From simon.esneault at gmail.com Mon Feb 2 10:24:41 2015 From: simon.esneault at gmail.com (Simon ESNEAULT) Date: Mon, 2 Feb 2015 16:24:41 +0100 Subject: [vtk-developers] [vtkusers] QVTKWidget issue and workaround for OSX, retina display and Qt5 Message-ID: Hello, There is a bug in QVTKWidget on OSX with a retina display, when build with Qt5. Any QVTKWidget will display only the bottom left quarter, like this : Here is a very simple program that reproduce the bug : ---------------------------------------------------------------------------------------------- *#include * *#include * *#include * *#include * *#include * *#include * *#include * *#include * *#define VTK_NEW(type, instance) \* * vtkSmartPointer instance = vtkSmartPointer::New();* *int main(int argc, char** argv)* *{* * QApplication app(argc, argv);* * QVTKWidget widget;* * VTK_NEW(vtkSphereSource, sphereSource);* * VTK_NEW(vtkPolyDataMapper, sphereMapper);* * VTK_NEW(vtkActor, sphereActor);* * VTK_NEW(vtkRenderWindow, renderWindow);* * VTK_NEW(vtkRenderer, renderer);* * sphereMapper->SetInputConnection(sphereSource->GetOutputPort());* * sphereActor->SetMapper(sphereMapper);* * renderWindow->AddRenderer(renderer);* * renderer->SetBackground(0.1, 0.3, 0.4);* * renderer->AddActor(sphereActor);* * widget.SetRenderWindow(renderWindow);* * widget.resize(256,256);* * widget.show();* * app.exec();* * return 0;* *}* ---------------------------------------------------------------------------------------------- and the CMakeLists.txt : ---------------------------------------------------------------------------------------------- *cmake_minimum_required(VERSION 3.1)* *if(POLICY CMP0053)* * cmake_policy(SET CMP0053 NEW)* *endif()* *project(QtImageViewer)* *find_package(VTK COMPONENTS vtkGUISupportQt )* *include(${VTK_USE_FILE})* *find_package(Qt5Core REQUIRED QUIET)* *add_executable(qtimageviewer **main.cxx**)* *qt5_use_modules(qtimageviewer Core Gui Widgets)* *target_link_libraries(qtimageviewer ${VTK_LIBRARIES})* ---------------------------------------------------------------------------------------------- To enable retina on a normal display, use this : http://cocoamanifest.net/articles/2013/01/turn-on-hidpi-retina-mode-on-an-ordinary-mac.html ---------------------------------------------------------------------------------------------- Now to get back to a normal behavior, like it was with Qt4, one need to call : *setWantsBestResolutionOpenGLSurface:NO *on the* NSView *of the QVTKWidget ! Here is an example on how to solve the problem on the small example above : main.cxx ---------------------------------------------------------------------------------------------- *#include * *#include * *#include * *#include * *#include * *#include * *#include * *#include * *#define VTK_NEW(type, instance) \* * vtkSmartPointer instance = vtkSmartPointer::New();* *#ifdef Q_OS_OSX* *#include "osxHelper.h"* *#endif* *int main(int argc, char** argv)* *{* * QApplication app(argc, argv);* * QVTKWidget widget;* *#ifdef Q_OS_OSX* * disableGLHiDPI(widget.winId());* *#endif* * VTK_NEW(vtkSphereSource, sphereSource);* * VTK_NEW(vtkPolyDataMapper, sphereMapper);* * VTK_NEW(vtkActor, sphereActor);* * VTK_NEW(vtkRenderWindow, renderWindow);* * VTK_NEW(vtkRenderer, renderer);* * sphereMapper->SetInputConnection(sphereSource->GetOutputPort());* * sphereActor->SetMapper(sphereMapper);* * renderWindow->AddRenderer(renderer);* * renderer->SetBackground(0.1, 0.3, 0.4);* * renderer->AddActor(sphereActor);* * widget.SetRenderWindow(renderWindow);* * widget.resize(256,256);* * widget.show();* * app.exec();* * return 0;* *}* ---------------------------------------------------------------------------------------------- osxHelper.h ---------------------------------------------------------------------------------------------- *#ifndef _OSX_HELPER_* *#define _OSX_HELPER_* *void disableGLHiDPI( long a_id );* *#endif* ---------------------------------------------------------------------------------------------- osxHelper.mm ---------------------------------------------------------------------------------------------- *#include * *#include "osxHelper.h"* *void disableGLHiDPI( long a_id ){* * NSView* view = reinterpret_cast( a_id );* * [view setWantsBestResolutionOpenGLSurface:NO];* *}* ---------------------------------------------------------------------------------------------- CMakeLists.txt ---------------------------------------------------------------------------------------------- *cmake_minimum_required(VERSION 3.1)* *if(POLICY CMP0053)* * cmake_policy(SET CMP0053 NEW)* *endif()* *project(QtImageViewer)* *find_package(VTK COMPONENTS vtkGUISupportQt )* *include(${VTK_USE_FILE})* *find_package(Qt5Core REQUIRED QUIET)* *find_library( LIB_COCOA cocoa )* *add_executable(qtimageviewer **main.cxx osxHelper.mm**)* *qt5_use_modules(qtimageviewer Core Gui Widgets)* *target_link_libraries(qtimageviewer ${VTK_LIBRARIES})* ---------------------------------------------------------------------------------------------- And there you go, problem solved ... I hope this is useful to others, maybe this can be integrated in the upcoming 6.2 release ! Have a nice day !! Simon Note on the topic if one day someone want to really support HiDPI for retina displays : The bug was probably introduced by this commit in Qt : https://qt.gitorious.org/qt/qtbase/commit/1caa0c023f4fa60446094e53f22ee79771130e2f Documentation : https://developer.apple.com/library/mac/documentation/GraphicsAnimation/Conceptual/HighResolutionOSX/CapturingScreenContents/CapturingScreenContents.html -- ------------------------------------------------------------------ Simon Esneault Rennes, France ------------------------------------------------------------------ -------------- next part -------------- An HTML attachment was scrubbed... URL: From ken.martin at kitware.com Mon Feb 2 11:26:24 2015 From: ken.martin at kitware.com (Ken Martin) Date: Mon, 2 Feb 2015 11:26:24 -0500 Subject: [vtk-developers] Some OpenGL2 Polygonal Updates etc Message-ID: Here are a few updates to the polygonal work we?ve been doing on the new OpenGL2 backend in VTK. 1) Created a vtkPointGaussianMapper [3] class for cosmology for quickly rendering lots of translucent Gaussian splats. This class will get refined a bit as it starts getting used more. Similar in concept to the PointSprite extensions that were added into Paraview by John Biddiscombe, Ugo Varetto and Stephane Ploix from CSCS and EDF. 2) Changed how the transformation matrices were handled in the vertex shader [7]. It was doing V? = VCDC * MCVC * V and a couple tests were failing. Instead we now compute MCDC on the CPU in double precision and instead do V? = MCDC * V in the shader which fixed the test and is generally a better way to do it. We often need the other two matrices as well which is why the original code just used those. 3) Added a rendering timing framework/executable [5] to make it easier to run test sequences across machines and see the results. Marcus and Rob have expanded on it and Aashish has added a volume rendering test. 4) Rewrote the vtkParametricFunctionSource [6] to be significantly faster (4x) and more memory efficient. We use this source to generate surfaces for rendering timings and it was a bit slow. 5) A bunch of fixes related to getting IceT to work with OpenGL2 [1] which led to some release graphics resource issues being fixed [2]. 6) ?Fixed? a longstanding failing test, TestChartXYZ [4] which was intermittently failing on some systems. 7) Marcus and Rob ran some rendering benchmarks showing upwards of 3 billion triangles per second on a 300 million triangle model. That is a big model and some solid performance. Currently I am wrapping up a change to the vtkCompositePolyDataMapper2 so that in some common circumstances it will render significantly faster (maybe 10x). This is targeted at helping apps that have lots of small polygonal parts that are not glyphed. Also adding in support for texture coordinate transformation matrices which are used by the GeoView classes in VTK. Tim Thirion is fixing up some iOS build issues and David Lonnie is working on removing an old text mapper that was causing some issues. JC has been working on updating Slicer to build against the current VTK with OpenGL2. Thanks Ken [1] http://review.source.kitware.com/#/t/5130/ [2] http://review.source.kitware.com/#/t/5171/ [3] http://review.source.kitware.com/#/t/5198/ [4] http://review.source.kitware.com/#/t/5248/ [5] http://review.source.kitware.com/#/t/5276/ [6] http://review.source.kitware.com/#/t/5302/ [7] http://review.source.kitware.com/#/t/5327/ Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 ken.martin at kitware.com 518 881-4901 (w) 518 371-4573 (f) This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From aashish.chaudhary at kitware.com Mon Feb 2 11:28:15 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Mon, 2 Feb 2015 11:28:15 -0500 Subject: [vtk-developers] independent components support Message-ID: VTK folks, As referred in this article (roadmap): http://www.kitware.com/source/home/post/154, support for independent components in GPU volume mapper is in master now. Currently we only have one test but we are in the midst of adding some more to check various use-cases. Please let us know if you have any questions. We are also planning to post another update on our accomplishments very soon. Thanks, -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.lorensen at gmail.com Mon Feb 2 16:11:23 2015 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Mon, 2 Feb 2015 16:11:23 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux Message-ID: Folks, I built VTK with the OpenGL2 backend. This example: http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget crashes when I hover over an actor. It is failing during the pick operation. My system is Fedoro 20. OPenGL version: OpenGL vendor string: VMware, Inc. OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) OpenGL version string: 2.1 Mesa 10.3.3 OpenGL shading language version string: 1.30 This is the reported error before the segfault: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, line 291 vtkShaderProgram (0x1340be0): 1: /*========================================================================= 2: 3: Program: Visualization Toolkit 4: Module: vtkglPolyDataFSHeadight.glsl 5: 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen 7: All rights reserved. 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for details. 9: 10: This software is distributed WITHOUT ANY WARRANTY; without even 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR 12: PURPOSE. See the above copyright notice for more information. 13: 14: =========================================================================*/ 15: // the lighting model for this shader is a Headlight 16: 17: // The following line handle system declarations such a 18: // default precisions, or defining precisions to null 19: #extension GL_EXT_gpu_shader4 : enable 20: #define highp 21: #define mediump 22: #define lowp 23: 24: 25: // all variables that represent positions or directions have a suffix 26: // indicating the coordinate system they are in. The possible values are 27: // MC - Model Coordinates 28: // WC - WC world coordinates 29: // VC - View Coordinates 30: // DC - Display Coordinates 31: 32: // VC positon of this fragment 33: varying vec4 vertexVC; 34: 35: // optional color passed in from the vertex shader, vertexColor 36: uniform float opacityUniform; // the fragment opacity 37: uniform vec3 ambientColorUniform; // intensity weighted color 38: uniform vec3 diffuseColorUniform; // intensity weighted color 39: uniform vec3 specularColorUniform; // intensity weighted color 40: uniform float specularPowerUniform; 41: 42: 43: // optional normal declaration 44: varying vec3 normalVCVarying; 45: 46: // Texture coordinates 47: //VTK::TCoord::Dec 48: 49: // picking support 50: uniform vec3 mapperIndex; 51: uniform int pickingAttributeIDOffset; 52: 53: // Depth Peeling Support 54: //VTK::DepthPeeling::Dec 55: 56: // clipping plane vars 57: //VTK::Clip::Dec 58: 59: void main() 60: { 61: //VTK::Clip::Impl 62: 63: vec3 ambientColor; 64: vec3 diffuseColor; 65: float opacity; 66: vec3 specularColor; 67: float specularPower; 68: ambientColor = ambientColorUniform; 69: diffuseColor = diffuseColorUniform; 70: opacity = opacityUniform; 71: specularColor = specularColorUniform; 72: specularPower = specularPowerUniform; 73: 74: 75: // Generate the normal if we are not passed in one 76: vec3 normalVC = normalize(normalVCVarying); 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } 78: 79: 80: // diffuse and specular lighting 81: float df = max(0.0, normalVC.z); 82: float sf = pow(df, specularPower); 83: 84: vec3 diffuse = df * diffuseColor; 85: vec3 specular = sf * specularColor; 86: 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); 88: //VTK::TCoord::Impl 89: 90: if (gl_FragColor.a <= 0.0) 91: { 92: discard; 93: } 94: 95: //VTK::DepthPeeling::Impl 96: 97: if (mapperIndex == vec3(0.0,0.0,0.0)) 98: { 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; 100: gl_FragColor = vec4(float(idx%256)/255.0, float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); 101: } 102: else 103: { 104: gl_FragColor = vec4(mapperIndex,1.0); 105: } 106: 107: } 108: 109: 110: 111: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, line 292 vtkShaderProgram (0x1340be0): 0:19(12): warning: extension `GL_EXT_gpu_shader4' unsupported in fragment shader 0:99(12): error: `gl_PrimitiveID' undeclared 0:99(12): error: operands to arithmetic operators must be numeric 0:99(12): error: operands to arithmetic operators must be numeric 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL ES 3.00 required) 0:100(22): error: cannot construct `float' from a non-numeric data type 0:100(22): error: operands to arithmetic operators must be numeric 0:100(17): error: cannot construct `vec4' from a non-numeric data type From aashish.chaudhary at kitware.com Mon Feb 2 17:39:09 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Mon, 2 Feb 2015 17:39:09 -0500 Subject: [vtk-developers] independent components support In-Reply-To: References: Message-ID: Folks, We will write a blog soon, but in the last few weeks, we have made several major improvements to the volume rendering code. Please see the list below (Note: We are still working on resolving some issues on bit older Intel graphics chipset). * Improved vtkTextureObject and now using it for the volume rendering (sharing code) * Added support for independent components * Fixed various mac issues * Improved interactive performance for large volumes * Fixed extents not getting updated issues in presence of other filters * Fixed window color adjustments issues. * Fixed issue with noise textures * Using the shader cache for volume rendering * Added volume rendering benchmarking tests * Added new volume tests. Now testing various features where were not tested before (52 volume tests now) * Fixed resources not releasing issues * 99 % tests are passing on Mac/Linux and Windows. Thanks, Aashish On Mon, Feb 2, 2015 at 11:28 AM, Aashish Chaudhary < aashish.chaudhary at kitware.com> wrote: > VTK folks, > > As referred in this article (roadmap): > http://www.kitware.com/source/home/post/154, support for independent > components in GPU volume mapper is in master now. Currently we only have > one test but we are in the midst of adding some more to check various > use-cases. Please let us know if you have any questions. We are also > planning to post another update on our accomplishments very soon. > > Thanks, > > -- > > > > *| Aashish Chaudhary | Technical Leader | Kitware Inc. * > *| http://www.kitware.com/company/team/chaudhary.html > * > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From aashish.chaudhary at kitware.com Mon Feb 2 17:47:15 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Mon, 2 Feb 2015 17:47:15 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: Ken will have more say but here is one of the reasons for the failure: GL_EXT_gpu_shader4 is required for gl_PrimitiveID to exist in the shader. It may be possible that what GLEW is reporting is not quite matching with what actually is supported. - Aashish On Mon, Feb 2, 2015 at 4:11 PM, Bill Lorensen wrote: > Folks, > > I built VTK with the OpenGL2 backend. This example: > http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget > crashes when I hover over an actor. It is failing during the pick > operation. > > My system is Fedoro 20. OPenGL version: > OpenGL vendor string: VMware, Inc. > OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) > OpenGL version string: 2.1 Mesa 10.3.3 > OpenGL shading language version string: 1.30 > > This is the reported error before the segfault: > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, > line 291 > vtkShaderProgram (0x1340be0): 1: > /*========================================================================= > 2: > 3: Program: Visualization Toolkit > 4: Module: vtkglPolyDataFSHeadight.glsl > 5: > 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen > 7: All rights reserved. > 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for > details. > 9: > 10: This software is distributed WITHOUT ANY WARRANTY; without even > 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A > PARTICULAR > 12: PURPOSE. See the above copyright notice for more information. > 13: > 14: > =========================================================================*/ > 15: // the lighting model for this shader is a Headlight > 16: > 17: // The following line handle system declarations such a > 18: // default precisions, or defining precisions to null > 19: #extension GL_EXT_gpu_shader4 : enable > 20: #define highp > 21: #define mediump > 22: #define lowp > 23: > 24: > 25: // all variables that represent positions or directions have a suffix > 26: // indicating the coordinate system they are in. The possible values > are > 27: // MC - Model Coordinates > 28: // WC - WC world coordinates > 29: // VC - View Coordinates > 30: // DC - Display Coordinates > 31: > 32: // VC positon of this fragment > 33: varying vec4 vertexVC; > 34: > 35: // optional color passed in from the vertex shader, vertexColor > 36: uniform float opacityUniform; // the fragment opacity > 37: uniform vec3 ambientColorUniform; // intensity weighted color > 38: uniform vec3 diffuseColorUniform; // intensity weighted color > 39: uniform vec3 specularColorUniform; // intensity weighted color > 40: uniform float specularPowerUniform; > 41: > 42: > 43: // optional normal declaration > 44: varying vec3 normalVCVarying; > 45: > 46: // Texture coordinates > 47: //VTK::TCoord::Dec > 48: > 49: // picking support > 50: uniform vec3 mapperIndex; > 51: uniform int pickingAttributeIDOffset; > 52: > 53: // Depth Peeling Support > 54: //VTK::DepthPeeling::Dec > 55: > 56: // clipping plane vars > 57: //VTK::Clip::Dec > 58: > 59: void main() > 60: { > 61: //VTK::Clip::Impl > 62: > 63: vec3 ambientColor; > 64: vec3 diffuseColor; > 65: float opacity; > 66: vec3 specularColor; > 67: float specularPower; > 68: ambientColor = ambientColorUniform; > 69: diffuseColor = diffuseColorUniform; > 70: opacity = opacityUniform; > 71: specularColor = specularColorUniform; > 72: specularPower = specularPowerUniform; > 73: > 74: > 75: // Generate the normal if we are not passed in one > 76: vec3 normalVC = normalize(normalVCVarying); > 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } > 78: > 79: > 80: // diffuse and specular lighting > 81: float df = max(0.0, normalVC.z); > 82: float sf = pow(df, specularPower); > 83: > 84: vec3 diffuse = df * diffuseColor; > 85: vec3 specular = sf * specularColor; > 86: > 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); > 88: //VTK::TCoord::Impl > 89: > 90: if (gl_FragColor.a <= 0.0) > 91: { > 92: discard; > 93: } > 94: > 95: //VTK::DepthPeeling::Impl > 96: > 97: if (mapperIndex == vec3(0.0,0.0,0.0)) > 98: { > 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; > 100: gl_FragColor = vec4(float(idx%256)/255.0, > float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); > 101: } > 102: else > 103: { > 104: gl_FragColor = vec4(mapperIndex,1.0); > 105: } > 106: > 107: } > 108: > 109: > 110: > 111: > > > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, > line 292 > vtkShaderProgram (0x1340be0): 0:19(12): warning: extension > `GL_EXT_gpu_shader4' unsupported in fragment shader > 0:99(12): error: `gl_PrimitiveID' undeclared > 0:99(12): error: operands to arithmetic operators must be numeric > 0:99(12): error: operands to arithmetic operators must be numeric > 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or > GLSL ES 3.00 required) > 0:100(22): error: cannot construct `float' from a non-numeric data type > 0:100(22): error: operands to arithmetic operators must be numeric > 0:100(17): error: cannot construct `vec4' from a non-numeric data type > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From aashish.chaudhary at kitware.com Mon Feb 2 18:18:33 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Mon, 2 Feb 2015 18:18:33 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: Another problem could be is that the shader does not define the version explicitly. If you can add a line like what I have shown below in the fragment shader, then it may solve your problem. I don't know on top of my head the file you have to modify. #version 130 On Mon, Feb 2, 2015 at 5:47 PM, Aashish Chaudhary < aashish.chaudhary at kitware.com> wrote: > Ken will have more say but here is one of the reasons for the failure: > > GL_EXT_gpu_shader4 is required for gl_PrimitiveID to exist in the shader. > It may be possible that what GLEW is reporting is not quite matching with > what actually is supported. > > - Aashish > > On Mon, Feb 2, 2015 at 4:11 PM, Bill Lorensen > wrote: > >> Folks, >> >> I built VTK with the OpenGL2 backend. This example: >> http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget >> crashes when I hover over an actor. It is failing during the pick >> operation. >> >> My system is Fedoro 20. OPenGL version: >> OpenGL vendor string: VMware, Inc. >> OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) >> OpenGL version string: 2.1 Mesa 10.3.3 >> OpenGL shading language version string: 1.30 >> >> This is the reported error before the segfault: >> ERROR: In >> /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, >> line 291 >> vtkShaderProgram (0x1340be0): 1: >> >> /*========================================================================= >> 2: >> 3: Program: Visualization Toolkit >> 4: Module: vtkglPolyDataFSHeadight.glsl >> 5: >> 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen >> 7: All rights reserved. >> 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for >> details. >> 9: >> 10: This software is distributed WITHOUT ANY WARRANTY; without even >> 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A >> PARTICULAR >> 12: PURPOSE. See the above copyright notice for more information. >> 13: >> 14: >> =========================================================================*/ >> 15: // the lighting model for this shader is a Headlight >> 16: >> 17: // The following line handle system declarations such a >> 18: // default precisions, or defining precisions to null >> 19: #extension GL_EXT_gpu_shader4 : enable >> 20: #define highp >> 21: #define mediump >> 22: #define lowp >> 23: >> 24: >> 25: // all variables that represent positions or directions have a suffix >> 26: // indicating the coordinate system they are in. The possible values >> are >> 27: // MC - Model Coordinates >> 28: // WC - WC world coordinates >> 29: // VC - View Coordinates >> 30: // DC - Display Coordinates >> 31: >> 32: // VC positon of this fragment >> 33: varying vec4 vertexVC; >> 34: >> 35: // optional color passed in from the vertex shader, vertexColor >> 36: uniform float opacityUniform; // the fragment opacity >> 37: uniform vec3 ambientColorUniform; // intensity weighted color >> 38: uniform vec3 diffuseColorUniform; // intensity weighted color >> 39: uniform vec3 specularColorUniform; // intensity weighted color >> 40: uniform float specularPowerUniform; >> 41: >> 42: >> 43: // optional normal declaration >> 44: varying vec3 normalVCVarying; >> 45: >> 46: // Texture coordinates >> 47: //VTK::TCoord::Dec >> 48: >> 49: // picking support >> 50: uniform vec3 mapperIndex; >> 51: uniform int pickingAttributeIDOffset; >> 52: >> 53: // Depth Peeling Support >> 54: //VTK::DepthPeeling::Dec >> 55: >> 56: // clipping plane vars >> 57: //VTK::Clip::Dec >> 58: >> 59: void main() >> 60: { >> 61: //VTK::Clip::Impl >> 62: >> 63: vec3 ambientColor; >> 64: vec3 diffuseColor; >> 65: float opacity; >> 66: vec3 specularColor; >> 67: float specularPower; >> 68: ambientColor = ambientColorUniform; >> 69: diffuseColor = diffuseColorUniform; >> 70: opacity = opacityUniform; >> 71: specularColor = specularColorUniform; >> 72: specularPower = specularPowerUniform; >> 73: >> 74: >> 75: // Generate the normal if we are not passed in one >> 76: vec3 normalVC = normalize(normalVCVarying); >> 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } >> 78: >> 79: >> 80: // diffuse and specular lighting >> 81: float df = max(0.0, normalVC.z); >> 82: float sf = pow(df, specularPower); >> 83: >> 84: vec3 diffuse = df * diffuseColor; >> 85: vec3 specular = sf * specularColor; >> 86: >> 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); >> 88: //VTK::TCoord::Impl >> 89: >> 90: if (gl_FragColor.a <= 0.0) >> 91: { >> 92: discard; >> 93: } >> 94: >> 95: //VTK::DepthPeeling::Impl >> 96: >> 97: if (mapperIndex == vec3(0.0,0.0,0.0)) >> 98: { >> 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; >> 100: gl_FragColor = vec4(float(idx%256)/255.0, >> float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); >> 101: } >> 102: else >> 103: { >> 104: gl_FragColor = vec4(mapperIndex,1.0); >> 105: } >> 106: >> 107: } >> 108: >> 109: >> 110: >> 111: >> >> >> ERROR: In >> /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cxx, >> line 292 >> vtkShaderProgram (0x1340be0): 0:19(12): warning: extension >> `GL_EXT_gpu_shader4' unsupported in fragment shader >> 0:99(12): error: `gl_PrimitiveID' undeclared >> 0:99(12): error: operands to arithmetic operators must be numeric >> 0:99(12): error: operands to arithmetic operators must be numeric >> 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or >> GLSL ES 3.00 required) >> 0:100(22): error: cannot construct `float' from a non-numeric data type >> 0:100(22): error: operands to arithmetic operators must be numeric >> 0:100(17): error: cannot construct `vec4' from a non-numeric data type >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: http://markmail.org/search/?q=vtk-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/vtk-developers >> >> > > > -- > > > > *| Aashish Chaudhary | Technical Leader | Kitware Inc. * > *| http://www.kitware.com/company/team/chaudhary.html > * > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From ken.martin at kitware.com Mon Feb 2 18:44:25 2015 From: ken.martin at kitware.com (Ken Martin) Date: Mon, 2 Feb 2015 18:44:25 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: Yup, known issue with Mesa. Right now picking with Mesa does not work. The mesa dashboard shows the same issue. We are slowly working on a fix which is basically making Mesa systems use a 3.2 context where they support the calls we need, but we have a couple more classes to convert to work with OpenGL 3.2 before that will work. Thanks Ken Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 ken.martin at kitware.com 518 881-4901 (w) 518 371-4573 (f) This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee.? Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message.? Thank you. -----Original Message----- From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of Bill Lorensen Sent: Monday, February 2, 2015 4:11 PM To: VTK Developers Subject: [vtk-developers] OpenGL2: Picking problems on Linux Folks, I built VTK with the OpenGL2 backend. This example: http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget crashes when I hover over an actor. It is failing during the pick operation. My system is Fedoro 20. OPenGL version: OpenGL vendor string: VMware, Inc. OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) OpenGL version string: 2.1 Mesa 10.3.3 OpenGL shading language version string: 1.30 This is the reported error before the segfault: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx x, line 291 vtkShaderProgram (0x1340be0): 1: /*======================================================================== = 2: 3: Program: Visualization Toolkit 4: Module: vtkglPolyDataFSHeadight.glsl 5: 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen 7: All rights reserved. 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for details. 9: 10: This software is distributed WITHOUT ANY WARRANTY; without even 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR 12: PURPOSE. See the above copyright notice for more information. 13: 14: =========================================================================* / 15: // the lighting model for this shader is a Headlight 16: 17: // The following line handle system declarations such a 18: // default precisions, or defining precisions to null 19: #extension GL_EXT_gpu_shader4 : enable 20: #define highp 21: #define mediump 22: #define lowp 23: 24: 25: // all variables that represent positions or directions have a suffix 26: // indicating the coordinate system they are in. The possible values are 27: // MC - Model Coordinates 28: // WC - WC world coordinates 29: // VC - View Coordinates 30: // DC - Display Coordinates 31: 32: // VC positon of this fragment 33: varying vec4 vertexVC; 34: 35: // optional color passed in from the vertex shader, vertexColor 36: uniform float opacityUniform; // the fragment opacity 37: uniform vec3 ambientColorUniform; // intensity weighted color 38: uniform vec3 diffuseColorUniform; // intensity weighted color 39: uniform vec3 specularColorUniform; // intensity weighted color 40: uniform float specularPowerUniform; 41: 42: 43: // optional normal declaration 44: varying vec3 normalVCVarying; 45: 46: // Texture coordinates 47: //VTK::TCoord::Dec 48: 49: // picking support 50: uniform vec3 mapperIndex; 51: uniform int pickingAttributeIDOffset; 52: 53: // Depth Peeling Support 54: //VTK::DepthPeeling::Dec 55: 56: // clipping plane vars 57: //VTK::Clip::Dec 58: 59: void main() 60: { 61: //VTK::Clip::Impl 62: 63: vec3 ambientColor; 64: vec3 diffuseColor; 65: float opacity; 66: vec3 specularColor; 67: float specularPower; 68: ambientColor = ambientColorUniform; 69: diffuseColor = diffuseColorUniform; 70: opacity = opacityUniform; 71: specularColor = specularColorUniform; 72: specularPower = specularPowerUniform; 73: 74: 75: // Generate the normal if we are not passed in one 76: vec3 normalVC = normalize(normalVCVarying); 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } 78: 79: 80: // diffuse and specular lighting 81: float df = max(0.0, normalVC.z); 82: float sf = pow(df, specularPower); 83: 84: vec3 diffuse = df * diffuseColor; 85: vec3 specular = sf * specularColor; 86: 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); 88: //VTK::TCoord::Impl 89: 90: if (gl_FragColor.a <= 0.0) 91: { 92: discard; 93: } 94: 95: //VTK::DepthPeeling::Impl 96: 97: if (mapperIndex == vec3(0.0,0.0,0.0)) 98: { 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; 100: gl_FragColor = vec4(float(idx%256)/255.0, float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); 101: } 102: else 103: { 104: gl_FragColor = vec4(mapperIndex,1.0); 105: } 106: 107: } 108: 109: 110: 111: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx x, line 292 vtkShaderProgram (0x1340be0): 0:19(12): warning: extension `GL_EXT_gpu_shader4' unsupported in fragment shader 0:99(12): error: `gl_PrimitiveID' undeclared 0:99(12): error: operands to arithmetic operators must be numeric 0:99(12): error: operands to arithmetic operators must be numeric 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL ES 3.00 required) 0:100(22): error: cannot construct `float' from a non-numeric data type 0:100(22): error: operands to arithmetic operators must be numeric 0:100(17): error: cannot construct `vec4' from a non-numeric data type _______________________________________________ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Search the list archives at: http://markmail.org/search/?q=vtk-developers Follow this link to subscribe/unsubscribe: http://public.kitware.com/mailman/listinfo/vtk-developers From aashish.chaudhary at kitware.com Tue Feb 3 00:01:32 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Tue, 3 Feb 2015 00:01:32 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: Ken, Just curious: Did we try setting the version number explicitly (I don't see it in the output shader string)? If a version is not set then by default GLSL assumes it is version 110 which is not what we want. More info here: https://www.opengl.org/wiki/Core_Language_%28GLSL%29 Also, based on the message here: 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL ES 3.00 required) It seems like it is falling back to 110 even though system supports upto 130. Thanks, Aashish On Mon, Feb 2, 2015 at 6:44 PM, Ken Martin wrote: > Yup, known issue with Mesa. Right now picking with Mesa does not work. The > mesa dashboard shows the same issue. We are slowly working on a fix which > is basically making Mesa systems use a 3.2 context where they support the > calls we need, but we have a couple more classes to convert to work with > OpenGL 3.2 before that will work. > > Thanks > Ken > > Ken Martin PhD > Chairman & CFO > Kitware Inc. > 28 Corporate Drive > Clifton Park NY 12065 > ken.martin at kitware.com > 518 881-4901 (w) > 518 371-4573 (f) > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you > are not the intended recipient, any disclosure, copying, distribution or > any action taken in reliance on it is prohibited and may be unlawful. If > you received this communication in error please notify us immediately and > destroy the original message. Thank you. > > -----Original Message----- > From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of > Bill Lorensen > Sent: Monday, February 2, 2015 4:11 PM > To: VTK Developers > Subject: [vtk-developers] OpenGL2: Picking problems on Linux > > Folks, > > I built VTK with the OpenGL2 backend. This example: > http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget > crashes when I hover over an actor. It is failing during the pick > operation. > > My system is Fedoro 20. OPenGL version: > OpenGL vendor string: VMware, Inc. > OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) > OpenGL version string: 2.1 Mesa 10.3.3 OpenGL shading language version > string: 1.30 > > This is the reported error before the segfault: > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx > x, > line 291 > vtkShaderProgram (0x1340be0): 1: > /*======================================================================== > = > 2: > 3: Program: Visualization Toolkit > 4: Module: vtkglPolyDataFSHeadight.glsl > 5: > 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen > 7: All rights reserved. > 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for > details. > 9: > 10: This software is distributed WITHOUT ANY WARRANTY; without even > 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A > PARTICULAR > 12: PURPOSE. See the above copyright notice for more information. > 13: > 14: > =========================================================================* > / > 15: // the lighting model for this shader is a Headlight > 16: > 17: // The following line handle system declarations such a > 18: // default precisions, or defining precisions to null > 19: #extension GL_EXT_gpu_shader4 : enable > 20: #define highp > 21: #define mediump > 22: #define lowp > 23: > 24: > 25: // all variables that represent positions or directions have a suffix > 26: // indicating the coordinate system they are in. The possible values > are > 27: // MC - Model Coordinates > 28: // WC - WC world coordinates > 29: // VC - View Coordinates > 30: // DC - Display Coordinates > 31: > 32: // VC positon of this fragment > 33: varying vec4 vertexVC; > 34: > 35: // optional color passed in from the vertex shader, vertexColor > 36: uniform float opacityUniform; // the fragment opacity > 37: uniform vec3 ambientColorUniform; // intensity weighted color > 38: uniform vec3 diffuseColorUniform; // intensity weighted color > 39: uniform vec3 specularColorUniform; // intensity weighted color > 40: uniform float specularPowerUniform; > 41: > 42: > 43: // optional normal declaration > 44: varying vec3 normalVCVarying; > 45: > 46: // Texture coordinates > 47: //VTK::TCoord::Dec > 48: > 49: // picking support > 50: uniform vec3 mapperIndex; > 51: uniform int pickingAttributeIDOffset; > 52: > 53: // Depth Peeling Support > 54: //VTK::DepthPeeling::Dec > 55: > 56: // clipping plane vars > 57: //VTK::Clip::Dec > 58: > 59: void main() > 60: { > 61: //VTK::Clip::Impl > 62: > 63: vec3 ambientColor; > 64: vec3 diffuseColor; > 65: float opacity; > 66: vec3 specularColor; > 67: float specularPower; > 68: ambientColor = ambientColorUniform; > 69: diffuseColor = diffuseColorUniform; > 70: opacity = opacityUniform; > 71: specularColor = specularColorUniform; > 72: specularPower = specularPowerUniform; > 73: > 74: > 75: // Generate the normal if we are not passed in one > 76: vec3 normalVC = normalize(normalVCVarying); > 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } > 78: > 79: > 80: // diffuse and specular lighting > 81: float df = max(0.0, normalVC.z); > 82: float sf = pow(df, specularPower); > 83: > 84: vec3 diffuse = df * diffuseColor; > 85: vec3 specular = sf * specularColor; > 86: > 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); > 88: //VTK::TCoord::Impl > 89: > 90: if (gl_FragColor.a <= 0.0) > 91: { > 92: discard; > 93: } > 94: > 95: //VTK::DepthPeeling::Impl > 96: > 97: if (mapperIndex == vec3(0.0,0.0,0.0)) > 98: { > 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; > 100: gl_FragColor = vec4(float(idx%256)/255.0, > float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); > 101: } > 102: else > 103: { > 104: gl_FragColor = vec4(mapperIndex,1.0); > 105: } > 106: > 107: } > 108: > 109: > 110: > 111: > > > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx > x, > line 292 > vtkShaderProgram (0x1340be0): 0:19(12): warning: extension > `GL_EXT_gpu_shader4' unsupported in fragment shader > 0:99(12): error: `gl_PrimitiveID' undeclared > 0:99(12): error: operands to arithmetic operators must be numeric > 0:99(12): error: operands to arithmetic operators must be numeric > 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL > ES 3.00 required) > 0:100(22): error: cannot construct `float' from a non-numeric data type > 0:100(22): error: operands to arithmetic operators must be numeric > 0:100(17): error: cannot construct `vec4' from a non-numeric data type > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From bill.lorensen at gmail.com Tue Feb 3 08:28:28 2015 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Tue, 3 Feb 2015 08:28:28 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: Ken, No hurry on this one. I can wait. I can test again when you are ready. Thanks, Bill On Mon, Feb 2, 2015 at 6:44 PM, Ken Martin wrote: > Yup, known issue with Mesa. Right now picking with Mesa does not work. The > mesa dashboard shows the same issue. We are slowly working on a fix which > is basically making Mesa systems use a 3.2 context where they support the > calls we need, but we have a couple more classes to convert to work with > OpenGL 3.2 before that will work. > > Thanks > Ken > > Ken Martin PhD > Chairman & CFO > Kitware Inc. > 28 Corporate Drive > Clifton Park NY 12065 > ken.martin at kitware.com > 518 881-4901 (w) > 518 371-4573 (f) > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you > are not the intended recipient, any disclosure, copying, distribution or > any action taken in reliance on it is prohibited and may be unlawful. If > you received this communication in error please notify us immediately and > destroy the original message. Thank you. > > -----Original Message----- > From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of > Bill Lorensen > Sent: Monday, February 2, 2015 4:11 PM > To: VTK Developers > Subject: [vtk-developers] OpenGL2: Picking problems on Linux > > Folks, > > I built VTK with the OpenGL2 backend. This example: > http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget > crashes when I hover over an actor. It is failing during the pick > operation. > > My system is Fedoro 20. OPenGL version: > OpenGL vendor string: VMware, Inc. > OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) > OpenGL version string: 2.1 Mesa 10.3.3 OpenGL shading language version > string: 1.30 > > This is the reported error before the segfault: > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx > x, > line 291 > vtkShaderProgram (0x1340be0): 1: > /*======================================================================== > = > 2: > 3: Program: Visualization Toolkit > 4: Module: vtkglPolyDataFSHeadight.glsl > 5: > 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen > 7: All rights reserved. > 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for > details. > 9: > 10: This software is distributed WITHOUT ANY WARRANTY; without even > 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A > PARTICULAR > 12: PURPOSE. See the above copyright notice for more information. > 13: > 14: > =========================================================================* > / > 15: // the lighting model for this shader is a Headlight > 16: > 17: // The following line handle system declarations such a > 18: // default precisions, or defining precisions to null > 19: #extension GL_EXT_gpu_shader4 : enable > 20: #define highp > 21: #define mediump > 22: #define lowp > 23: > 24: > 25: // all variables that represent positions or directions have a suffix > 26: // indicating the coordinate system they are in. The possible values > are > 27: // MC - Model Coordinates > 28: // WC - WC world coordinates > 29: // VC - View Coordinates > 30: // DC - Display Coordinates > 31: > 32: // VC positon of this fragment > 33: varying vec4 vertexVC; > 34: > 35: // optional color passed in from the vertex shader, vertexColor > 36: uniform float opacityUniform; // the fragment opacity > 37: uniform vec3 ambientColorUniform; // intensity weighted color > 38: uniform vec3 diffuseColorUniform; // intensity weighted color > 39: uniform vec3 specularColorUniform; // intensity weighted color > 40: uniform float specularPowerUniform; > 41: > 42: > 43: // optional normal declaration > 44: varying vec3 normalVCVarying; > 45: > 46: // Texture coordinates > 47: //VTK::TCoord::Dec > 48: > 49: // picking support > 50: uniform vec3 mapperIndex; > 51: uniform int pickingAttributeIDOffset; > 52: > 53: // Depth Peeling Support > 54: //VTK::DepthPeeling::Dec > 55: > 56: // clipping plane vars > 57: //VTK::Clip::Dec > 58: > 59: void main() > 60: { > 61: //VTK::Clip::Impl > 62: > 63: vec3 ambientColor; > 64: vec3 diffuseColor; > 65: float opacity; > 66: vec3 specularColor; > 67: float specularPower; > 68: ambientColor = ambientColorUniform; > 69: diffuseColor = diffuseColorUniform; > 70: opacity = opacityUniform; > 71: specularColor = specularColorUniform; > 72: specularPower = specularPowerUniform; > 73: > 74: > 75: // Generate the normal if we are not passed in one > 76: vec3 normalVC = normalize(normalVCVarying); > 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } > 78: > 79: > 80: // diffuse and specular lighting > 81: float df = max(0.0, normalVC.z); > 82: float sf = pow(df, specularPower); > 83: > 84: vec3 diffuse = df * diffuseColor; > 85: vec3 specular = sf * specularColor; > 86: > 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); > 88: //VTK::TCoord::Impl > 89: > 90: if (gl_FragColor.a <= 0.0) > 91: { > 92: discard; > 93: } > 94: > 95: //VTK::DepthPeeling::Impl > 96: > 97: if (mapperIndex == vec3(0.0,0.0,0.0)) > 98: { > 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; > 100: gl_FragColor = vec4(float(idx%256)/255.0, > float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); > 101: } > 102: else > 103: { > 104: gl_FragColor = vec4(mapperIndex,1.0); > 105: } > 106: > 107: } > 108: > 109: > 110: > 111: > > > ERROR: In > /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx > x, > line 292 > vtkShaderProgram (0x1340be0): 0:19(12): warning: extension > `GL_EXT_gpu_shader4' unsupported in fragment shader > 0:99(12): error: `gl_PrimitiveID' undeclared > 0:99(12): error: operands to arithmetic operators must be numeric > 0:99(12): error: operands to arithmetic operators must be numeric > 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL > ES 3.00 required) > 0:100(22): error: cannot construct `float' from a non-numeric data type > 0:100(22): error: operands to arithmetic operators must be numeric > 0:100(17): error: cannot construct `vec4' from a non-numeric data type > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers -- Unpaid intern in BillsBasement at noware dot com From ken.martin at kitware.com Tue Feb 3 09:21:43 2015 From: ken.martin at kitware.com (Ken Martin) Date: Tue, 3 Feb 2015 09:21:43 -0500 Subject: [vtk-developers] OpenGL2: Picking problems on Linux In-Reply-To: References: Message-ID: <64d0ed48eb768de0fb162c7a590314e3@mail.gmail.com> I believe I tried that a month or two ago and Mesa will fail saying that you cannot use shader version 150 with a 2.1 context. If we create a 3.2 context then you can use a shader version that includes what we need. Thanks Ken Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 ken.martin at kitware.com 518 881-4901 (w) 518 371-4573 (f) This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. *From:* Aashish Chaudhary [mailto:aashish.chaudhary at kitware.com] *Sent:* Tuesday, February 3, 2015 12:02 AM *To:* Ken Martin *Cc:* Bill Lorensen; VTK Developers *Subject:* Re: [vtk-developers] OpenGL2: Picking problems on Linux Ken, Just curious: Did we try setting the version number explicitly (I don't see it in the output shader string)? If a version is not set then by default GLSL assumes it is version 110 which is not what we want. More info here: https://www.opengl.org/wiki/Core_Language_%28GLSL%29 Also, based on the message here: 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL ES 3.00 required) It seems like it is falling back to 110 even though system supports upto 130. Thanks, Aashish On Mon, Feb 2, 2015 at 6:44 PM, Ken Martin wrote: Yup, known issue with Mesa. Right now picking with Mesa does not work. The mesa dashboard shows the same issue. We are slowly working on a fix which is basically making Mesa systems use a 3.2 context where they support the calls we need, but we have a couple more classes to convert to work with OpenGL 3.2 before that will work. Thanks Ken Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 ken.martin at kitware.com 518 881-4901 (w) 518 371-4573 (f) This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. -----Original Message----- From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of Bill Lorensen Sent: Monday, February 2, 2015 4:11 PM To: VTK Developers Subject: [vtk-developers] OpenGL2: Picking problems on Linux Folks, I built VTK with the OpenGL2 backend. This example: http://www.vtk.org/Wiki/VTK/Examples/Cxx/Widgets/BalloonWidget crashes when I hover over an actor. It is failing during the pick operation. My system is Fedoro 20. OPenGL version: OpenGL vendor string: VMware, Inc. OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.4, 128 bits) OpenGL version string: 2.1 Mesa 10.3.3 OpenGL shading language version string: 1.30 This is the reported error before the segfault: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx x, line 291 vtkShaderProgram (0x1340be0): 1: /*======================================================================== = 2: 3: Program: Visualization Toolkit 4: Module: vtkglPolyDataFSHeadight.glsl 5: 6: Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen 7: All rights reserved. 8: See Copyright.txt or http://www.kitware.com/Copyright.htm for details. 9: 10: This software is distributed WITHOUT ANY WARRANTY; without even 11: the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR 12: PURPOSE. See the above copyright notice for more information. 13: 14: =========================================================================* / 15: // the lighting model for this shader is a Headlight 16: 17: // The following line handle system declarations such a 18: // default precisions, or defining precisions to null 19: #extension GL_EXT_gpu_shader4 : enable 20: #define highp 21: #define mediump 22: #define lowp 23: 24: 25: // all variables that represent positions or directions have a suffix 26: // indicating the coordinate system they are in. The possible values are 27: // MC - Model Coordinates 28: // WC - WC world coordinates 29: // VC - View Coordinates 30: // DC - Display Coordinates 31: 32: // VC positon of this fragment 33: varying vec4 vertexVC; 34: 35: // optional color passed in from the vertex shader, vertexColor 36: uniform float opacityUniform; // the fragment opacity 37: uniform vec3 ambientColorUniform; // intensity weighted color 38: uniform vec3 diffuseColorUniform; // intensity weighted color 39: uniform vec3 specularColorUniform; // intensity weighted color 40: uniform float specularPowerUniform; 41: 42: 43: // optional normal declaration 44: varying vec3 normalVCVarying; 45: 46: // Texture coordinates 47: //VTK::TCoord::Dec 48: 49: // picking support 50: uniform vec3 mapperIndex; 51: uniform int pickingAttributeIDOffset; 52: 53: // Depth Peeling Support 54: //VTK::DepthPeeling::Dec 55: 56: // clipping plane vars 57: //VTK::Clip::Dec 58: 59: void main() 60: { 61: //VTK::Clip::Impl 62: 63: vec3 ambientColor; 64: vec3 diffuseColor; 65: float opacity; 66: vec3 specularColor; 67: float specularPower; 68: ambientColor = ambientColorUniform; 69: diffuseColor = diffuseColorUniform; 70: opacity = opacityUniform; 71: specularColor = specularColorUniform; 72: specularPower = specularPowerUniform; 73: 74: 75: // Generate the normal if we are not passed in one 76: vec3 normalVC = normalize(normalVCVarying); 77: if (gl_FrontFacing == false) { normalVC = -normalVC; } 78: 79: 80: // diffuse and specular lighting 81: float df = max(0.0, normalVC.z); 82: float sf = pow(df, specularPower); 83: 84: vec3 diffuse = df * diffuseColor; 85: vec3 specular = sf * specularColor; 86: 87: gl_FragColor = vec4(ambientColor + diffuse + specular, opacity); 88: //VTK::TCoord::Impl 89: 90: if (gl_FragColor.a <= 0.0) 91: { 92: discard; 93: } 94: 95: //VTK::DepthPeeling::Impl 96: 97: if (mapperIndex == vec3(0.0,0.0,0.0)) 98: { 99: int idx = gl_PrimitiveID + 1 + pickingAttributeIDOffset; 100: gl_FragColor = vec4(float(idx%256)/255.0, float((idx/256)%256)/255.0, float(idx/65536)/255.0, 1.0); 101: } 102: else 103: { 104: gl_FragColor = vec4(mapperIndex,1.0); 105: } 106: 107: } 108: 109: 110: 111: ERROR: In /home/lorensen/ProjectsGIT/VTKGerrit/Rendering/OpenGL2/vtkShaderProgram.cx x, line 292 vtkShaderProgram (0x1340be0): 0:19(12): warning: extension `GL_EXT_gpu_shader4' unsupported in fragment shader 0:99(12): error: `gl_PrimitiveID' undeclared 0:99(12): error: operands to arithmetic operators must be numeric 0:99(12): error: operands to arithmetic operators must be numeric 0:100(28): error: operator '%' is reserved in GLSL 1.10 (GLSL 1.30 or GLSL ES 3.00 required) 0:100(22): error: cannot construct `float' from a non-numeric data type 0:100(22): error: operands to arithmetic operators must be numeric 0:100(17): error: cannot construct `vec4' from a non-numeric data type _______________________________________________ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Search the list archives at: http://markmail.org/search/?q=vtk-developers Follow this link to subscribe/unsubscribe: http://public.kitware.com/mailman/listinfo/vtk-developers _______________________________________________ Powered by www.kitware.com Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Search the list archives at: http://markmail.org/search/?q=vtk-developers Follow this link to subscribe/unsubscribe: http://public.kitware.com/mailman/listinfo/vtk-developers -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.maynard at kitware.com Tue Feb 3 10:56:43 2015 From: robert.maynard at kitware.com (Robert Maynard) Date: Tue, 3 Feb 2015 10:56:43 -0500 Subject: [vtk-developers] Some OpenGL2 Polygonal Updates etc In-Reply-To: References: Message-ID: I want to quickly expand on the performance section to show what kind of performance you should expect with different kinds of graphics cards. On Mobile graphics cards with 1GB of memory you should expect at least 30-60fps up to 15 million triangles when rendering at 1080p. Above 15 million we start to see less that 30fps and by 30 million we reach 15fps. Above 30 million we approach the memory limits of the graphics card. On current generation midrange workstation graphics cards you should expect at least 60fps up to 20 million triangles when rendering at 1080p. Above 30 million we start to see ~30fps, by 50 million we reach 15fps, and at 100 million we get 10fps. On current generation highend workstation graphics cards you should expect at least 60fps up to 50 million triangles when rendering at 1080p. Above 75 million we start to see ~30fps, by 200 million we reach 15fps, and at 300 million we get 10fps. On Mon, Feb 2, 2015 at 11:26 AM, Ken Martin wrote: > Here are a few updates to the polygonal work we?ve been doing on the new > OpenGL2 backend in VTK. > > > > 1) Created a vtkPointGaussianMapper [3] class for cosmology for quickly > rendering lots of translucent Gaussian splats. This class will get refined > a bit as it starts getting used more. Similar in concept to the PointSprite > extensions that were added into Paraview by John Biddiscombe, Ugo Varetto > and Stephane Ploix from CSCS and EDF. > > > > 2) Changed how the transformation matrices were handled in the vertex > shader [7]. It was doing V? = VCDC * MCVC * V and a couple tests were > failing. Instead we now compute MCDC on the CPU in double precision and > instead do V? = MCDC * V in the shader which fixed the test and is generally > a better way to do it. We often need the other two matrices as well which is > why the original code just used those. > > > > 3) Added a rendering timing framework/executable [5] to make it easier > to run test sequences across machines and see the results. Marcus and Rob > have expanded on it and Aashish has added a volume rendering test. > > > > 4) Rewrote the vtkParametricFunctionSource [6] to be significantly > faster (4x) and more memory efficient. We use this source to generate > surfaces for rendering timings and it was a bit slow. > > > > 5) A bunch of fixes related to getting IceT to work with OpenGL2 [1] > which led to some release graphics resource issues being fixed [2]. > > > > 6) ?Fixed? a longstanding failing test, TestChartXYZ [4] which was > intermittently failing on some systems. > > > > 7) Marcus and Rob ran some rendering benchmarks showing upwards of 3 > billion triangles per second on a 300 million triangle model. That is a big > model and some solid performance. > > > > > > Currently I am wrapping up a change to the vtkCompositePolyDataMapper2 so > that in some common circumstances it will render significantly faster (maybe > 10x). This is targeted at helping apps that have lots of small polygonal > parts that are not glyphed. Also adding in support for texture coordinate > transformation matrices which are used by the GeoView classes in VTK. Tim > Thirion is fixing up some iOS build issues and David Lonnie is working on > removing an old text mapper that was causing some issues. JC has been > working on updating Slicer to build against the current VTK with OpenGL2. > > > > Thanks > > Ken > > > > > > [1] http://review.source.kitware.com/#/t/5130/ > > [2] http://review.source.kitware.com/#/t/5171/ > > [3] http://review.source.kitware.com/#/t/5198/ > > [4] http://review.source.kitware.com/#/t/5248/ > > [5] http://review.source.kitware.com/#/t/5276/ > > [6] http://review.source.kitware.com/#/t/5302/ > > [7] http://review.source.kitware.com/#/t/5327/ > > > > > > Ken Martin PhD > > Chairman & CFO > > Kitware Inc. > > 28 Corporate Drive > > Clifton Park NY 12065 > > ken.martin at kitware.com > > 518 881-4901 (w) > > 518 371-4573 (f) > > > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you are > not the intended recipient, any disclosure, copying, distribution or any > action taken in reliance on it is prohibited and may be unlawful. If you > received this communication in error please notify us immediately and > destroy the original message. Thank you. > > > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From aashish.chaudhary at kitware.com Tue Feb 3 15:40:21 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Tue, 3 Feb 2015 15:40:21 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: <20150129210101.GA16304@megas.kitwarein.com> References: <20150129210101.GA16304@megas.kitwarein.com> Message-ID: David, Thanks to Ken, we were able to reproduce the problem when Ken switched to use Intel driver on his Windows machine. We worked on a fix that possibly fix issues you were seeing (it worked ok Ken's machine). The good thing is that it became ~10 times faster on Windows machine. I will push a fix momentarily and hopefully it will get merged tonight. - Aashish On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel wrote: > On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: > > " I must admit, I'm a little bit disappointed that some of the commits > > in what was merged to VTK master don't even compile... I would hope we > > could be a bit less sloppy, and hold ourselves to a higher standard. > > Rushing something in before it's ready for prime time serves no one." > > > > I want to make sure I understand what you are saying. Are you saying in > this > > case code was merged into master that did not compile? Or are you saying > > that someone did multiple commits, some of which did not compile, but > they > > were not pushed or merged until it did compile? > > FWIW, I have a script I use to ensure this for long-lived branches: > > ======= > #!/bin/sh > > base="${1:-master}" > shift > > GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" > export GIT_EDITOR > > exec git rebase -i "$( git merge-base HEAD "$base" )" > ======= > > A 'build.sh' script (provided by you) is run after each commit and stops > on any commit which fails to build. > > --Ben > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From DLRdave at aol.com Wed Feb 4 07:04:43 2015 From: DLRdave at aol.com (David Cole) Date: Wed, 4 Feb 2015 07:04:43 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: References: <20150129210101.GA16304@megas.kitwarein.com> Message-ID: Thanks very much -- we'll find out tonight when the next nightly dashboard runs on "mug.neocisinc" I'll ping again if there's anything further. D On Tue, Feb 3, 2015 at 3:40 PM, Aashish Chaudhary wrote: > David, > > Thanks to Ken, we were able to reproduce the problem when Ken switched to > use Intel driver on his Windows machine. We worked on > a fix that possibly fix issues you were seeing (it worked ok Ken's machine). > The good thing is that it became ~10 times faster on Windows machine. > > I will push a fix momentarily and hopefully it will get merged tonight. > > - Aashish > > > > > On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel > wrote: >> >> On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: >> > " I must admit, I'm a little bit disappointed that some of the commits >> > in what was merged to VTK master don't even compile... I would hope we >> > could be a bit less sloppy, and hold ourselves to a higher standard. >> > Rushing something in before it's ready for prime time serves no one." >> > >> > I want to make sure I understand what you are saying. Are you saying in >> > this >> > case code was merged into master that did not compile? Or are you saying >> > that someone did multiple commits, some of which did not compile, but >> > they >> > were not pushed or merged until it did compile? >> >> FWIW, I have a script I use to ensure this for long-lived branches: >> >> ======= >> #!/bin/sh >> >> base="${1:-master}" >> shift >> >> GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" >> export GIT_EDITOR >> >> exec git rebase -i "$( git merge-base HEAD "$base" )" >> ======= >> >> A 'build.sh' script (provided by you) is run after each commit and stops >> on any commit which fails to build. >> >> --Ben > > > > > -- > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > | http://www.kitware.com/company/team/chaudhary.html From dave.demarle at kitware.com Wed Feb 4 10:11:32 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Wed, 4 Feb 2015 10:11:32 -0500 Subject: [vtk-developers] ready to branch for 6.2? Message-ID: Does anyone out that have partially completed work in progress that we should delay the release branch point for? We are hoping to get 6.2 started in a day or two. thanks, David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 -------------- next part -------------- An HTML attachment was scrubbed... URL: From deevankshu096 at gmail.com Wed Feb 4 18:04:15 2015 From: deevankshu096 at gmail.com (DEEVANKSHU GARG) Date: Thu, 5 Feb 2015 04:34:15 +0530 Subject: [vtk-developers] header intact Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From aashish.chaudhary at kitware.com Thu Feb 5 09:45:59 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Thu, 5 Feb 2015 09:45:59 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: References: <20150129210101.GA16304@megas.kitwarein.com> Message-ID: Looks like we are back in business -:) (-22) https://open.cdash.org/viewTest.php?onlyfailed&buildid=3682020 Thanks, - Aashish On Wed, Feb 4, 2015 at 7:04 AM, David Cole wrote: > Thanks very much -- we'll find out tonight when the next nightly > dashboard runs on "mug.neocisinc" > > I'll ping again if there's anything further. > > > D > > > On Tue, Feb 3, 2015 at 3:40 PM, Aashish Chaudhary > wrote: > > David, > > > > Thanks to Ken, we were able to reproduce the problem when Ken switched to > > use Intel driver on his Windows machine. We worked on > > a fix that possibly fix issues you were seeing (it worked ok Ken's > machine). > > The good thing is that it became ~10 times faster on Windows machine. > > > > I will push a fix momentarily and hopefully it will get merged tonight. > > > > - Aashish > > > > > > > > > > On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel > > wrote: > >> > >> On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: > >> > " I must admit, I'm a little bit disappointed that some of the commits > >> > in what was merged to VTK master don't even compile... I would hope we > >> > could be a bit less sloppy, and hold ourselves to a higher standard. > >> > Rushing something in before it's ready for prime time serves no one." > >> > > >> > I want to make sure I understand what you are saying. Are you saying > in > >> > this > >> > case code was merged into master that did not compile? Or are you > saying > >> > that someone did multiple commits, some of which did not compile, but > >> > they > >> > were not pushed or merged until it did compile? > >> > >> FWIW, I have a script I use to ensure this for long-lived branches: > >> > >> ======= > >> #!/bin/sh > >> > >> base="${1:-master}" > >> shift > >> > >> GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" > >> export GIT_EDITOR > >> > >> exec git rebase -i "$( git merge-base HEAD "$base" )" > >> ======= > >> > >> A 'build.sh' script (provided by you) is run after each commit and stops > >> on any commit which fails to build. > >> > >> --Ben > > > > > > > > > > -- > > | Aashish Chaudhary > > | Technical Leader > > | Kitware Inc. > > | http://www.kitware.com/company/team/chaudhary.html > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From ken.martin at kitware.com Thu Feb 5 09:47:09 2015 From: ken.martin at kitware.com (Ken Martin) Date: Thu, 5 Feb 2015 09:47:09 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: References: <20150129210101.GA16304@megas.kitwarein.com> Message-ID: <1e5cbf417008990602b0edef8da96a4e@mail.gmail.com> Awesome! Ken Martin PhD Chairman & CFO Kitware Inc. 28 Corporate Drive Clifton Park NY 12065 ken.martin at kitware.com 518 881-4901 (w) 518 371-4573 (f) This communication, including all attachments, contains confidential and legally privileged information, and it is intended only for the use of the addressee. Access to this email by anyone else is unauthorized. If you are not the intended recipient, any disclosure, copying, distribution or any action taken in reliance on it is prohibited and may be unlawful. If you received this communication in error please notify us immediately and destroy the original message. Thank you. *From:* Aashish Chaudhary [mailto:aashish.chaudhary at kitware.com] *Sent:* Thursday, February 5, 2015 9:46 AM *To:* David Cole *Cc:* Ben Boeckel; Ken Martin; vtkdev *Subject:* Re: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard Looks like we are back in business -:) (-22) https://open.cdash.org/viewTest.php?onlyfailed&buildid=3682020 Thanks, - Aashish On Wed, Feb 4, 2015 at 7:04 AM, David Cole wrote: Thanks very much -- we'll find out tonight when the next nightly dashboard runs on "mug.neocisinc" I'll ping again if there's anything further. D On Tue, Feb 3, 2015 at 3:40 PM, Aashish Chaudhary wrote: > David, > > Thanks to Ken, we were able to reproduce the problem when Ken switched to > use Intel driver on his Windows machine. We worked on > a fix that possibly fix issues you were seeing (it worked ok Ken's machine). > The good thing is that it became ~10 times faster on Windows machine. > > I will push a fix momentarily and hopefully it will get merged tonight. > > - Aashish > > > > > On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel > wrote: >> >> On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: >> > " I must admit, I'm a little bit disappointed that some of the commits >> > in what was merged to VTK master don't even compile... I would hope we >> > could be a bit less sloppy, and hold ourselves to a higher standard. >> > Rushing something in before it's ready for prime time serves no one." >> > >> > I want to make sure I understand what you are saying. Are you saying in >> > this >> > case code was merged into master that did not compile? Or are you saying >> > that someone did multiple commits, some of which did not compile, but >> > they >> > were not pushed or merged until it did compile? >> >> FWIW, I have a script I use to ensure this for long-lived branches: >> >> ======= >> #!/bin/sh >> >> base="${1:-master}" >> shift >> >> GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" >> export GIT_EDITOR >> >> exec git rebase -i "$( git merge-base HEAD "$base" )" >> ======= >> >> A 'build.sh' script (provided by you) is run after each commit and stops >> on any commit which fails to build. >> >> --Ben > > > > > -- > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > | http://www.kitware.com/company/team/chaudhary.html -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From marcus.hanwell at kitware.com Thu Feb 5 11:06:28 2015 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Thu, 5 Feb 2015 11:06:28 -0500 Subject: [vtk-developers] ready to branch for 6.2? In-Reply-To: References: Message-ID: On Wed, Feb 4, 2015 at 10:11 AM, David E DeMarle wrote: > Does anyone out that have partially completed work in progress that we > should delay the release branch point for? > > We are hoping to get 6.2 started in a day or two. > Let's do it, I haven't heard of anything, it is time. Marcus From DLRdave at aol.com Thu Feb 5 12:34:44 2015 From: DLRdave at aol.com (David Cole) Date: Thu, 5 Feb 2015 12:34:44 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: <1e5cbf417008990602b0edef8da96a4e@mail.gmail.com> References: <20150129210101.GA16304@megas.kitwarein.com> <1e5cbf417008990602b0edef8da96a4e@mail.gmail.com> Message-ID: Yes!! Thank you, thank you, thank you. Very much! The code change that actually fixed it looks like a reasonably small bit of code. You may also want to try this with the original OpenGL backend, and see if there is a similarly simple fix to be done there. This machine had similar failures before I switched to OpenGL2, which might be just as simple to fix. Thanks again, D On Thu, Feb 5, 2015 at 9:47 AM, Ken Martin wrote: > Awesome! > > > > Ken Martin PhD > > Chairman & CFO > > Kitware Inc. > > 28 Corporate Drive > > Clifton Park NY 12065 > > ken.martin at kitware.com > > 518 881-4901 (w) > > 518 371-4573 (f) > > > > This communication, including all attachments, contains confidential and > legally privileged information, and it is intended only for the use of the > addressee. Access to this email by anyone else is unauthorized. If you are > not the intended recipient, any disclosure, copying, distribution or any > action taken in reliance on it is prohibited and may be unlawful. If you > received this communication in error please notify us immediately and > destroy the original message. Thank you. > > > > From: Aashish Chaudhary [mailto:aashish.chaudhary at kitware.com] > Sent: Thursday, February 5, 2015 9:46 AM > To: David Cole > Cc: Ben Boeckel; Ken Martin; vtkdev > Subject: Re: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard > > > > Looks like we are back in business -:) (-22) > https://open.cdash.org/viewTest.php?onlyfailed&buildid=3682020 > > > > Thanks, > > - Aashish > > > > On Wed, Feb 4, 2015 at 7:04 AM, David Cole wrote: > > Thanks very much -- we'll find out tonight when the next nightly > dashboard runs on "mug.neocisinc" > > I'll ping again if there's anything further. > > > D > > > > On Tue, Feb 3, 2015 at 3:40 PM, Aashish Chaudhary > wrote: >> David, >> >> Thanks to Ken, we were able to reproduce the problem when Ken switched to >> use Intel driver on his Windows machine. We worked on >> a fix that possibly fix issues you were seeing (it worked ok Ken's >> machine). >> The good thing is that it became ~10 times faster on Windows machine. >> >> I will push a fix momentarily and hopefully it will get merged tonight. >> >> - Aashish >> >> >> >> >> On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel >> wrote: >>> >>> On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: >>> > " I must admit, I'm a little bit disappointed that some of the commits >>> > in what was merged to VTK master don't even compile... I would hope we >>> > could be a bit less sloppy, and hold ourselves to a higher standard. >>> > Rushing something in before it's ready for prime time serves no one." >>> > >>> > I want to make sure I understand what you are saying. Are you saying in >>> > this >>> > case code was merged into master that did not compile? Or are you >>> > saying >>> > that someone did multiple commits, some of which did not compile, but >>> > they >>> > were not pushed or merged until it did compile? >>> >>> FWIW, I have a script I use to ensure this for long-lived branches: >>> >>> ======= >>> #!/bin/sh >>> >>> base="${1:-master}" >>> shift >>> >>> GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" >>> export GIT_EDITOR >>> >>> exec git rebase -i "$( git merge-base HEAD "$base" )" >>> ======= >>> >>> A 'build.sh' script (provided by you) is run after each commit and stops >>> on any commit which fails to build. >>> >>> --Ben >> >> >> >> >> -- >> | Aashish Chaudhary >> | Technical Leader >> | Kitware Inc. >> | http://www.kitware.com/company/team/chaudhary.html > > > > > > -- > > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > > | http://www.kitware.com/company/team/chaudhary.html From aashish.chaudhary at kitware.com Thu Feb 5 12:36:21 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Thu, 5 Feb 2015 12:36:21 -0500 Subject: [vtk-developers] 22 new failing GPU tests on OpenGL2 dashboard In-Reply-To: References: <20150129210101.GA16304@megas.kitwarein.com> <1e5cbf417008990602b0edef8da96a4e@mail.gmail.com> Message-ID: On Thu, Feb 5, 2015 at 12:34 PM, David Cole wrote: > Yes!! > > Thank you, thank you, thank you. Very much! > > The code change that actually fixed it looks like a reasonably small > bit of code. You may also want to try this with the original OpenGL > backend, and see if there is a similarly simple fix to be done there. > This machine had similar failures before I switched to OpenGL2, which > might be just as simple to fix. > Sure, I can have a look. Thanks for reporting this. - Aashish > > > Thanks again, > D > > > > On Thu, Feb 5, 2015 at 9:47 AM, Ken Martin wrote: > > Awesome! > > > > > > > > Ken Martin PhD > > > > Chairman & CFO > > > > Kitware Inc. > > > > 28 Corporate Drive > > > > Clifton Park NY 12065 > > > > ken.martin at kitware.com > > > > 518 881-4901 (w) > > > > 518 371-4573 (f) > > > > > > > > This communication, including all attachments, contains confidential and > > legally privileged information, and it is intended only for the use of > the > > addressee. Access to this email by anyone else is unauthorized. If you > are > > not the intended recipient, any disclosure, copying, distribution or any > > action taken in reliance on it is prohibited and may be unlawful. If you > > received this communication in error please notify us immediately and > > destroy the original message. Thank you. > > > > > > > > From: Aashish Chaudhary [mailto:aashish.chaudhary at kitware.com] > > Sent: Thursday, February 5, 2015 9:46 AM > > To: David Cole > > Cc: Ben Boeckel; Ken Martin; vtkdev > > Subject: Re: [vtk-developers] 22 new failing GPU tests on OpenGL2 > dashboard > > > > > > > > Looks like we are back in business -:) (-22) > > https://open.cdash.org/viewTest.php?onlyfailed&buildid=3682020 > > > > > > > > Thanks, > > > > - Aashish > > > > > > > > On Wed, Feb 4, 2015 at 7:04 AM, David Cole wrote: > > > > Thanks very much -- we'll find out tonight when the next nightly > > dashboard runs on "mug.neocisinc" > > > > I'll ping again if there's anything further. > > > > > > D > > > > > > > > On Tue, Feb 3, 2015 at 3:40 PM, Aashish Chaudhary > > wrote: > >> David, > >> > >> Thanks to Ken, we were able to reproduce the problem when Ken switched > to > >> use Intel driver on his Windows machine. We worked on > >> a fix that possibly fix issues you were seeing (it worked ok Ken's > >> machine). > >> The good thing is that it became ~10 times faster on Windows machine. > >> > >> I will push a fix momentarily and hopefully it will get merged tonight. > >> > >> - Aashish > >> > >> > >> > >> > >> On Thu, Jan 29, 2015 at 4:01 PM, Ben Boeckel > >> wrote: > >>> > >>> On Thu, Jan 29, 2015 at 14:48:00 -0500, Ken Martin wrote: > >>> > " I must admit, I'm a little bit disappointed that some of the > commits > >>> > in what was merged to VTK master don't even compile... I would hope > we > >>> > could be a bit less sloppy, and hold ourselves to a higher standard. > >>> > Rushing something in before it's ready for prime time serves no one." > >>> > > >>> > I want to make sure I understand what you are saying. Are you saying > in > >>> > this > >>> > case code was merged into master that did not compile? Or are you > >>> > saying > >>> > that someone did multiple commits, some of which did not compile, but > >>> > they > >>> > were not pushed or merged until it did compile? > >>> > >>> FWIW, I have a script I use to ensure this for long-lived branches: > >>> > >>> ======= > >>> #!/bin/sh > >>> > >>> base="${1:-master}" > >>> shift > >>> > >>> GIT_EDITOR="sed -i -e '/^pick/p;/^pick/cexec ./build.sh'" > >>> export GIT_EDITOR > >>> > >>> exec git rebase -i "$( git merge-base HEAD "$base" )" > >>> ======= > >>> > >>> A 'build.sh' script (provided by you) is run after each commit and > stops > >>> on any commit which fails to build. > >>> > >>> --Ben > >> > >> > >> > >> > >> -- > >> | Aashish Chaudhary > >> | Technical Leader > >> | Kitware Inc. > >> | http://www.kitware.com/company/team/chaudhary.html > > > > > > > > > > > > -- > > > > | Aashish Chaudhary > > | Technical Leader > > | Kitware Inc. > > > > | http://www.kitware.com/company/team/chaudhary.html > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dan.lipsa at kitware.com Fri Feb 6 15:12:56 2015 From: dan.lipsa at kitware.com (Dan Lipsa) Date: Fri, 6 Feb 2015 15:12:56 -0500 Subject: [vtk-developers] vtkDataSetAttributes::COPYTUPLE for GlobalIds and appending data Message-ID: Hello all, We would like the feedback of the community for the following problem: In an append operation, the GLOBALIDS array is not copied to the output because, by default, vtkDataSetAttributes::COPYTUPLE is 0 for the vtkDataSetAttributes::GLOBALIDS attribute. This behavior makes sense if appending datasets coming from different sources, but it does work in a ParaView use case. In ParaView running with a remote pvserver configuration, data is processed on MPI nodes, sent to node 0 and appended together there (using vktAppendCompositeDataLeaves) and then sent to the client. In this case, the GLOBALIDS arrays is valid, and should be preserved. We are thinking of two possible/alternate solutions to this problem: 1. Change vtkDataSetAttributes such that if COPYTUPLE for an attribute is 0, data is still copied but it does not set that attribute as an active attribute (using vtkDataSetAttributes::SetActiveAttribute). 2. Provide API for vktAppendCompositeDataLeaves (and vtkAppendFilter, vtkAppendPolyData) that specifies that the GLOBALIDS should be copied. Do you have any feedback on this? Thank you, Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From DLRdave at aol.com Fri Feb 6 15:33:10 2015 From: DLRdave at aol.com (David Cole) Date: Fri, 6 Feb 2015 15:33:10 -0500 Subject: [vtk-developers] vtkDataSetAttributes::COPYTUPLE for GlobalIds and appending data In-Reply-To: References: Message-ID: When I read "if COPYTUPLE for an attribute is 0, data is still copied" I think to myself, well that doesn't seem to make sense... I think (2) sounds like a better idea. 2 cents, D On Fri, Feb 6, 2015 at 3:12 PM, Dan Lipsa wrote: > Hello all, > We would like the feedback of the community for the following problem: > > In an append operation, the GLOBALIDS array is not copied to the output > because, by default, vtkDataSetAttributes::COPYTUPLE is 0 for the > vtkDataSetAttributes::GLOBALIDS attribute. > > This behavior makes sense if appending datasets coming from different > sources, but it does work in a ParaView use case. > > In ParaView running with a remote pvserver configuration, data is processed > on MPI nodes, sent to node 0 and appended together there (using > vktAppendCompositeDataLeaves) and then sent to the client. In this case, the > GLOBALIDS arrays is valid, and should be preserved. > > We are thinking of two possible/alternate solutions to this problem: > > 1. Change vtkDataSetAttributes such that if COPYTUPLE for an attribute is 0, > data is still copied but it does not set that attribute as an active > attribute (using vtkDataSetAttributes::SetActiveAttribute). > > > 2. Provide API for vktAppendCompositeDataLeaves (and vtkAppendFilter, > vtkAppendPolyData) that > specifies that the GLOBALIDS should be copied. > > Do you have any feedback on this? > > Thank you, > Dan > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From jcplatt at dsl.pipex.com Mon Feb 9 06:26:22 2015 From: jcplatt at dsl.pipex.com (John Platt) Date: Mon, 09 Feb 2015 11:26:22 +0000 Subject: [vtk-developers] vtkDataSetAttributes::COPYTUPLE for GlobalIds and appending data In-Reply-To: References: Message-ID: <54D8995E.3090000@dsl.pipex.com> Hi, This problem also arises when an unstructured grid must be split into parts because the field data is not continuous, for example, stresses between two different materials. Appending the parts again for node labelling loses the global Ids (node numbers) . vtkClipPolyData does not copy point global Ids which is reasonable because the data would typically need to be interpolated. But if the poly data consists solely of vertices, it is safe to copy them. vtkShrinkFilter does not copy cell global Ids; vtkUnstructuredGridGeometryFilter does not copy point global Ids (VTK 5.10.1). I would definitely go for an API on the algorithm which allows specification of attribute data to be copied. Thanks. John. On 06/02/2015 20:33, David Cole via vtk-developers wrote: > When I read "if COPYTUPLE for an attribute is 0, data is still copied" > I think to myself, well that doesn't seem to make sense... > > I think (2) sounds like a better idea. > > > 2 cents, > D > > > > On Fri, Feb 6, 2015 at 3:12 PM, Dan Lipsa wrote: >> Hello all, >> We would like the feedback of the community for the following problem: >> >> In an append operation, the GLOBALIDS array is not copied to the output >> because, by default, vtkDataSetAttributes::COPYTUPLE is 0 for the >> vtkDataSetAttributes::GLOBALIDS attribute. >> >> This behavior makes sense if appending datasets coming from different >> sources, but it does work in a ParaView use case. >> >> In ParaView running with a remote pvserver configuration, data is processed >> on MPI nodes, sent to node 0 and appended together there (using >> vktAppendCompositeDataLeaves) and then sent to the client. In this case, the >> GLOBALIDS arrays is valid, and should be preserved. >> >> We are thinking of two possible/alternate solutions to this problem: >> >> 1. Change vtkDataSetAttributes such that if COPYTUPLE for an attribute is 0, >> data is still copied but it does not set that attribute as an active >> attribute (using vtkDataSetAttributes::SetActiveAttribute). >> >> >> 2. Provide API for vktAppendCompositeDataLeaves (and vtkAppendFilter, >> vtkAppendPolyData) that >> specifies that the GLOBALIDS should be copied. >> >> Do you have any feedback on this? >> >> Thank you, >> Dan >> >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: http://markmail.org/search/?q=vtk-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/vtk-developers >> >> > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From aashish.chaudhary at kitware.com Wed Feb 11 09:11:57 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Wed, 11 Feb 2015 09:11:57 -0500 Subject: [vtk-developers] tracking VTK changes Message-ID: Hi all, I am trying to figure out some changes that resulted in problems parsing the VTK packages (Please see the message below). Is it possible that wrapping has changed in the last 4-5 weeks? Thanks, signatures = get_method_signature(method) File "/Users/aashish/tools/uvcdat/build_debug/install/vistrails/vistrails/packages/vtk/init.py", line 161, in get_method_signature tmptmp = doc.split('\n') AttributeError: 'NoneType' object has no attribute 'split' -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From DLRdave at aol.com Wed Feb 11 09:38:25 2015 From: DLRdave at aol.com (David Cole) Date: Wed, 11 Feb 2015 09:38:25 -0500 Subject: [vtk-developers] tracking VTK changes In-Reply-To: References: Message-ID: We took a snapshot of VTK on Oct. 30, 2014 - the git hash for nightly-master that day was ea28caf4. This command shows more than 1100 commits made since then: gitk ea28caf4.. & This command shows ~400 merge commits made since then: gitk --merges ea28caf4.. & This command shows the less than 30 commits made since then that touch any files in CMake or Wrapping or the CMakeLists.txt file: gitk --merges ea28caf4.. -- Wrapping CMake CMakeLists.txt & You can probably find any change affecting wrapping in this last set of commits. HTH, David C. On Wed, Feb 11, 2015 at 9:11 AM, Aashish Chaudhary wrote: > Hi all, > > I am trying to figure out some changes that resulted in problems parsing the > VTK packages (Please see the message below). Is it possible that wrapping > has changed in the last 4-5 weeks? > > Thanks, > > > signatures = get_method_signature(method) > File > "/Users/aashish/tools/uvcdat/build_debug/install/vistrails/vistrails/packages/vtk/init.py", > line 161, in get_method_signature > tmptmp = doc.split('\n') > AttributeError: 'NoneType' object has no attribute 'split' > > -- > | Aashish Chaudhary > | Technical Leader > | Kitware Inc. > | http://www.kitware.com/company/team/chaudhary.html > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From aashish.chaudhary at kitware.com Wed Feb 11 09:43:58 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Wed, 11 Feb 2015 09:43:58 -0500 Subject: [vtk-developers] tracking VTK changes In-Reply-To: References: Message-ID: Thanks Dave. I was hoping that someone who has been actively working or worked on the wrapping can provide a quick summary of the changes that could possibly cause the __doc__ slot to be missing before I dig down deep into the individual commits (which I was planning to do later today). - Aashish On Wed, Feb 11, 2015 at 9:38 AM, David Cole wrote: > We took a snapshot of VTK on Oct. 30, 2014 - the git hash for > nightly-master that day was ea28caf4. > > This command shows more than 1100 commits made since then: > gitk ea28caf4.. & > > This command shows ~400 merge commits made since then: > gitk --merges ea28caf4.. & > > This command shows the less than 30 commits made since then that touch > any files in CMake or Wrapping or the CMakeLists.txt file: > gitk --merges ea28caf4.. -- Wrapping CMake CMakeLists.txt & > > You can probably find any change affecting wrapping in this last set of > commits. > > > HTH, > David C. > > > On Wed, Feb 11, 2015 at 9:11 AM, Aashish Chaudhary > wrote: > > Hi all, > > > > I am trying to figure out some changes that resulted in problems parsing > the > > VTK packages (Please see the message below). Is it possible that wrapping > > has changed in the last 4-5 weeks? > > > > Thanks, > > > > > > signatures = get_method_signature(method) > > File > > > "/Users/aashish/tools/uvcdat/build_debug/install/vistrails/vistrails/packages/vtk/init.py", > > line 161, in get_method_signature > > tmptmp = doc.split('\n') > > AttributeError: 'NoneType' object has no attribute 'split' > > > > -- > > | Aashish Chaudhary > > | Technical Leader > > | Kitware Inc. > > | http://www.kitware.com/company/team/chaudhary.html > > > > _______________________________________________ > > Powered by www.kitware.com > > > > Visit other Kitware open-source projects at > > http://www.kitware.com/opensource/opensource.html > > > > Search the list archives at: > http://markmail.org/search/?q=vtk-developers > > > > Follow this link to subscribe/unsubscribe: > > http://public.kitware.com/mailman/listinfo/vtk-developers > > > > > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Thu Feb 12 10:39:16 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 12 Feb 2015 10:39:16 -0500 Subject: [vtk-developers] Google Summer of Code organization application 2015 Message-ID: Greetings developers, We are putting together another proposal for VTK to be a Google Summer Of Code organization. If VTK is accepted, Google will fund summer interns to work with on VTK related topics, with help from some of us as volunteer mentors. We have until next Friday February 20th to come up with a list of mentors and proposed topic ideas. As a mentor there is a significant time commitment, but it can also be very rewarding and it is a great way to get students involved in open source over the summer. If you are willing to mentor and have ideas for topics, please fill in a section or two of http://vtk.org/Wiki/VTK/GSOC_2014. thanks! David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 -------------- next part -------------- An HTML attachment was scrubbed... URL: From marcus.hanwell at kitware.com Thu Feb 12 10:53:36 2015 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Thu, 12 Feb 2015 10:53:36 -0500 Subject: [vtk-developers] Google Summer of Code organization application 2015 In-Reply-To: References: Message-ID: On Thu, Feb 12, 2015 at 10:39 AM, David E DeMarle wrote: > Greetings developers, > > We are putting together another proposal for VTK to be a Google Summer Of > Code organization. If VTK is accepted, Google will fund summer interns to > work with on VTK related topics, with help from some of us as volunteer > mentors. > > We have until next Friday February 20th to come up with a list of mentors > and proposed topic ideas. As a mentor there is a significant time > commitment, but it can also be very rewarding and it is a great way to get > students involved in open source over the summer. > > If you are willing to mentor and have ideas for topics, please fill in a > section or two of > http://vtk.org/Wiki/VTK/GSOC_2014. > I think the correct link is http://vtk.org/Wiki/VTK/GSoC_2015, note that they will work remotely and all interaction is using the available tools for online communication. I think our previous two years went really well, and am looking forward to seeing what students we get this year. Marcus From kimtaikee at gmail.com Fri Feb 13 04:12:39 2015 From: kimtaikee at gmail.com (kimtaikee at gmail.com) Date: Fri, 13 Feb 2015 17:12:39 +0800 Subject: [vtk-developers] How to use QVTKWidget & vtkChartXYZ correctly ? Message-ID: <201502131708129088071@gmail.com> Hey Guys,? ? I want to display vtkChartXYZ in a QVTKWidget, ?how could I accomplish that ? I tried the following code, but I get some errors, but the 3D graph could be shown. // Code:int TestLinePlot3D(int argc, char * argv[]){ // Create the data. QApplication app(argc, argv); QVTKWidget* widget = new QVTKWidget; vtkNew varXSolution; vtkNew arrX0; arrX0->SetName("X"); varXSolution->AddColumn(arrX0.GetPointer()); vtkNew arrX1; arrX1->SetName("Y"); varXSolution->AddColumn(arrX1.GetPointer()); vtkNew arrX2; arrX2->SetName("Z"); varXSolution->AddColumn(arrX2.GetPointer()); const unsigned int numberOfTimePoints = 1000; varXSolution->SetNumberOfRows(numberOfTimePoints); float varX[3]; varX[0] = 0.0f; varX[1] = 1.0f; varX[2] = 1.05f; float varXDerivative[3]; const float deltaT = 0.01f; for (unsigned int ii = 0; ii < numberOfTimePoints; ++ii) { varXSolution->SetValue(ii, 0, varX[0]); varXSolution->SetValue(ii, 1, varX[1]); varXSolution->SetValue(ii, 2, varX[2]); lorenz(varX, varXDerivative); varX[0] += varXDerivative[0] * deltaT; varX[1] += varXDerivative[1] * deltaT; varX[2] += varXDerivative[2] * deltaT; } // Set up a 3D scene and add an XYZ chart to it. vtkNew view; view->GetRenderWindow()->SetSize(400, 300); vtkNew chart; chart->SetGeometry(vtkRectf(10.0, 20.0, 250, 260)); view->GetScene()->AddItem(chart.GetPointer()); // Add a line plot. vtkNew plot; plot->SetInputData(varXSolution.GetPointer()); plot->GetPen()->SetColorF(0.1, 0.2, 0.8, 1.0); chart->AddPlot(plot.GetPointer()); // Finally render the scene and compare the image to a reference image. view->GetRenderWindow()->SetMultiSamples(0); view->GetRenderer()->SetBackground(0, 110, 0); chart->SetFitToScene(true);// widget->SetRenderWindow(view->GetRenderWindow());// view->SetInteractor(widget->GetInteractor()); view->GetInteractor()->Initialize();// view->GetInteractor()->Start(); // shouldn't be calling this view->SetRenderWindow(widget->GetRenderWindow()); widget->show(); return app.exec();} // Errors:ERROR: In E:\open\vtk\vtkvc32\VTK-6.1.0\Rendering\Context2D\vtkOpenGLContextDevice2D.cxx, line 1244 vtkOpenGL2ContextDevice2D (074DB1B8): failed after PopMatrix 1 OpenGL errors detected 0 : (1284) Stack underflow ERROR: In E:\open\vtk\vtkvc32\VTK-6.1.0\Rendering\Context2D\vtkOpenGLContextDevice2D.cxx, line 177 vtkOpenGL2ContextDevice2D (074DB1B8): failed after End 1 OpenGL errors detected 0 : (1284) Stack underflow Any help would be appreciated. P.S. I am using Qt5.4.0 & VTK 6.1.0 Thanks.Jason Zhang. -------------- next part -------------- An HTML attachment was scrubbed... URL: From will.schroeder at kitware.com Sun Feb 15 18:28:37 2015 From: will.schroeder at kitware.com (Will Schroeder) Date: Sun, 15 Feb 2015 18:28:37 -0500 Subject: [vtk-developers] vtkCheckerboardSplatter Message-ID: *Summary:* In our continuing effort to speed up VTK and add in new parallel computing support, we just implemented a new version of vtkGaussianSplatter. The new class, vtkCheckerboardSplatter, is much faster in serial mode, and if you enable threading (e.g., use TBB through vtkSMPTools) it is >10x faster depending on the number of threads and particulars of the input data. While I was at it, I also multithreaded a portion of vtkGaussianSplatter so this class is a bit faster too when threading is enabled. The code is undergoing review in gerrit: http://review.source.kitware.com/#/c/19269/ *Background:* Surprisingly vtkGaussianSplatter is still in widespread use, mainly to convert unstructured data (by sampling with points), or data without topology, into a volumetric dataset. The resulting volume can then be isocontoured, volume rendered, etc. The problem with this class is that it was originally written over 20 years ago as a demonstration and has lots of other cobwebs in evidence. Most importantly, this original implementation does not lend itself to effective shared memory parallel processing. As a result, we created the vtkCheckerboardSplatter that threads across the splat footprint, as well as splatting multiple points simultaneously. The basic idea is to use a 8-way checkerboarding to splat points in safely separated regions of the volume so that the splat operations can write into the output volume without write contention. Think of creating blocks consisting of checkerboard squares (like an octree) each one of eight colors, where the points resident in one of the colors/squares can be simultaneously splatted in parallel. The splat operation, for big enough splat footprints, is threaded as well. There were some other modifications to the API, mainly the splat footprint is separated from the Gaussian radius. Also, the accumulation process is slightly different, and the input and output types have been expanded to include combinations of (float,double). If you are interested, please try it out and provide feedback. I'm also looking for candidates for other classes to multithread / speed up, so please let me know via separate email. At some point I'll write up a paper but I'm having too much fun coding now to be slowed down by that sort of thing :-) -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.amaclean at gmail.com Sun Feb 15 22:25:52 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Mon, 16 Feb 2015 14:25:52 +1100 Subject: [vtk-developers] CMP0020 warning in Tutorial Example 6... why? Message-ID: When I build the examples in VTK why do I get a CMP0020 warning in Examples/Tutorial/Step6/Cxx/CMakeLists.txt but not for any of the other examples. It would be easily fixed by adding: if(POLICY CMP0020) cmake_policy(SET CMP0020 NEW) endif() to the CMakeLists.txt file. However I don't see why this particular example needs it. It doesn't use QT and I don't see why the other examples don't need this fix. Can anyone provide some insight here? I am building with CMake 3.1.3, QT5.4, OpenGL2 using MSVC 3013. This warning occurs multiple times when building VTK: ------------------------------------ CMake Warning (dev) in Examples/Tutorial/Step6/Cxx/CMakeLists.txt: Policy CMP0020 is not set: Automatically link Qt executables to qtmain target on Windows. Run "cmake --help-policy CMP0020" for policy details. Use the cmake_policy command to set the policy and suppress this warning. This warning is for project developers. Use -Wno-dev to suppress it. ------------------------------------ Thanks for any insight. Regards Andrew -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.maynard at kitware.com Mon Feb 16 08:12:08 2015 From: robert.maynard at kitware.com (Robert Maynard) Date: Mon, 16 Feb 2015 08:12:08 -0500 Subject: [vtk-developers] CMP0020 warning in Tutorial Example 6... why? In-Reply-To: References: Message-ID: Have you tried removing the find_package(VTK) on line 19? On Sun, Feb 15, 2015 at 10:25 PM, Andrew Maclean wrote: > When I build the examples in VTK why do I get a CMP0020 warning in > Examples/Tutorial/Step6/Cxx/CMakeLists.txt but not for any of the other > examples. > It would be easily fixed by adding: > if(POLICY CMP0020) > cmake_policy(SET CMP0020 NEW) > endif() > to the CMakeLists.txt file. However I don't see why this particular example > needs it. It doesn't use QT and I don't see why the other examples don't > need this fix. > > Can anyone provide some insight here? > > I am building with CMake 3.1.3, QT5.4, OpenGL2 using MSVC 3013. > > This warning occurs multiple times when building VTK: > ------------------------------------ > CMake Warning (dev) in Examples/Tutorial/Step6/Cxx/CMakeLists.txt: > Policy CMP0020 is not set: Automatically link Qt executables to qtmain > target on Windows. Run "cmake --help-policy CMP0020" for policy details. > Use the cmake_policy command to set the policy and suppress this warning. > This warning is for project developers. Use -Wno-dev to suppress it. > ------------------------------------ > > Thanks for any insight. > > Regards > Andrew > > -- > ___________________________________________ > Andrew J. P. Maclean > > ___________________________________________ > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From aashish.chaudhary at kitware.com Mon Feb 16 11:25:27 2015 From: aashish.chaudhary at kitware.com (Aashish Chaudhary) Date: Mon, 16 Feb 2015 11:25:27 -0500 Subject: [vtk-developers] vtkCheckerboardSplatter In-Reply-To: References: Message-ID: Will, This is an excellent news! And will be useful for some of the projects I am involved with. We will come up with some numbers for your information. Thanks, - Aashish On Sun, Feb 15, 2015 at 6:28 PM, Will Schroeder wrote: > *Summary:* > In our continuing effort to speed up VTK and add in new parallel computing > support, we just implemented a new version of vtkGaussianSplatter. The new > class, vtkCheckerboardSplatter, is much faster in serial mode, and if you > enable threading (e.g., use TBB through vtkSMPTools) it is >10x faster > depending on the number of threads and particulars of the input data. > > While I was at it, I also multithreaded a portion of vtkGaussianSplatter > so this class is a bit faster too when threading is enabled. > > The code is undergoing review in gerrit: > http://review.source.kitware.com/#/c/19269/ > > > *Background:* > Surprisingly vtkGaussianSplatter is still in widespread use, mainly to > convert unstructured data (by sampling with points), or data without > topology, into a volumetric dataset. The resulting volume can then be > isocontoured, volume rendered, etc. The problem with this class is that it > was originally written over 20 years ago as a demonstration and has lots of > other cobwebs in evidence. Most importantly, this original implementation > does not lend itself to effective shared memory parallel processing. As a > result, we created the vtkCheckerboardSplatter that threads across the > splat footprint, as well as splatting multiple points simultaneously. > > The basic idea is to use a 8-way checkerboarding to splat points in safely > separated regions of the volume so that the splat operations can write into > the output volume without write contention. Think of creating blocks > consisting of checkerboard squares (like an octree) each one of eight > colors, where the points resident in one of the colors/squares can be > simultaneously splatted in parallel. The splat operation, for big enough > splat footprints, is threaded as well. > > There were some other modifications to the API, mainly the splat footprint > is separated from the Gaussian radius. Also, the accumulation process is > slightly different, and the input and output types have been expanded to > include combinations of (float,double). > > If you are interested, please try it out and provide feedback. I'm also > looking for candidates for other classes to multithread / speed up, so > please let me know via separate email. At some point I'll write up a paper > but I'm having too much fun coding now to be slowed down by that sort of > thing :-) > -- *| Aashish Chaudhary | Technical Leader | Kitware Inc. * *| http://www.kitware.com/company/team/chaudhary.html * -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Mon Feb 16 16:43:12 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Mon, 16 Feb 2015 16:43:12 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready Message-ID: The VTK developement team is happy to announce that VTK 6.2 has entered the release candidate stage! You can find the source, data, and new vtkpython binary packages here: http://www.vtk.org/VTK/resources/software.html#latestcand Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.2.0 final. The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview: -- VTK?s use of the parallelism available in modern architectures continues to mature. To set the stage for this move, we have refactored how pieces and extents are handled in the pipeline. The pipeline used to inject extent translators into the pipeline to obtain structured sub-extents for each requested unstructured piece. This approach had flaws so now the translation responsibility falls to the readers which are best able to make the computation and parallel filters are now responsible for dealing gracefully with unexpected extents. Next we have we have updated to the latest version of the Dax toolkit and begun to lay the groundwork for adopting the vtk-m project as a successor to the Piston, Dax, and EAVL efforts. There have been a series of minor improvements to the SMP framework and vtkSMP filters too as we get ready to adopt them in a large swath of VTK?s algorithms. In related work, note that we have begun to support Xeon Phi (MIC) chips. For information about this please refer to: h ttp://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29 . Similarly VTK?s support for the Web continues to advance in this release. VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and updated examples of using the web launcher to start vtkweb applications. vtkWeb applications now support http-only server/client configurations for situations when websockets are not acceptable. There were also improvements made to VTK to support the creation and use of Cinema and Workbench applications (in-situ deferred visualization) that are described in a SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ Visualization and Analysis? Wrapped languages have gotten a share of the attention in this release. There have been several fixes for ActiViz.Net which was recently updated from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples have finally been upgraded following modularization and note that it is now mandatory to invoke the method Start on the instance of vtkRenderInteractor from Tcl. In Java, vtkPanel?s behavior was changed to better support advanced class loader system like OSGI. VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven. Python has received the heaviest dose of updates. Wrapped namespaces and enum types are now available in Python. We?ve also made significant improvements to the VTK-numpy integration by introducing a number of Python modules that provide numpy-compatible interfaces to VTK data structures. See http://www.kitware.com/blog/home/post/723 for details. We?ve also introduced vtkPythonAlgorithm, which makes it easier than ever to quickly extend VTK with filters that are written directly in Python. See http://www.kitware.com/blog/home/post/737 for details. VTK?s rendering is making both evolutionary and revolutionary advances in this release. The evolutionary changes include the usual number of incremental improvements. These include such things like advanced color and display controls, "sticky" axes mode for vtkCubeAxesActor, out of range color assignments, and indexed color lookups for vtkStringArrays. There are also a number of text rendering improvements such as better multiline, rotated, and aligned text as well as BackgroundColor and BackgroundOpacity options. The revolutionary changes can be found in the OpenGL2 modules. These are under active development at the moment, and the API will be subject to change in the next release, but early adopters are encouraged to try it out and report their experiences with it on the mailing list. OpenGL2 is a rewrite of VTK?s rendering backend that brings VTK up to date with modern OpenGL programming practices. By replacing antiquated rendering techniques with modern ones, we have increased rendering performance by orders of magnitude in some situations. This work, funded by the NIH VTK Maintenance grant touches both surface and volume rendering techniques. You can read about OpenGL2 at: http://www.kitware.com/source/home/post/144 and http://www.kitware.com/source/home/post/154 Besides the above progress some of the most notable changes in VTK 6.2 include - deprecated InfovisParallel - vtkIOXdmf3, an interface to ARL?s greatly improved interface to HDF5 backed data storage - added a reader and writer for NIfTI files, including the 64-bit NIfTI-2 format - added support for SpaceMouse devices - external rendering support for immersive environments http://kitware.com/blog/home/post/688 - removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS - Rewrote the OS X Cocoa mouse event handling code to make it more robust. We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- Aashish Chaudhary (261): Fixed headlight not working right when using the ext renderer Fixed missing display, window, and context id not being set Fixed splitviewport stereo rendering Added new volume rendering code Fixed texture state being dirty Use correct local bounds Fixed rotation bug Fixed style Fixed geometry beign missing after volume render Fixed indent Silenced cast warnings Fixed style and naming conventions Added copyright Embedded volume shaders into C++ units Fixed broken volume rendering because of uninitialized variable Fixed code document style a bit Added method to validate volume rendering Enable sample distance control Enabled additive and composite blend mode Use depth buffer to terminate ray Fixed incorrect mix geometry and volume rendering Added string replacement code First pass on compositing shader for the rendering Formalizing the grammar for the shader replacement Updated volume code to use new compositing mode Updating shader to be fully programmable Fixed loose ends Using appropriate variable names Rebuild shader if necessary Rename Helper to ShaderComposer as thats best suited Added tests for the new volume mapper Consolidating shading and color accumulation Got maximum intensity working Pass mapper to composer calls as its needed to query volume params Simplified the code a bit Fixed minor calculation bug Got rid of hard-coded maximum number of samples Removed references to vtkgl Separated lookup for clarity Compute and store actual max value Fixed bad gl state at the start of volume rendering Implemented cropping feature Using a better amplitude for the noise texture Use boolean to skip or consider a pixel for rendering Build cropping shader dynamically Condensed preprocessor tags Perform close bounds check Updated tests Got clipping working Passing size and data for clip planes in a single uniform array Implemented clipping via substitution Fixed clipping for multiple clip planes Fixed normals not getting passed correctly Fixed crash on resizing Updated clip test planes to reasonable values Following consistent way of checking for uninitiliazed values Removed hard-coded sample distance Adding minimum intensity blend capability Fixed wrong initial value of the min value Fixed biased not being passed to the shader Some more minor fixes Fixed rendering of short and some other types Added support for double (and other 8 bytes) datatypes Adding alpha blending Fixed issues with additive blending Renaming so that we can replace the old one with the new one Porting new volume mapper to proper vtk Got new mapper working Renamed the new volume module Fixed cropping Some more fixes Removed experimental code on cropping flags Updated lighting parameters Adding gradient opacity Implementing gradient opacity Added gradient opacity table to the mapper Added some more pieces for the gradient opacity Enabled auto-adjust sample distance Fixed style Added new gradient opacity test Added debug code Adding gradient opacity calculation Fixing / cleaing variables that are not required Fixed access from the texture Fixing the scaling of the gradients Fixing long data types volumes Using GLSL version 120 Making gradient opacity optional Removed check for gl error for now Fixed shader issues on MAC Fix white edge showing up Fixed wrong invert matrix being passed Revert "Removed experimental code on cropping flags" Fixed cropping test Fixing gradient vector computation Added support for 4 components Adding support for mask type Fixing shading Adding support for label mask type Fixing issues Restructured the code a bit more to make it flexible Fixed some more issues Fixed lighting issues Clean up some code Improvig code readability VTK style fixes Adding support for parallel projection Got parallel projection working Disabled the gradient opacity test for now Fixed cpp issues Silenced compiler warnings Implemented release graphics method For now set reduction ratio to 1.0 Fixed texture units Fixed wrong bounds used to pass bound values to shader Few style fixes Updated vtk point picker Added support for rendering when clipping by near plane Fixed volume geomety not updated Using smallest possible delta to offset precision issues Fixed release graphics resources Re-added gradient opacity Wrapped long lines Organzied the code bit better Disabled the gradient opacity test for now Fixed failing tests Added missing API to set uniform Updating to use vtk shader API Removed calls to old API Fixed volume not showing up Removed unused files Fixed licensing Removed unused file from cmake Updated to use vtkTextureObject Use vtkTextureObject if set Updated ray cast image display helper to use GL2 Removed unused code Fixed casting to wrong type; copy-paste error Updated headers Fixed test failing by restoring defaults Updated logic for testing for translucency Updated logic to check for external texture object Updated display helper Improved code by removing redundant code Fixed style issues Added more volume tests that are passing now Removed unused header files Set referencee to render window for external texture object Fixed typo Mark as translucent if 1,2 or 4 components With Ken, fixed resources not release issue Added notes, comments Updated volume rendering to use shader cache Added error checks for volume render calls Using render window shader cache Fixed volume tests failing because of shader cache ShaderCache should release its graphics resources Removed unnecessary call and added notes Let shader cache invoke release graphics resources on shader Attempt to fix failing test by removing calls to interactor Added second baseline to fix failing test on some systems Adding support for light kit Passing required parameters to the shaders Handling more lighting cases Fixing headlight Fixed gradient opacity computation Implementing positional light Adding test for testing light kit and fixed shader issues Enabling reduction factor for large volumes Updated so that we can compute the time to draw correctly Updating tests for consistent dashboard results Silenced compiler warnings Updated baseline and test for light kit Removed dead code Fixing lighting computation for lightkit Added convenient method to check for gradient opacity Updated code to use check on availability of gradient opacity Removed dead code Check for two sided lighting Attempt to fix failing tests Added a note on the workaround as well Commented out debug code Renamed for consistency Fixed positional light calculation Fixing positional light bugs Clean up code Compute appropriate matrix for normal transformation Fixed variable not found Fixing variable names Fixed style issues Fixed various lighting and formatting issues Silenced compiler warnings Fixed smart volume mapper crashes because of gradient opacity Fixed smart volume mapper level test Added GDAL based raster reader Fixed various minor issues Added baseline and data for the testing Addressed minor code clean up issues Fixed various cleanup issues Fixed bad sampling issues when dealing with large sample distances Fixed incorrect name used for the test file Update texture with the data Fixed data not showing up Use last texture units for 3D textures Added ability to set depth mode formats Added ability to use texture modes Set texture parameters before creating the texture Adding support for 3D textures Fixed check for depth texture mode Apply scale and bias in the shader Added ability to set texture attributes directly Make sure to release graphics resources Updating volume masks Handle large data type textures Fixing various issues Updated and improved texture object implementation Allow reuse of texture units Removed debug messages Fixing MIP issues Fixing some more issues Fixed memory leaks and release of textures Removed dead code Removed code for depth texture mode as they are deprecated Style fixes Fixed loose ends Removed extra flag to force re-computation of texture format Updated code so that applications will still work in OpenGL2 Fixed when mappers try to reuse the same texture Re-added volume benchmark test Typedef the volume test Updated volume test Tweaked opacity function to force more computation Clean up cmake a bit Adding GPU independent component feature Updated test to highlight mappings Adding support for multiple components in GPU RayCast Remove limiting check for independent components Adding support for multi RGB tables Fixed erroneous logic Updating shader code to support independent components Adding support for independent components for opacity and color function Some more updates to support independent components Fixed composite blend mode Added independent support for min and max projections Pass component weights to the shader Fixed testing Renaming classes for consistency Added support for independent components for gradient opacity Improved the test a bit Reinitialize transfer functions if the property changes Independent component is not supported in OpenGL GPU mapper Fixed MIP independent shader Fixed independent component computing for min intensity projection Fixing volume tests not passing on intel machine Removed unused code and check Do not assume initial min and max values Attempt to fix failing independent component tests GLSL 120 does not support sampler arrays Added missing header Fixed errors found using valgrind leak check Silenced compiler warnings Andrew Bauer (27): Need to check for static targets before adding abi visibility flags. Composite data writers weren't passing down the HeaderType info. Passing missing parallel writer options to piece writers. Fixing overflow issue that appeared with GCC compilers. Fix the way point and cell data arrays are marked for reading. Adding in arctan2 through numpy. Adding the ability to not delete the passed in arrays. Fixed memory leak in test and added in comments for clarification. Adding in a filter to append structured grids together. Adding the ability to merge structured grids for composite data sets. Fixed issue with ghost cells not being deleted properly. Removing temporary file that accidentally got added. Fixed issue of possibly modifying other polydatas. Removing Tcl tests the duplicate existing Python testing. Fixed incorrect cell data output. Making some cell locators more efficient. Giving a warning if trying to create a subcontroller with 0 procs. Temporary fixes for finding ADIOS. Fixing flow paths correct injection time and ability to add extra arrays. Renaming CurrentTime to CurrentTimeValue due to windows mangling issues. Fixing particle paths to have proper SimulationTime output. Adding in SimulationTimeStep output for particle paths. Getting rid of clang analyzer warnings. Skipping the wasteful and unnecessary inverse sin computation. Cosmetic change to be consistent with file codeing convention. Fix parallel flow paths filters. Improving testing of deleting cells from a polydata. Andrew Maclean (13): Remove WindowLevelInterface.tcl and WindowLevelInterface.py Revert Remove WindowLevelInterface.tcl and WindowLevelInterface.py Sphere widget behaves better when scaled down to zero ENH: VTK Parametric surfaces - modernised the code. ENH: Remove the need to manually regenerate the hill data. DOC: Update vtkRandomHills documentation. BUG: Fix by adding a new keyword. ENH: Add option to return an ordinal lookup table. Make the generation of random hills consistent across all platforms. Add a missing header in vtkCompositePolyDataMapper2 ENH: Restore / convert Infovis examples Lots of minor changes for Examples to build with OpenGL2 This fixes the c2039 error in VS2013. Antonio Cervone (1): Fixed the BiQuadraticQuad interpolator Arnaud Gelas (1): Handle properties are not properly set up when starting vtkBoxwidget2 Ben Boeckel (254): STLWriter: Raise an error for non-triangles STLReader: Close the file when an error occurs ExodusIIReader: Add the title to the field data STLReader: Fix error signaling when reading files AVIWriter: Don't overflow the fourcc storage AVIWriter: Fix a comment typo Exodus: Always attach the title to the output fields Rewrite the testing macros Update the Cxx test call sites Update the MPI test call sites Update the Python test call sites Update Tcl test call sites Minor CMake logic simplification Copy Statistics tests to StatisticsGnuR Infovis: Add an include for newer Boost versions ParallelMPI: Add missing return statement Allow arguments to be passed directly to Python Skip adding the executable if no tests are created Allow for overriding the data directory in tests Allow overriding the baseline directory in tests Build the VTK test data directory properly Support custom baselines in tests Store VTK's ExternalData in its source directory Support VTKExternalData_OBJECT_STORES as well Support custom ExternalData targets for tests Support flags to tests common at the module level Support JUST_VALID for Python tests Support Python+MPI testing Use VTK_TEST_OUTPUT_DIR rather than rebuilding the path examples: Fix typo in testing calls ParallelGeometry: Fix Grid symbol scoping testing: Use vtk-example as a fallback for prefix Fix test function signatures testing: Don't look for tests when making an executable FiltersGeometry: Remove unnecessary links Filters/Geometry: Remove some unused functions from tests Filters/Geometry: Hide WriteGrid if ENABLE_IO isn't on testing: add some missing TEST_DEPENDS python: use typedef names python: add reference counting to the object map FindPythonModules: use a function FindPythonModules: search if not found pythoncore: extract python objects from arguments wrapping: add function to check for Python objects wrapping: wrap PyObject* in python code python: accept any argument for PyObject* arguments vtkAlgorithm: fix a typo FiltersPython: create a vtkPythonAlgorithm class python: fix leaks when setting constants in wrappers vtkPythonAlgorithm: support older pythons vtkpython: add site-packages to the path vtkFiltersPython: Disable without Python wrapping wrapping: add variables to exclude from language-specific wrapping vtkFiltersPython: exclude from Java and Tcl wrapping java: get the list of java modules properly testing: append the tests to the given variable pythonalg: use keyword arguments rather than members pythonalg: raise an error in RequestData by default wrapping: remove the ${module}_WRAP_PYTHON variable vtkStructuredImplicitConnectivity: fix msvc warnings typo: fix some typos unused: remove unused functions mpi4py: add module information mpi4py: link to CMAKE_DL_LIBS testing: skip test executables with no tests IOVPIC: include MPI headers as SYSTEM headers cmake: set policy 0053 if available cmake: set policy 0053 in thirdparty cmake mpi4py: add a class to convert communicators wrapping: check for Tcl wrapping exclusion for warnings vtkAMRUtilities: split out vtkParallelAMRUtilities hierarchy: use a .txt extension GenerateExportHeader: support object libraries as well vtk_module: support a KIT argument modules: add modules to "kits" linking: wrap tll() with a module-aware function export: OBJECT libraries may not be exported modules: output kit information in the module tree modules: topo-sort modules with kits as well modules: build kits from the member modules module: depend on the dependent module object library python: add an option to disable vtkpython testing: check for pvtkpython before using it vtkpython: don't install if not enabled vtkAVIWriter: make a more informative error message IO/Geometry: add obj index tests FiltersParallelGeometry: move functions to the sources vtkMPICommunicator: avoid deprecated MPI functions vtkInstantiator: remove instantiator macro from main sources typo: remove duplicate source listing vtkCompositePolyDataMapper2: cache opacity check vtkCompositePolyDataMapper2: include for std::max vtkCompositePolyDataMapper2: avoid a NULL dereference OpenGL: use STATUS for the OSMesa message vtkPStreamTracer: conditionalize functions vtkPTemporalStreamTracer: remove useless code unused: add a macro to mark functions as "unused" vtkChartBox: initialize segmentIndex nit: remove extra semicolon vtkMPI: install so that vtk_mpi_link is available python: install packages properly opengl2: use STATUS for the OSMesa message Maintenance: recognize the KIT keyword in scripts vtkmpi4py: empty out Py_{UN,}BLOCK_THREADS module: add comment to update maintenance scripts too Maintenance: recognize language exclusion keywords warnings: fix narrowing warnings TestSurfacePlot: loop using vtkIdType vtkQuaternionInterpolator: call the constructor properly warnings: remember values with the proper type warnings: use static casts to point out precision loss warnings: use floating point intermediate values warnings: resolve unused variables vtkOpenGLImageMapper: fix unintentional shadowing warnings: fix signed/unsigned comparison warnings vtkOpenGLProperty: remove unused function unused: use the vtkNotUsed macro Q4VTKWidgetPlugin: only define qDebug if not already defined mpi4py: use the output, not the result cmake: remove unnecessary condition copies vtkSmartPyObject: ignore warning C4127 on MSVC mpi4py: use old code with Python 2.6 as well web: support Image from PIL in testing web: print the traceback from errors as well vtkPStreamTracer: remove unused function vtkModuleMacros: use STREQUAL rather than MATCHES cmake: use functions instead of macros cmake: minor style nits vtkPythonAlgorithm: print out all the error information tk: update to support 8.6 vtkEnSightReader: use memmove for overlapping regions vtkFFMPEGWriter: use av_malloc to match the av_free vtkChartXY: check the scene before use otherByteSwap: fix calls for byteswapping netcdf: don't use the cache to set variables in the parent scope HeaderTesting: skip non-vtk* files HeaderTests: only run if VTK_SOURCE_DIR is set vtkChartXYZ: use dummy bounds rather for bounds calculation vtkChartXYZ: don't copy the points out RenderingOpenGL: remove unused file cmake: fix CMP0054 warnings cmake: remove variable dereferences in if statements windows: use the proper format specifier for DWORD windows: put parentheses around mixed && and || operators vtkWin32OpenGLRenderWindow: avoid narrowing warnings vtkWin32Header: support 64-bit properly on non-MSVC vtkWin32Header: always use {Get,Set}WindowLongPtr hdf5: avoid CMP0054 warning vtkPythonPackages: use IN LISTS syntax in foreach vtkPythonPackages: avoid CMP0054 warnings vtkOpenGLTexture: don't check for a mapped window cmake: remove arguments to else and endfoo commands cmake: remove arguments from else and endfoo in alglib cmake: remove arguments from else and endfoo in exodusII cmake: remove arguments from else and endfoo in expat cmake: remove arguments from else and endfoo in freerange cmake: remove arguments from else and endfoo in freetype cmake: remove arguments from else and endfoo in ftgl cmake: remove arguments from else and endfoo in gl2ps cmake: remove arguments from else and endfoo in hdf5 cmake: remove arguments from else and endfoo in jpeg cmake: remove arguments from else and endfoo in libproj4 cmake: remove arguments from else and endfoo in libxml2 cmake: remove arguments from else and endfoo in mpi4py cmake: remove arguments from else and endfoo in netcdf cmake: remove arguments from else and endfoo in oggtheora cmake: remove arguments from else and endfoo in png cmake: remove arguments from else and endfoo in sqlite cmake: remove arguments from else and endfoo in TclTk cmake: remove arguments from else and endfoo in tiff cmake: remove arguments from else and endfoo in verdict cmake: remove arguments from else and endfoo in VPIC cmake: remove arguments from else and endfoo in xdmf2 cmake: remove arguments from else and endfoo in xdmf3 cmake: remove arguments from else and endfoo in zlib cmake: remove arguments to else and endfoo commands cmake: remove @var@ expansions FindTBB: fix up typo'd logic cmake: fix up some version checks zlib: remove empty if statements vtkInformationDataObjectKey: move to CommonDataModel numpy_support: always support [u]int64 numpy_support: raise an error if a type cannot be found Revert "vtkInformationDataObjectKey: move to CommonDataModel" numpy_support: support using dtype array types cmake: use COPYONLY rather than COPY_ONLY vtkWrap: some uses of COPY_ONLY weren't even wrong numpy_support: don't convert array types if possible numpy_support: deep copy for type mismatched array data numpy: test numpy_to_vtk when requesting specific types vtkParticleTracerBase: remove unused function vtkImageCanvasSource2D: add missing newline at eof vtkKdNode, vtkOctreePointLocatorNode: don't assign after a short-circuit vtkPolyhedron: initialize the iterator HyperTree example: use the iterator before invalidating it Widgets example: fix up logic around planeSource vtkYoungsMaterialInterface: plug memory leaks vtkUnstructuredGridGeometryFilter: plug memory leak vtkPKdTree: plug memory leak vtkDistributedDataFilter: plug memory leaks vtkKMeansStatistics: plug memory leak vtkAMRFlashReaderInternal: don't return a pointer to temporary memory vtkExodusIIWriter: use a larger buffer TestIncrementalOctreePointLocator: close file handles vtkCGMWriter: fix typo vtkNIFTIImageReader: plug memory leaks vtkSLCReader: plug memory leak vtkPLY: handle realloc errors properly vtkEnSightWriter: terminate buffers properly vtkImagePlaneWidget: avoid divide-by-zero vtkResliceCursorRepresentation: avoid divide-by-zero vtkParseMain: handle realloc errors properly vtkParseMain: close the input file vtkParsePreprocess: handle realloc errors properly vtkParsePreprocess: fix up some free() logic for params vtkWrapHierarchy: handle realloc errors properly TestIncrementalOctreePointLocator: use malloc vtkParse: regenerate lex.yy.c vtkParse: address cppcheck messages vtkReebGraph: add a comment about realloc usage vtkOpenGLPolyDataMapper2D: comment possible divide-by-zero TestIncrementalOctreePointLocator: remove dead code vtkWin32OpenGLRenderWindow: use size_t for integer storage of pointers vtkWin32OpenGLRenderWindow: wrap assignments in while() in parens hdf5: check for the right policy python: remove echo from custom target FindMPI: search for msmpi as well FindMPI: fix a CMP0054 warning for CMake FindMPI: add in the SDK path FindMPI: support newer MSMPI hdf5: remove conflict markers FindMPI: convert to a safer path representation. vtkDataArrayTemplate: fix support for compilers without explicit instantiation Copyright: bump the year vtkRenderingVolumeOpenGL2: remove copy_shaders target IOXdmf3: remove excess MPI stuff install: install missing module-related files netcdf: support a system netcdf cmake: reformat java compilation commands java: add an option to set the target for java compiles wxVTKRenderWindow: support wxPython 3.0 wrapping: find superclasses that are templates examples: add shebang lines doxygen: remove obsolete options doxygen: add an option to disable CHM files release: add scripts for release maintenance python: don't export Python modules as targets maint: use "git checkout" rather than reset maint: ignore css files for ctags as well metaio: remove stray reject file vtkModuleTop: handle UsrMove-like setups vtkWrap: use size_t for sizes vtkPistonMapper: fix a overload-virtual warning vtkWrapTcl: hide dString when there are no methods Berk Geveci (64): Removed priority based streaming and fast path. Removed unused code from vtkModelMetaData and vtkExodusModel COMP: Fixed compiler error - needed to remove functions from .h. Improved test to increase vtkDistributedDataFilter's coverage. Cleaned up the Exodus metadata. Fixed minor typo - case sensitivity issue. Removed verts from streaklines - lines are enough. Removed unnecessary recursion in pipeline streaming. Filter was modifying self in request. Fixed. Fixed minor typo in documentation. Added test for vtkTableBasedClipDataSet. Added caching to temporal interpolator. Fixed minor bug in the ensight gold reader. Added EnSight tests to increase coverage. Added initial ghost points support to vtkDataSetSurfaceFilter. PERF: Parallel streamline was slow. Fixed. Refactored how pieces and extents are handled. Update the XML readers and writers to work with pipeline changes. Cleaned up and fixed transmit filters for structured data. Fixed bug in send structured data. Fixed legacy parallel reader and writer. BUG: Was using array of wrong size. Fixed compiler warnings. Changed synchronized templates class for new pipeline behavior. Removed unused keys. Updated filters and sources to work with new pipeline logic. Moved key. Cleaned up extent translators. Removed unnecessary reference to extent translator ExtractSelection was not managing pieces corretly. Fixed. Filters should silently handle empty input Fixed multiple issues with resample/probe filters. Fixed contour filters Fixed issues in the group filter. Fixed crash in the empty input test. Fixed vtkXdmfReader for parallel structured data. Parallel Exodus writer was not asking for pieces. Fixed. Added new meta-data and request capability to the pipeline. PERF: Removed unnecessary function call. Added an ensemble source. Added documentation and testing to vtkPythonAlgorithm. Fixed no newline at end of file warnings. Executive was setting update extent initialized falsely. PERF: Removed unnecessary garbage collection. Fixed indentation and style to match VTK's. BUG: Filter was overwriting input scalars. Fixed. Added modules that create a numpy type interface. Certain MPI types were not being recognized by mpi4py. Improved TestNumpyInterface to cover various append implementations. Added sum and mean per block to Python algorithms. Added support for CompositeDataSet.Points. Fixed innaccurate comments. Added a SeedIds array to the output of stream tracer. Added support for specifying split mode. Added VTKPythonAlgorithmBase - base class for Python algorithms. Fixed vtkImageGradient in parallel. Added Python support for making new information keys. Removed mrmpi ThirdParty, which was not used. Fixed quadratic quad and biquadratic quad derivatives. Added a new example demonstrating developing SMP algorithms. Added another SMP example. Fixed bug in vtkExecutive::CopyDefaultInformation. Fixed bug in vtkMaskPoints that caused corrupt output. Improved unstructured grid support in dataset_adapter. Bill Hoffman (2): Fix build with mingw 4.8.2 and qt 5.3.2, bug #14735 Detect win32 or pthread model in a mingw gcc build. Bill Lorensen (39): COMP: Test driver array bounds error BUG0013829: vtkAssembly::GetBounds adds bounds which are uninitialized BUG0014331: Filter skips VTK_CUBIC_LINE COMP: Restore TestTessellator test ENH: Added tests that were removed during modularization COMP: Refactored TestDataObjectIO COMP: test require NO_VALID modifier ENH: Add removed tests COMP: Improve coverage of vtkMeshQuality COMP: Missing NO_VALID ENH: Improve code coverage script BUG: Flawed logic for NO_OUTPUT COMP: Python unittest requies NO_OUTPUT COMP: Performance warning: int to bool COMP: Remove QtInitialization from otherPrint test ENH: Remove ThirdParty code from coverage results COMP: Remove bogus line info from generated files ENH: Unit test for ParametricSpline ENH: Coverage for MultiThreshold. BUG: CreateDefaultLookupTable crashes if Input is NULL. ENH: Add some more coverage for MultiThreshold ENH: Improve code coverage for UnicodeStringArray ENH: Unit test for vtkMath COMP: Adjust tolerances COMP: Guard against infinite loop in vtkClearOpenGLErrors ENH: Provide PlaybackFile for recorded interaction events BUG14463 vtkMy does not work for vtk6 BUG: 0013057: bad xml input to XML Reader's causes exception BUG14527: Subdivision fails for non-manifold data BUG: ImageAccumulate overflows for large images BUG: Some types of files are not closed after processing COMP: A more robust way to check file descriptor leaks ENH: Add support to read tiled tiff images COMP: Add execption for vtkJSONWriter ENH: Restore missing Infovis classes ENH: Unit Test FunctionParser COJMP: More robust fuzzy results checking BUG: Memory leaks detectd by valgrind COMP: Prefix HDF5 cmake messages with HDF5:. Brad King (37): COMP: Drop invalid custom commands from HDF5 COMP: Drop missing dependencies of target 'vtkpython_pyc' vtkhdf5: Remove extra calls to cmake_minimum_required Accelerators/Dax: Remove extra call to cmake_minimum_required Refactor top-level CMake Policy settings as loop Examples: Set required CMake version and policies before project() Set CMake Policies CMP0025 and CMP0042 as necessary Rendering/Volume: Convert new baseline to ExternalData content link Wrapping/Tcl: Fix CMake Policy CMP0050 warnings IO/XML: Remove unnecessary include IO/XML: Add missing include IO/XML: Fix HeaderTest for this module vtkOStreamWrapper: Support std::string Revert "vtkOStreamWrapper: Support std::string" vtkOStreamWrapper: Support std::string Revert "vtkOStreamWrapper: Support std::string" vtkOStreamWrapper: Support std::string Allow custom TestTM3DLightComponents timeout Allow custom TestProp3DFollower timeout Honor VTK_INSTALL_NO_DEVELOPMENT for archive libraries Use vtkMPI.cmake helper for all module MPI tests Set MACOSX_RPATH property default consistently across CMake versions ExternalData: Add backtrace to missing file warning Export the locations of Qt5 packages on which modules depend Allow configuration of external module dependency locations Views/Infovis: Add another baseline for TestConeLayoutStrategy Allow custom Interaction/Widgets test timeouts Fix ExternalData pre-commit hook on msysGit Add option to exclude VTKData target from default build mpi4py: Suppress warnings in third-party code vtkFFMPEGWriter: Fix build with libavcodec55 Exclude classes requiring a QApplication from VTK smoke tests module: fix export macros with kits vtkglew: Install glew header files IO/ADIOS: Exclude module from VTK_BUILD_ALL_MODULES ENH: Use if(DEFINED) to simplify conditions COMP: Include "vtk_glew.h" instead of "GL/glew.h" Burlen Loring (3): surface LIC - stl and NULL vtkweb: buffering issue vtkweb: connect to existing session issue Casey Goodlett (4): Ignore files in project.xcworkspace Fix names of texture and normal arrays in the OBJ reader Fix libpng "Not recognizing known sRGB profile that has been edited" Skip install of .pdb files for hdf5 on windows Chris Harris (27): Add failure filter to session.auth(...) deferred Add error function to vtk:mouseInteraction RPC vtkWeb: Add newline to end of each proxy map entry vtkWeb: Teach launcher.py to wait for ready_line vtkWeb: Add appropriate error codes for launch requests vtkWeb: Don't reset viewer to null on updateViewer(...) vtkWeb: Propagate mouse events to VGL vtkWeb: Fix calls to old VGL API vtkWeb: Rename ogs.vgl => vgl Rename md5 => sceneMD5 to avoid confusion Update canvas size before creating viewer Prevent mouse event handlers be bound multiple times Update VGL to master Update version of VGL Prevent scene and clipping range from being reset Update VGL to lastest version vtkWeb: Refactor fetchObject(..) to use cached objects Add code to clean of VGL actor cache Add missing initialization of member Fix typo introduce by 06aaa805 session => m_session Update to modern CMake style Allow location of OSMesa install to be specified vtkWeb: parseObject now returns an array of actors Export same scalar bar title as is displayed by actor Update VGL version vtkWeb: Use unbuffered file read/write for stdout vtkWeb: Cleanup python processes that timeout on startup Christoph Kolb (1): Add a method to enable multisampling in the QVTKWidget2 Chuck Atkins (22): Initial ADIOS readers and writers Remove leftover debug messages Expose WriteAllTimeSteps as a setable parameter for ADIOS Properly handle WriteAllTimeSteps = false for vtkADIOSWriter vtkIOADIOS: Fix incorrect multi-block multi-step scalar values. vtkIOADIOS: Ensure time values always get written, even if just the step. vtkIOADIOS: Collection of small fixes jsoncpp: Update to use the seperate-file version vtkIOGeometry: Correct typos in older jsoncpp API calls jsoncpp: Fix vs7 build error from missing C99 functions COMP: Suppress warnings on PGI with commonly used patterns COMP: Remove unreachable break; statements following returns BUG: Replace a few previously removed exit conditions BUG 14681: Allow for early exit when no data can be decoded. BUG 14681: Close leftover file handles on exit XMLReader: Prevent streams from being closed when they don't exist. ADIOS: Resize the write buffer to the known required size. ADIOS: Update FindADIOS to match upstream ADIOS: Add an initial polydata write-readback test HDF5: Fix build errors for PGI on Linux ADIOS: Properly handle variable sized data. ADIOS: Fix memory leaks by adding deletes Cory Quammen (34): BUG: Fix build failure in RenderingParallelLIC on Intel compiler Added "sticky" axes mode to vtkCubeAxesActor Fixed linkage problem in Windows Improved comment handling KW00001404: Major overhaul of vtkAppendFilter Fixed memory leak and improper erasing of std::set elements Fixed error in appending cell data Removed unused types and member variables, fixed formatting Rename local functions and embed in anonymous namespace Change clamping range for texture coordinates Turn on edge clamping Fixed errors exposed by changes to clamping of texture coordinates Replaced buggy and redundant implementation of GetRGBPoints() Updated baseline for when test uses alpha blending Improved comments and made some formatting improvements Removed code mistakenly introduced during a merge Updated test baseline Removed unneeded function argument Enabled easier specification of custom test name Re-enabled some tests overlooked by vtk_add_test_* Fixed some file extensions in documentation 6657: Add out-of-range colors to color maps CMake was complained about unknown arguments to find_package BUG: Fixed how the header size is computed BUG 10708: Fix vtkPolyDataMapper2D color mapping Avoid insertion of extra newlines when adding comments BUG: Added missing build of lookup table Handle coloring by field data Enable coloring by indexed lookup for vtkStringArrays Added previous signature back for MapScalars() method Fixed memory errors in newly added tests Fixed test to work on older graphics hardware Added alternative baseline image for TestActor2D Fixed memory leak when RGBPoints are read Dan Lipsa (38): Report an error when writing a file with changing topology. Update documentation. Remove debugging printouts. Name enum LEFT, BOTTOM, ... Location. Add component-wise set functions for Border and Gutter (vtkChartMatrix) Set Bold and Italic axis label properties for ScatterPlotMatrix. vtkMultiProcessStream: Add insert and extract operators for bool Fix Windows compilation warning. Forces the compiler to convert a char* to a string rather than a bool. FIX: Get rid of dependent properties. FIX: Set the hue range for the default lookup table. FIX: NULL data in RequestInformation causes a segfault. Fix compilation after upgrade to hdf5-1.8.13 Build static hdf5 and build on VS 2003 7.1 Fix mingw release build. snprintf from HDF5 conflicts with the one from GMVReader Make hdf5 properties advanced. Add VTK_COLOR_MODE_DIRECT_SCALARS. See vtkScalarsToColors::MapScalars. Fix KdTree::GenerateRepresentation and add test. Add a test for VTK_COLOR_MODE_DIRECT_SCALARS Fix hdf5 library names. Remove lib prefix from windows libraries and mangle additional symbols. Fix hdf5 library names for VTK (and ParaView). FIX: We don't select 2D annotations. Add data_index to Exodus driver FIX: remove warning declaration of ?? shadows a previous local Prefix BUILD_STATIC_EXECS with HDF5 and make it advanced. Add missing numpy functions previously available in ParaView 4.1. Rename data_index to mode_shape in Exodus reader. BUG: Writing an Exodus file in parallel segfaults. Stop all processes if one process stops. Reshape VTKArrays so that they can be combined through operators. BUG: Account for gradient G + mean(G,0) Fix conflicts in cmake files from hd5 upgrade. Accumulate strings printed to console by Python and send them in one piece. Fix style problems. BUG: Buffer output only for RunSimpleString. BUG: Fix mispelled excluded class and print out all exclusions. Dave DeMarle (55): Increment version to VTK 6.2.0 Add a test for the VPIC reader document arguments to test macros move dashboard management scripts into better place fix link errors that happened when I turned on InfovisParallel deprecate this module fix a compilation warning make rubber band look correct on scenes with textures optimize geojson writer more optimization of geojson writer fix variety of dashboard problems Put libxdmf into a namespace so that libxdmf3 can coincide with it. Adapt libxdmf to build within VTK. add cmakescript infrastructure to make libxdmf a VTK module Add a reader and writer for the new xdmf library. Fixes for xdmf3 regression test data. Fix issues the dashboards turned up. Fix more issues the dashboards turned up. Really fix on of the issues the dashboards turned up Fix a regression in the xdmf bump STYLE: improve comments and organize internal classes fix a case of discarded arrays Remove stray debugf. bring back subsets put array reuse in Fix dashboard warnings and test failure. disable a balky test until we can fix it Add file series reading and top level spatial partitioning fill in time varying graphs STYLE: pull helpers out and format all for vtkstyle move strict partition multipiece constuction out to reader level disable oggtheora on cray too Fix problems that the dashboards turned up. remove MetaIO to replace with updated content exclude matlab from the "all" modules set Add render mode in which values are drawn as recoverable 24bit colors. fix a compilation warning Fix test failure Fix test failure Revert "Fix test failure" Revert "Fix test failure" now 000 is reserved as the no value (background) color fix dashboards that lack multisampling make area picker respect prop transformations prevent possibly uninitialized ivar use java install shouldn't delete /usr/bin ! shrink size of xdmf3 written files fix a bug in value painting for cell arrays fix a bug in triangle frustum intersection COMP: fix a valgrind leak in new test update cdash scrapers to use https COMP: fix unused warning edge clipping fallback path needs correct edges COMP: fix macro redefinition comp warning per bug 14586 fix external vtkLocal David C. Lonie (67): Tie vtkUnstructuredGridBase into the dataset type system. Update baseline Handle empty array initialization in numpy_support. Don't reset traversal location in each iteration of box clip filter. Process facestream in vtkExtractUnstructuredGridPiece::RequestData. Handle polyhedra in vtkUnstructuredGridWriter. Disable FP frame buffer when GL_ARG_texture_float is not present. Return early if ProjectedTetrahedra volume rendering is unsupported. Add new baseline for rendering without floating point buffers. Add new baseline for dashlin1. Pass std::strings through the OStreamWrapper. Revert "Pass std::strings through the OStreamWrapper." Return early in TransformPoints if input is NULL. Update system font test baseline. Check if the selection array exists before dereferencing it. Ignore blanked cells in vtkCellDataToPointData. Error if attribute arrays are incorrectly sized in vtkGlyph2D. Remove unused variable. Don't attempt to render NULL strings. Add DebugTextures to FreeTypeTools to show texture background. Use unrotated faces to determine text height metrics. Move templated methods to private visiblity. Style cleanup in vtkTextActor3D. Remove pointer tests for ImageActor in vtkTextActor3D. Strip newlines from GL2PS comments. Always upload test image if a regression test fails. Improve aligned and rotated text. Update tests and baselines. Account for frame width in ScalarBarActor title position. Use Round vs Floor(+.5). Update baselines due to text rendering changes. Fix shadowed variable warning. Cache number of points used in asserts. Simplify and inline much of the logic in vtkStructuredData. Update vtkExtractStructuredGridHelper to remove data descriptions args. Batch copy point data. Add API to copy ranges of tuples efficiently. Copy ranges of points during VOI extraction, if possible. Move InsertTuples implementation to helper class. Fix "uninitialized usage" warnings. Make a clear distinction between extent indices and values. Add helper functions for computing partitioned structured data. Fix bugs in vtkStructuredImplicitConnectivity. Fix vtkPExtractRectilinearGrid extent issues. Fix vtkPExtractVOI extent issues. Fix vtkPExtractStructuredGrid extent issues. Fix a buffer overrun in vtkExtractStructuredGridHelper. Remove redeclared variable. Fix bool -> int conversion. Clamp global VOI to the output whole extent in subset filters. More warning cleanup from nightly dashboards. Rearrange inline methods in vtkStructuredData.h. Move vtkTextActor and vtkTextActor3D into Rendering/Core. Add unary minus (negation) to vtkVectorOperators.h. Fix c/v qualifier on vtkVector::Normalized and Cross methods. Manually mark text buffer images as modified. Add a vtkLabeledContourMapper for producing inline labels on isolines. Added ability to turn label visibility ON/OFF Add vtkTextPropertyCollection. Allow multiple text properties to be used in vtkLabeledContourMapper. Reenable GL2PS MathText tests. Add background color/opacity settings to vtkTextProperty. Use the new background color rendering for the labeled contour test. Fix an error in setting an array length. Fix conic point identification in the path renderer. Fix warning in vtkLabeledContourMapper.cxx Allow the initial GL2PS buffer size to be set. David Cole (14): Add a vtkMath method to compute the angle between two vectors Tests: Re-activate the old TestResliceCursorWidget tests Tests: Add test for the vtkCornerAnnotation class - Follow on commit to use the right style #include ("", not <>) BUG: Add missing header files to enable try_run tests to run without crashing Testing: Call SetMultiSamples(0) from tests Rendering: Use AdjustWindowRect to ensure pixel perfect client size Rendering: Retrieve the full screen size in GetScreenSize Rendering: Use AdjustWindowRect to ensure pixel perfect client size vtkTesting: Refactor RegressionTest, eliminating hard-coded cout vtkTesting: If test fails with back buffer, try front buffer too VRMLImporter: Fix memory leaks and crashes VRMLImporter: Fix crash. Avoid dereferencing NULL pointer (#1624) vtkVRMLImporter: if non-NULL, delete CurrentTransform in destructor David Gobbi (93): 14196: Allow unicode path in vtkLoadPythonTkWidgets.py. In wrappers, guard against #including dirs. Fix cell picker for images with negative spacing. Fix uninitialized array warnings. Fix unreachable code warning in otherCellTypes.cxx. Fix a crash when deleting vtk-python objects. Fix bad SetColor/GetColor interaction in vtkProperty. Fix the extent for tiled tiff files. Two fixes for certain lsm (zeiss) tiff images. Add reader/writer for NIFTIv1 and NIFTIv2 files. Add SetSlabModeToSum() to vtkImageResliceMapper. Clean up vtkMedicalImageProperties PrintSelf. NIFTI makes IOImage module depend on vtkzlib. Fix warning about trailing comma in enum list. Bad null pointer check in vtkBSplineTransform. Fix segfault if vtkImageWeightedSum's inputs don't match. Rewrite the comments for vtkDataArray::GetRange. Add generic PyObject support to vtkPythonOverload. Provide a method to set the sample spacing for slab modes. Add wrap hints for vtkPoints2D and vtkRenderWindow. Reactivate the RenderingImage tests. Replace python mutable with its value in GetValue(). Clean up cocoa input event handling. Interpret OS X Command key as Ctrl instead of Alt. Initialize interactor in Start() if not done yet. Fix illegal extent request in ImageThresholdConnectivity. 14445: Fix vtkImageStencilData IsInside upper bound. 14552: Check python interp and lib versions. 12098: Mangle void_p like other swig pointers. Exit the Cocoa event loop when the window closes. Readability improvements to vtkCocoaGLView event code. 14999: Python Tk widgets fail to load on Tcl 8.6. Remove InstallMessageProc flag from OS X interactors. Reduce code duplication for Start() method. Ensure python header version matches lib version. Remove the deprecated PYTHON_INCLUDE_PATH variable. Fix Initialize() and Start() interactor documentation. Revert "Remove the deprecated PYTHON_INCLUDE_PATH variable." Remove obsolete PYTHON_INCLUDE_PATH from CMakeLists. Do not cache deprecated PYTHON_INCLUDE_PATH. Make streaming optional for the 3D image mappers. 14042: Numerical stability of vtkPolyDataToImageStencil. BlackmanHarris4 fourth coefficient was ignored. Default parameter for Kaiser window is wrong. Fix copy-paste parameter name. Add missing newline to end of file. Fix handling of tolerance in vertical direction. Deprecate the obsolete InsertLine method signature. Add test for LassoStencilSource orientations. Re-enable vtkPolyDataToImageStencil tolerance. Python init config used COPYONLY instead of ONLY. Fix unreachable-code-break warnings. Check whether volume property is null. Before c99, variable decls go before other statements. Before c99, variable decls go before other statements. Before c99, variable decls go before other statements. Before c99, variable decls go before other statements. Refactor vtkWrapPython into smaller source files. Wrap enum constant members for non-vtkObject types. Add an enum type to the python wrappers. Fix two previously undetected wrapper bugs. Wrap namespaces in python. Allow use of all enum types as method parameters. Move VTK_BUILD_SHARED_LIBS check to generated code. Wrap constants more efficiently. Avoid creating duplicate wrapped namespaces. Only wrap namespaces that have useful contents. Update the readme file for new wrapping capabilities. Fix compile warnings for vtkParse.tab.c. Use a wider integer for the parser stack. Fix warnings due to anonymous namespace. Revert "Use a wider integer for the parser stack." Allow enum types to be used as method parameters. Fix "conversion from size_t to int" warning. Remove NULL check on reference. Fix potential vtkUnicodeStringArray instability. Fix warnings due to anonymous namespace. Fix type-punned pointer dereference warning. Fix warning about comma at end of enum list. Fix incorrect index types for loops. NumberOfArrays is int, so use int as counter type. Fix warning for possible uninitialized array access. Fix a bug caused by recent tolerance fixes. Fix error when kernel size is greater than slab thickness. Remove some unecessary casts. Revert "Remove some unecessary casts." Allow absence of i64 suffix when __int64 exists. Increase tolerance for UnitTestParametricSpline. Increase numerical tolerance in UnitTestFunctionParser. A "Set" method in the constructor caused a UMC. Revert last commit to UnitTestFunctionParser.cxx. Make wrappers ignore scoped class definitions. Fix arg conversion of const ref arg via constructor. David Stoup (1): Add multi-layer support to the vgl interface. David Thompson (8): Bug 14713: Read LS-Dyna point coordinates properly. Add a regression test for bug 14713. Small fixes to docs, comments, and formatting. Add support for reading polyhedra. Test reading a polyhedral Exodus dataset. Add test data with multiple polyhedra. Remove MD5 test-data MD5 file. Fix SetAllArrayStatus on the ExodusII reader. D?enan Zuki? (1): VS2013 compile fix Eli Kahn (4): Adding mousewheel support to Java GUI components. Fix java code formatting. Add java file to generated jar file. In java code, handle situation when wheel rotation value is zero. Georg Hammerl (1): Add caching for time steps in EnSightGoldReader binary reader. George Zagaris (18): Disable I/O in TestStructuredAMRGridConnectivity ENH: DuplicateNodes property to RCB partitioners BUGFIX: Fix integer overflow in vtkStructuredData ENH: FieldData deserialization to a subextent ENH: Add blanking to vtkXMLUniformGridAMRReader ENH: Implicit structured data connectivity COMP: Update vtkFiltersParallelMPI tests ENH: Rename nodes to points in vtkStructuredData BUGFIX: Add guards for NULL ivars in PrintSelf BUGFIX: Fix issue with unigrid enzo datasets COMP: Correct ParallelMPI module test dependencies COMP: Remove HyperTreeGridGeometry::UpdateExtent() ENH: Update extract structured data filters ENH: Use vtkImageData in implicit connectivity ENH: RectilinearGrid implicit connectivity support ENH: MPI structured data extraction filters COMP: Fix a few compiler warnings ENH: Added broadcast implementation for RMI triggers HDF Group (1): hdf5-1.8.13-r25462-reduced Hans Johnson (1): DOC: Remove documentation that is not relevant to VTK6 Jacob Becker (1): Fix bad check for incomplete polygons, fix div-by-zero. Jameson Merkow (1): Add files, and made modifications for MatlabMex to wotk in vtk 6 Jean-Christophe Fillion-Robin (8): vtkPythonCommand - Add support for additional call data types vtkPythonCommand - Fix indent and variable names Fix regression ensuring macro VTK_MAKE_INSTANTIATOR3 configure files Fix component names in "vtkhdf5-hl" install rules. Fixes #15282 Fix MacOSX build error related to OpenGL2/vtkOpenGL.h. See #15285 vtkRenderer: Fix null pointer crash in ViewToWorld() with no active camera vtkhdf5: Fix build error when building on system with glic >= 2.19 per bug 14826 make find_package vtk more lenient Jeffrey Baumes (1): Fixing array bounds issue Joachim Pouderoux (40): Introduce Robust PCA. Introduce box plot and the corresponding chart. Introduce Compute Quartiles filter to generate box plots tables. Enhance bag plot API to allow specify line and points properties. Fix bag plot legends. Fix black box plots issue. Fix double to unsigned char cast error on clang compiler. Fix bag plot and functional bag plots tests. Fix and enhance the box plot and chart Fix and enhance bag plots. Fix bug in picking of invisible props in an assembly. Clean an enhance hypertree grids. Fix potential compiler error with ternary operator. Replace a verbose cout by a vtkDebugMacro() Fix clang errors on array initialization Fix and clean vtkWin32OpenGLRenderWindow class. Fix functional bag plot. Clean and uniformize vtkPoints & vtkPoints2D. Fix crash on normal computation when polygon is empty. Fix code formating & style to VTK coding-style rules. Add algorithm header for std::lower_bound function call. Fix non Delaunay internal edges added by RecoverEdge(). Add missing NoBlockSend(vtkIdType) method. Remove compilation warnings on some platforms. Add support for EDGEFLAG attributes in legacy reader & writer. Add a new test for EDGEFLAGS point attributes. Fix vtkHyperOctree copy functions. Disable edgeflag test if Mesa driver is detected. Fix and clean VRML importer code. Make vtkContextScene ButtonPress/ReleaseEvent invoke a button event. Fix a memory leak that occures when Uniform Variables are updated Fix Xdmf3 CMake file to export symbols in DLL. Add functions to get data bounds and number of plotted bars. Fix Xdmf2 writer: array type was not respected. Also save block name. Enhance Xdmf3 writer to save block names. Fix a warning (variable shadows a parameter) Optimize data array range computation functions. Clean and fix STL reader. Small fixes to allow empty cells. Fix EnSight Reader BUG 0015268 Johan Andruejol (1): Refactor parts of vtkLookupTable John Stark (2): Add a test for vtkImageImport / vtkImageExport Fix pipeline update issues in vtkImageExport John Tourtellott (7): Fix memory leak in vtkGDALRasterReader::RequestData() Set geo-origin and -spacing in RequestInformation() call Add vtkGDAL class & MAP_PROJECTION key for pipeline data and field data Change missing geotransform from error to warning. Fix ctest errors Turn off wrapping for vtkGDAL.h Declare vtkGDAL constructor & destructor private, since class is static Jorge Perez (1): Upgrading the Tcl examples wrt. VTK 6.x Julien Finet (11): Prevent error message when deep copying vtkGridTransform Change GetInputConnection error message into a debug warning message Fix vtkAbstractPolygonalHandleRepresentation3D visibility Fix crash in vtkFixedSliceHandleRepresentation3D Add vtkPlot::SelectionPen and vtkPlot::SelectionBrush vtkPlotPoints selections must be sorted Add vtkChart::SELECTION_COLUMNS Add vtkAbstractContextItem::StackAbove and StackUnder Reorder the columns of vtkExtractFunctionBagPlot Apply selection pen opacity when drawing selected points Fix crash when vtkPlotPoints has no Axis Julien Jomier (10): ENH: Fixes for ActiViz.Net BUG: Wrong transformation used to restore original parameters ENH: Added back support for TDx COMP: GLEW_STATIC should be defined with MSVC2010 when building static COMP: Missing type for vtkWindow GetScreenSize() COMP: Q_OS_X11 doesn't exist with Qt4 replacing it back to Q_WS_X11 ENH: Added support for ClampToBorder (as it was in OpenGL) ENH: Added function to initialize volume BUG: AreaPicker doesn't consider position/orientation of assembly BUG: RenderedAreaPicker was not using the PickFromList KWSys Robot (7): KWSys 2014-03-12 (dd873734) KWSys 2014-08-11 (32023afd) KWSys 2014-09-08 (80e852f6) KWSys 2014-09-19 (6aa1f800) KWSys 2014-09-25 (29ffaf43) KWSys 2014-10-31 (88c8cc7f) KWSys 2014-11-12 (5843f590) Ken Martin (308): some rough changes to the VBO polydata mapper some more cleanup added light override compute cam matrix, add posiitonal light in progress updates to support positional lights change how object factory is done some fixes to normals etc added ambient color in some module changes override actor change bunny color changes in normal calcs moved property over and started gutting it fix normals on test minor changes some bug fixes and start on mapper2D allow testing using old mapper bug fix added cxx for Helper duh many updates to 2d mapper bug fix lots of rough changes moved renderer over, moved init code in progress in progress cell scalars some cleanup and fixes lots of changes in progress more fixes fix array bounds issue and memory leak fix performance issue add fast path for float points no col no norm and fix bug started on texture mapping first cut a tmap support add textured backgrounds and fix issue with gradients bug fixes clean up a bit change opacity clean some texture stuff minor fixes fix crash on test fix crash on test add fragment shader lighting of lines support textured luts some cleanup some support for resolving coincident polys ambient color fix some minor fixes start working on image slice mapper add baseine images updated baseline images updated baseline image minor backwards compatability change basic ImageSliceMapper conversion some more fixes moved render window over bug fix in old VTK bunch of optimization more to come bad optimizations and missing classes' doh forgot really key line bug fix and minor cleanup fix a glyph bug and minor cleanup another bug fix better picking support some cleanup move interface closer to old Rendering API convert ImageMapper some bug fixes another valid image, remove unused test fix SetPixelData and fix picking issue add depth peeling make wider so we can see errors some win32 fixes rename VBO mapper cleaned up the mapper a bit minor cleanup minor lighting fix for intel handle wireframe tstrips some bug fixes etc more fixes more fixes add support for edge visibility modify API to be a bit more future proof fix some picking issues some fixes for andriod some fixes for andriod remove some old code some fixes for andriod more andriod glew changes more andriod glew changes many changes to support glew and android many changes to support glew and android many changes to support glew and android first working in quotes android app first working in quotes android app comment out more modukles by default some warning fixes and minor cleanups fix some more warnings fix some more warnings make vtkOpenGLTexture use vtkTextureObject convert depth peeling to use more texture objects a tad more depth peeling cleanup add android output window better debugging out for shader programs clean up some shader code new android interactor and cleanup some shader cleanup some android fixes in progress improved release graphics resources improved release graphics resources minor change to be more compatible with some opengl hardware fix android resize and lingering glew issue added multitouch framework and support on windows and android fix java issue in progress in progress more reasonable first cut at java android clean up some shader code and classes removed unused includes fix some warnings dashboard fixes and debug leak some fixes for macs fix glyph memory leak and issue on intel cleanup reorg and fix some memory leaks minor doc fix fix unit memory issue in progress ios more iOS changes and depth peeling fix some minor cleanups to wrapped classes lighting fix for generated normals remove unimplemented methods fix a number of dashbaord issues fix for lines that are exactly horizontal or vert plus cleanup fix for lines that are exactly horizontal or vert plus cleanup add support for cell normals add support for cell normals fix a bad test minor cleanup to a test forgot to update the python version yank test for a feature we do not plan to support fix an initialization issue with image reslice fix incorrect filename added valid image for line lighting in opengl2 removed broken test form opengl corrected for opengl2 more ios changes some good cleanup ** working ** app and some cleanup fix for iphone test minor fix to cover vtkcompiletools remove some extra files that we do not need make sure modules are really turned off in progress need rebase turn off more modules by default in progress working clean prepping for opengl instancing some minor cleanups really turn shared libs off Fix compile warning in TextureObject update toolchains to ios8 Fix scalar colors using textures Fix image mapper positions Fix the tcl and python cells test Fix otherPrint and TestEmptyInput tests fix for failing mac dashbaords Remove setting of ambient to 1.0 Add support for backface properties use GL_EXT_gpu_shader4 when picking Rename TextureUnitManager to not have the OpenGL Rename texture unit manager to not have OpenGL in it Forgot to uncomment line on prior commit Add support for OpenGL32 instancing in GlyphMapper Had to rebase as master had comflicts fix opengl es issues with glyphing code Try adding support for ARB_instanced_arrays Handle case with no lights on Remove extra call to set the divisor Forgot to add the image hash Replace some valid images lost with the big merge and a pick bug Fix error in glyphing and extend to mac Fix fast path on apple Fix cleanup issue Fix two glyph issues trivial compiler warning fix Forgot a file dangit Exclude freetype form iOS and andriod Remove EdgeFlags test from OpenGL2 Support edge flags. Fix for the 1 0 1 edge flag case. Minor cleanup Another attempt. I think this one should work. Update the ios example to use GLKit View and put most logic in controller temporarily turn of mesa test until Joachim ets a chance to look at it Add in support for clipping planes Add back in the edge test Some suggested changes form Casey and remove user files A first cut at a CompositePolyDataMapper2 Fix compiler warning and increase test threshold Cache camera and actor matrices for performance Remove the unused old code Fix compiler error Fix issue with mesa and gl_FrontFacing make it so that this test works on the old OpenGL backend Fix a MTime check in the matrix cache Fix header test warning Minor fix for glyphing using CurrentInput Fix float to integer comparison in shader code Cleanup some indentation Convert Projected Tetrahedra class to OpenGL2 backend Fix up compiler warning of shadowed variables Fixing ray cast image display helper Fix a few compiler warnings Cleanup a C style cast to keep it real plus a bug fix Fix a compile warning Remove bad transpose of normal matrix Remove some unused headers and code Updated API to force use of vtkShaderCache Remove some extra code that isnt required Get render passes working in OpenGL2 Restore file that was accidentally changed Fix some compiler errors that show up on other compilers Add in depth peeling pass and a couple other passes A number of Parallel and MPI fixes for OpenGL2 Fix use of undefined vector behavior. Fix for out of range memcpy Fix three dashbaord issues Fix clearzpass on opengl ES 2 Fix edge color with composite polydata mapper 2 Added some ifdef needed for OpenGL ES 20 Fix a bad merge Fix RenderingParallel when wrapped and fix a compiler warning Make the old freetype code work with OpenGL2 Minor fix for clearing the zbuffer Add support for imposters for molecular rendering Make it so that fragment shaders can modify vertexVC Fix some OpenGLES compile issues Significantly better performance with EdgeVisibilityOn Add two more valid images Fix failing cursor2D test and some minor cleanup Fix bas valid image size and a couple NULL copies in SLACReader Fix Context2D to clear the current shader before drawing. Fix lighting overdrive on ImagePlaneWidget Camera was caching when shared between multiple renderers Fix unnormalized interpolated normals Update the Composite Mapper to reflect changing in Edge Rendering Fix for shadowed variable warning Fail gracefully if depth peeling is not supported Fix 2d transparent annotation after depth peeling Fix an coordinate issue with CopyToFrameBuffer Minor compiler cleanup make sure glsl extensions are requested at the start of a shader Fixes for wrong case in include file name Fixes for composite poly data mapper2 Make sure scale is included in the normal matrix Make sure molecule mapper works with depth peeling Fix lighting to use Blinn-Phong specular model Reduce the size of the test it was too big for my macbook Add valid image for other systems (not AA etc) Add in support for OpenGL ES 3.0 trivial fix for es 2.0 Fixes so the recent scalar changes compile on OpenGL2 Improve picking support in OpenGL2 Minor compile fix for ES Fix a few issues with parallel passes in VTK remove some leftover debugging code A fix for rendering coincident polygons or lines. Remove old references to vtkOpenGL.h and fix comment Enable IO/Export (minus gl2ps) for OPenGL2 Fix SynchronizedRenderers to work with use OpenGL2 Fix gcc warnings and build error with gl2ps Fix a bad array reference More changes needed to get iceT working and cleanup Fix compiler warning Add valid image due to different z buffer Fix some release graphics resource issues Ifdef out some code for OpenGLES systems Add a point gaussian mapper for cosmology and cleanup Remove test I added from old OpenGL backend Fix up some dashboard issues Minor fix to the picking code Fix floating point issue with chart test Add a timing framework in for testing rendering performance Filename case issue Add another valid image as this test is sensitive to zbuffer Fix a simple compiler warning. Break the UpdateVBO method into smaller pieces. Fix missing virtual destructors Improve the performance of the parametric function source Add a molecule test Uh maybe fixing a cmake issue Cleanup the CMake build a bit Add aother valid image Added another valid image Fix a couple long lines Add virtual destructors again Fix dashbaord compile error and some long lines Add valid image for OpenGL2 Use MCDCMatrix in shaders Use MCDCMatrix in shaders Fix a compiler warning Add two valid images Add a subclass for faster composite dataset mapping Fix compile issues for es2 Expose more methods in opengles3 Fix GLES3 Add support for vtkTextures Transform ivar Fix some dashboard warnings Fix for systems with spaces in the path Fix memory leak in helper class holding input data Fix two compiler warnings Some updates to the timing test Fix molecular rendering in parallel projection Fix an issue with mapper requiring a polydata input Update to a cleaner way of solving the issue Kenneth Moreland (6): Dax marching cubes now uses input array not active scalars. Allow Dax marching cubes and threshold to work with doubles. Change vtkDaxMarchingCubes to vtkDaxContour Add compute scalars to Dax contour filter. Support Dax marching tetrahedra Support passing point field data in Dax threshold filter. Lisandro Dalcin (1): import mpi4py 1.3 Marco Cecchetti (5): vtkColorString - Helper class for defining a color through a string. Merged vtkColorString into vtkNamedColors. I added a RGBAToHTMLColor and made some minor changes. Tentative fix for Windows build issue. Tentative fix for big-endian machine issue. Marcus D. Hanwell (123): Added a test for biquadratic quad interpolation Exclude the vtkgl.* files (generated code) from coverage Make consistent with vtkBiQuadraticQuad Added back the QtSQL test for the database class Removed misleading comment - test is built Added back various rendering tests Added back the named components test in FiltersCore Removed a few remaining Cg related files Refactored the TIFF reader, add support for floats Added progress updates when reading a volume Remove redundant checks for old JPEG compression Copy simple TIFF images into buffers more directly First pass at a new render widget Implement a basic VBO based poly data mapper Basic support for color from property Add the find module for Eigen 3 Added a hack to override OpenGL overrides with GL2 Roughly mapping the colors/scalars, updates to shaders Normalize the colors, and put the light the other side... Added the dragon model (already in external data store) Moved the VBO and IBO creation to helper functions Use the vtkPolyDataNormals filter to calculate normals Added a points/lines test, added point/line rendering Added in a program for lines, change Moved the replace function into the helper header Refactoring a little, movig IBO stuff into structs Added a polyline, and basic support for rendering it Refactor uniforms a little, enable point rendering again Small changes for point size and line width Minor clean up for includes in the mappers Ensure we set the active shaders Removed commented code/classes from the test Added in the case for multiple renders/VBO updates Refactor the shader API to comply with VTK style Refactor API, comply with VTK style Refactor the shader program API to comply with VTK style Added a Vertex Attribute Object (VAO) class Enable textures now that the vtkTexture class doesn't do it Set the z-spacing in a TIFF to match x-spacing Early attempt at getting the 3D glyph mapper working Hackish, map the colors for the glyphs Keep the color mapper around, shallow copy for settings Fixed the normal matrix, now molecules look much better Make it compile on Linux/X again Bring over the Cocoa classes for Mac Intel GLSL compiler is much fussier about type... Removed the vtkFiltersCore dep Removed vtkRenderingOpenGL as a dep now too! Made vtkRenderingOpenGL2 independent of vtkRenderingOpenGL The PainterDeviceAdaptor is abstract, can be NULL Rename vtkOpenGL2* -> vtkOpenGL* Very judicious use of sed, vtkOpenGL2* -> vtkOpenGL* Added a concept of backends, added OpenGL and OpenGL2 File moves necessary to create a context backend Context OpenGL and OpenGL2 backends Some minor tweaks to get the tests passing again Some modules only in groups if correct backend selected Missed these OpenGL2 variants with the mass rename Brought more of the OpenGL module logic over OpenGL and OpenGL2 are mutually exclusive If enabled, always override in OpenGL 2 Unused include for a file that was deleted Disable multitexturing in Geovis class Moving rendering tests to more appropriate places More test moves into more appropriate locations Removed the TDx classes from OpenGL 2 Only enable backends for interface modules being built Disable test for render widget Changed the default CDash URL for VTK Move several TEST_DEPENDS to switch on backend Create object factory CMake file, move vtk_add_override Now generate the object factory files in a function Revert "updated baseline images" Revert "updated baseline image" Some dependency updates now vtkRenderingContext2D is abstract Don't exclude the module from wrapping - initialization Made the label tests pass again! Make TCL wrapping happy - no numbers in module names Added include for std::string Save the rendering backend used to the config file Avoid early termination when collecting actors vtkOpenGLContextActor should subclass vtkContextActor One of the tests needs a context device to render Removed unused variable - used in subclasses These classes are abstract, fix smoke tests Fix for Windows mangling issue Must be marked WRAP_EXCLUDE too for Tcl smoke tests Reduce the number of renders (avoid time outs) Move device adapter initialization to the OpenGL module BUG: Make selections more permanent, linked to object STYLE: Some basic style fixes Removed sorts as the inputs were already sorted Added a test to exercise bar graph selection Remove the wrap exclusion at the module level More exclusions, avoid setting ABSTRACT twice TCL doesn't like numbers in module names Comment out the animated PDB renders CMake's exec_program is deprecated, revert back Added some code to fix Tcl and its use of OpenGL backends Added excludes for OpenGL2 headers Enable the Java rendering test for OpenGL2 too Actually define GLX_GLXEXT_LEGACY to prevent the include Update to the upstream GLEW 1.11.0 release Added logic for OSMesa contexts Add support for OSMesa builds of OpenGL2 Added colored bar charts, along with a test Expand support for the OpenGL2 backend Restore test deps for matplotlib tests Disable some widget tests building for OpenGL2 BUG: Fix bug #14378, ensure observer is removed in dtor Added a non-const form of vtkTesting::AddArguments Trivial fixes for API change in OpenGL2 shaders BUG: Fixed error in passing position of key press Fix compile failures seen with latest FreeType Make the VBO update more selective for picking Convert the file name to a const char* Minor fixes for code style, indentation, use vtkNew Benchmarking triangles, chart that updates Made a benchmark module, moved the benchmark tools Added a --timeout option, default to 1 second Call Write on the delimited text writer Just make the benchmark module depend on volume rendering Use the resolution variable to set the cube size Matt McCormick (1): Reset VTK_MODULES_REQUESTED to all modules with multiple calls. Matthew Woehlke (2): Fix rotation handling in vtkTransformInterpolator. Fix operation order in vtkTransformInterpolator. MetaIO developers (1): MetaIO 2014-07-09 (53aa417f) Nicolas Gallego (1): IntersectWithLine method documentation update Orion Poplawski (1): thirdparty: support defaulting all third party library sources Patric Schmitz (1): Include vtkPythonPackages in Web/JavaScript/CMakeLists.txt Patrick O'Leary (1): added paraviewweb widget for catalyst workbench Paul Edwards (3): Fix for polyhedron cells. Bug fix for dataset surface filter. Fixing how polydata cells are deleted. Robert Maynard (17): Update Dax Accelerator to use the new tll signature. Update the Dax Accelerator to use the new dispatcher classes. Correct warnings about negative unsigned constant values. Properly implement the factory for the dax filters. EnSight Binary uses 64bit longs so explicitly use those instead of long. Netcdf now handles different directories for c and cxx bindings. Improve the performance of vtkDataArray::ComputeScalarRange. Improve the performance of vtkDataArray::ComputeVectorRange. Move all free functions in vtkDataArray into an anonymous namespace. Update ComputeVectorRange based on gerrit feedback. Update DataArray and DataArrayTemplate to use vtkTypeTraits. Correct ComputeScalarRange when dealing with multiple components. Improve ComputeScalarRange when we have a single component. Correct an issue with converting dax data to vtk polydata. Refactor GetRange to compute all ranges at the same time. Correctly detect that we are compiling with MSVC. Enable the output of the number of triangles in the benchmark. Roger Bramon (1): Fix QVTKWidget problem on Windows with Aero off and Qt5 Ronald R?mer (1): bugfix 14459: Iterating through polydata cells incorrectly. Sandeep Menon (1): vtkOBJReader: accept relative indices in OBJ files Sankhesh Jhaveri (119): BUG: Fix failing EnSightBlow5ASCII tests BUG: Turned multisampling OFF for certain tests failing on linux with ATI Fixed multisampling issues for some tests on ATI ENH: Script to generate API differences between two git revisions ENH: Added test for vtkImageCroppingRegionsWidget to improve coverage ENH: Enable selection of translucent geometry BUG: VTK_JAVA_SOURCE_VERSION to accomodate different java versions Re-enabling tests disabled during modularization for Interaction/Widgets BUG: Duplicate class definition in TestSeedWidget2 BUG: Duplicate class definition of vtkSeedCallback2 Additional baseline for TestAreaSelections BUG: Border visibility from underlying vtkScalarBar actor ENH: New RenderingExternal module ENH: Added custom renderer and camera classes for external support Variable renamed to maintain consistent syntax ENH: Cleaned up the ExternalRenderWindow Set light transform matrix irrespective of light type ENH: Added test for vtkRenderingExternal Replace baseline with MD5 sum Fix lighting for the external rendering module ENH: Fetch viewport size for the existing OpenGL context Cleaned debugging code Updated baseline for TestGLUTRenderWindow FIX: vtkRenderingExternal module is OFF by default ENH: Synchronize camera focalpoint ENH: External renderer preserves buffers by default Using the external renderwindow always uses current context ENH: Add ExternalVTKWidget STYLE: Fixed indentation as per VTK style guidelines ENH: Added interactor to ExternalVTKWidget ENH: Compute the window size when initializing from context Moved the eye determination code to Start ENH: Make sure VTK does stereo rendering Use vtkGenericOpenGLRenderWindow Attempt at fixing light issues across multiple screens Renamed module to vtkRenderingVolumeOpenGLNew ENH: Changed dependency from glew to vtkglew FIX: Build error due to misplaced colon and style fixes ENH: Implement PrintSelf STYLE: Fix lines larger than 80 chars FIX: Failing HeaderTest and rename vtkGLSLShader Bring back deleted files FIX: VertexArray issue on APPLE machines ENH: Additional tests for new volume mapper ENH: Added baselines based generated using old volume mapper FIX: Exclude the vtkRenderingOpenGLNew module from BUILD_ALL_MODULES FIX: Use RegressionTestImage instead of InteractorEventLoop FIX: Disable new volume rendering tests for old mapper FIX: Exclude vtkRenderingVolumeOpenGLNew from vtkOpenGL kit FIX: Reset clipping planes FIX: Remove all vtkgl dependencies for vtkRenderingVolumeOpenGLNew Fix unused variable warning ENH: Added new vtkOpenGLGPUVolumeRayCastMapper to VolumeOpenGL2 ENH: Add all GPURayCast tests for OpenGL2 backend Remove unused shader files Fix cmake string issue STYLE: Rename shader variables to maintain consistency ENH: Initial take at SmartVolumeMapper support ENH: Removed all vtkVolumeTextureMapper3D refs from OpenGL2 module Port warning and style fixes from VolumeOpenGL2 ENH: Sorted out python and tcl tests for new volume rendering Make SmartVolumeMapper available for VolumeAMR ENH: Volume Mapper benchmark test FIX: Make sure context is present FIX: Bypass testing if OpenGL extensions not supported STYLE: Ensure change follows VTK style FIX: Python testing was saving foo_1.valid.png style files Removed use of unrequired variable FIX: Disable antialiasing in new gpu volume tests FIX: Nice axis bounds calculation when flipped FIX: Tooltips not showing up when axis inverted. ENH: Tests for inverted axis functionality Test new volume mapper against positional lights BUG: Representation not updating when volume input changed Fix crash when running VolumeUpdate test Test for large sample distance in volume mapper Test against geometry rendering Changed the noise generator amplitude to 0.1 Add baseline for new sample distance test Convert RGB Table to use vtkTextureObject ENH: Ability to load create alpha textures from raw data Minor semantic changes ENH: Use texture object for opacity tables ENH: Use vtkTextureObject for gradient opacity table ENH: Make sure the new API for opacity tables is supported by mapper Initial null value for pointer ENH: Convert noise and depth textures to use texture object Initial work to reformat the window level test Started working on gradient opacity test for new mapper Added new gradient opacity test Test the new mapper for gradient opacity support BUG: Fix an issue in the shader code that supports gradient opacity Pre-compute gradient function to reduce computation overhead Replaced old window level test for a better one Check if extensions are supported for the new window level test Turn off auto adjust sample distances Set parameters before activating texture Free the texture object memory at destruction time Safety checks before deleting objects Safeguard against bad memory access Bring back the smart volume mapper window level test ENH: Use generic render window to support multiple platforms Fix GLUT API include header for Mac OSX and Windows Maintain consistent style across classes Baseline for SmartVolumeMapper window level test Additional baseline for failing test Prevent releasing graphics resources on the same memory twice Removed unwanted commented out code If check against bad memory access Support multiple renderers in the ExternalVTKWidget Fix segfault due to invalid pointer address Fix viewport size change when renderwindow resizes Removed redundant commented out code Exclude vtkRenderingExternal from BUILD_ALL_MODULES Improved style and documentation Deleted unrequired commented debug code GLUT test as an example for ExternalVTKWidget use Disable vtkRenderingExternal module for OpenGL2 backend Remove debugging code Scott Wittenburg (20): Added code to draw current selection rectangle in viewport. Added code to support a cache option for filebrowser widget. Added canvas based image zooming and panning capabilities. Fixed a null reference by surrounding the call with a check. Support faster configuration times by checking dependencies at runtime. Changed to make tests run serially, async nature was causing collisions. Replace an import which disappeared and broke all dashboard web tests. Updating to Autobahn python version 0.8.13 to get http long poll endpoints. Added latest Autobahn JS (0.9.4) to support pvweb authentication. Properly override ApplicationSession c-tor, supports Crossbar.io. This should fix problems with paraviewweb and python 2.6. Added alternate jQuery catalyst view constructor. Added an rpc method to exit after a delay instead of immediately. Update vtkweb loader to support addition of scalar opacity widget. Use "-dr" in lauch examples to encourage disabling registry for pvweb. In the vtkweb loader, add css specific to the proxy editor widget. Changes to viewports supporting capturing screenshots for pvweb. Support vtkweb and pvweb servers using http only (no websockets). Added missing import which broke http-only servers not serving content. Updating launcher to include a complete set of example profiles. Sean McBride (141): bug #14418: Remove addition of "-Wno-deprecated" compiler flag Fix clang analyzer warning about passing null to strcmp Fixed clang analyzer warning about passing null to memcpy Fixed clang analyzer warning about reading uninitialized memory Fixed clang analyzer waring about a dead store. Fixed clang analyzer warning about dead store Fixed clang analyzer warning about dead store. Fixed clang analyzer warning about dead store. Fixed clang analyzer warning about dead store. Fixed clang analyzer warning about dead store. Fixed clang analyzer warning about 2 dead stores. Fixed clang analyzer warning about dead store. Fixed clang analyzer warning about reading past an array. Fixed several clang analyzer warnings. Fixed clang analyzer warning about reading uninitialized memory. Fixed clang analyzer warning about dead store to 'pd'. Fixed clang analyzer warning about passing null to memcpy. Fixed clang analyzer warning about dead store to 'index'. Fixed clang analyzer warning about dead store to 'bestTime'. Workaround clang analyzer warning by creating temp variable. Fixed clang analyzer warnings about dead stores. Misc cleanup of code nearby clang analyzer warnings. Fixed clang analyzer warnings about dead stores. Fixed clang analyzer warning about dead store at very end of method. Fixed clang analyzer warning about dead store to 'itr'. Fixed clang analyzer warning about dead store to 'cptr'. Fixed null deref found by clang analyzer Fixed clang analyzer warnings about dereferences Suppress clang analyzer warning Fixed clang analyzer warning about null deref Fixed clang analyzer warning about reading uninitialized data. Fixed infinite recursion in vtkBitArray::RemoveFirstTuple() Improved code coverage of vtkScalarsToColors by adding more tests Added tests for a few uncovered vtkMatrix3x3 methods Fixed error in vtkFiltersCoreCxx-TestClipPolyData test Added comments about the danger of not clamping. Removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS Various Cocoa improvements and fixes Fixed a bunch of clang -Wabsolute-value warnings Removed all uses of deprecated NSOpenGLPFACompliant Assume UTF8 for vtkCarbonRenderWindow::SetWindowName Remove unnessary 'if' before 'delete[]' Add #error when trying to build as Cocoa ARC Fixed confusing indentation, no behaviour change Fixed unused variable warning in release Rewrote vtkCoreGraphicsGPUInfoList to not use deprecated API Removed checks for antique Borland versions Changed public API of vtkGPUInfo to return vtkTypeUInt64 for memory sizes. Fix clang warning about comparision that's always true. Added TODOs to indicate source of TestQuadricLODActor failure Change #include style from <> to "" for some vtk files Changed #include to replaced #include with replaced most with includes changed some to changed some to Fix gcc warning about possible use of uninitialized variable Added check for null before dereference, for bug #14671 cleanup all uses of fclose() in vtkSTLReader, for bug #14515 Removed dead code in test by using #if 0 Moved unused var suppression to suppressed dead code warning Removed unneeded break statements in switches Fixed inconsistent #include style, from <> to "" Fixed inconsistent #include style, changed <> to "" Fixed SimpleCocoaVTK Xcode project to build properly. Fixed bug #14266: fix observation of NSView size change Fixed unused const var warning Fixed a backwards ?if? test, and memory leak Eliminated unneeded null checks before using delete/free Fixed indentation level Fixed ?null deref? warning Simplified memory deallocation Removed useless assignment Fixed warning about using null with memcpy Removed useless assignments Misc cleanup and refactoring Bail early on null parameter Sync up Cocoa code between OpenGL and OpenGL2 Added partial/initial support for Cocoa ARC memory management Remove dead store to fix warning. Added some consts in a few places Put initial assignment at declaration Fixed analyzer warning by removing null check Fixed analyzer warning about possible null deref Fixed a few dead store warnings Fixed dead store warning by removing redundant 'break' Fixed dead store warning by reformulating loop Fixed null dereference Fixed false positive warning about null deref Fixed null dereference Fixed obvious null dereferences Hopefully supress dashboard compiler warning in QT header Fixed use of memory after its deallocation Fixed incorrect indentation Prevent null deref of ?null? Prevent null deref of ?composite? fn pointer Init memory to prevent it being used uninitialized Removed dead store to ?j? Removed dead stores Suppress numerous warnings about using uninitialized memory Remove a likely dead store Prevent null dereferences Fixed null deref and whitespace/indentation Fixed obvious dead stores Fixed null deref by returning upon error Fixed memory leaks and malloc(0) Removed minor dead stores, to fix warnings Refactor to make intent clear to clang analyzer Fixed warning about possible null deref Fix clang analyzer warning by adding temp variable Fixed warning about using ?volBounds? uninitialized Only react to NSView frame changes if VTK itself created the NSView Suppress clang analyzer warning Suppress warning about dead store Suppress warning about ?x? used uninitialized Move warning check to prevent ?fd? null deref Return after warning to prevent ?dsa? null deref Simplify and exit quickly if input point ids empty Another try to suppress warnings from Qt headers Advance null check to before cast Fix possible null deref of ?dsa? Added vtkErrorMacro to code paths that give null ptrs Removed harmless-looking dead code Another attempt to silence warning from Qt header Fixed -Wstring-conversion warnings Whitespace/spelling Fix warning about use-after-free Fix possible null deref warnings Made some copy-pasted code more self-similar Fixed types of some copy-pasted code Move some declarations closer to use. Removed useless null check Return early to avoid null to strcmp() Removed dubious #undef of keywords Refactoring and cleanup of Cocoa NSWindow and NSView observations Remove double underscore in header guards (.txx & __vtk*_txx form) Remove double underscore in header guards (.in & __vtk*_h form) Remove double underscore in header guards (__vtk*_cxx form) Manual tweak of .hxx include guard naming Remove double underscore in header guards (.h & __vtk*_h form) Manual search and cleanup of "__vtk" Sebastien Jourdain (60): Add catalyst web widget into VTK Add NaN support in web chart rendering Add a resampling filter Add JSON writer for ImageData Extend vtkImageMapToColors to support Mask color and Non Active Scalar input Use / for path separator to prevent Windows related issue with escape char Improve implementation of Catalyst viewer Add image resampler catalyst web widget Remove test dependency to data when not needed Remove usage of the keyword with to allow old python Add new set of catalyst web widgets Attempt to handle some system raise condition Add catalyst web composite widget Better composite catalyst web widget better UI for catalyst re-sampler web widget Various VTK-web improvement Fix vtkweb catalyst css Add cost information in workbench analysis Improve web compositing for dynamic interaction Properly handle background selector visibility Fix composite rendering widget for search Add sorting capabilities Refresh workbench UI and workflow Fix Composite Javascript for WebKit Improve ParaView cinema UI Improve basic search in web-catalyst Fix catalyst web - cost time computation Improve catalyst web UI Catalyst web: Minor Web UI change Catalyst-web: Improve composite search experience Add image width column Add cost estimate for Catalyst Web Catalyst-Web estimate UI fix Change catalyst-web analysis API to expect full URL Fix center of rotation for WebGL renderer Fix WebGL background orientation Fix basic catalyst viewer with theta Add runtime libraries on Windows for Java binary package Fix several catalyst web UI issue Add configuration pass to Catalyst PVW workflow Add support for mobile device interaction Catalyst Web - Composite improvement Catalyst web Add new rendering class API based on vtkPanel class Remove @Override annotation to prevent compilation issue on old Java compiler Upgrade autobahn version to 0.8.9 Upgrade Twisted version to 14.0.0 Upgrade ZopeInterface version to 4.1.1 Add SixPython module for Autobahn dependency Update Autobahn/JS library version Update VTK Web Python code to handle WAMP v2 Update Web applications to support new protocol API Fix python for vtkweb Allow arbitrary WebSocket endpoint Add bottstrap3 + pv lib in loader Fix import with latest Autobahn update Add missing call to properly handle webgl rendering Remove loading of native lib from vtk classes Add Get/SetTuple6 methods Add additional information for Java package build Shawn Waldon (20): Add a smartpointer for PyObjects Format comments according to VTK stlye guide Fix the vtkPolyPlane implicit function Change vtkProbeFilter to expose the 'in' threshold Move static data in vtkVRMLImporter to local struct Assume block request produced what was requested Add a documentation module that contains information keys Add block amount of detail information key Block opacity and color are now more independent Add a baseline for block opacity with no depth peeling Adding a key to indicate the current process has the block Legacy reader/writers for composite data now save field data Added comment on required invocation order Prefer vtkDoubleArray::GetValue to GetTuple for single component arrays Made vtkGaussianSplatter handle composite data input Made sure composite data iterators were deleted properly Made vtkPointGaussianMapper handle NULL arrays Add an option to set default radius of point gaussian Fix triangle size for sprites with radius > 1 Made Point Gaussian mapper handle scale arrays not in dataset Shinya Onogi (1): fix_GPU_resource_handling Sujin Philip (6): Fix for buffer overflow in vtkCubeAxesActor class Toolchain files for MICs on stampede Fix compile warnings in SMP under Windows Additions to SMP API Fix SMP examples warnings VTK SMP implementation of SynchronizedTemplates3D Thomas Vaughan (1): Set origin and spacing in vtkSurfaceReconstructionFilter. Ref: Mantis 0002826. Tim Thirion (14): Prefer xcrun when querying for iOS-related SDK elements Add CMake option to select OpenGL ES 2.0 or 3.0 Add x86_64 to list of archs for iOS simulator Surfaces example ported to iOS Add CMake script to allow user to `make framework` for iOS Remove shell script for creating VTK framework; update iOS readme Update CMakeLists for iOS examples Create iOS framework in one pass through CMake config Clean up iOS build script Make vtkVersion available sooner in CMakeLists.txt iOS build: check install directories before any compilation iOS build: pass OpenGL ES version string to child builds Update Android toolchain file (now support NDKs r5 through r10d) Add one-pass Android build Tristan Coulange (5): BUG: fix some xdmf writer crashes BUG: Add a missing include for VS 2013 Add support for original ids for generic dataset BUG: fix xdmf test failure BUG: Fix the quads evaluate location Utkarsh Ayachit (56): Adding a new test for TestSocketCommunicator. Added test to test vtkSynchronizedRenderers. Deprecate vtkOpenGLPolyDataMapper. Fix segfault with TestSetGet. Fix build issues with deprecated vtkOpenGLPolyDataMapper. Add a delay between process launches. Fixed mismatches cell arrays in vtkPStreamTracer. Added ability to probe filter to allow disabling of passing field data. Add support for empty strinngs in SetInputArrayToProcess. Make it possible to override ClampPos. Remove OpenGL error checks that were causing errors. BUG #14244: Fix stereo image captures. Add helper method to convert from SelectionField to AttributeType. Fix appending of scalar arrays. Fix DataSet reference in when input array is passed to output. Avoid exceptions on StringArrays. Fixed exceptions with older numpy version. Cleanup MPI controller properly. vtkExtractSelectedRows only works with IdTypeArray. BUG #14828: Keep surface color from interacting with scalar color. Handle append scalar in CompositeDataSetAttributes. BUG: Fixed opacity mapping when scalar array is vtkUnsignedChar. BUG #14813: Fix range reported for mode shapes. BUG #14599: Handle double values properly. Fix issue with empty arrays with numpy 1.2.1. Don't glyph masked points in vtkUniformGrid. BUG: Fix invalid return value check in vtkClientSocket. vtkScalarBarRepresentation doesn't call Superclass. Fixed logic to work when an empty vtkSelection is passed. Fix handling of empty selection on ParallelCoordinates. Fixed issue with selection frame line style. Refactoring CMake code for Python modules. BUG #14971. Fix invalid sprintf() in vtkPNGWriter. Add missing include. Needed for std::max/min. Handle case when make_vector is called with NumPy arrays. Make vtkScalarsToColorsPainter respect ScalarMaterialMode. Update TestScalarMaterialMode test image size. BUG #14779. Fixes for vtkDistancePolyDataFilter. BUG 14693: Fixed vtkAssignAttribute for unnamed arrays. BUG #12753: Fixed support for large images. BUG #11607: Fixed support for large images. Add support for vtkTable to vtkExtractSelectedThresholds. Fixed warnings with shadowed variables. BUG #15046, BUG #15020: Fix issues with OpenGL 1.2. BUG #15058: Fixes unclipped grid. BUG #14809: Use title size when placing title in vtkChartXY. vtkPlotGrid draws grid even when axis is invisible. BUG #15132: Fix incorrect flag check. Disable OpenGL error checks in release builds. Adding alternate baseline for TestLinePlotAxisFonts. Change invalid extent error to a debug message. Fixed typo in commit 031f7f6371. Fix inverted ifdef NDEBUG in commit f1a90fd2. Forward InteractionEvent to Interactor. Miscellaneous vtkContextInteractorStyle fixes. Fix missing headers. Will Schroeder (13): Disabling translation and scaling now works. Added the ability to specify a clip tolerance. Fixed bug causing plane to disappear on boundary. InsertNextCell and related method now return the proper vtkIdType. Create vtkPolyData instead of vtkUnstructuredGrid when passing points. Extended filter to support culling unused points. Added ability to control alpha shape output Added hints for vtkBox.GetBounds() Updated documentation Select internal tetra tessellation based on scalar change New data file for QuadraticTetra Alternative test image Progress reporting was broken XDMF Developers (3): export xdmf for import into vtk export xdmf for import into vtk export xdmf for import into vtk Xabi Riobe (1): BUG: The method vtkClipPlanesPainter::UpdateBounds must not be implemented. Zack Galbreath (25): fix lingering off-by-one error new writer: vtkPhyloXMLTreeWriter attempt to silence "possible loss of data" warning improve the thoroughness of TestPhyloXMLTreeWriter add baseline directory get the baseline file's path from the command line add option to write XML to string instead of file fix failing tests revealed by gerrit fix the format of PhyloXML references fix blank newline in XML output vtkXMLReader can now read from an input string new class: vtkPhyloXMLTreeReader consolidate tests for PhyloXML silence unused parameter warning change signatures for wrapping copy row names from R to vtkTable do not assume 1st column holds row names fix failing SetGet test forvtkHeatmapItem preserve R data.frame row names break vtkRCalculatorFilter's input/output symmetry style cleanup add SSL support to VTKWeb server fix potential segmentation fault change the way we shut down the R interface fix failing R tests reported by CDash xabi riobe (4): Fix memory leak in vtkPNGWriter when OutOfDiskSpaceError Initialize row_pointers Fix numerical precision problem in vtkCellLocator::IntersectWithLine Add default value for argument From andrew.amaclean at gmail.com Mon Feb 16 17:13:02 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Tue, 17 Feb 2015 09:13:02 +1100 Subject: [vtk-developers] CMP0020 warning in Tutorial Example 6... why? In-Reply-To: References: Message-ID: Yes, that was the problem. Thankyou for spotting it. I have submitted a change for review: http://review.source.kitware.com/#/c/19278/ Regards Andrew On Tue, Feb 17, 2015 at 12:12 AM, Robert Maynard wrote: > Have you tried removing the find_package(VTK) on line 19? > > On Sun, Feb 15, 2015 at 10:25 PM, Andrew Maclean > wrote: > > When I build the examples in VTK why do I get a CMP0020 warning in > > Examples/Tutorial/Step6/Cxx/CMakeLists.txt but not for any of the other > > examples. > > It would be easily fixed by adding: > > if(POLICY CMP0020) > > cmake_policy(SET CMP0020 NEW) > > endif() > > to the CMakeLists.txt file. However I don't see why this particular > example > > needs it. It doesn't use QT and I don't see why the other examples don't > > need this fix. > > > > Can anyone provide some insight here? > > > > I am building with CMake 3.1.3, QT5.4, OpenGL2 using MSVC 3013. > > > > This warning occurs multiple times when building VTK: > > ------------------------------------ > > CMake Warning (dev) in Examples/Tutorial/Step6/Cxx/CMakeLists.txt: > > Policy CMP0020 is not set: Automatically link Qt executables to qtmain > > target on Windows. Run "cmake --help-policy CMP0020" for policy > details. > > Use the cmake_policy command to set the policy and suppress this > warning. > > This warning is for project developers. Use -Wno-dev to suppress it. > > ------------------------------------ > > > > Thanks for any insight. > > > > Regards > > Andrew > > > > -- > > ___________________________________________ > > Andrew J. P. Maclean > > > > ___________________________________________ > > > > _______________________________________________ > > Powered by www.kitware.com > > > > Visit other Kitware open-source projects at > > http://www.kitware.com/opensource/opensource.html > > > > Search the list archives at: > http://markmail.org/search/?q=vtk-developers > > > > Follow this link to subscribe/unsubscribe: > > http://public.kitware.com/mailman/listinfo/vtk-developers > > > > > -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerrick.Bivins at halliburton.com Mon Feb 16 17:10:49 2015 From: Gerrick.Bivins at halliburton.com (Gerrick Bivins) Date: Mon, 16 Feb 2015 22:10:49 +0000 Subject: [vtk-developers] [EXTERNAL] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: ?VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven? This is awesome! Are there plans to upload these to maven central? Or even something like Jcenter: https://bintray.com/bintray/jcenter https://bintray.com/howbintrayworks Gerrick From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of David E DeMarle Sent: Monday, February 16, 2015 3:43 PM To: vtkdev; vtkusers at vtk.org; Kitware Corporate Communications Subject: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready The VTK developement team is happy to announce that VTK 6.2 has entered the release candidate stage! You can find the source, data, and new vtkpython binary packages here: http://www.vtk.org/VTK/resources/software.html#latestcand Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.2.0 final. The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview: -- VTK?s use of the parallelism available in modern architectures continues to mature. To set the stage for this move, we have refactored how pieces and extents are handled in the pipeline. The pipeline used to inject extent translators into the pipeline to obtain structured sub-extents for each requested unstructured piece. This approach had flaws so now the translation responsibility falls to the readers which are best able to make the computation and parallel filters are now responsible for dealing gracefully with unexpected extents. Next we have we have updated to the latest version of the Dax toolkit and begun to lay the groundwork for adopting the vtk-m project as a successor to the Piston, Dax, and EAVL efforts. There have been a series of minor improvements to the SMP framework and vtkSMP filters too as we get ready to adopt them in a large swath of VTK?s algorithms. In related work, note that we have begun to support Xeon Phi (MIC) chips. For information about this please refer to: http://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29. Similarly VTK?s support for the Web continues to advance in this release. VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and updated examples of using the web launcher to start vtkweb applications. vtkWeb applications now support http-only server/client configurations for situations when websockets are not acceptable. There were also improvements made to VTK to support the creation and use of Cinema and Workbench applications (in-situ deferred visualization) that are described in a SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ Visualization and Analysis? Wrapped languages have gotten a share of the attention in this release. There have been several fixes for ActiViz.Net which was recently updated from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples have finally been upgraded following modularization and note that it is now mandatory to invoke the method Start on the instance of vtkRenderInteractor from Tcl. In Java, vtkPanel?s behavior was changed to better support advanced class loader system like OSGI. VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven. Python has received the heaviest dose of updates. Wrapped namespaces and enum types are now available in Python. We?ve also made significant improvements to the VTK-numpy integration by introducing a number of Python modules that provide numpy-compatible interfaces to VTK data structures. See http://www.kitware.com/blog/home/post/723 for details. We?ve also introduced vtkPythonAlgorithm, which makes it easier than ever to quickly extend VTK with filters that are written directly in Python. See http://www.kitware.com/blog/home/post/737 for details. VTK?s rendering is making both evolutionary and revolutionary advances in this release. The evolutionary changes include the usual number of incremental improvements. These include such things like advanced color and display controls, "sticky" axes mode for vtkCubeAxesActor, out of range color assignments, and indexed color lookups for vtkStringArrays. There are also a number of text rendering improvements such as better multiline, rotated, and aligned text as well as BackgroundColor and BackgroundOpacity options. The revolutionary changes can be found in the OpenGL2 modules. These are under active development at the moment, and the API will be subject to change in the next release, but early adopters are encouraged to try it out and report their experiences with it on the mailing list. OpenGL2 is a rewrite of VTK?s rendering backend that brings VTK up to date with modern OpenGL programming practices. By replacing antiquated rendering techniques with modern ones, we have increased rendering performance by orders of magnitude in some situations. This work, funded by the NIH VTK Maintenance grant touches both surface and volume rendering techniques. You can read about OpenGL2 at: http://www.kitware.com/source/home/post/144 and http://www.kitware.com/source/home/post/154 Besides the above progress some of the most notable changes in VTK 6.2 include ? deprecated InfovisParallel ? vtkIOXdmf3, an interface to ARL?s greatly improved interface to HDF5 backed data storage ? added a reader and writer for NIfTI files, including the 64-bit NIfTI-2 format ? added support for SpaceMouse devices ? external rendering support for immersive environments http://kitware.com/blog/home/post/688 ? removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS ? Rewrote the OS X Cocoa mouse event handling code to make it more robust. We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 ---------------------------------------------------------------------- This e-mail, including any attached files, may contain confidential and privileged information for the sole use of the intended recipient. Any review, use, distribution, or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive information for the intended recipient), please contact the sender by reply e-mail and delete all copies of this message. -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Tue Feb 17 08:52:24 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Tue, 17 Feb 2015 08:52:24 -0500 Subject: [vtk-developers] [EXTERNAL] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: On Mon, Feb 16, 2015 at 5:10 PM, Gerrick Bivins < Gerrick.Bivins at halliburton.com> wrote: > *?**VTK?s dashboards now automatically build redistributable packages on > Windows, Linux and Mac that will allow us to feed Maven?* > > This is awesome! > > Agreed! > > > Are there plans to upload these to maven central? > I believe that is the plan. Sebastien will have the details. > Or even something like Jcenter: > > https://bintray.com/bintray/jcenter > > https://bintray.com/howbintrayworks > > > > Gerrick > > > > > > *From:* vtk-developers [mailto:vtk-developers-bounces at vtk.org] *On Behalf > Of *David E DeMarle > *Sent:* Monday, February 16, 2015 3:43 PM > *To:* vtkdev; vtkusers at vtk.org; Kitware Corporate Communications > *Subject:* [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release > candidate 1 is ready > > > > The VTK developement team is happy to announce that VTK 6.2 has entered > the release candidate stage! > > > > You can find the source, data, and new vtkpython binary packages here: > > http://www.vtk.org/VTK/resources/software.html#latestcand > > > > Please try this version of VTK and report any issues to the list or the > bug tracker so that we can try to address them before VTK 6.2.0 final. > > > > The official release notes will be available when VTK final is released in > the next few weeks. In the meantime here is a preview: > > > > -- > > > > VTK?s use of the parallelism available in modern architectures continues > to mature. To set the stage for this move, we have refactored how pieces > and extents are handled in the pipeline. The pipeline used to inject extent > translators into the pipeline to obtain structured sub-extents for each > requested unstructured piece. This approach had flaws so now the > translation responsibility falls to the readers which are best able to make > the computation and parallel filters are now responsible for dealing > gracefully with unexpected extents. Next we have we have updated to the > latest version of the Dax toolkit and begun to lay the groundwork for > adopting the vtk-m project as a successor to the Piston, Dax, and EAVL > efforts. There have been a series of minor improvements to the SMP > framework and vtkSMP filters too as we get ready to adopt them in a large > swath of VTK?s algorithms. In related work, note that we have begun to > support Xeon Phi (MIC) chips. For information about this please refer to: h > ttp://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29 > . > > > > Similarly VTK?s support for the Web continues to advance in this release. > VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and > updated examples of using the web launcher to start vtkweb applications. > vtkWeb applications now support http-only server/client configurations for > situations when websockets are not acceptable. There were also improvements > made to VTK to support the creation and use of Cinema and Workbench > applications (in-situ deferred visualization) that are described in a > SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ > Visualization and Analysis? > > > > Wrapped languages have gotten a share of the attention in this release. > There have been several fixes for ActiViz.Net which was recently updated > from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples > have finally been upgraded following modularization and note that it is now > mandatory to invoke the method Start on the instance of vtkRenderInteractor > from Tcl. In Java, vtkPanel?s behavior was changed to better support > advanced class loader system like OSGI. VTK?s dashboards now automatically > build redistributable packages on Windows, Linux and Mac that will allow us > to feed Maven. > > > > Python has received the heaviest dose of updates. Wrapped namespaces and > enum types are now available in Python. We?ve also made significant > improvements to the VTK-numpy integration by introducing a number of Python > modules that provide numpy-compatible interfaces to VTK data structures. > See http://www.kitware.com/blog/home/post/723 for details. We?ve also > introduced vtkPythonAlgorithm, which makes it easier than ever to quickly > extend VTK with filters that are written directly in Python. See > http://www.kitware.com/blog/home/post/737 for details. > > > > VTK?s rendering is making both evolutionary and revolutionary advances in > this release. The evolutionary changes include the usual number of > incremental improvements. These include such things like advanced color and > display controls, "sticky" axes mode for vtkCubeAxesActor, out of range > color assignments, and indexed color lookups for vtkStringArrays. There are > also a number of text rendering improvements such as better multiline, > rotated, and aligned text as well as BackgroundColor and BackgroundOpacity > options. > > > > The revolutionary changes can be found in the OpenGL2 modules. These are > under active development at the moment, and the API will be subject to > change in the next release, but early adopters are encouraged to try it out > and report their experiences with it on the mailing list. OpenGL2 is a > rewrite of VTK?s rendering backend that brings VTK up to date with modern > OpenGL programming practices. By replacing antiquated rendering techniques > with modern ones, we have increased rendering performance by orders of > magnitude in some situations. This work, funded by the NIH VTK Maintenance > grant touches both surface and volume rendering techniques. You can read > about OpenGL2 at: http://www.kitware.com/source/home/post/144 and > http://www.kitware.com/source/home/post/154 > > > > Besides the above progress some of the most notable changes in VTK 6.2 > include > > ? deprecated InfovisParallel > > ? vtkIOXdmf3, an interface to ARL?s greatly improved interface to > HDF5 backed data storage > > ? added a reader and writer for NIfTI files, including the 64-bit > NIfTI-2 format > > ? added support for SpaceMouse devices > > ? external rendering support for immersive environments > http://kitware.com/blog/home/post/688 > > ? removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS > > ? Rewrote the OS X Cocoa mouse event handling code to make it > more robust. > > > > We hope you enjoy this release of VTK! As always contact Kitware and the > mailing lists for assistance. > > > > David E DeMarle > Kitware, Inc. > R&D Engineer > 21 Corporate Drive > Clifton Park, NY 12065-8662 > Phone: 518-881-4909 > ------------------------------ > This e-mail, including any attached files, may contain confidential and > privileged information for the sole use of the intended recipient. Any > review, use, distribution, or disclosure by others is strictly prohibited. > If you are not the intended recipient (or authorized to receive information > for the intended recipient), please contact the sender by reply e-mail and > delete all copies of this message. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerrick.Bivins at halliburton.com Tue Feb 17 09:22:25 2015 From: Gerrick.Bivins at halliburton.com (Gerrick Bivins) Date: Tue, 17 Feb 2015 14:22:25 +0000 Subject: [vtk-developers] [EXTERNAL] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: Excellent! You guys are really making great strides with VTK. I?ve been holding off on upgrading for some time but this feature list (not just the maven bits) is too tempting! Great work! From: David E DeMarle [mailto:dave.demarle at kitware.com] Sent: Tuesday, February 17, 2015 7:52 AM To: Gerrick Bivins; Sebastien Jourdain Cc: VTK Developers Subject: Re: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready On Mon, Feb 16, 2015 at 5:10 PM, Gerrick Bivins > wrote: ?VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven? This is awesome! Agreed! Are there plans to upload these to maven central? I believe that is the plan. Sebastien will have the details. Or even something like Jcenter: https://bintray.com/bintray/jcenter https://bintray.com/howbintrayworks Gerrick From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of David E DeMarle Sent: Monday, February 16, 2015 3:43 PM To: vtkdev; vtkusers at vtk.org; Kitware Corporate Communications Subject: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready The VTK developement team is happy to announce that VTK 6.2 has entered the release candidate stage! You can find the source, data, and new vtkpython binary packages here: http://www.vtk.org/VTK/resources/software.html#latestcand Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.2.0 final. The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview: -- VTK?s use of the parallelism available in modern architectures continues to mature. To set the stage for this move, we have refactored how pieces and extents are handled in the pipeline. The pipeline used to inject extent translators into the pipeline to obtain structured sub-extents for each requested unstructured piece. This approach had flaws so now the translation responsibility falls to the readers which are best able to make the computation and parallel filters are now responsible for dealing gracefully with unexpected extents. Next we have we have updated to the latest version of the Dax toolkit and begun to lay the groundwork for adopting the vtk-m project as a successor to the Piston, Dax, and EAVL efforts. There have been a series of minor improvements to the SMP framework and vtkSMP filters too as we get ready to adopt them in a large swath of VTK?s algorithms. In related work, note that we have begun to support Xeon Phi (MIC) chips. For information about this please refer to: http://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29. Similarly VTK?s support for the Web continues to advance in this release. VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and updated examples of using the web launcher to start vtkweb applications. vtkWeb applications now support http-only server/client configurations for situations when websockets are not acceptable. There were also improvements made to VTK to support the creation and use of Cinema and Workbench applications (in-situ deferred visualization) that are described in a SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ Visualization and Analysis? Wrapped languages have gotten a share of the attention in this release. There have been several fixes for ActiViz.Net which was recently updated from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples have finally been upgraded following modularization and note that it is now mandatory to invoke the method Start on the instance of vtkRenderInteractor from Tcl. In Java, vtkPanel?s behavior was changed to better support advanced class loader system like OSGI. VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven. Python has received the heaviest dose of updates. Wrapped namespaces and enum types are now available in Python. We?ve also made significant improvements to the VTK-numpy integration by introducing a number of Python modules that provide numpy-compatible interfaces to VTK data structures. See http://www.kitware.com/blog/home/post/723 for details. We?ve also introduced vtkPythonAlgorithm, which makes it easier than ever to quickly extend VTK with filters that are written directly in Python. See http://www.kitware.com/blog/home/post/737 for details. VTK?s rendering is making both evolutionary and revolutionary advances in this release. The evolutionary changes include the usual number of incremental improvements. These include such things like advanced color and display controls, "sticky" axes mode for vtkCubeAxesActor, out of range color assignments, and indexed color lookups for vtkStringArrays. There are also a number of text rendering improvements such as better multiline, rotated, and aligned text as well as BackgroundColor and BackgroundOpacity options. The revolutionary changes can be found in the OpenGL2 modules. These are under active development at the moment, and the API will be subject to change in the next release, but early adopters are encouraged to try it out and report their experiences with it on the mailing list. OpenGL2 is a rewrite of VTK?s rendering backend that brings VTK up to date with modern OpenGL programming practices. By replacing antiquated rendering techniques with modern ones, we have increased rendering performance by orders of magnitude in some situations. This work, funded by the NIH VTK Maintenance grant touches both surface and volume rendering techniques. You can read about OpenGL2 at: http://www.kitware.com/source/home/post/144 and http://www.kitware.com/source/home/post/154 Besides the above progress some of the most notable changes in VTK 6.2 include ? deprecated InfovisParallel ? vtkIOXdmf3, an interface to ARL?s greatly improved interface to HDF5 backed data storage ? added a reader and writer for NIfTI files, including the 64-bit NIfTI-2 format ? added support for SpaceMouse devices ? external rendering support for immersive environments http://kitware.com/blog/home/post/688 ? removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS ? Rewrote the OS X Cocoa mouse event handling code to make it more robust. We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 ________________________________ This e-mail, including any attached files, may contain confidential and privileged information for the sole use of the intended recipient. Any review, use, distribution, or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive information for the intended recipient), please contact the sender by reply e-mail and delete all copies of this message. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gerrick.Bivins at halliburton.com Wed Feb 18 09:11:25 2015 From: Gerrick.Bivins at halliburton.com (Gerrick Bivins) Date: Wed, 18 Feb 2015 14:11:25 +0000 Subject: [vtk-developers] [EXTERNAL] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: Sebastien, Just thinking out loud here. ?something similar to what is done in JOGL?. Do you guys(Kitware) have a feeling for the usage of VTK/Java, meaning is the typical usage trying to leverage VTK in an OSGi type environment or just in a standalone java application? I think that would have some implications on how to set this up. Example of some typical issues when dealing with JNI in a ?modular runtime?: http://wiki.osgi.org/wiki/Dependencies_In_Native_Code In fact, it would probably necessitate 2 different solutions to target the different runtime configurations. Side note, I?ve often wondered if CMake could help here by generating the poms from the dependency hierarchy. Gerrick From: Sebastien Jourdain [mailto:sebastien.jourdain at kitware.com] Sent: Tuesday, February 17, 2015 9:46 AM To: Gerrick Bivins Cc: David E DeMarle Subject: Re: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready Hi Gerrick, I've been excited about the Maven part too. It is not perfect yet and still need some love but it is definitely a step forward. Maybe with a little bit of help/guidance, we will be able to improve that even more. The loading of the native library is not yet supported from the "native" jars, but if someone is willing to contribute something similar to what is done in JOGL, that would be great. So far, we are just generating the native code and packaging inside a jar along with a POM. Moreover, pushing to a central maven repo, is definitely my goal. Although, I will need some guidance on how to do so. Good links explaining that process would be great. Thanks, Seb On Tue, Feb 17, 2015 at 7:22 AM, Gerrick Bivins > wrote: Excellent! You guys are really making great strides with VTK. I?ve been holding off on upgrading for some time but this feature list (not just the maven bits) is too tempting! Great work! From: David E DeMarle [mailto:dave.demarle at kitware.com] Sent: Tuesday, February 17, 2015 7:52 AM To: Gerrick Bivins; Sebastien Jourdain Cc: VTK Developers Subject: Re: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready On Mon, Feb 16, 2015 at 5:10 PM, Gerrick Bivins > wrote: ?VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven? This is awesome! Agreed! Are there plans to upload these to maven central? I believe that is the plan. Sebastien will have the details. Or even something like Jcenter: https://bintray.com/bintray/jcenter https://bintray.com/howbintrayworks Gerrick From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of David E DeMarle Sent: Monday, February 16, 2015 3:43 PM To: vtkdev; vtkusers at vtk.org; Kitware Corporate Communications Subject: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready The VTK developement team is happy to announce that VTK 6.2 has entered the release candidate stage! You can find the source, data, and new vtkpython binary packages here: http://www.vtk.org/VTK/resources/software.html#latestcand Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.2.0 final. The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview: -- VTK?s use of the parallelism available in modern architectures continues to mature. To set the stage for this move, we have refactored how pieces and extents are handled in the pipeline. The pipeline used to inject extent translators into the pipeline to obtain structured sub-extents for each requested unstructured piece. This approach had flaws so now the translation responsibility falls to the readers which are best able to make the computation and parallel filters are now responsible for dealing gracefully with unexpected extents. Next we have we have updated to the latest version of the Dax toolkit and begun to lay the groundwork for adopting the vtk-m project as a successor to the Piston, Dax, and EAVL efforts. There have been a series of minor improvements to the SMP framework and vtkSMP filters too as we get ready to adopt them in a large swath of VTK?s algorithms. In related work, note that we have begun to support Xeon Phi (MIC) chips. For information about this please refer to: http://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29. Similarly VTK?s support for the Web continues to advance in this release. VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and updated examples of using the web launcher to start vtkweb applications. vtkWeb applications now support http-only server/client configurations for situations when websockets are not acceptable. There were also improvements made to VTK to support the creation and use of Cinema and Workbench applications (in-situ deferred visualization) that are described in a SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ Visualization and Analysis? Wrapped languages have gotten a share of the attention in this release. There have been several fixes for ActiViz.Net which was recently updated from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples have finally been upgraded following modularization and note that it is now mandatory to invoke the method Start on the instance of vtkRenderInteractor from Tcl. In Java, vtkPanel?s behavior was changed to better support advanced class loader system like OSGI. VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven. Python has received the heaviest dose of updates. Wrapped namespaces and enum types are now available in Python. We?ve also made significant improvements to the VTK-numpy integration by introducing a number of Python modules that provide numpy-compatible interfaces to VTK data structures. See http://www.kitware.com/blog/home/post/723 for details. We?ve also introduced vtkPythonAlgorithm, which makes it easier than ever to quickly extend VTK with filters that are written directly in Python. See http://www.kitware.com/blog/home/post/737 for details. VTK?s rendering is making both evolutionary and revolutionary advances in this release. The evolutionary changes include the usual number of incremental improvements. These include such things like advanced color and display controls, "sticky" axes mode for vtkCubeAxesActor, out of range color assignments, and indexed color lookups for vtkStringArrays. There are also a number of text rendering improvements such as better multiline, rotated, and aligned text as well as BackgroundColor and BackgroundOpacity options. The revolutionary changes can be found in the OpenGL2 modules. These are under active development at the moment, and the API will be subject to change in the next release, but early adopters are encouraged to try it out and report their experiences with it on the mailing list. OpenGL2 is a rewrite of VTK?s rendering backend that brings VTK up to date with modern OpenGL programming practices. By replacing antiquated rendering techniques with modern ones, we have increased rendering performance by orders of magnitude in some situations. This work, funded by the NIH VTK Maintenance grant touches both surface and volume rendering techniques. You can read about OpenGL2 at: http://www.kitware.com/source/home/post/144 and http://www.kitware.com/source/home/post/154 Besides the above progress some of the most notable changes in VTK 6.2 include ? deprecated InfovisParallel ? vtkIOXdmf3, an interface to ARL?s greatly improved interface to HDF5 backed data storage ? added a reader and writer for NIfTI files, including the 64-bit NIfTI-2 format ? added support for SpaceMouse devices ? external rendering support for immersive environments http://kitware.com/blog/home/post/688 ? removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS ? Rewrote the OS X Cocoa mouse event handling code to make it more robust. We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 ________________________________ This e-mail, including any attached files, may contain confidential and privileged information for the sole use of the intended recipient. Any review, use, distribution, or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive information for the intended recipient), please contact the sender by reply e-mail and delete all copies of this message. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sebastien.jourdain at kitware.com Wed Feb 18 22:55:05 2015 From: sebastien.jourdain at kitware.com (Sebastien Jourdain) Date: Wed, 18 Feb 2015 20:55:05 -0700 Subject: [vtk-developers] [EXTERNAL] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: Certainly both. We have a templated Pom that cmake use to generate the Java package. Sent from my iPad > On Feb 18, 2015, at 07:11, Gerrick Bivins wrote: > > Sebastien, > > Just thinking out loud here. > > ?something similar to what is done in JOGL?. > Do you guys(Kitware) have a feeling for the usage of VTK/Java, > meaning is the typical usage trying to leverage VTK in an OSGi type > environment or just in a standalone java application? > I think that would have some implications on how to set this up. > Example of some typical issues when dealing with JNI > in a ?modular runtime?: > http://wiki.osgi.org/wiki/Dependencies_In_Native_Code > > In fact, it would probably necessitate 2 different solutions > to target the different runtime configurations. > > Side note, I?ve often wondered if CMake could help here by > generating the poms from the dependency hierarchy. > > Gerrick > From: Sebastien Jourdain [mailto:sebastien.jourdain at kitware.com] > Sent: Tuesday, February 17, 2015 9:46 AM > To: Gerrick Bivins > Cc: David E DeMarle > Subject: Re: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready > > Hi Gerrick, > > I've been excited about the Maven part too. It is not perfect yet and still need some love but it is definitely a step forward. > Maybe with a little bit of help/guidance, we will be able to improve that even more. > > The loading of the native library is not yet supported from the "native" jars, but if someone is willing to contribute something similar to what is done in JOGL, that would be great. So far, we are just generating the native code and packaging inside a jar along with a POM. > > Moreover, pushing to a central maven repo, is definitely my goal. Although, I will need some guidance on how to do so. Good links explaining that process would be great. > > Thanks, > > Seb > > On Tue, Feb 17, 2015 at 7:22 AM, Gerrick Bivins wrote: > Excellent! > You guys are really making great strides with VTK. > I?ve been holding off on upgrading for some time but this > feature list (not just the maven bits) is too tempting! > Great work! > > > From: David E DeMarle [mailto:dave.demarle at kitware.com] > Sent: Tuesday, February 17, 2015 7:52 AM > To: Gerrick Bivins; Sebastien Jourdain > Cc: VTK Developers > Subject: Re: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready > > > On Mon, Feb 16, 2015 at 5:10 PM, Gerrick Bivins wrote: > ?VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven? > This is awesome! > > Agreed! > > > Are there plans to upload these to maven central? > > I believe that is the plan. Sebastien will have the details. > > Or even something like Jcenter: > https://bintray.com/bintray/jcenter > https://bintray.com/howbintrayworks > > Gerrick > > > From: vtk-developers [mailto:vtk-developers-bounces at vtk.org] On Behalf Of David E DeMarle > Sent: Monday, February 16, 2015 3:43 PM > To: vtkdev; vtkusers at vtk.org; Kitware Corporate Communications > Subject: [EXTERNAL] [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready > > The VTK developement team is happy to announce that VTK 6.2 has entered the release candidate stage! > > You can find the source, data, and new vtkpython binary packages here: > http://www.vtk.org/VTK/resources/software.html#latestcand > > Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.2.0 final. > > The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview: > > -- > > VTK?s use of the parallelism available in modern architectures continues to mature. To set the stage for this move, we have refactored how pieces and extents are handled in the pipeline. The pipeline used to inject extent translators into the pipeline to obtain structured sub-extents for each requested unstructured piece. This approach had flaws so now the translation responsibility falls to the readers which are best able to make the computation and parallel filters are now responsible for dealing gracefully with unexpected extents. Next we have we have updated to the latest version of the Dax toolkit and begun to lay the groundwork for adopting the vtk-m project as a successor to the Piston, Dax, and EAVL efforts. There have been a series of minor improvements to the SMP framework and vtkSMP filters too as we get ready to adopt them in a large swath of VTK?s algorithms. In related work, note that we have begun to support Xeon Phi (MIC) chips. For information about this please refer to: http://www.paraview.org/Wiki/ParaView_and_VTK_on_Xeon_Phi_%28KNC%29. > > Similarly VTK?s support for the Web continues to advance in this release. VTK-Web has migrated to WAMP 2.0 in 6.2 and there are now complete and updated examples of using the web launcher to start vtkweb applications. vtkWeb applications now support http-only server/client configurations for situations when websockets are not acceptable. There were also improvements made to VTK to support the creation and use of Cinema and Workbench applications (in-situ deferred visualization) that are described in a SuperComputing 2015 paper ?An Image-based Approach to Extreme Scale In Situ Visualization and Analysis? > > Wrapped languages have gotten a share of the attention in this release. There have been several fixes for ActiViz.Net which was recently updated from 5.8 to VTK 6.1 and will soon be updated again to 6.2. The Tcl examples have finally been upgraded following modularization and note that it is now mandatory to invoke the method Start on the instance of vtkRenderInteractor from Tcl. In Java, vtkPanel?s behavior was changed to better support advanced class loader system like OSGI. VTK?s dashboards now automatically build redistributable packages on Windows, Linux and Mac that will allow us to feed Maven. > > Python has received the heaviest dose of updates. Wrapped namespaces and enum types are now available in Python. We?ve also made significant improvements to the VTK-numpy integration by introducing a number of Python modules that provide numpy-compatible interfaces to VTK data structures. See http://www.kitware.com/blog/home/post/723 for details. We?ve also introduced vtkPythonAlgorithm, which makes it easier than ever to quickly extend VTK with filters that are written directly in Python. See http://www.kitware.com/blog/home/post/737 for details. > > VTK?s rendering is making both evolutionary and revolutionary advances in this release. The evolutionary changes include the usual number of incremental improvements. These include such things like advanced color and display controls, "sticky" axes mode for vtkCubeAxesActor, out of range color assignments, and indexed color lookups for vtkStringArrays. There are also a number of text rendering improvements such as better multiline, rotated, and aligned text as well as BackgroundColor and BackgroundOpacity options. > > The revolutionary changes can be found in the OpenGL2 modules. These are under active development at the moment, and the API will be subject to change in the next release, but early adopters are encouraged to try it out and report their experiences with it on the mailing list. OpenGL2 is a rewrite of VTK?s rendering backend that brings VTK up to date with modern OpenGL programming practices. By replacing antiquated rendering techniques with modern ones, we have increased rendering performance by orders of magnitude in some situations. This work, funded by the NIH VTK Maintenance grant touches both surface and volume rendering techniques. You can read about OpenGL2 at: http://www.kitware.com/source/home/post/144 and http://www.kitware.com/source/home/post/154 > > Besides the above progress some of the most notable changes in VTK 6.2 include > ? deprecated InfovisParallel > ? vtkIOXdmf3, an interface to ARL?s greatly improved interface to HDF5 backed data storage > ? added a reader and writer for NIfTI files, including the 64-bit NIfTI-2 format > ? added support for SpaceMouse devices > ? external rendering support for immersive environments http://kitware.com/blog/home/post/688 > ? removed -fobjc-gc from VTK_REQUIRED_OBJCXX_FLAGS > ? Rewrote the OS X Cocoa mouse event handling code to make it more robust. > > We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance. > > David E DeMarle > Kitware, Inc. > R&D Engineer > 21 Corporate Drive > Clifton Park, NY 12065-8662 > Phone: 518-881-4909 > This e-mail, including any attached files, may contain confidential and privileged information for the sole use of the intended recipient. Any review, use, distribution, or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive information for the intended recipient), please contact the sender by reply e-mail and delete all copies of this message. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sean at rogue-research.com Thu Feb 19 15:04:44 2015 From: sean at rogue-research.com (Sean McBride) Date: Thu, 19 Feb 2015 15:04:44 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: Message-ID: <20150219200444.933334771@mail.rogue-research.com> On Mon, 16 Feb 2015 16:43:12 -0500, David E DeMarle said: >The VTK developement team is happy to announce that VTK 6.2 has entered the >release candidate stage! Great work all! I've updated our app and our QA will be trying it. How many release candidates are planned? On what schedule? Approx final release date? Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From dave.demarle at kitware.com Thu Feb 19 15:16:54 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 19 Feb 2015 15:16:54 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150219200444.933334771@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> Message-ID: On Thu, Feb 19, 2015 at 3:04 PM, Sean McBride wrote: > > How many release candidates are planned? On what schedule? Approx final > release date? > > Aiming for an rc each Monday. The number of rcs depends on the severity of things that people find. So far Ben and I have our eyes on three topics, none of which are severe. Most likely those will fold into release tonight and become 6.2.0 final come Monday. If something else turns up we'll do rc2 then instead and reevaluate later next week. Patch releases are not time consuming, so if something slips through we'll do a patch release in a month or so. > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sean at rogue-research.com Thu Feb 19 16:25:11 2015 From: sean at rogue-research.com (Sean McBride) Date: Thu, 19 Feb 2015 16:25:11 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> Message-ID: <20150219212511.1505792326@mail.rogue-research.com> On Thu, 19 Feb 2015 15:16:54 -0500, David E DeMarle said: >Aiming for an rc each Monday. The number of rcs depends on the severity of >things that people find. > >So far Ben and I have our eyes on three topics, none of which are severe. >Most likely those will fold into release tonight and become 6.2.0 final >come Monday. If something else turns up we'll do rc2 then instead and >reevaluate later next week. > >Patch releases are not time consuming, so if something slips through we'll >do a patch release in a month or so. That seems awfully fast. :( Isn't the goal of a release candidate phase to give a chance for people to test before releasing a dud? 7 days is pretty short. Releasing a dud and requiring a 6.2.1 a couple of days later would look bad. Personally, I'd prefer to see something like ITK or CMake or llvm, which release a few rc's over a few weeks. ex: searching my email history: 2014-12-08 ITK 4.7rc1 2014-12-16 ITK 4.7rc2 2014-12-19 ITK 4.7 Taking my own case: rc1 announced Monday, Tuesday I had time to build it, Wednesday I tested a little, today I merged into our build process, tonight's nightly will include it, and only tomorrow will our QA take a crack at it. Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From marcus.hanwell at kitware.com Thu Feb 19 16:36:12 2015 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Thu, 19 Feb 2015 16:36:12 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150219212511.1505792326@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> Message-ID: On Thu, Feb 19, 2015 at 4:25 PM, Sean McBride wrote: > On Thu, 19 Feb 2015 15:16:54 -0500, David E DeMarle said: > >>Aiming for an rc each Monday. The number of rcs depends on the severity of >>things that people find. >> >>So far Ben and I have our eyes on three topics, none of which are severe. >>Most likely those will fold into release tonight and become 6.2.0 final >>come Monday. If something else turns up we'll do rc2 then instead and >>reevaluate later next week. >> >>Patch releases are not time consuming, so if something slips through we'll >>do a patch release in a month or so. > > That seems awfully fast. :( > > Isn't the goal of a release candidate phase to give a chance for people to test before releasing a dud? 7 days is pretty short. Releasing a dud and requiring a 6.2.1 a couple of days later would look bad. > > Personally, I'd prefer to see something like ITK or CMake or llvm, which release a few rc's over a few weeks. ex: searching my email history: > > 2014-12-08 ITK 4.7rc1 > 2014-12-16 ITK 4.7rc2 > 2014-12-19 ITK 4.7 > > Taking my own case: rc1 announced Monday, Tuesday I had time to build it, Wednesday I tested a little, today I merged into our build process, tonight's nightly will include it, and only tomorrow will our QA take a crack at it. It isn't that short, and we know not everyone will get time to test the RCs. Frankly I have seen plenty of releases with extended RC cycles, only to discover a critical issue days after release and make a prompt bug fix release. Why make more than one RC if there are no critical flaws discovered? It looks like ITK might have rolled a final in 8 days had they not had to make a second RC. If you have feedback we would welcome it, Monday is still a ways off and it sounds like you will have the RC with QA tomorrow. We do a lot of QA continuously, and test with various projects based upon VTK too. Marcus From sean at rogue-research.com Thu Feb 19 16:47:49 2015 From: sean at rogue-research.com (Sean McBride) Date: Thu, 19 Feb 2015 16:47:49 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> Message-ID: <20150219214749.727843881@mail.rogue-research.com> On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: >It isn't that short I guess we'll agree to disagree. :) >Why make more than one RC if there are no >critical flaws discovered? No disagreement there. But if the 1 rc period lasts only 4 business days (it was announced late Monday), there's *insufficient time* to look for flaws. VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From DLRdave at aol.com Thu Feb 19 18:23:12 2015 From: DLRdave at aol.com (David Cole) Date: Thu, 19 Feb 2015 18:23:12 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150219214749.727843881@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: The 1 week does seem too short when compared to a 13 month life span. It would be closer to appropriate if the releases were more frequent, say quarterly, like CMake's. I'm closer to Sean on this one. On Thu, Feb 19, 2015 at 4:47 PM, Sean McBride wrote: > On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: > >>It isn't that short > > I guess we'll agree to disagree. :) > >>Why make more than one RC if there are no >>critical flaws discovered? > > No disagreement there. But if the 1 rc period lasts only 4 business days (it was announced late Monday), there's *insufficient time* to look for flaws. > > VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > From marcus.hanwell at kitware.com Fri Feb 20 09:09:22 2015 From: marcus.hanwell at kitware.com (Marcus D. Hanwell) Date: Fri, 20 Feb 2015 09:09:22 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150219214749.727843881@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: On Thu, Feb 19, 2015 at 4:47 PM, Sean McBride wrote: > On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: > >>Why make more than one RC if there are no >>critical flaws discovered? > > No disagreement there. But if the 1 rc period lasts only 4 business days (it was announced late Monday), there's *insufficient time* to look for flaws. Agreed, I have been really busy with other things and didn't appreciate this. Hopefully we can extend it out a day or two - would that offer enough time? > > VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? > These things are always a balancing act, and we would like to increase the release frequency so that we avoid this 13 month cycles. Thanks for your feedback, and all of your contributions to VTK. Marcus From will.schroeder at kitware.com Sat Feb 21 07:45:29 2015 From: will.schroeder at kitware.com (Will Schroeder) Date: Sat, 21 Feb 2015 07:45:29 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: There is also the pointy-haired management concern. That is, we are coming up to the end of Year 2 NIH funding for VTK Maintenance. We have a report due in mid-March and I want to make it as strong as possible so we can continue doing other great work along the lines of OpenGL2, interaction, community, and medical data integration. Being able to talk about a release makes me very happy, and it will honor the investment that NIH is making into VTK. W On Fri, Feb 20, 2015 at 9:09 AM, Marcus D. Hanwell < marcus.hanwell at kitware.com> wrote: > On Thu, Feb 19, 2015 at 4:47 PM, Sean McBride > wrote: > > On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: > > > >>Why make more than one RC if there are no > >>critical flaws discovered? > > > > No disagreement there. But if the 1 rc period lasts only 4 business > days (it was announced late Monday), there's *insufficient time* to look > for flaws. > > Agreed, I have been really busy with other things and didn't > appreciate this. Hopefully we can extend it out a day or two - would > that offer enough time? > > > > VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? > > > These things are always a balancing act, and we would like to increase > the release frequency so that we avoid this 13 month cycles. Thanks > for your feedback, and all of your contributions to VTK. > > Marcus > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > -- William J. Schroeder, PhD Kitware, Inc. 28 Corporate Drive Clifton Park, NY 12065 will.schroeder at kitware.com http://www.kitware.com (518) 881-4902 -------------- next part -------------- An HTML attachment was scrubbed... URL: From c.kolb at dkfz-heidelberg.de Sat Feb 21 14:45:27 2015 From: c.kolb at dkfz-heidelberg.de (Christoph Kolb) Date: Sat, 21 Feb 2015 20:45:27 +0100 Subject: [vtk-developers] Delayed mouse interaction due to event-loop bug in Qt5 Message-ID: <54E8E057.1060409@dkfz-heidelberg.de> Dear vtk developers, when I am using the QVTKWidget in VTK 6.1 with Qt5, the mouse interactions are severely delayed. I can reproduce this issue by starting the SimpleView example and rotating the text. I think that this might be caused by a bug in Qt5, where the mousemove events are flooding the event queue: https://bugreports.qt.io/browse/QTBUG-40889 However, other Linux distributions and Windows seem to be unnaffected by this. I have only observed it with Archlinux (Qt 5.4.0) so far. Does anyone else have this problem, or do you know what else i could check? Thank you! Regards Christoph From DLRdave at aol.com Sat Feb 21 20:19:53 2015 From: DLRdave at aol.com (David Cole) Date: Sat, 21 Feb 2015 20:19:53 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: Just some real-world data to consider: For the past three VTK releases (5.10, 6.0 and 6.1), here are the stats, based on the tags in the git repo. average number of release candidates before a release's "final" (x.x.0) release: 3 average number of days between consecutive release candidates: 16 average number of days from rc1 to final release: 40 average number of days between x-rc1 and (x+1)-rc1: 357 I would think such readily available data ought to be considered when determining the schedule for a set of release candidates. Even by those whose hair is geometrically challenged. ;-) Based on historical data, I would expect the final release of VTK 6.2.0 to land around Feb. 16th + 40, or approximately March 28th. Thanks for the efforts. I am particularly looking forward to the OpenGL2 backend becoming a reality, even if it is still experimental for this release. Thx, David C. On Sat, Feb 21, 2015 at 7:45 AM, Will Schroeder wrote: > There is also the pointy-haired management concern. That is, we are coming > up to the end of Year 2 NIH funding for VTK Maintenance. We have a report > due in mid-March and I want to make it as strong as possible so we can > continue doing other great work along the lines of OpenGL2, interaction, > community, and medical data integration. Being able to talk about a release > makes me very happy, and it will honor the investment that NIH is making > into VTK. > > W > > On Fri, Feb 20, 2015 at 9:09 AM, Marcus D. Hanwell > wrote: >> >> On Thu, Feb 19, 2015 at 4:47 PM, Sean McBride >> wrote: >> > On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: >> > >> >>Why make more than one RC if there are no >> >>critical flaws discovered? >> > >> > No disagreement there. But if the 1 rc period lasts only 4 business >> > days (it was announced late Monday), there's *insufficient time* to look for >> > flaws. >> >> Agreed, I have been really busy with other things and didn't >> appreciate this. Hopefully we can extend it out a day or two - would >> that offer enough time? >> > >> > VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? >> > >> These things are always a balancing act, and we would like to increase >> the release frequency so that we avoid this 13 month cycles. Thanks >> for your feedback, and all of your contributions to VTK. >> >> Marcus >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: http://markmail.org/search/?q=vtk-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/vtk-developers >> > > > > -- > William J. Schroeder, PhD > Kitware, Inc. > 28 Corporate Drive > Clifton Park, NY 12065 > will.schroeder at kitware.com > http://www.kitware.com > (518) 881-4902 > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From david.lonie at kitware.com Mon Feb 23 16:59:36 2015 From: david.lonie at kitware.com (David Lonie) Date: Mon, 23 Feb 2015 16:59:36 -0500 Subject: [vtk-developers] Python wrapped void pointers Message-ID: A project I work on is facing a regression related to the vtkRenderWindow::SetDisplayId(void*) method. It apparently used to work by passing a string similar to: '_f02100_void_p' as the argument, but now they're getting TypeErrors: TypeError('SetDisplayId argument 1: value is _f02100_void_p, required type is p_void',) For context, the failure is happening here: https://github.com/UV-CDAT/VisTrails/blob/cb2220ba88e957006c3dcc098ef29c5c08461117/vistrails/packages/vtk/vtkcell.py#L356 It looks like the python wrappers might have changed how they handle the void* type -- does anyone have an idea what's going on here? Thanks, Dave -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Mon Feb 23 17:09:56 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Mon, 23 Feb 2015 15:09:56 -0700 Subject: [vtk-developers] Python wrapped void pointers In-Reply-To: References: Message-ID: This is a result of one of the bugs that I fixed during last year's hackathon: http://www.vtk.org/Bug/view.php?id=12098 http://review.source.kitware.com/#/c/17351/ On Mon, Feb 23, 2015 at 2:59 PM, David Lonie wrote: > A project I work on is facing a regression related to the > vtkRenderWindow::SetDisplayId(void*) method. > > It apparently used to work by passing a string similar to: > > '_f02100_void_p' > > as the argument, but now they're getting TypeErrors: > > TypeError('SetDisplayId argument 1: value is _f02100_void_p, required type > is p_void',) > > For context, the failure is happening here: > > > https://github.com/UV-CDAT/VisTrails/blob/cb2220ba88e957006c3dcc098ef29c5c08461117/vistrails/packages/vtk/vtkcell.py#L356 > > It looks like the python wrappers might have changed how they handle the > void* type -- does anyone have an idea what's going on here? > > Thanks, > Dave > -------------- next part -------------- An HTML attachment was scrubbed... URL: From shawn.waldon at kitware.com Mon Feb 23 18:39:32 2015 From: shawn.waldon at kitware.com (Shawn Waldon) Date: Mon, 23 Feb 2015 18:39:32 -0500 Subject: [vtk-developers] 64 Bit Integer Data Array Message-ID: Hello all, What is the correct type of vtkDataArray to use for 64 bit integers? I have a file format that I am reading in that has a 64 bit id and while there seem to be several options, the right one isn't clear. I would like something that just works across platforms and various builds of VTK without #ifdefs throughout the code. -vtkLongLongArray - long long is at least 64 bits, so I am using this one right now. But it apparently doesn't exist in a vtk builds? -vtk__Int64Array - this sounds more like what I want since I don't care about the underlying datatype, but this also does not always exist. -directly use the template data array with int64_t? Will other VTK code handle this well since it is not one of the built-in data array types? -vtkLongArray? - On some platforms long is a 64 bit type, but this is platform dependant. However, it is what ParaView defaults to when loading data defined as Int64 from a .vtu file on my machine. Thank you, Shawn -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Mon Feb 23 18:45:32 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Mon, 23 Feb 2015 16:45:32 -0700 Subject: [vtk-developers] 64 Bit Integer Data Array In-Reply-To: References: Message-ID: vtkTypeInt64Array with type vtkTypeInt64 Always 64-bit, and always built on every platform. On Mon, Feb 23, 2015 at 4:39 PM, Shawn Waldon wrote: > Hello all, > > What is the correct type of vtkDataArray to use for 64 bit integers? I > have a file format that I am reading in that has a 64 bit id and while > there seem to be several options, the right one isn't clear. I would like > something that just works across platforms and various builds of VTK > without #ifdefs throughout the code. > > -vtkLongLongArray - long long is at least 64 bits, so I am using this one > right now. But it apparently doesn't exist in a vtk builds? > > -vtk__Int64Array - this sounds more like what I want since I don't care > about the underlying datatype, but this also does not always exist. > > -directly use the template data array with int64_t? Will other VTK code > handle this well since it is not one of the built-in data array types? > > -vtkLongArray? - On some platforms long is a 64 bit type, but this is > platform dependant. However, it is what ParaView defaults to when loading > data defined as Int64 from a .vtu file on my machine. > > Thank you, > Shawn > -------------- next part -------------- An HTML attachment was scrubbed... URL: From DLRdave at aol.com Mon Feb 23 19:41:22 2015 From: DLRdave at aol.com (David Cole) Date: Mon, 23 Feb 2015 19:41:22 -0500 Subject: [vtk-developers] Python wrapped void pointers In-Reply-To: References: Message-ID: So can callers now pass "_f02100_p_void" instead of the formerly accepted "_f02100_void_p" ... ? On Mon, Feb 23, 2015 at 5:09 PM, David Gobbi wrote: > This is a result of one of the bugs that I fixed during last year's > hackathon: > > http://www.vtk.org/Bug/view.php?id=12098 > http://review.source.kitware.com/#/c/17351/ > > > On Mon, Feb 23, 2015 at 2:59 PM, David Lonie > wrote: >> >> A project I work on is facing a regression related to the >> vtkRenderWindow::SetDisplayId(void*) method. >> >> It apparently used to work by passing a string similar to: >> >> '_f02100_void_p' >> >> as the argument, but now they're getting TypeErrors: >> >> TypeError('SetDisplayId argument 1: value is _f02100_void_p, required type >> is p_void',) >> >> For context, the failure is happening here: >> >> >> https://github.com/UV-CDAT/VisTrails/blob/cb2220ba88e957006c3dcc098ef29c5c08461117/vistrails/packages/vtk/vtkcell.py#L356 >> >> It looks like the python wrappers might have changed how they handle the >> void* type -- does anyone have an idea what's going on here? >> >> Thanks, >> Dave > > > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > From sean at rogue-research.com Mon Feb 23 19:56:00 2015 From: sean at rogue-research.com (Sean McBride) Date: Mon, 23 Feb 2015 19:56:00 -0500 Subject: [vtk-developers] 64 Bit Integer Data Array In-Reply-To: References: Message-ID: <20150224005600.1976953912@mail.rogue-research.com> I guess vtk__Int64Array predates vtkTypeInt64Array? Is there any reason to not deprecate it at this point? Sean On Mon, 23 Feb 2015 16:45:32 -0700, David Gobbi said: >vtkTypeInt64Array with type vtkTypeInt64 > >Always 64-bit, and always built on every platform. > >On Mon, Feb 23, 2015 at 4:39 PM, Shawn Waldon >wrote: > >> Hello all, >> >> What is the correct type of vtkDataArray to use for 64 bit integers? I >> have a file format that I am reading in that has a 64 bit id and while >> there seem to be several options, the right one isn't clear. I would like >> something that just works across platforms and various builds of VTK >> without #ifdefs throughout the code. >> >> -vtkLongLongArray - long long is at least 64 bits, so I am using this one >> right now. But it apparently doesn't exist in a vtk builds? >> >> -vtk__Int64Array - this sounds more like what I want since I don't care >> about the underlying datatype, but this also does not always exist. >> >> -directly use the template data array with int64_t? Will other VTK code >> handle this well since it is not one of the built-in data array types? >> >> -vtkLongArray? - On some platforms long is a 64 bit type, but this is >> platform dependant. However, it is what ParaView defaults to when loading >> data defined as Int64 from a .vtu file on my machine. From david.gobbi at gmail.com Mon Feb 23 20:07:25 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Mon, 23 Feb 2015 18:07:25 -0700 Subject: [vtk-developers] Python wrapped void pointers In-Reply-To: References: Message-ID: On Mon, Feb 23, 2015 at 5:41 PM, David Cole wrote: > So can callers now pass "_f02100_p_void" instead of the formerly > accepted "_f02100_void_p" ... ? Yes. The whole point of the "void_p" was compatibility with SWIG, but SWIG itself switched to "p_void" many years ago. VTK is just catching up... -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Mon Feb 23 20:13:46 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Mon, 23 Feb 2015 18:13:46 -0700 Subject: [vtk-developers] 64 Bit Integer Data Array In-Reply-To: <20150224005600.1976953912@mail.rogue-research.com> References: <20150224005600.1976953912@mail.rogue-research.com> Message-ID: On Mon, Feb 23, 2015 at 5:56 PM, Sean McBride wrote: > I guess vtk__Int64Array predates vtkTypeInt64Array? Is there any reason > to not deprecate it at this point? > I believe that they were both added to VTK at the same time. On UNIX, vtkTypeInt64Array is a sublcass of vtkLongLongArray. On Win32, vtkTypeInt64Array is (was?) a subclass of vtk__Int64Array. Of course, this was due to the fact that "long long" was not a valid type on many of the compilers that VTK supported, hence 64-bit arrays were handled with __int64 or a 64-bit long wherever "long long" was unavailable. - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.lonie at kitware.com Tue Feb 24 08:59:37 2015 From: david.lonie at kitware.com (David Lonie) Date: Tue, 24 Feb 2015 08:59:37 -0500 Subject: [vtk-developers] Python wrapped void pointers In-Reply-To: References: Message-ID: On Mon, Feb 23, 2015 at 8:07 PM, David Gobbi wrote: > On Mon, Feb 23, 2015 at 5:41 PM, David Cole wrote: > >> So can callers now pass "_f02100_p_void" instead of the formerly >> accepted "_f02100_void_p" ... ? > > > Yes. The whole point of the "void_p" was compatibility with SWIG, but SWIG > itself switched to "p_void" many years ago. VTK is just catching up... > Ok, thanks for the quick answer! I'll pass this along. Dave -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Tue Feb 24 13:32:58 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Tue, 24 Feb 2015 13:32:58 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150219214749.727843881@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: How goes the testing Sean? Have you come across anything serious yet? David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Thu, Feb 19, 2015 at 4:47 PM, Sean McBride wrote: > On Thu, 19 Feb 2015 16:36:12 -0500, Marcus D. Hanwell said: > > >It isn't that short > > I guess we'll agree to disagree. :) > > >Why make more than one RC if there are no > >critical flaws discovered? > > No disagreement there. But if the 1 rc period lasts only 4 business days > (it was announced late Monday), there's *insufficient time* to look for > flaws. > > VTK 6.1 was released 13 months ago. Why the rush to declare 6.2 final? > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From luis.vieira at vektore.com Wed Feb 25 10:23:42 2015 From: luis.vieira at vektore.com (Luis Vieira) Date: Wed, 25 Feb 2015 12:23:42 -0300 Subject: [vtk-developers] vtkSphereSource Wireframe Message-ID: <54EDE8FE.8050606@vektore.com> Hello vtkusers, I have been developing a stereographic representation using vtkSphereSource. Everything is ok; however, I am struggling to show only the external latitude and longitude edges. I want turn off the internal edges when I apply a clip and look inside the sphere without edges. Have to show only on the external surface. Any ideas? Thank you very much. From sean at rogue-research.com Wed Feb 25 11:52:55 2015 From: sean at rogue-research.com (Sean McBride) Date: Wed, 25 Feb 2015 11:52:55 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> Message-ID: <20150225165255.967467100@mail.rogue-research.com> On Tue, 24 Feb 2015 13:32:58 -0500, David E DeMarle said: >How goes the testing Sean? >Have you come across anything serious yet? No serious issues found. Maybe found a mistake I made though: I think some of the vtkErrorMacros I added should maybe not be there, as I see them running my app where I don't believe I saw them before. Then again, they are maybe flagging a real issue that was not flagged before. Haven't had time to dig... Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From dave.demarle at kitware.com Wed Feb 25 11:59:26 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Wed, 25 Feb 2015 11:59:26 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150225165255.967467100@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> Message-ID: Keep us informed. If it needs fixing we're still making fixes and taking patches based off anywhere upstream of the release branch point and tested in master. We'll make a decision on turning the release branch into either rc2 or final on Monday some time. thanks! David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Wed, Feb 25, 2015 at 11:52 AM, Sean McBride wrote: > On Tue, 24 Feb 2015 13:32:58 -0500, David E DeMarle said: > > >How goes the testing Sean? > >Have you come across anything serious yet? > > No serious issues found. > > Maybe found a mistake I made though: > > > > I think some of the vtkErrorMacros I added should maybe not be there, as I > see them running my app where I don't believe I saw them before. Then > again, they are maybe flagging a real issue that was not flagged before. > Haven't had time to dig... > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Wed Feb 25 12:42:40 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Wed, 25 Feb 2015 10:42:40 -0700 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> Message-ID: It looks like the added vtkErrorMacro() will be called whenever Squeeze() is called on an empty array, which is a common occurrence and should definitely not generate an error. I think the patch should be reverted (though I was the reviewer... oops!). - David On Wed, Feb 25, 2015 at 9:59 AM, David E DeMarle wrote: > Keep us informed. If it needs fixing we're still making fixes and taking > patches based off anywhere upstream of the release branch point and tested > in master. > > We'll make a decision on turning the release branch into either rc2 or > final on Monday some time. > > On Wed, Feb 25, 2015 at 11:52 AM, Sean McBride > wrote: >> >> >> No serious issues found. >> >> Maybe found a mistake I made though: >> >> >> >> I think some of the vtkErrorMacros I added should maybe not be there, as >> I see them running my app where I don't believe I saw them before. Then >> again, they are maybe flagging a real issue that was not flagged before. >> Haven't had time to dig... > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Wed Feb 25 13:15:15 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Wed, 25 Feb 2015 11:15:15 -0700 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> Message-ID: As a follow up, here is some vtkpython code that generates a spurious error: a = vtkVariantArray() a.SetNumberOfValues(1) a.SetNumberOfValues(0) a.Squeeze() ERROR: In Common/Core/vtkVariantArray.cxx, line 730 vtkVariantArray (0x7fe1bbca8cb0): Memory size must be positive On Wed, Feb 25, 2015 at 10:42 AM, David Gobbi wrote: > It looks like the added vtkErrorMacro() will be called whenever Squeeze() > is called on an empty array, which is a common occurrence and should > definitely not generate an error. > > I think the patch should be reverted (though I was the reviewer... oops!). > > - David > > > On Wed, Feb 25, 2015 at 9:59 AM, David E DeMarle > wrote: > >> Keep us informed. If it needs fixing we're still making fixes and taking >> patches based off anywhere upstream of the release branch point and tested >> in master. >> >> We'll make a decision on turning the release branch into either rc2 or >> final on Monday some time. >> >> On Wed, Feb 25, 2015 at 11:52 AM, Sean McBride >> wrote: >>> >>> >>> No serious issues found. >>> >>> Maybe found a mistake I made though: >>> >>> >>> >>> I think some of the vtkErrorMacros I added should maybe not be there, as >>> I see them running my app where I don't believe I saw them before. Then >>> again, they are maybe flagging a real issue that was not flagged before. >>> Haven't had time to dig... >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From sean at rogue-research.com Wed Feb 25 17:12:06 2015 From: sean at rogue-research.com (Sean McBride) Date: Wed, 25 Feb 2015 17:12:06 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> Message-ID: <20150225221206.951172400@mail.rogue-research.com> On Wed, 25 Feb 2015 10:42:40 -0700, David Gobbi said: >I think the patch should be reverted (though I was the reviewer... oops!). Agreed. Is there any git magic to be done, or do I just manually remove the stuff I added and commit as usual? Cheers, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From dave.demarle at kitware.com Wed Feb 25 17:18:51 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Wed, 25 Feb 2015 17:18:51 -0500 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150225221206.951172400@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> <20150225221206.951172400@mail.rogue-research.com> Message-ID: I'ld keep it simple and go with a new commit. Started off of release branch or or higher. That way, once merged into master I can also merge the same commit into release. David E DeMarle Kitware, Inc. R&D Engineer 21 Corporate Drive Clifton Park, NY 12065-8662 Phone: 518-881-4909 On Wed, Feb 25, 2015 at 5:12 PM, Sean McBride wrote: > On Wed, 25 Feb 2015 10:42:40 -0700, David Gobbi said: > > >I think the patch should be reverted (though I was the reviewer... oops!). > > Agreed. Is there any git magic to be done, or do I just manually remove > the stuff I added and commit as usual? > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.gobbi at gmail.com Wed Feb 25 17:21:09 2015 From: david.gobbi at gmail.com (David Gobbi) Date: Wed, 25 Feb 2015 15:21:09 -0700 Subject: [vtk-developers] announce: vtk 6.2.0 release candidate 1 is ready In-Reply-To: <20150225221206.951172400@mail.rogue-research.com> References: <20150219200444.933334771@mail.rogue-research.com> <20150219212511.1505792326@mail.rogue-research.com> <20150219214749.727843881@mail.rogue-research.com> <20150225165255.967467100@mail.rogue-research.com> <20150225221206.951172400@mail.rogue-research.com> Message-ID: Something like this: git checkout release git checkout -b revert-array-error git revert 42815afe git gerrit-push On Wed, Feb 25, 2015 at 3:12 PM, Sean McBride wrote: > On Wed, 25 Feb 2015 10:42:40 -0700, David Gobbi said: > > >I think the patch should be reverted (though I was the reviewer... oops!). > > Agreed. Is there any git magic to be done, or do I just manually remove > the stuff I added and commit as usual? > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng sean at rogue-research.com > Rogue Research www.rogue-research.com > Mac Software Developer Montr?al, Qu?bec, Canada > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.amaclean at gmail.com Thu Feb 26 00:21:55 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Thu, 26 Feb 2015 16:21:55 +1100 Subject: [vtk-developers] A note regarding: std::min and std::max. Message-ID: For those of you using std::min and std::max in VTK. Visual Studio 2013 requires the header. I think (but don't quote me on this) that the older MSVC compilers used to automatically include a MSVC specific header called xutility where these were defined. With respect to other compilers e.g. gcc, it is possible that one of the other includes is adding the algorithm header in. I have recently picked up two tests that failed in VS 2013 but were passing in GCC, Clang and earlier versions of MSVC.. Would it be possible to set up one of the CDash machines to use VS2013? I am not testing the parallel implementations so there could be problems there that I haven't detected. However if someone can walk me through setting up MPI on a laptop with VS2013 Community Edition I would be glad to test the Parallel implementations. Regards Andrew -- ___________________________________________ Andrew J. P. Maclean ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From shawn.waldon at kitware.com Thu Feb 26 10:38:19 2015 From: shawn.waldon at kitware.com (Shawn Waldon) Date: Thu, 26 Feb 2015 10:38:19 -0500 Subject: [vtk-developers] 64 Bit Integer Data Array In-Reply-To: References: <20150224005600.1976953912@mail.rogue-research.com> Message-ID: I have pushed a doc branch[1] to gerrit to attempt to make this clearer. Did I miss any types I should have updated? Shawn [1]: http://review.source.kitware.com/#/t/5507 On Mon, Feb 23, 2015 at 8:13 PM, David Gobbi wrote: > On Mon, Feb 23, 2015 at 5:56 PM, Sean McBride > wrote: > >> I guess vtk__Int64Array predates vtkTypeInt64Array? Is there any reason >> to not deprecate it at this point? >> > > I believe that they were both added to VTK at the same time. > > On UNIX, vtkTypeInt64Array is a sublcass of vtkLongLongArray. > On Win32, vtkTypeInt64Array is (was?) a subclass of vtk__Int64Array. > > Of course, this was due to the fact that "long long" was not a valid type > on many of the compilers that VTK supported, hence 64-bit arrays were > handled with __int64 or a 64-bit long wherever "long long" was unavailable. > > - David > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dave.demarle at kitware.com Thu Feb 26 11:48:16 2015 From: dave.demarle at kitware.com (David E DeMarle) Date: Thu, 26 Feb 2015 11:48:16 -0500 Subject: [vtk-developers] A note regarding: std::min and std::max. In-Reply-To: References: Message-ID: > > > Would it be possible to set up one of the CDash machines to use VS2013? I > am not testing the parallel implementations so there could be problems > there that I haven't detected. > > Already on my todo list. Pester me if if doesn't get done soon. -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.amaclean at gmail.com Thu Feb 26 14:57:53 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Fri, 27 Feb 2015 06:57:53 +1100 Subject: [vtk-developers] A note regarding: std::min and std::max. In-Reply-To: References: Message-ID: Will do, thanks David. On 27/02/2015 3:48 AM, "David E DeMarle" wrote: > >> Would it be possible to set up one of the CDash machines to use VS2013? I >> am not testing the parallel implementations so there could be problems >> there that I haven't detected. >> >> > Already on my todo list. Pester me if if doesn't get done soon. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From linyufly at gmail.com Thu Feb 26 23:29:16 2015 From: linyufly at gmail.com (Mingcheng Chen) Date: Thu, 26 Feb 2015 22:29:16 -0600 Subject: [vtk-developers] [ITK Community] What is the algorithm of itk::WatershedImageFilter? Message-ID: Hello, I know it may be a hierarchical watershed method, but I wish to know what paper it is. Thanks a lot! -Mingcheng -- Research Assistant in Graphics Group University of Illinois at Urbana-Champaign http://mingchengchen.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From jhlegarreta at vicomtech.org Fri Feb 27 03:03:27 2015 From: jhlegarreta at vicomtech.org (Jon Haitz Legarreta) Date: Fri, 27 Feb 2015 09:03:27 +0100 Subject: [vtk-developers] [ITK Community] What is the algorithm of itk::WatershedImageFilter? In-Reply-To: References: Message-ID: Dear Mingcheng, having a look at the ITK Software Guide [1], section 4.2, may be helpful on this. The papers by Eberly [2] and Koenderink et al. [3] (reference 31 on the ITK Software Guide) may have been an inspiration for the implementation. HTH, JON HAITZ [1] http://www.itk.org/ItkSoftwareGuide [2] David Eberly. Ridges in Image and Data Analysis . Kluwer Academic Publishers, Dordrecht, 1996 [3] J. Koenderink and A. van Doorn. Local features of smooth shapes: Ridges and courses. SPIE Proc. Geometric Methods in Computer Vision II, 2031:2?13, 1993 P.S.: if it is strictly an ITK question, it'd be advisable to address the question exclusively to the ITK community. This answer is addressed to all original recipients just for the records. On 27 February 2015 at 05:29, Mingcheng Chen wrote: > Hello, > > I know it may be a hierarchical watershed method, but I wish to know what > paper it is. > > Thanks a lot! > > -Mingcheng > > -- > Research Assistant in Graphics Group > University of Illinois at Urbana-Champaign > http://mingchengchen.org > > _______________________________________________ > Powered by www.kitware.com > > Visit other Kitware open-source projects at > http://www.kitware.com/opensource/opensource.html > > Search the list archives at: http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From c.kolb at dkfz-heidelberg.de Fri Feb 27 08:59:51 2015 From: c.kolb at dkfz-heidelberg.de (Christoph Kolb) Date: Fri, 27 Feb 2015 14:59:51 +0100 Subject: [vtk-developers] Delayed mouse interaction due to event-loop bug in Qt5 In-Reply-To: <54E8E057.1060409@dkfz-heidelberg.de> References: <54E8E057.1060409@dkfz-heidelberg.de> Message-ID: <54F07857.4060302@dkfz-heidelberg.de> I am not sure if the Qt5 bug is ever going to be fixed, because the report (and patch) is very old. It could even be intentional... In the worst case, it might be necessary to compress the events in VTK. Or to prioritize them somehow. But I'm still very curious why this issue is showing only on some machines. Is it possible that the rendering is just slower in my case? I have compared my test execution times with those of https://open.cdash.org/buildSummary.php?buildid=3711375 and they seem to be equal. Is the described problem also showing there? Regards Christoph On 02/21/2015 08:45 PM, Christoph Kolb wrote: > Dear vtk developers, > > when I am using the QVTKWidget in VTK 6.1 with Qt5, the mouse > interactions are severely delayed. I can reproduce this issue by > starting the SimpleView example and rotating the text. > > I think that this might be caused by a bug in Qt5, where the mousemove > events are flooding the event queue: > https://bugreports.qt.io/browse/QTBUG-40889 > > However, other Linux distributions and Windows seem to be unnaffected by > this. I have only observed it with Archlinux (Qt 5.4.0) so far. > Does anyone else have this problem, or do you know what else i could check? > > Thank you! > > Regards > Christoph > _______________________________________________ > Powered bywww.kitware.com > > Visit other Kitware open-source projects athttp://www.kitware.com/opensource/opensource.html > > Search the list archives at:http://markmail.org/search/?q=vtk-developers > > Follow this link to subscribe/unsubscribe: > http://public.kitware.com/mailman/listinfo/vtk-developers > From linyufly at gmail.com Fri Feb 27 10:23:31 2015 From: linyufly at gmail.com (Mingcheng Chen) Date: Fri, 27 Feb 2015 09:23:31 -0600 Subject: [vtk-developers] [ITK Community] What is the algorithm of itk::WatershedImageFilter? In-Reply-To: References: Message-ID: Hi Jon, Thank you very much for your response! I am sorry for posting it everywhere. -Mingcheng On Fri, Feb 27, 2015 at 2:03 AM, Jon Haitz Legarreta < jhlegarreta at vicomtech.org> wrote: > Dear Mingcheng, > having a look at the ITK Software Guide [1], section 4.2, may be helpful > on this. > The papers by Eberly [2] and Koenderink et al. [3] (reference 31 on the > ITK Software Guide) may have been an inspiration for the implementation. > > HTH, > JON HAITZ > > [1] http://www.itk.org/ItkSoftwareGuide > [2] David Eberly. Ridges in Image and Data Analysis . Kluwer Academic > Publishers, Dordrecht, 1996 > [3] J. Koenderink and A. van Doorn. Local features of smooth shapes: > Ridges and courses. SPIE Proc. Geometric Methods in Computer Vision II, > 2031:2?13, 1993 > > P.S.: if it is strictly an ITK question, it'd be advisable to address the > question exclusively to the ITK community. This answer is addressed to all > original recipients just for the records. > > > > On 27 February 2015 at 05:29, Mingcheng Chen wrote: > >> Hello, >> >> I know it may be a hierarchical watershed method, but I wish to know what >> paper it is. >> >> Thanks a lot! >> >> -Mingcheng >> >> -- >> Research Assistant in Graphics Group >> University of Illinois at Urbana-Champaign >> http://mingchengchen.org >> >> _______________________________________________ >> Powered by www.kitware.com >> >> Visit other Kitware open-source projects at >> http://www.kitware.com/opensource/opensource.html >> >> Search the list archives at: http://markmail.org/search/?q=vtk-developers >> >> Follow this link to subscribe/unsubscribe: >> http://public.kitware.com/mailman/listinfo/vtk-developers >> >> >> > -- Research Assistant in Graphics Group University of Illinois at Urbana-Champaign http://mingchengchen.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben.boeckel at kitware.com Fri Feb 27 10:32:17 2015 From: ben.boeckel at kitware.com (Ben Boeckel) Date: Fri, 27 Feb 2015 10:32:17 -0500 Subject: [vtk-developers] A note regarding: std::min and std::max. In-Reply-To: References: Message-ID: <20150227153217.GB6233@megas.kitwarein.com> On Thu, Feb 26, 2015 at 16:21:55 +1100, Andrew Maclean wrote: > Would it be possible to set up one of the CDash machines to use VS2013? I > am not testing the parallel implementations so there could be problems > there that I haven't detected. nemesis runs VS2013: https://open.cdash.org/viewBuildError.php?buildid=3711838 > However if someone can walk me through setting up MPI on a laptop with > VS2013 Community Edition I would be glad to test the Parallel > implementations. MSMPI is here: https://msdn.microsoft.com/en-us/library/bb524831%28v=vs.85%29.aspx IIRC, it was just an install-and-go setup (now that VTK's FindMPI can find it ;) ). --Ben From andrew.amaclean at gmail.com Fri Feb 27 14:14:32 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Sat, 28 Feb 2015 06:14:32 +1100 Subject: [vtk-developers] A note regarding: std::min and std::max. In-Reply-To: <20150227153217.GB6233@megas.kitwarein.com> References: <20150227153217.GB6233@megas.kitwarein.com> Message-ID: Thats really great, thanks! Andrew. On 28/02/2015 2:32 AM, "Ben Boeckel" wrote: > On Thu, Feb 26, 2015 at 16:21:55 +1100, Andrew Maclean wrote: > > Would it be possible to set up one of the CDash machines to use VS2013? I > > am not testing the parallel implementations so there could be problems > > there that I haven't detected. > > nemesis runs VS2013: > > https://open.cdash.org/viewBuildError.php?buildid=3711838 > > > However if someone can walk me through setting up MPI on a laptop with > > VS2013 Community Edition I would be glad to test the Parallel > > implementations. > > MSMPI is here: > > https://msdn.microsoft.com/en-us/library/bb524831%28v=vs.85%29.aspx > > IIRC, it was just an install-and-go setup (now that VTK's FindMPI can > find it ;) ). > > --Ben > -------------- next part -------------- An HTML attachment was scrubbed... URL: From andrew.amaclean at gmail.com Fri Feb 27 14:43:34 2015 From: andrew.amaclean at gmail.com (Andrew Maclean) Date: Sat, 28 Feb 2015 06:43:34 +1100 Subject: [vtk-developers] For those testing VTK 6.2.0 RC 1 Message-ID: Just letting those testing VTK 6.2 RC1 that issues with the following files have been fixed: Filters/SMP/Testing/Cxx/TestThreadedSynchronizedTemplatesCutter3D.cxx Rendering/FreeType/Testing/Cxx/BenchmarkFreeTypeRendering.cxx Examples/Tutorial/Step6/Cxx/CMakeLists.txt See: http://review.source.kitware.com/#/c/19326/ http://review.source.kitware.com/#/c/19306/ http://review.source.kitware.com/#/c/19278/ Andrew -------------- next part -------------- An HTML attachment was scrubbed... URL: