this issue and opened a bug for it:
> http://www.vtk.org/Bug/view.php?id=15859
>
> I'll try to have this fixed soon, but in the meantime, it looks like
> adding some whitespace between the $$ (e.g. "$x $") will workaround the bug.
>
> Thanks,
> Dave
>
> On Tue, Nov 24, 2015 at 12:19 PM, Hahn, Steven E. <hah...@ornl.gov> wrote:
>
>> Hi,
>>
>> I am trying to generate a plot using ParaView v5.0.0-RC1 where the axes
>> contain mathtext labels. If the label is 2 or more characters long, then
>> everything is OK. If my label is one character only, it plots the $ signs.
>>
>> Here is a simple way to reproduce this issue:
>> - Use a sphere source.
>> - Show the axes
>> - Edit axes grid
>> - At the end of X Title string add ³$x$²
>> - At the end of Y Title string add ³$yy$²
>>
>> The X axis title is ³X Axis $x$². The Y axis title is ³Y Axis yy² where yy
>> is italic.
>>
>> Best,
>> Steven Hahn
>>
>> ___
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview
>>
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Labels using mathtext notation

[David Lonie]

On Thu, Dec 3, 2015 at 5:42 PM, Hahn, Steven E.  wrote:

> Hi Dave,
>
> Thank you for looking into and quickly fixing this issue. Will this bugfix
> be included in ParaView 5?
>

Yes, this should be fixed in the next release.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Labels using mathtext notation

[David Lonie]

Hi Steven,

Thanks for the report. I've confirmed this issue and opened a bug for it:
http://www.vtk.org/Bug/view.php?id=15859

I'll try to have this fixed soon, but in the meantime, it looks like adding
some whitespace between the $$ (e.g. "$x $") will workaround the bug.

Thanks,
Dave

On Tue, Nov 24, 2015 at 12:19 PM, Hahn, Steven E.  wrote:

> Hi,
>
> I am trying to generate a plot using ParaView v5.0.0-RC1 where the axes
> contain mathtext labels. If the label is 2 or more characters long, then
> everything is OK. If my label is one character only, it plots the $ signs.
>
> Here is a simple way to reproduce this issue:
> - Use a sphere source.
> - Show the axes
> - Edit axes grid
> - At the end of X Title string add ³$x$²
> - At the end of Y Title string add ³$yy$²
>
> The X axis title is ³X Axis $x$². The Y axis title is ³Y Axis yy² where yy
> is italic.
>
> Best,
> Steven Hahn
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] antialiasing in export scene, vector graphics, font selection

[David Lonie]

On Fri, Jun 3, 2016 at 5:03 PM, Mark Lohry  wrote:

> 1) Assuming "export scene" is the preferred way to output images, how do I
> get anti-aliasing of the produced image? I see no option for it. Failing
> that -- if I choose to not rasterize the image, I'm losing some color
> information whereas the save screenshot function correctly outputs the
> color (see attached figures).
>

The raster-format exports tend to just grab the image directly from the
OpenGL framebuffer, but unfortunately anti-aliased rendering is not
currently enabled in ParaView as it interferes with some of the other
rendering passes. There is some work-in-progress on adding DPI options to
paraview exports that will help with this issue in a future release.

I'm not sure what's going on with the color in the vector export. The GL2PS
library used to generate the vector graphics files is somewhat fickle with
some coloring methods, you may be able to play with the rendering settings
to get that to come out right. I know that, for instance, the "Interpolate
Colors Before Mapping" option will often cause issues.

Right now your best bet for getting an anti-aliased export from ParaView is
to emulate SSAA by making a screenshot with a custom resolution. By setting
the resolution in File > Save Screenshot to a larger resolution than your
desired image size and then shrinking the image externally you should get a
smoother result.


>
> 2) In the color legend font properties, only Arial, Courier, and Times are
> available. Is there a way to use some other font from my system?
>

This is currently not possible in ParaView. There is support for custom
fonts in VTK (the library that drives ParaView's rendering), but it has not
been exposed in ParaView.

Hope this helps,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Minor bug in OPENFOAM import filter

[David Lonie]

On Wed, Jun 22, 2016 at 8:04 AM, Eugene de Villiers 
wrote:

> Hi,
>
>
>
> We have come across a minor, but annoying bug in the Paraview OPENFOAM
> reader.
>

Hi Eugene,

I'll take a look at this. I'm not terribly familiar with openfoam files, so
if you could send me that small example to reproduce the bug I'd appreciate
it :)

Thanks,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Opacity Issues

[David Lonie]

On Fri, Jun 24, 2016 at 11:15 AM, Paluszek, Lukasz <
lukasz.palus...@airbus.com> wrote:

> Hello,
>
>
>
> In Paraview 5.1.0 with OpenGL2, I have black slices throughout the model
> when opacity is on in parallel mode (see attached picture). When switching
> depth peeling off, the black slices disappear but the image gets blurred
> and some pixels glow with random colours. I can’t reproduce it with 5.0.0,
> my NVidia driver is 361.42. Does anyone else experience similar behaviour?
>

Interesting, I haven't seen this before. Does it happen for all depth
peeled transparent rendering, or just certain scenes? If it's only certain
scenes, can you provide a state file or details on the scene?

Also, which OS is this?

Thanks,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Minor bug in OPENFOAM import filter

[David Lonie]

Hi Eugene,

On Wed, Jun 22, 2016 at 11:45 AM, Eugene de Villiers  wrote:

> Attached find a small case
>

Thanks -- I've loaded this up and can reproduce the error. The error occurs
in vtkFoamEntryValue::ReadList, which contains the following comment:

// general-purpose list reader - guess the type of the list and read

// it. only supports ascii format and assumes the preceding '(' has

// already been thrown away.  the reader supports nested list with

// variable lengths (e. g. `((token token) (token token token)).'

// also *supports compound of tokens and lists (e. g. `((token token)*

// *token)') only if a list comes as the first value.*


So the list (1.1 (1 2 3)) is being parsed as a list, and our reader
only supports lists

containing mixed tokens / lists if the lists precede the tokens. And
indeed, changing

file to read ((1 2 3) 1.1) will eliminate the error. The current
parser assumes that the

remainder of the list will also be scalars and attempts to read them
in -- then it

chokes when it encounters a '(' instead of another scalar.


Again, I'm not familiar with the openFOAM format, but would writing
the list with the

nested list first be feasible on your end? I'm not sure if this is a
format restriction or

an implementation detail of our reader.


Alternatively, I saw some discussion around skipping every entry in
boundaryField

specifications other than 'value'. This would likely be quite easy to
implement. My

only question is, would it break anything for the visualization?


Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Opacity Issues

[David Lonie]

On Fri, Jun 24, 2016 at 1:10 PM, Paluszek, Lukasz <
lukasz.palus...@airbus.com> wrote:

> It doesn’t happen all the time, I suppose you need a fairly large surface
> dataset. I managed to reproduce the problem on a wavelet source (attached
> state file), try increasing the number of elements if you do not see any
> artifacts. I have bult Paraview from source on OpenSuse 13.1 but also
> cross-checked with the Linux binaries from Paraview.org and had same
> issues. There was no problem with the osmesa-llvm flavor though.
>

I couldn't reproduce this here on arch linux with a newer nVidia driver
(367.27), so I'm suspicious that this is a driver bug that's affecting the
new depth peeling implementation.

Since you're building from source, can you try editing
VTK/Rendering/OpenGL2/vtkOpenGLRenderer.cxx, and change line 322

if (dualDepthPeelingSupported)

to

if (false)

in DeviceRenderTranslucentPolygonalGeometry, recompile using the OpenGL2
backend, and let me know if the issue goes away? I suspect it will since
both the OpenGL1 backend and OpenGL2 mesa/llvm do not use the new code.

I'll put together a patch to force the older peeling implementation via an
environment variable to make the work-around easier to apply.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Opacity Issues

[David Lonie]

This patch fixes the remote-rendering issue in the new peeling
implementation:

https://gitlab.kitware.com/vtk/vtk/merge_requests/1589

Cheers,
Dave

On Mon, Jun 27, 2016 at 6:06 AM, Paluszek, Lukasz <
lukasz.palus...@airbus.com> wrote:

> Switching to the old peeling implementation did the trick, no more issues
> and yes I am using remote rendering via pvserver.
>
>
>
> [image: Beschreibung: Beschreibung: Description :
> cid:image001.png@01CF0B8F.DB7831F0]
>
> *Łukasz Paluszek*
>
> Airbus Helicopters
>
> Aerodynamics Department (ETGA)
>
> 86607 Donauwörth
>
> Germany
>
> Tel.: +49 (90) 671 8231
>
> Fax.: +49 (90) 671 9122
>
> lukasz.palus...@airbus.com
>
>
>
> *From:* David Lonie [mailto:david.lo...@kitware.com]
> *Sent:* 24 June 2016 21:22
> *To:* Paluszek, Lukasz
> *Cc:* paraview@paraview.org
> *Subject:* Re: [Paraview] Opacity Issues
>
>
>
> On Fri, Jun 24, 2016 at 2:15 PM, David Lonie <david.lo...@kitware.com>
> wrote:
>
> On Fri, Jun 24, 2016 at 1:10 PM, Paluszek, Lukasz <
> lukasz.palus...@airbus.com> wrote:
>
> It doesn’t happen all the time, I suppose you need a fairly large surface
> dataset. I managed to reproduce the problem on a wavelet source (attached
> state file), try increasing the number of elements if you do not see any
> artifacts. I have bult Paraview from source on OpenSuse 13.1 but also
> cross-checked with the Linux binaries from Paraview.org and had same
> issues. There was no problem with the osmesa-llvm flavor though.
>
>
>
> I couldn't reproduce this here on arch linux with a newer nVidia driver
> (367.27), so I'm suspicious that this is a driver bug that's affecting the
> new depth peeling implementation.
>
>
>
> Since you're building from source, can you try editing
> VTK/Rendering/OpenGL2/vtkOpenGLRenderer.cxx, and change line 322
>
>
>
> if (dualDepthPeelingSupported)
>
>
>
> to
>
>
>
> if (false)
>
>
>
> in *DeviceRenderTranslucentPolygonalGeometry*, recompile using the
> OpenGL2 backend, and let me know if the issue goes away? I suspect it will
> since both the OpenGL1 backend and OpenGL2 mesa/llvm do not use the new
> code.
>
>
>
> I'll put together a patch to force the older peeling implementation via an
> environment variable to make the work-around easier to apply.
>
>
>
> Patch is here: https://gitlab.kitware.com/vtk/vtk/merge_requests/1589
>
>
>
> By the way, are you running paraview with remote rendering via pvserver? I
> was just made aware of a similar issue with remote rendering and depth
> peeling.
>
>
>
> Dave
>
> AIRBUS HELICOPTERS DEUTSCHLAND GmbH
>
> Sitz der Gesellschaft / Registered Office: Donauwörth
>
> Registergericht / Registration Court: Amtsgericht Augsburg HRB 16508
>
> Vorsitzender des Aufsichtsrates / Chairman of the Supervisory Board:
> Guillaume Faury
>
> Geschäftsführung / Board of Management: Dr. Wolfgang Schoder, Vorsitzender
> / CEO; Thomas Hundt; Johannes Kleidorfer; Dr. Klaus Przemeck; Martin
> Schübel; Stefan Thomé
>
>
> The information in this e-mail is confidential. The contents may not be 
> disclosed or used by anyone other than the addressee. Access to this e-mail 
> by anyone else is unauthorised.
> If you are not the intended recipient, please notify Airbus immediately and 
> delete this e-mail.
> Airbus cannot accept any responsibility for the accuracy or completeness of 
> this e-mail as it has been sent over public networks. If you have any 
> concerns over the content of this message or its Accuracy or Integrity, 
> please contact Airbus immediately.
> All outgoing e-mails from Airbus are checked using regularly updated virus 
> scanning software but you should take whatever measures you deem to be 
> appropriate to ensure that this message and any attachments are virus free.
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Opacity Issues

[David Lonie]

On Fri, Jun 24, 2016 at 2:15 PM, David Lonie <david.lo...@kitware.com>
wrote:

> On Fri, Jun 24, 2016 at 1:10 PM, Paluszek, Lukasz <
> lukasz.palus...@airbus.com> wrote:
>
>> It doesn’t happen all the time, I suppose you need a fairly large surface
>> dataset. I managed to reproduce the problem on a wavelet source (attached
>> state file), try increasing the number of elements if you do not see any
>> artifacts. I have bult Paraview from source on OpenSuse 13.1 but also
>> cross-checked with the Linux binaries from Paraview.org and had same
>> issues. There was no problem with the osmesa-llvm flavor though.
>>
>
> I couldn't reproduce this here on arch linux with a newer nVidia driver
> (367.27), so I'm suspicious that this is a driver bug that's affecting the
> new depth peeling implementation.
>
> Since you're building from source, can you try editing
> VTK/Rendering/OpenGL2/vtkOpenGLRenderer.cxx, and change line 322
>
> if (dualDepthPeelingSupported)
>
> to
>
> if (false)
>
> in DeviceRenderTranslucentPolygonalGeometry, recompile using the OpenGL2
> backend, and let me know if the issue goes away? I suspect it will since
> both the OpenGL1 backend and OpenGL2 mesa/llvm do not use the new code.
>
> I'll put together a patch to force the older peeling implementation via an
> environment variable to make the work-around easier to apply.
>

Patch is here: https://gitlab.kitware.com/vtk/vtk/merge_requests/1589

By the way, are you running paraview with remote rendering via pvserver? I
was just made aware of a similar issue with remote rendering and depth
peeling.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] [EXTERNAL] Re: Rendering issue with cross-sections in 3D

[David Lonie]

Has anyone tested OpenGL1 vs. OpenGL2 backends on master?

It may be an issue with the recent changes to depth peeling. OpenGL1 will
still use the old implementation, OpenGL2 (not on mesa) will use the new
version.

On Fri, Jun 17, 2016 at 5:34 AM, Cory Quammen 
wrote:

> Folks,
>
> The bad news: I can reproduce this in ParaView 5.0.1 on a Mac given
> the data files Craig shared.
>
> The good news: I cannot reproduce it in my development version of
> ParaView 5.1 RC1. Mind downloading that version of ParaView and giving
> it a shot? I would guess there might have been a rendering bug in the
> OpenGL2 backend in VTK that has since been fixed.
>
> Thanks,
> Cory
>
>
>
> On Fri, Jun 17, 2016 at 3:15 AM, Scott, W Alan  wrote:
> > Steve,
> >
> > I worry it sounds a bit like a bug.
> >
> > Can you provide us with a toy dataset (or non-toy, if necessary) that
> replicates the issue?  Also, really, really good directions (+Y, roll mouse
> wheel in 4 times)?
> >
> > Alan
> >
> > -Original Message-
> > From: ParaView [mailto:paraview-boun...@paraview.org] On Behalf Of
> Steve Lamont
> > Sent: Thursday, June 16, 2016 7:12 PM
> > To: paraview@paraview.org
> > Subject: [EXTERNAL] Re: [Paraview] Rendering issue with cross-sections
> in 3D
> >
> >> I'm using Paraview v5.0.1 to visualize some geophysical data, but I'm
> >> running into a rendering issue. When I zoom in to my data sets, data
> >> in the background randomly jumps in front of cross-sections in the
> >> foreground.
> >
> > I'm running into a similar issue.
> >
> > I have three cross sections, along the three major axes, and they do not
> seem to occlude one another as one would expect.
> >
> > I've turned on depth peeling and have experimented with setting the
> maximum number of peels to no avail.
> >
> > I know this is a classic OpenGL depth sorting issue since I've run into
> myself back when I was coding such things.
> >
> > I've updated my nVidia drivers to what passes for the Latest and
> Greatest for Ubuntu Linux and I have a fairly recent graphics card (Quadro
> K2000D).  I'll be happy to supply any other system information I can if
> it'll help.
> >
> > Is there something else I need to switch on in order to get ParaView to
> operate in the "expected" manner?
> >
> > Thanx.
> >
> > spl
> >
> > ___
> > Powered by www.kitware.com
> >
> > Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
> >
> > Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
> >
> > Search the list archives at: http://markmail.org/search/?q=ParaView
> >
> > Follow this link to subscribe/unsubscribe:
> > http://public.kitware.com/mailman/listinfo/paraview
> > ___
> > Powered by www.kitware.com
> >
> > Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
> >
> > Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
> >
> > Search the list archives at: http://markmail.org/search/?q=ParaView
> >
> > Follow this link to subscribe/unsubscribe:
> > http://public.kitware.com/mailman/listinfo/paraview
>
>
>
> --
> Cory Quammen
> R Engineer
> Kitware, Inc.
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] ParaView Binary Installer 5.0 - Linux 64 bit - Scene Export : SVG, EPS and PDF formats missing

[David Lonie]

On Tue, Feb 9, 2016 at 11:59 AM, Tomislav Maric  wrote:
> I am using ParaView 5.0 from the 64 bit binary installer on on Arch
> Linux. When I click on File->Export Scene, I can't find the SVG, EPS and
> PDF formats.

Due to differences in the OpenGL features available to the VTK OpenGL2
backend (in particular, the deprecation of the feedback render mode),
the GL2PS exporter was not compatible with builds of ParaView that
used the new backend, such as the 5.0 release.

I've recently made some changes to gl2ps and our exporter that allow
vector export to work again, although with reduced functionality. The
new (OpenGL2) GL2PS exporter will only export the following as vector
graphics:
- Charts, etc (anything drawn via the vtkContext2D interface)
- 2D Text (as native text objects or generic path objects; see
vtkGL2PSExporter::TextAsPath)
- 2D MathText (always as path objects)
- 3D Text (always as path objects)

All other rendered objects (including polygonal data previously
exported as vectors by the old backend) will be rasterized to the
image's background.

This should be available in the next paraview release. In the
meantime, builds of ParaView that use the OpenGL VTK backend will
still have the old implementation of the exporter available.

Hope this helps,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Paraview 5.0.0 pdf, eps, svg export

[David Lonie]

Just to chime in, a new vector graphics backend is implemented for OpenGL2
and should be available in the new release. It's functionality is limited
compared to the old one (most things are just rasterized to the image
background with vectorized text, etc overlayed), but it will produce output
of the proper format.

Cheers,
Dave

On Wed, Feb 24, 2016 at 4:38 AM, Buesing, Henrik <
hbues...@eonerc.rwth-aachen.de> wrote:

> Dear Joachim,
>
>
>
> I will use Paraview 4.4 then. Thank you!
>
> Henrik
>
>
>
>
>
> *Von:* Joachim Pouderoux [mailto:joachim.pouder...@kitware.com]
> *Gesendet:* 24 February 2016 10:13
> *An:* Buesing, Henrik
> *Cc:* paraview@paraview.org
> *Betreff:* Re: [Paraview] Paraview 5.0.0 pdf, eps, svg export
>
>
>
> Henrik,
>
> As you might know, distributed binaries for ParaView 5.0.0 are based on
> the new VTK 'OpenGL2' rendering backend which provides better rendering
> efficiency. However, the VTK module for exporting visualization to a PDF or
> PS file is highly dependent on the rendering engine... and the adaptation
> of this module this new rendering backend is still in progress.
>
> To enable this feature again you will have to compile ParaView 5.0 with
> the legacy OpenGL(1) rendering backend, or use a previous version of the
> software (4.4).
>
> Best,
>
>
> *Joachim Pouderoux*
>
> *PhD, Technical Expert*
> *Kitware SAS *
>
>
>
> 2016-02-24 9:36 GMT+01:00 Buesing, Henrik  >:
>
> Dear all,
>
>
>
> I’m trying out Paraview 5.0.0. I would like to export vector graphics. In
> Paraview 4.3.1 I had “Export Scene” and could export pdf, ps, eps, svg etc.
> Now I only have pov, vrml, webgl, x3d, x3db (which say nothing to me). So
> how do I get my pdf, eps or svg?
>
> Thank you!
>
> Henrik
>
>
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Paraview 5.0.1: Export Scene to pdf -- appears in white

[David Lonie]

On Thu, May 19, 2016 at 10:16 PM, Utkarsh Ayachit <
utkarsh.ayac...@kitware.com> wrote:

> + Not sure what's the deal with 4.0.1, but it's too old at this point
> so let's ignore that.
> + When you downloaded ParaView 5.0.1 from paraview.org, starting with
> 5.0.1, we are using a new implementation of the rendering backend for
> our binaries [1]. This rendering backend, let's call OpenGL2, doesn't
> support exporting to PDF, at least, not in 5.0.1. It's been fixed in
> git "master".
> + I suspect ParaView 5.0.1 from Ubuntu is using the old rendering
> backend by default. The title bar should say something like "ParaView
> 5.0.1 (Legacy Rendering Backend)". This backend still support the old
> export to PDF code and hence should have that option.
> + When you compiled 5.0.1 from source, I suspect you are getting the
> old rendering backend too (unless you changed the CMake variable
> VTK_RENDERING_BACKEND to OpenGL2). Not entirely sure why export would
> fail in that case. There isn't any special flag to enable PDF export.
>
> However, the best option to try out Export to PDF would be build the
> latest "master" from source with VTK_RENDERING_BACKEND set to OpenGL2.
>
> (Dave, do correct me if I am misspoke).
>

This sounds about right. The gl2ps library we use to generate the pdfs was
not compatible with the new OpenGL2 backend, and I had to tweak it a bit to
get it working again. Looking at the git log, the modifications didn't make
it into the 5.0.1 release, so export would only work using the old backend
with that release.

The current development branch supports pdf export on both backends.

I'm curious about the white exported image, though. IIRC, the
vector-graphics export options shouldn't be shown on the versions of
paraview that don't support gl2ps, so if you're able to export, it should
be working. It sounds like you may have also brushed up against one of the
limitations of the gl2ps library -- for instance, it can't handle textured
objects (they'll just appear a solid color, rather than patterned). This
can be problematic as some of the color mapping techniques use textures.

If you're seeing incorrect output, try playing with the "rasterize 3D
geometry" and sort options to see if you can get a better representation of
the scene.

HTH,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Paraview 5.0.1: Export Scene to pdf -- appears in white

[David Lonie]

On Fri, May 20, 2016 at 10:19 AM, Ramiro James Rebolledo Cormack <
ramre...@gmail.com> wrote:
>
> I compile paraview in my laptop (ubuntu 16.04) and my desktop (ubuntu
> 14.04) with exactly the same options in ccmake, in particular: OpenGL.
> In my laptop 'Export to pdf' works perfect.
> Is this related this the nvidia driver? In my desktop I have installed
> nvidia driver, but in my destop I have not installed nvidia driver.
>

Hmm, the opengl driver shouldn't matter on the OpenGL1 backend, unless
there was a bug in their feedback rendering mode. Are you trying to export
the same scene on both systems?


> + When you compiled 5.0.1 from source, I suspect you are getting the
>> old rendering backend too (unless you changed the CMake variable
>> VTK_RENDERING_BACKEND to OpenGL2). Not entirely sure why export would
>> fail in that case. There isn't any special flag to enable PDF export.
>>
>
> Now, in my desktop I'm trying compiling with OpenGL2 option (the option
> doesn't comes hiting the enter button, instead of my laptop, but I chosed
> writting it), and crossing my fingers.
>
> I appreciate any ideas or suggestions (I don't have more), to find a way
> for fix this 'little' problem: to have the export to pdf option.
>

I'd suggest just playing with the export options and see if there's a
solution there. If you keep having trouble, send me a small state file that
reproduces the issue and I'll take a look. There are certain cases where
the gl2ps exporter will fail to capture all of the details in a scene,
though it should produce more than an empty white image.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] a lot of Generic Warning for vtkBuffer in PV5.1

[David Lonie]

On Wed, Aug 10, 2016 at 9:48 AM, claire.guilb...@cea.fr
 wrote:
> When running ParaView 5.1 in debug with DEBUG_LEAKS=ON, I have several times
> the following message :
>
> Generic Warning: In …/vtkDebugLeaks.cxx, line 285:
>   Deleting unknown object: vtkBuffer
>
> When I close PV, there are a lot of vtkBuffer instances still
> around.

It looks like the vtkDebugLeaks reference manager is getting confused
by the class template. It appears that vtkBuffer is being registered
as 'vtkBuffer' and unregistered as 'vtkBuffer'.

The good news is that you can likely ignore these for now -- they
aren't actually leaking, but the object tracker is getting confused.
I'll see about fixing this soon.

Thanks for the report!
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Depth rendering issues on ParaView Desktop 5.1.0

[David Lonie]

This looks like the front faces of the surface have been culled. In the
properties tab, enable advanced properties (the gear button) and locate the
"Backface representation" setting. It's usually set to "Follow Frontface",
but "Cull Frontface" will give results like you're seeing.

Does changing this value help?

On Wed, Aug 3, 2016 at 10:53 PM, Júlio Hoffimann 
wrote:

> Dear all,
>
> I have a depth rendering issue that is now compromising my work on a Arch
> Linux laptop. This is a simple Dell XPS 13 with an integrated Intel(R) HD
> Graphics Skylake ULT GT2 GPU.
>
> The attached figure shows a 3D volume with incorrect depth rendering:
>
>
> ​
> Is this an issue with the graphics card driver? How can I solve it?
>
> I appreciate any guidance.
>
> Sincerely,
> -Júlio
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] a lot of Generic Warning for vtkBuffer in PV5.1

[David Lonie]

Oops, I sent that too quickly -- the vtkBuffer.h line should actually be:

   vtkTemplateTypeMacro(vtkBuffer, vtkObject)

with the template argument.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Minor bug in OPENFOAM import filter

[David Lonie]

On Sat, Jun 25, 2016 at 1:02 PM, Eugene de Villiers 
wrote:

> Hi Takuya, David,
>
> That would be perfect. What kind of timeline are we looking at to see this
> in a release version?
>

It's difficult to say for sure. It might be out in the next patch release,
but it might have to wait for the next major/minor release. The openFOAM
reader is a complicated bit of code, I'm still investigating the best way
to go about implementing this. I'll update this thread once I know more.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Minor bug in OPENFOAM import filter

[David Lonie]

On Tue, Jun 28, 2016 at 3:06 PM, David Lonie <david.lo...@kitware.com>
wrote:

> On Sat, Jun 25, 2016 at 1:02 PM, Eugene de Villiers <
> e.devilli...@engys.com> wrote:
>
>> Hi Takuya, David,
>>
>> That would be perfect. What kind of timeline are we looking at to see
>> this in a release version?
>>
>
> It's difficult to say for sure. It might be out in the next patch release,
> but it might have to wait for the next major/minor release. The openFOAM
> reader is a complicated bit of code, I'm still investigating the best way
> to go about implementing this. I'll update this thread once I know more.
>

Good news -- I found an easy fix. Instead of failing outright when

(scalar (scalar scalar))

is found because the internal data structures can't handle nested lists
following scalar data, I patched the reader to simply read the nested list,
discard it, print a warning, and continue reading the file.

You'll still see a warning pop up in paraview, but the file should still
read successfully as long as the unsupported list does not appear in a
field used for visualization.

The patch is here: https://gitlab.kitware.com/vtk/vtk/merge_requests/1598

If this is agreeable to everyone, we'll get it merged and incorporated into
a future release.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Depth rendering issues on ParaView Desktop 5.1.0

[David Lonie]

The only other thing I can think of is that it might be an issue with
the new depth peeling code, which is used to blend layers of
translucent geometry. However, from the look of the screenshots, it
appears that all of the geometry is opaque, in which case depth
peeling wouldn't be used.

Still, you can try setting the environment variable
"VTK_USE_LEGACY_DEPTH_PEELING" to force the older depth peeling code.
This was added for 5.1.2, so you'll need to test that on the new
release.

Other than that...I'm out of ideas, unfortunately. Maybe try an older
4.x version of paraview?

Maybe someone else will have some idea what may be happening. It
appears that the depth test has been reversed on these builds somehow.

On Fri, Aug 5, 2016 at 9:04 AM, Júlio Hoffimann
 wrote:
> Thank you David, I tried with Mesa and the result is the same. I'm gonna try
> the newer release, but I suspect it won't solve the issue.
>
> Any information that I can export and that can help figure out the root of
> the problem?
>
> -Júlio
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Depth rendering issues on ParaView Desktop 5.1.0

[David Lonie]

That is very odd. I use arch linux as well and don't see this here.

I'm suspecting a graphics driver issue at this point. Try running paraview
with the "--mesa-llvm" command line argument and see if it fares better.
That will force paraview to use a software-based OpenGL implementation and
bypass the GPU and system OpenGL drivers.

You may also want to try the slightly newer 5.1.2 release here:
http://www.paraview.org/download/

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Depth rendering issues on ParaView Desktop 5.1.0

[David Lonie]

On Mon, Aug 8, 2016 at 9:55 PM, Júlio Hoffimann
 wrote:
> According to the Arch Linux documentation
> (https://wiki.archlinux.org/index.php/xorg#Driver_installation), the Intel
> driver has priority over the others, is it being picked up in the log?

I'm afraid this is going beyond my ability to help, I'm not terribly
familiar with these details. But it does look to me like the drivers
are loaded correctly in the log.

> I updated ParaView from AUR, could you confirm that 5.x.x releases are
> working with Arch? I only have another machine with Ubuntu to test on.

I updated my arch box yesterday, and can confirm the the 5.x branches
are working correctly, both using the proprietary nVidia drivers and
the bundled mesa-llvm implementation. I don't have an intel card to
test on, but the mesa-llvm test rules out the hardware driver.

The only thing left I can think of would be to bisect the source, but
that can be a very time consuming process. If you locate the commit
that's causing the error, perhaps it would shed some light on what's
happening.

I'm out of ideas at this point -- maybe someone else will have some suggestions?

Good luck,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Depth rendering issues on ParaView Desktop 5.1.0

[David Lonie]

Hi Julio,

On Sun, Aug 7, 2016 at 2:01 PM, Júlio Hoffimann
 wrote:
> Hi David,
>
> I downloaded all 4.x.x releases and everything works as expected. I
> downloaded ParaView 5.0.1 and ParaView 5.1.2, both give me the following
> error message about the driver:
>
> libGL error: unable to load driver: i965_dri.so
> libGL error: driver pointer missing
> libGL error: failed to load driver: i965
> libGL error: unable to load driver: i965_dri.so
> libGL error: driver pointer missing
> libGL error: failed to load driver: i965
> libGL error: unable to load driver: swrast_dri.so
> libGL error: failed to load driver: swrast
> Segmentation fault (core dumped)
>
> Based on that outcome, I confirm that this is a bug introduced somewhere in
> between the 4.x.x and 5.x.x releases. How can we go and investigate this
> further? I can print more information about my system in case it is needed.

Are those drivers installed correctly? Looks like they're provided by
the intel-dri and mesa packages on archlinux:

$ pkgfile i965_dri.so
extra/intel-dri
$ pkgfile swrast_dri.so
extra/mesa

Those errors, and that the 5.x releases are working on other archlinux
installs, make me think this is a configuration issue with this
specific machine. The X log file /var/log/Xorg.0.log may have some
clues if there's a problem initializing the driver support needed for
the newer paraview. I'm not convinced this is a bug in ParaView just
yet -- the 5.x versions require different OpenGL contexts than the 4.x
versions, so it may be that something is not quite right for the newer
contexts.

Have you tried building from source and setting VTK_RENDERING_BACKEND
to OpenGL (instead of OpenGL2) in cmake? That would build paraview
with the older backend used by the 4.x line, and might work better for
you.

The only other idea that comes to mind would be to grab the
development branch from git and use 'git bisect' to identify the
commit that introduced the functionality that's failing on this
installation:

Getting the sources from git:
https://gitlab.kitware.com/paraview/paraview/blob/master/Documentation/dev/git/develop.md

Git bisect:
http://stackoverflow.com/questions/4713088/how-to-use-git-bisect

Hope this helps,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Opacity Issues

[David Lonie]

On Wed, Jun 29, 2016 at 3:48 AM, Paluszek, Lukasz <
lukasz.palus...@airbus.com> wrote:

> I tested the patch by simply replacing the two modified files and the new
> peeling worked with parallel rendering.
>

Great!


> I noticed, however, that I cannot load any state files saved before the
> patch application (or even from previous versions), Paraview segfaulted
> each time. I did not investigate the matter any further but perhaps you
> could double check that there is no backward compatibility issue with older
> state files that do have opacity information.
>

The changes only affected shader code in the rendering engine, this
shouldn't make any difference in the ParaView state. Do they still segfault
in master without patching the files for the opacity fix?

In addition the peeling, regardless whether it’s the old or new
> implementation, takes significantly more time with parallel rendering than
> in the serial mode. Although, it’s expectable, the overhead is suspiciously
> high, I tried on 8 processes and a geometry which, in serial mode can be
> made opaque dynamically with no visible delay, took 34 seconds to be
> rendered. Most of the overhead comes from parallel distribution for
> compositing.
>

Was it 34 seconds for the initial frame, or for every frame? If the dataset
is large a longer first render would be expected as the data propagates.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] a lot of Generic Warning for vtkBuffer in PV5.1

[David Lonie]

Hi Claire,

Sure, if you can send me the files I'll push them for review/merging.

Thanks,
Dave

On Tue, Sep 6, 2016 at 3:53 AM, GUILBAUD Claire <claire.guilb...@cea.fr> wrote:
> Hi David,
>
> The trouble happens when ParaView is compiled with
> VTK_ALL_NEW_OBJECT_FACTORY=ON.
> It seems there is a naming problem when using factories: the name used in
> the New method is not the same as the one used in the destructor.
> One of my colleague fixed it: in VTK/Common/Core, he modified
> vtkObjectFactory.h and .cxx, vtkPoints.cxx (New method), vtkPoints2D.cxx
> (New method).
> If you are interested, I can send you those 3 files.
>
> Have a good day.
>
> Claire
>
> On 08/12/16 15:21, David Lonie wrote:
>
> Oops, I sent that too quickly -- the vtkBuffer.h line should actually be:
>
>vtkTemplateTypeMacro(vtkBuffer, vtkObject)
>
> with the template argument.
>
> Dave
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

[Paraview] At last -- Anti-aliasing in ParaView!

[David Lonie]

I'm happy to announce that ParaView now includes an option for FXAA
anti-aliasing, which works quite nicely for both serial and parallel
rendering. It is available in the current master branch on gitlab/github,
but is disabled by default so you'll need to check some boxes in the
Settings to use it.

You can read more about it on the Kitware blog:

https://blog.kitware.com/new-fxaa-anti-aliasing-option-in-paraviewvtk/

Big thanks to Sandia National Labs for supporting this work.

Cheers,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] VTK File Format

[David Lonie]

That's correct -- the cell information specifies the points by index. You
can find more information on the VTK file formats here:

http://www.vtk.org/wp-content/uploads/2015/04/file-formats.pdf

HTH,
Dave

On Wed, Oct 5, 2016 at 9:19 AM, Sellmann, Kai Asaad <
asaad.sellm...@rwth-aachen.de> wrote:

> Hello everyone!
>
> I have created a vtk File with point coordinates, and cell definitions. As
> I am getting distorted surfaces I was wondering what it could be.
>
> The coordinates are nodal coordinates taken from a FE Model. The elements
> are defined by the IDs of their corresponding nodes.
>
>
> So it looks kinda like this
>
>
> ...
>
> CELLS 59256 651816
> 10 21615 1 32020 21623 2 3 4 21625 5 6
> 10 21623 21615 1 21624 21625 2 5 21627 21626 7
> 10 8 21623 21612 21615 9 21631 10 11 21625 21621
> 10 34096 12 34106 1 13 14 34110 15 16 17
> 10 22735 18 32680 32663 19 20 21 22 23 32682
>
> ...
>
>
> The 10 indicating the number of information "coming up".
>
> So first of all I am little bit confused and would like to know if the
> cells Information is refering to the prior defined Nodes and their IDs?
>
> For example the first line "10 21615 1 32020 21623 2 3 4 21625 5 6" means
> that the first cell consists of the Nodes 21615, 1, 32020, 21623, 2, 3, 4,
> 21625, 5 and 6, right? And these nodes are defined like this:
>
>
> ...
>
> POINTS 104667 FLOAT
> 0 0 0
> -0.00133054 -0.0651073 0.00296884
> -0.00366651 -0.0661655 0.00448453
> -0.00430551 -0.0647424 0.00148457
> -0.00664148 -0.0658006 0.00300025
> -0.00257653 -0.0641143 0.00448451
> -0.0055515 -0.0637484 0.00300024
> -0.00147854 -0.0663074 0.00448453
> -0.00974145 -0.0646966 0.00299095
>
> ...
>
>
> So the second node (1) of the first element would have the coordinates
> -0.00133054, -0.0651073 and 0.00296884, correct?
>
> If someone has an idea I could send you the .vtk-file
>
>
> Thnaks,
>
> asaad
>
>
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Paraview 4.4.0 crashes when exporting scene without rasterization of 3D objects

[David Lonie]

On Fri, Oct 28, 2016 at 2:46 PM, Simon Couture 
wrote:

> I try to export scene with the cylinder source as I described in my first
> e-mail, I save it to .svg format, unchecking the Rasterize 3D geometry.
> This time Paraview does not crash, however upon opening the .svg file I
> find that the content is an embedded image, not a vector graphic, that is,
> unchecking the Rasterize 3D geometry option gives the same result as when
> this option is checked.
>

Yes, unfortunately this is correct. Currently, with the OpenGL2 backend,
all 3D geometry will be rasterized, and only things like text and 2D charts
are vectorized. It would be possible to fix this, but it's a significant
effort that hasn't found funding yet.

I think your best option at this point to get vectorized 3D geometry
exported is to try building ParaView yourself with the OpenGL1 backend
using the superbuild:

https://gitlab.kitware.com/paraview/paraview-superbuild/

This can be tricky to get right, and I'm not sure how well documented the
build process for the superbuild is on windows. Alternatively you could try
an older version of paraview than 4.4. I think the vector format export was
added around v4.0.

HTH,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Vectorized data assignment

[David Lonie]

On Mon, Nov 7, 2016 at 9:00 AM, Christian Gabriel <
cgabr...@matrix-solutions.com> wrote:

> Good day.
>
>
>
> I found this example code
> 
> :
>
>
>
> # Create the points
>
> points = vtk.vtkPoints()
>
> points.InsertNextPoint(0.0, 0.0, 0.0)
>
> points.InsertNextPoint(1.0, 0.0, 0.0)
>
> points.InsertNextPoint(1.0, 1.0, 0.0)
>
> points.InsertNextPoint(0.0, 1.0, 0.0)
>
> … … …
>
>
>
> which had me wonder if there is a way to assign data in a vectorized
> fashion instead of point by point?
>
>
>
> Let’s say I can read in 3 1D arrays (vectors) with X, Y, Z data or one 2D
> array (matrix, dataframe) with the same data in columns, is there a way to
> assign that data to the points all at once instead of iteratively?
>
>
>
> points = vtk.vtkPoints()
>
> points.InsertNextPoint(X, Y, Z)- just saying; this is obviously
> not correct!
>
>
>
> Obviously, any data could be combined and reshaped first to whatever
> format required ….
>
>
>
> Given that ParaView/VTK was designed to work with (very) large data sets
> I’d be surprised if there were no vectorized operations implemented.
>

In C++, there are a number of ways to do this by modifying the data array
directly (instead of going through the vtkPoints interface). You can
extract the data array from the points by calling points.GetData(). You can
also create a points array separately and assign it to the vtkPoints
instance by calling points.SetData(myArray).

There are two types of arrays in VTK, AoS (Array-of-Structs, the default
used for e.g. vtkFloatArray, etc), and SoA (Struct-of-Arrays, still
somewhat experimental at this point):

These doxygen pages describe the specific APIs for them:

http://www.vtk.org/doc/nightly/html/classvtkAOSDataArrayTemplate.html
http://www.vtk.org/doc/nightly/html/classvtkSOADataArrayTemplate.html

There are several ways to do low-level copying of data using methods like
AoS::SetArray, SoA::SetComponentArray, AoS::GetPointer,
SoA::GetComponentArrayPointer, etc, which will allow you to set the array's
memory buffer explicitly, or just get low-level access to it for for
modifications.

However, I'm not sure how much of this is exposed in the python/numpy
layers, so it might take some experimentation. Hopefully someone more
familiar with the python bindings will chime in.

HTH,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] ParaView 5.2 Release Candidate 3 available for download

[David Lonie]

On Fri, Nov 4, 2016 at 10:18 AM, Armin Wehrfritz  wrote:

> Hi,
>
> I tested 5.2 RC3 and encountered an issue when reading OpenFOAM cases.
>
> The problem only occurs when the OpenFOAM data (i.e. the mesh or fields)
> are written in binary format. I setup two test cases that are otherwise
> identical, where for the one the data is in ascii and the other one in
> binary format.
> With ParaView 5.1.2 I can read both cases without any error, but with
> ParaView 5.2 RC3 the cases for which the data is in binary format leads
> to an error (see below for the exact error message).
>
> The case is taken directly from the official OpenFOAM 4.x tutorials, and
> here is a link to the case files that can be directly used for debugging:
> https://www.dropbox.com/s/l0cb8vd6pjkb1zz/debug_PV_OF.tar.gz?dl=0
>
> All tests were done with the ParaView binaries distributed by Kitware.
>
> Let me know if you need more information or if I can help debugging this
> further.
>

Thanks for the bug report! I have a patch for this here:

https://gitlab.kitware.com/vtk/vtk/merge_requests/2147

This should be able to make it into the upcoming release.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Paraview 4.4.0 crashes when exporting scene without rasterization of 3D objects

[David Lonie]

I just grabbed the 4.4 release on linux and it works fine, so it sounds
like something specific to the windows builds.

4.4 is quite dated at this point. Do you still see this issue on newer
versions of paraview?

http://www.paraview.org/download/

Dave

On Fri, Oct 28, 2016 at 12:27 AM, Simon Couture 
wrote:

> Hello all,
>
>
> I want to export a paraview scene to a vector graphic format but for some
> reason Paraview crashes during the operation.
>
>
> The version that I am running is from the installation file
> ParaView-4.4.0-Qt4-Windows-64bit.exe. I have tried the export operation
> on two different machines, one with Windows 10, the other with Windows 7,
> with the same result.
>
>
> The most straightforward way to reproduce my problem is the following:
> from the Sources menu I create a cylinder with default height 1 and radius
> 0.5. I then go to File -> Export scene..., I type a filename to save to, I
> tried selecting formats .eps, .svg and .pdf all with the same result, then
> in the "Export Options" window that pops up, I uncheck the Rasterize 3D
> geometry option, click save, and I get the "paraview.exe" has stopped
> working message (actually I just tried to save with the Rasterize 3D
> geometry option checked, and that also crashes paraview).
>
>
> Has anyone encountered similar behavior? Any help is appreciated.
>
>
> Simon
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Can't export to vector graphics in ParaView 5.0

[David Lonie]

The GL2PS library that we use to generate the vector exports didn't support
our new rendering backend, so there was a brief period of time where
VTK/ParaView applications using the OpenGL2 backend had to have the vector
graphics export disabled. This included the 5.0 release. Since then we've
modified GL2PS to work around some of these issues and new versions of
ParaView have restored support for it.

HTH,
Dave

On Tue, Oct 11, 2016 at 2:05 PM, Nicolás Guarín-Zapata  wrote:

> Hello,
>
> In my laptop with Windows 10, ParaView 5.0 can't export to vector graphics
> formats. You can see a screenshot here:
>
> https://twitter.com/nicoguaro/status/785890202166321152?s=09
>
> Nevertheless, I just downloaded and installed version 5.2 RC1 and the
> options appear. Is this a known problem?
>
> Thanks,
> Nicolás
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Can't export to vector graphics in ParaView 5.0

[David Lonie]

On Tue, Oct 11, 2016 at 2:54 PM, Nicolás Guarín-Zapata  wrote:

> It might be known in the mailing list. But it is not written in the wiki
>
> http://www.paraview.org/Wiki/ParaView/Vector_Graphics_Export


Updated.
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] [Paraview-developers] ParaView 5.2 Release Candidate 1 available for download

[David Lonie]

On Thu, Oct 13, 2016 at 7:18 AM, Gena Bug via ParaView <
paraview@paraview.org> wrote:

> On 11.10.2016 18:15, Utkarsh Ayachit wrote:
>
>> Folks,
>>
>> Among several new features, this release includes anti-aliasing
>> support [1]. Currently, it's off by default. I'd encourage all testers
>> and early adopters to turn it on and report any issues you notice. If
>> we don't hear of any major obstacles, we would anti-aliasing on my
>> default in the final release (or subsequent RCs)
>>
> I attached two screenshots of a simple 2d cartesian mesh, one with
> anti-aliasing enabled and one without it. For me the one with AA looks
> strange -- cells look like circles not squares. Is it expected behavior?
>

Unfortunately it is. FXAA is an approximate anti-aliasing solution that's
applied as a post-processing pass over the rendered scene -- since ParaView
often does distributed rendering, this is really the only way we can
reliably get anti-aliasing without introducing major artifacts at processor
boundaries.

Because it's only working with the rendered pixel data (instead of actual
geometric primitives), there are some occasional issues like this one. In
this case, it's detecting the intersections as aliased edges and attempting
to smooth them out, fattening the line near the intersection as a result.

HTH,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] ParaView 5.2 Release Candidate 1 available for download

[David Lonie]

Hi Gena,

I can confirm this, and it looks like there's an open bug report for the
issue:

https://gitlab.kitware.com/paraview/paraview/issues/15724

That should get updated when this gets addressed. In the meantime, changing
the background color manually (via the "Background" options in the property
panel) still seems to work properly in the screen shots. You can use this
to get a white background if desired.

HTH,
Dave

On Thu, Oct 13, 2016 at 6:57 AM, Gena Bug via ParaView <
paraview@paraview.org> wrote:

> Hi again!
>
> I'm slightly confused with background color for the screenshots when
> trying to override palette. If I set Background to a Single color in the
> Properties tab (gray by default) then nothing can change it -- background
> will always be gray on the screenshots. If I set Gradient for the
> Background then:
> 1. current palette: works as expected
> 2. black bg: works
> 3. default bg: don't know what's default but it looks like 1... OK
> 4. gradient bg: like 1.
> 5. gray bg: like 1.
> 6. print bg: it looks like Color 1 is set to white in Background settings
> 7. white bg: same as 6.
>
> So now there is no way to set white background for the screenshots
> (previously it was possible when Single Color background was used but not
> Gradient).
>
> On 11.10.2016 16:50, Cory Quammen wrote:
>
>> On behalf of the ParaView development team, I am pleased to announce
>> that ParaView 5.2 Release Candidate 1 binaries and source tarballs are
>> now available for download at
>>
>> http://www.paraview.org/download/
>>
>> Please let us know if you encounter any problems with this release
>> candidate.
>>
>> Thanks,
>> Cory
>>
>> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/opensou
> rce/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Custom Glyph

[David Lonie]

cc'ing this reply to the list.

On Wed, Jan 4, 2017 at 12:36 PM, Bob Flandard  wrote:

> Hi Dave,
>
> That sounds like an exciting feature. Will it be able to use a multiblock
> source from the pipeline?
>

Yes -- there will be a drop-down menu on the property panel that lets you
select arbitrary inputs from the pipeline browser the produce the supported
data types.


> That would be useful because then a single glyph can easily be made to
> have regions of different colors - easily edited using the excellent
> multiblock inspector - without the need for something hacky, like multiple
> glyph filters (one for each color part).
>

Unfortunately, it looks like there will still be some hoop-jumping needed
for multicolored glyphs. The glyph mapping engine doesn't support
multicolored glyphs at the moment, since it doesn't fit well with the
current coloring logic in vtk/paraview's glyphing.

Cheers,
Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Custom Glyph

[David Lonie]

I'm currently working on adding the ability to use custom glyphs with
ParaView's glyph mapper. It will allow you to select a pipeline connection
for the glyph type (rather than the static list of arrow, sphere, etc).

Should be available in the next few weeks.

Dave

On Tue, Jan 3, 2017 at 4:41 AM, Bob Flandard  wrote:

> Thanks David,
>
> I did try that first and was unable to select the "Glyph Type" and gave
> up. Luckily your post encouraged me to RTM where it states that vtkPolyData
> should be used for the glyph. A quick extract surface filter fixed that, so
> thanks very much for your help.
>
> Bob
>
> On 2 January 2017 at 19:49, David E DeMarle 
> wrote:
>
>> Glyph with Custom source filter can do it. Note that it works in CPU/RAM
>> instead of in a GPU shader so memory can be an issue for all the replicated
>> geometry.
>>
>> To use it, manufacture a mesh that appears the way you want it and save
>> it in a common mesh format. Load that into the pipeline and hide it, select
>> the object you want to apply glyphs to and apply the Glyph with Custom
>> Source filter. Choose the shape you read in as the "Glyph Type" input.
>>
>>
>>
>>
>> David E DeMarle
>> Kitware, Inc.
>> R Engineer
>> 21 Corporate Drive
>> Clifton Park, NY 12065-8662
>> Phone: 518-881-4909 <(518)%20881-4909>
>>
>> On Mon, Jan 2, 2017 at 2:35 PM, Bob Flandard  wrote:
>>
>>> Hi,
>>>
>>> Any links or advice on how to create a custom glyph?
>>>
>>> I'd like a sphere divided into 8 equal sized/shaped patches and colored
>>> alternate black and white, to represent the centroid of a point mass. I
>>> have lots of these, so a glyph would be best.
>>>
>>> Thanks, Bob
>>>
>>> ___
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ParaView Wiki at:
>>> http://paraview.org/Wiki/ParaView
>>>
>>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview
>>>
>>>
>>
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Custom source glyph error

[David Lonie]

I believe the issue here is that the glyph source must be polydata at the
moment. Does a dataset surface filter (or similar) help?

I mentioned on the other thread that I'm working on improving the glyph
representation so you won't need to use the filter to get custom glyphs.
This patch will also add the ability to use a composite dataset (with
polydata leaves) as the glyph source.

Cheers,
Dave

On Wed, Jan 4, 2017 at 8:02 AM, Bob Flandard  wrote:

> Hi,
>
> I'm getting an error message (see end) using the "glyph with custom source
> filter" for the case where the input points have associated results data
> sets and the custom glyph source is of multiblock type.
>
> I've attached two small sample state files that demonstrate the issue.
>
> The custom glyph should be two hemispheres, one black and one white. One
> state file shows the desired result, but the point source data has no
> results data sets associated with it. The other state file has point
> sources with two associated results data sets (mode1 and mode2) contained
> in the attached Ensight *.case files. This second file gives the error
> message below when loading.
>
> Is this a bug or is there a filter that implements "CopyStructure", that I
> should use?
>
> Thanks for any suggestions,
>
> Bob
>
>
> The error message is:
>
> ERROR: In 
> C:\bbd\df0abce0\source-paraview\VTK\Common\DataModel\vtkDataObjectTree.cxx,
> line 306
> vtkMultiBlockDataSet (17C45180): Structure does not match. You
> must use CopyStructure before calling this method.
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Newer object always in front in Render View

[David Lonie]

The cmake config looks ok to me.

What sort of graphics set up are you using? Is this mesa, or on a GPU? It
sounds like a driver bug or something may be causing the depth test to
break, or maybe there's a misconfigured framebuffer. Does an earlier
version of ParaView work on the same system?

Dave

On Wed, Jan 4, 2017 at 10:41 AM, Robert Sawko  wrote:

> David,
>
> Thanks for a quick reply. Tried with -dr and it's the same. I am attaching
> my
> state file for spheres. It may help someone to reproduce.
>
> I am running it on Arch Linux and paraview is taken from AUR. I am just
> looking at the way the binary was generated and I see these options in
> cmake
>
>   cmake \
> -DBUILD_SHARED_LIBS:BOOL=ON \
> -DBUILD_TESTING:BOOL=OFF \
> -DCMAKE_BUILD_TYPE=Release \
> -DCMAKE_C_COMPILER=mpicc \
> -DCMAKE_CXX_COMPILER=mpicxx \
> -DCMAKE_INSTALL_PREFIX:PATH=/usr \
> -DCMAKE_VERBOSE_MAKEFILE:BOOL=OFF \
> -DOSPRAY_INSTALL_DIR:PATH=/usr \
> -DPARAVIEW_ENABLE_CGNS:BOOL=OFF \
> -DPARAVIEW_ENABLE_FFMPEG:BOOL=ON \
> -DPARAVIEW_ENABLE_PYTHON:BOOL=ON \
> -DPARAVIEW_PYTHON_VERSION=2 \
> -DPARAVIEW_QT_VERSION=5 \
> -DPARAVIEW_USE_MPI:BOOL=ON \
> -DPARAVIEW_USE_VISITBRIDGE:BOOL=ON \
> -DPARAVIEW_USE_OSPRAY:BOOL=ON \
> -DVISIT_BUILD_READER_CGNS:BOOL=OFF \
> -DVTK_PYTHON_VERSION=2 \
> -DVTK_QT_VERSION=5 \
> -DVTK_RENDERING_BACKEND:STRING=OpenGL2 \
> -DVTK_SMP_IMPLEMENTATION_TYPE:STRING=OpenMP \
> -DVTK_USE_SYSTEM_HDF5:BOOL=OFF \
> ${VTK_USE_SYSTEM_LIB} \
> ../ParaView-v${_pkgver}
>
> Does any of this look suspicious?
>
> Robert
> --
> Newton's bucket
> http://en.wikipedia.org/wiki/Bucket_argument
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/
> opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Custom source glyph error

[David Lonie]

I dug into the error a bit and at a glance, it seems like it might be an
issue with the vtkCompositeDataPipeline (pinging Berk for this since he
knows more about composite pipelines).

The stack for the error is:

1   vtkDataObjectTree::SetDataSetFrom vtkDataObjectTree.cxx:305
2   vtkDataObjectTree::SetDataSet vtkDataObjectTree.cxx:242
3   vtkCompositeDataPipeline::ExecuteEach vtkCompositeDataPipeline.cxx:310
4   vtkCompositeDataPipeline::ExecuteSimpleAlgorithm
vtkCompositeDataPipeline.cxx:398
5   vtkCompositeDataPipeline::ExecuteData vtkCompositeDataPipeline.cxx:167
...

Since the glyph filter is not equipped to handle composite glyphs, the
pipeline iterates through all of the datasets in the composite glyph source
and executes the filter for each of them, combining the outputs into
another composite dataset. But for some reason, the output composite
dataset doesn't seem to match the input composite dataset's structure.

The odd thing I see is that ExecuteSimpleAlgorithm does call CopyStructure
before iterating through the glyphs at vtkCompositeDataPipeline:369. Not
sure why things are mismatched when the output is updated.

On Wed, Jan 4, 2017 at 12:46 PM, Bob Flandard <bfland...@gmail.com> wrote:

> Hi Dave,
>
> In the state file, the glyph source is already polyData - created using
> the sphere source (sensibly the gui doesn't allow selection of a non
> polyData glyph source).
>
> The composite dataset feature will be very useful for multi colored glyphs.
>
> Thanks, Bob
>
> On 4 January 2017 at 17:25, David Lonie <david.lo...@kitware.com> wrote:
>
>> I believe the issue here is that the glyph source must be polydata at the
>> moment. Does a dataset surface filter (or similar) help?
>>
>> I mentioned on the other thread that I'm working on improving the glyph
>> representation so you won't need to use the filter to get custom glyphs.
>> This patch will also add the ability to use a composite dataset (with
>> polydata leaves) as the glyph source.
>>
>> Cheers,
>> Dave
>>
>> On Wed, Jan 4, 2017 at 8:02 AM, Bob Flandard <bfland...@gmail.com> wrote:
>>
>>> Hi,
>>>
>>> I'm getting an error message (see end) using the "glyph with custom
>>> source filter" for the case where the input points have associated results
>>> data sets and the custom glyph source is of multiblock type.
>>>
>>> I've attached two small sample state files that demonstrate the issue.
>>>
>>> The custom glyph should be two hemispheres, one black and one white. One
>>> state file shows the desired result, but the point source data has no
>>> results data sets associated with it. The other state file has point
>>> sources with two associated results data sets (mode1 and mode2) contained
>>> in the attached Ensight *.case files. This second file gives the error
>>> message below when loading.
>>>
>>> Is this a bug or is there a filter that implements "CopyStructure", that
>>> I should use?
>>>
>>> Thanks for any suggestions,
>>>
>>> Bob
>>>
>>>
>>> The error message is:
>>>
>>> ERROR: In 
>>> C:\bbd\df0abce0\source-paraview\VTK\Common\DataModel\vtkDataObjectTree.cxx,
>>> line 306
>>> vtkMultiBlockDataSet (17C45180): Structure does not match. You
>>> must use CopyStructure before calling this method.
>>>
>>> ___
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ParaView Wiki at:
>>> http://paraview.org/Wiki/ParaView
>>>
>>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://public.kitware.com/mailman/listinfo/paraview
>>>
>>>
>>
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] Export Scene in Vector Graphic Format not Working Correctly

[David Lonie]

Hi Faraz,

The vector graphic export in VTK/ParaView has always been tricky and,
unfortunately, somewhat flaky. Many objects, most notably those that
have textures applied for color, do not come out correctly when
exporting through the GL2PS library, which VTK/ParaView are currently
using. These objects can only be exported as a raster image, since
GL2PS has no way of getting texture information out of the OpenGL
feedback buffer.

With newer versions of OpenGL, the feedback buffer used by GL2PS has
been deprecated, greatly reducing the usability of the export library.
Because of this, all 3D geometry is currently rasterized when using
the new OpenGL2 backend to VTK/ParaView.

Right now, the best you can reliably expect is a rasterized 3D
geometry, with vectorized text. 2D charts in ParaView should export
well, but any general renderview scenes will be mostly rasterized.

We're hoping to improve this in the future, but no work has been
scheduled to improve 3D geometry exports.

One thing that *might* help is to play with the the "Interpolate
Scalars Before Mapping" option on older versions of ParaView while
exporting. This can prevent texturing and make it possible for GL2PS
to extract the necessary information, but it will not make a
difference when the OpenGL2 backend is used.

HTH,
Dave


On Wed, Apr 26, 2017 at 9:48 PM, Faraz Arbabi  wrote:
> Hi all,
>
> I am in trouble exporting a 2D contour (slice) containing streamlines in
> vector graphic form. When I uncheck Rasterize 3D Image (in .eps, or .pdf
> format export for example), it produces an empty file in Paraview versions
> before 4.4, and produces rasterized (bitmap) form in Paraview 5.3 (when I
> zoom-in the quality drops just like a .jpg file).
> I have gone through all the threads available online and could not find any
> solutions. Has anyone faced this problem? Is there a way to export using
> python script and get a real .eps image? Any solutions?
>
> Thank you,
> Faraz
>
> ___
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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

Re: [Paraview] [Paraview-developers] ParaView 5.4.0 Release Candidate 1 binaries are available for download

[David Lonie]

On Tue, May 16, 2017 at 9:22 AM, Cory Quammen  wrote:
>> I have a laptop with an NVidia 840M and an integrated GPU (Intel HD 5500). I 
>> have to manually enable the NVidia card for ParaView to get rid of this 
>> behaviour, so it looks like the Intel card is at fault.
>
> I have seen the same thing on my PC with an NVIDIA card and Intel
> integrated GPU. Obviously, FXAA isn't working with some Intel
> integrated GPUs. I, too, had to specifically enable my NVIDIA card.
> But that's what ParaView should ideally be using anyway.
>
> ParaView 5.4.0 RC-2 has some improvements to ParaView's Qt widget for
> rendering that *may* improve the situation. That should be out later
> today (US east coast time) or tomorrow.

Interesting, I hadn't heard of this problem. I don't have an intel
card that reproduces this, but if the new version has the same issue
with FXAA and I can find hardware to debug it I'll be happy to take a
look.

Dave
___
Powered by www.kitware.com

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

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

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