Skip to content

Conversation

@adamrankin
Copy link

Integrating VTK9 generation fixes (and others)

jcfr and others added 30 commits November 29, 2017 19:27
List of changes:

$ git shortlog 06379c2..fc16632 --no-merges
Hina Shah (1):
      COMP Adding VTK_OVERRIDE keyword to the overrident functions

Jean-Christophe Fillion-Robin (6):
      COMP: vtkIGTLToMRMLTrajectory: Fix deprecated declaration warning
      COMP: vtkMRMLIGTLConnectorNode: Fix sign compare warning
      STYLE: Fix unused variable warnings
      ENH: Add support for Qt5
      COMP: Update minimum required CMake version to match Slicer requirements
      COMP: Improve support for building as a "Slicer Remote module"

Longquan Chen (7):
      EHN: Extend the module to use meta info to specify the mrmlnode.
      EHN: 1: Image message working now regarding the user specified mrmlnode from meta info. 2: GUI element improved
      EHN: Refactoring in progress
      EHN: Sending version2 openigtlink message in progress
      EHN: 1.Converters are created for each connector. simultaneously IGTLToMRML convension for two  devices with the same types is now possible.2. A converter factory class is added to make the code cleaner.
      EHN: improving the handling of version 2 igtlmessage without meta information
      EHN: Updated according to the new API of polydatamessage

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26661 3bd1e089-480b-0410-8dfb-8563597acbee
In addition, a feature is added to warn the user about issues with DICOM export. Many users have asked why they could not export a segmentation, and the answer always was that because it was not under a patient+study. A check is now done when clicking Export to DICOM in subject hierarchy, and a question is shown asking whether default patient and study should be created and referenced datasets (by DICOM or MRML) moved in it for export. If the user chooses yes then those defaults are created, unless a referenced item is in a study, because then the exported node and all other references are moved under that.

With MRML references involved in SH, in addition to DICOM references, MRML references are also highlighted.
Highlight added also for referencing items with a slightly different color.

Fixes https://issues.slicer.org/view.php?id=4313

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26662 3bd1e089-480b-0410-8dfb-8563597acbee
List of changes:

git shortlog 66db91f..4f25a8a --no-merges
Hans J. Johnson (5):
      Moved documentation to the public wiki page
      DOC:  Added usecase documentation for super resolution.
      ENH: Add the ability to deface using BAW results.
      ENH: Improve cmake to force new behaviors with CMake 3.5
      COMP: Improve compilation behaviors for CMake > 3.8.2

Isaiah Norton (1):
      BUG: fix FSLToNrrd mode from Slicer

Jean-Christophe Fillion-Robin (3):
      COMP: Export BRAINSCommonLib target for build tree
      COMP: Consistently exclude test if BRAINSTools_DISABLE_TESTING is ON
      COMP: Support integration with Slicer setting LAUNCH_EXE using SEM_LAUNCH_COMMAND

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26663 3bd1e089-480b-0410-8dfb-8563597acbee
It turns out that the version Slicer/OpenIGTLinkIF@fc16632 integrated in
r26661 also required a newer version of OpenIGTLink the was only
supporting C++11.

Waiting the Slicer trunk transition to C++11, this commit reverts to an
older version of OpenIGTLinkIF including the following new changes:

$ git shortlog 06379c2..0266aa2 --no-merges
Jean-Christophe Fillion-Robin (2):
      COMP: Update minimum required CMake version to match Slicer requirements
      COMP: Improve support for building as a "Slicer Remote module"

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26664 3bd1e089-480b-0410-8dfb-8563597acbee
…ectories

This commit simplifies handling of remote module, remove extension and
extension directories.

It also updates BRAINSTools to the expected hash omitted in r26663:

$ git shortlog 42ac3c3..4f25a8a --no-merges
Jean-Christophe Fillion-Robin (1):
      COMP: Support integration with Slicer setting LAUNCH_EXE using SEM_LAUNCH_COMMAND

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26665 3bd1e089-480b-0410-8dfb-8563597acbee
The function provides easy access to voxels of a segment using numpy.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26666 3bd1e089-480b-0410-8dfb-8563597acbee
This commit ensures that remote modules are not included if associated
options are disabled.

For example, this will avoid re-build error if python support is
disabled after CompareVolumes was first enabled and built.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26668 3bd1e089-480b-0410-8dfb-8563597acbee
This commit removes the ROi module originally disabled in 2011
in r15893 (ENH: Disable ROI module as vtkMRMLROINode is obsolete)

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26669 3bd1e089-480b-0410-8dfb-8563597acbee
Now it is possible to plot a single column of a table.
Also fixed a couple of minor issues with getting/setting column names.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26671 3bd1e089-480b-0410-8dfb-8563597acbee
DICOM scalar volume importer creates acquisition transform to compensate missing slices or tilted acquisition plane.
When multiple such volumes were loaded, it used the same node name "Acquisition Transform" for all of them, which caused all the nodes overwritten by the last one.

Fixed by generating unique node names for each acquisition transform.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26672 3bd1e089-480b-0410-8dfb-8563597acbee
It is now possible to export a video that is playing the animation forward and backwards.
Also, the animation can be repeated multiple times, which allows playing short videos more conveniently on players where looped playback is restricted (e.g., YouTube and some media players).

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26673 3bd1e089-480b-0410-8dfb-8563597acbee
ExportAllSegmentsToModelHierarchy returned with True but did not export anything as ExportSegmentsToModelHierarchy method did not export all segments when the received segment list was empty.

Fixed it to work the same way as labelmap exporter: when segment list is empty then all segments are exported.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26674 3bd1e089-480b-0410-8dfb-8563597acbee
Thanks: Isaiah Norton <inorton@bwh.harvard.edu>

References:
* BRAINSia/BRAINSTools#356
* https://discourse.slicer.org/t/dti-nifti-data-import-via-dwi-convert-module/1460/4

List of changes:

$ git shortlog 42ac3c3..c1289e6 --no-merges
Hans J. Johnson (1):
      ENH: Need to add allowLossy for file types other than short.

Isaiah Norton (2):
      BUG: ITK is row-major; fix NRRD output with mismatched (x,y) axis sizes
      BUG: FSL converter should check reader status

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26676 3bd1e089-480b-0410-8dfb-8563597acbee
…for Qt5

This commit fixes the following error reported when loading the DICOM
module:

File "/path/to/Slicer-build/lib/Slicer-4.9/qt-scripted-modules/DICOMLib/DICOMWidgets.py", line 519, in promptForDatabaseDirectory
    documentsLocation = qt.QDesktopServices.DocumentsLocation
AttributeError: QDesktopServices has no attribute named 'DocumentsLocation'

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26678 3bd1e089-480b-0410-8dfb-8563597acbee
Also reduced an error message to debug in a recent subject hierarchy function

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26679 3bd1e089-480b-0410-8dfb-8563597acbee
The call to run that is made in cli.runSync does not pass along the update_display parameter, even though runSync accepts it as a parameter. This fix passes the value of update_display to the run method.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26680 3bd1e089-480b-0410-8dfb-8563597acbee
Remove incorrect "PYTHON_EXECUTABLE" argument. This has never been a
problem because the variable was never set in the current scope.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26681 3bd1e089-480b-0410-8dfb-8563597acbee
This commit teaches "ExternalProject_Execute" function to understand
the variable EP_EXECUTE_DISABLE_CAPTURE_OUTPUTS. If set to 1, output
is directly reported and not captured into files.

This is particularly useful when debugging issue with building and installing
of python packages.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26682 3bd1e089-480b-0410-8dfb-8563597acbee
This commit updates the build system so that it:
* ensures that python files are packaged
* an error is reported if "PYTHON_DIR" variable is not set in SlicerBlockInstallPython

Symptom leading to this fix was the following error reported when
starting the application:

  ImportError: No module named site

Co-authored-by: Dzenan Zukic <dzenan.zukic@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26683 3bd1e089-480b-0410-8dfb-8563597acbee
…r.sln

Since the top-level solution file is always named "Slicer.sln", this commit
removes the use of Slicer_MAIN_PROJECT_APPLICATION_NAME variable.

Suggested-by: Dzenan Zukic <dzenan.zukic@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26684 3bd1e089-480b-0410-8dfb-8563597acbee
This change includes a fix so that exported grid
transforms are vtkOrientedGridTransform and can
be saved in .h5 format.

From: Steve Pieper <pieper@isomics.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26685 3bd1e089-480b-0410-8dfb-8563597acbee
User may have carefully set up view, so resetting it when switching to Segment Editor module would cause inconvenience.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26686 3bd1e089-480b-0410-8dfb-8563597acbee
In case special characters (e.g. in the patient name) ended up being saved in the MRML file, loading the MRML failed due to an XML reading error. These special characters were valid in Slicer, as they were produced by slicer.util.toVTKString. If the character set is specified in the MRML file, then loading is successful.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26687 3bd1e089-480b-0410-8dfb-8563597acbee
This commit ensures that packages installed in directory like "~/.local/"
are not imported before Slicer own packages.

References:
* https://docs.python.org/2/library/site.html#site.getusersitepackages
* https://www.python.org/dev/peps/pep-0370/

Reported-by: Dzenan Zukic <dzenan.zukic@kitware.com>
Reported-by: Forrest Li <forrest.li@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26688 3bd1e089-480b-0410-8dfb-8563597acbee
…AGS_INIT

This allows use of the module in extension building python packages.

Indeed, since regular CMAKE_(C|CXX)_FLAGS can contain flags incompatible
with python packages (e.g "-Wl,--no-undefined"), this commit removes
usage of the variable to initialize CFLAGS and CXXFLAGS variables.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26689 3bd1e089-480b-0410-8dfb-8563597acbee
…on build dir

It also adds "UserWarning:" as an error exception.

See AIM-Harvard/SlicerRadiomics#23 (comment)

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26690 3bd1e089-480b-0410-8dfb-8563597acbee
jcfr and others added 30 commits December 13, 2017 17:53
…led.

Reported-by: Dzenan Zukic <dzenan.zukic@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26717 3bd1e089-480b-0410-8dfb-8563597acbee
This commit updates the build system to ensure vtk can be imported from
a build or an install tree.

It fixes a regression introduced in r26697 where a newer version VTK9
is used. The Slicer build system is updated to adapt to changes
introduced the commit Kitware/VTK@867d93c (Cleanup VTK's Python package.).

It also updates VTK9 to include the following patch:

$ git shortlog Slicer/VTK@27c57a1..Slicer/VTK@887c506 --no-merges
Jean-Christophe Fillion-Robin (1):
      cmake: re-associate RuntimeLibraries component with python install rules

also contributed upstream as https://gitlab.kitware.com/vtk/vtk/merge_requests/3708

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26719 3bd1e089-480b-0410-8dfb-8563597acbee
Added options to application settings:

- Acquisition geometry regularization: enable/disable adding regularization transform.
  A default option is added for those who don't know/don't care. For now, it is disabled by default.
  When an issue is found then Slicer displays a warning and the user may enable regularization.
  In the future we may add a "resample" option as well, to not just apply a transform but also use
  it to resample the volume.

- Allow loading subseries by time: if disabled (default) then loadables are not offered by content or trigger time.
  This reduces clutter and increases chance that users try alternative loadables, grouped for example by image orientation.

Minor improvements:
- Simpler node names are generated for subseries. Instead of including actual image orientation and timestamps (which are long, cryptic
strings), we use just indexes (imageOrientationPatient 1, imageOrientationPatient 2, ...).
- Acquisition transform is added under the same subject hierarchy folder as the volume.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26720 3bd1e089-480b-0410-8dfb-8563597acbee
…ake module

The following warning was reported when directly invoking the module
from the command line.

For example:

$ cmake \
  -DCMAKE_INSTALL_PREFIX:PATH=/path/to/SlicerVMTK-build/inner-build/_CPack_Packages/Darwin/TGZ/26720-macosx-amd64-SlicerVMTK-git0efbf09-2017-12-18 \
  -P SlicerExtensionBundle/SlicerExtensionCPackBundleFixup.cmake

CMake Warning (dev) in /path/to/Slicer/CMake/BundleUtilitiesWithRPath.cmake:
  Policy CMP0011 is not set: Included scripts do automatic cmake_policy PUSH
  and POP.  Run "cmake --help-policy CMP0011" for policy details.  Use the
  cmake_policy command to set the policy and suppress this warning.

  The included script

    /path/to/Slicer/CMake/BundleUtilitiesWithRPath.cmake

  affects policy settings.  CMake is implying the NO_POLICY_SCOPE option for
  compatibility, so the effects are applied to the including context.
Call Stack (most recent call first):
  SlicerExtensionBundle/SlicerExtensionCPackBundleFixup.cmake:2 (include)
This warning is for project developers.  Use -Wno-dev to suppress it.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26721 3bd1e089-480b-0410-8dfb-8563597acbee
Seg files have one bit per pixel and itk/gdcm will not
read them so an exception is thrown.

Before this fix a generic exception was caught in the
main application suggesting the user quit and restart.

Now user gets a specific dialog that the file is not readadble.

https://issues.slicer.org/view.php?id=4483

From: Steve Pieper <pieper@isomics.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26814 3bd1e089-480b-0410-8dfb-8563597acbee
If Histogram section in Volumes module is not open then now the histogram is not computed to avoid unnecessary (and potentially lengthy) computation.

Also updated bin count computation. For values with discrete values, number of values in range will be used as bin count (but max. 1000). For floating-point volumes bin count is always 1000.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26815 3bd1e089-480b-0410-8dfb-8563597acbee
MRBrainT1 and MRBrainT2 data sets are added to sample data sets.
They belong to the same patient as CTBrain, and they are all spatially registered.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26816 3bd1e089-480b-0410-8dfb-8563597acbee
Node name was not set and the empty name resulted in an exception that prevented loading of the sequence.

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26817 3bd1e089-480b-0410-8dfb-8563597acbee
Editor extensions that needed to hook into the undo/redo
functionality needed to connect to button signals, which
is fragile when the editor is embedded in another module.

See discussion here:
QIICR/PETTumorSegmentation#10

With this change the extension effect directly observes the
undo/redo class.  Multiple observers can be used.

The older callback API is left in place for backwards compatibility.

The PETTumorSegmentationEffect is being updated to use these new calls.

From: Steve Pieper <pieper@isomics.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26818 3bd1e089-480b-0410-8dfb-8563597acbee
This allows clients to query extension metadata without knowing the
extension ID ahead of time. The existing API is designed for the web
interface, which already has access to the extension ID and can pass
the ID back directly.

From: Isaiah Norton <inorton@bwh.harvard.edu>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26820 3bd1e089-480b-0410-8dfb-8563597acbee
- Runs installer directly
- Includes more useful information

From: Isaiah Norton <inorton@bwh.harvard.edu>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26821 3bd1e089-480b-0410-8dfb-8563597acbee
Ancient versions of CMake required else(), endif(), and similar block
termination commands to have arguments matching the command starting the
block. This is no longer the preferred style.

Code updated using https://github.com/jcfr/cmake-auto-style/blob/master/docs/cmake-commands-remove-block-end-arguments.md#cmake-commands-remove-block-end-arguments

From: Hans Johnson <hans-johnson@uiowa.edu>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26822 3bd1e089-480b-0410-8dfb-8563597acbee
Ancient CMake versions required upper-case commands. Later
command names became case-insensitive. Now the preferred style
is lower-case

Code updated using https://github.com/jcfr/cmake-auto-style/blob/master/docs/cmake-commands-to-lower-case.md#cmake-commands-to-lower-case

From: Hans Johnson <hans-johnson@uiowa.edu>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26823 3bd1e089-480b-0410-8dfb-8563597acbee
Due to the Qt issue https://bugreports.qt.io/browse/QTBUG-19683, the
checkable comboboxes do not appear anymore with Qt5.
Overloading the style option SH_ComboBox_Popup from the qSlicerStyle solves
this.

From: Johan Andruejol <johan.andruejol@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26825 3bd1e089-480b-0410-8dfb-8563597acbee
From: Johan Andruejol <johan.andruejol@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26826 3bd1e089-480b-0410-8dfb-8563597acbee
Base style should use the new Qt fusion style as QCleanlooksStyle is not
supported anymore.
This solves issue #4456 (https://issues.slicer.org/view.php?id=4456).

From: Johan Andruejol <johan.andruejol@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26827 3bd1e089-480b-0410-8dfb-8563597acbee
QCleanlooksStyle was introduced in 06e21d2. As it is no longer supported
and we are using the QFusionStyle in Slicer (commit f5f9325), we can
remove it.

From: Johan Andruejol <johan.andruejol@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26828 3bd1e089-480b-0410-8dfb-8563597acbee
The neew Qt5 always returns lower case keys for each style.
Adding a method to convert the style to camel case makes sure the behavior
stays the same as with Qt4.
Also, make sure that when looking up a style name, we are doing it without
checking the case as qt would not either.

From: Johan Andruejol <johan.andruejol@kitware.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26829 3bd1e089-480b-0410-8dfb-8563597acbee
List of changes:

$ git shortlog 887c506..db38b04  --no-merges
Allison Vacanti (3):
      Deprecate vtkHyperOctreeFractalSource.
      Deprecate everything that touches vtkHyperOctree.
      Ensure that vtkMath::Min/Max consistently handle NaN inputs.

Alvaro Sanchez (9):
      Fix cell-data volume rendering when loading large data types.
      Support vtkSmartVolumeMapper::MAGNITUDE for cell-data.
      Added support for multiple volume inputs in GPUVolumeRayCastMapper
      Use 32F internal texture format for VTK_FLOAT.
      Fix container-overflow.
      Remove GetBounds(port) from vtkAbstractVolume.
      Fix pessimizing-move warnings.
      Remove unnecessary variable.
      Remove commented code.

Andrew Bauer (1):
      Avoid using dlopen() and dlsym() in glew when linking against OSMesa

Ben Boeckel (88):
      wrapping: remove the WRAP_EXCLUDE property
      vtkmlib/PortalTraits: fix an include guard comment
      vtkCocoaGLView: hide from the wrappers
      Accelerators/Vtkm: hide vtkm headers from wrappers
      Accelerators/Vtkm: hide from wrapper sources too
      IO/LSDyna: remove the `private` directory
      IO/LSDyna: hide unwrappable classes from wrappers
      IO/ADIOS: hide unwrappable classes from wrappers
      IO/ADIOS: hide from the wrapper sources as well
      wrapping: hide internal classes from the wrappers
      vtkWrapJava: only wrap headers which exist
      wrapping: drop the WRAP_EXCLUDE_PYTHON property
      wrapping: don't set WRAP_EXCLUDE_PYTHON
      tests: add missing test depends (used from Python)
      IO/SegY: exclude private classes from wrapping
      vtkWrapHierarchy: declare that the hierarchy file is an output
      vtkWrapHierarchy: remove the stamp file
      vtkWrapHierarchy: help non-Ninja generator out
      vtkWrapHierarchy: fix fallout from the stamp.txt removal
      cmake: use lowercase command names
      cmake: remove old conditional block
      cmake: set CMAKE_MODULE_PATH earlier
      cmake: move compiler checks into its own module
      cmake: initialize the build type earlier
      cmake: initialize VTK's version earlier
      cmake: move VTK_FORBID_DOWNLOADS to its own module
      cmake: move testing logic to its own module
      cmake: move build and install directories to a module
      cmake: move mobile device logic to its own module
      cmake: remove old compile check
      cmake: move cross compiling logic into its own module
      cmake: move Apple-specific logic to its own module
      cmake: move utility macros to a separate module
      cmake: add a function for deprecating an old option
      Common/Core: set atomic detection flags as narrowly as possible
      cmake: remove unused variable
      cmake: move socket detection features to be narrower
      cmake: remove conditionals for Borland
      cmake: move third-party settings to a single module
      cmake: remove an old setting
      cmake: move threading model detection to its own module
      cmake: move wrapping settings to its own module
      cmake: rename vtkCompilerExtras to vtkCompilerExtraFlags
      cmake: rename vtkDetermineCompilerFlags to vtkCompilerPlatformFlags
      TopologicalSort: fix a typo
      TopologicalSort: the variable case needs to match the node names
      Common/Core: sort source listings
      cmake: pass arguments rather than configuring vtkUnicodeString.cmake
      cmake: clean up the array dispatch list creation
      cmake: clean up type list creation
      cmake: move SMP selection logic to its own module
      cmake: move type array logic to its own module
      cmake: move a single-use function to its usage location
      cmake: move endian-ness test to its usage location
      vtkPeriodicDataArray: move to Common/DataModel
      wrapping: depend on the target for non-Ninja generators
      vtkAbstractTransform: remove unnecessary semicolons
      vtkSetGet: use VTK_SIZEHINT
      Common/Transforms: use VTK_SIZEHINT
      Rendering/Core: use VTK_SIZEHINT
      Common/ExecutionModel: use VTK_SIZEHINT
      Interaction/Widgets: use VTK_SIZEHINT
      Rendering/Annotation: use VTK_SIZEHINT
      Rendering/VolumeAMR: use VTK_SIZEHINT
      Common/Core: use VTK_SIZEHINT
      Common/DataModel: use VTK_SIZEHINT
      Rendering/OpenGL2: use VTK_SIZEHINT
      Common/Math: use VTK_SIZEHINT
      Common/Misc: use VTK_SIZEHINT
      Common/Transforms: use VTK_SIZEHINT
      Filters/Core: use VTK_SIZEHINT
      Filters/Extraction: use VTK_SIZEHINT
      Filters/General: use VTK_SIZEHINT
      Filters/Geometry: use VTK_SIZEHINT
      IO/Image: use VTK_SIZEHINT
      IO/MINC: use VTK_SIZEHINT
      Imaging/Core: use VTK_SIZEHINT
      Imaging/General: use VTK_SIZEHINT
      Imaging/Sources: use VTK_SIZEHINT
      Imaging/Statistics: use VTK_SIZEHINT
      Interaction/Image: use VTK_SIZEHINT
      Rendering/Volume: use VTK_SIZEHINT
      hints: remove the hints file
      Examples: remove hint file
      vtkWrapPython: depend on all module hierarchy files for kits
      vtkParseHierarchy: fail if hierarchy files are not available
      testing: use the module name in test module cmake code
      vtkModuleMacros: add the hierarchy dependency after wrapping

Brad King (9):
      Testing: Drop unused Install/VIT project
      hdf5: add script to import from upstream
      hdf5: Honor CMAKE_MODULE_PATH when including check modules
      hdf5: Enable use of emulator for build-time steps executed on host
      hdf5: Skip embedding of build settings into library with Emscripten
      OpenGL: Configure for build on Emscripten
      Filters/MomentInvariants: Fix compilation
      CTestCustom: Suppress warnings in Eigen headers
      KWSys: Backport macOS utimensat availability check to VTK 8.1

Cory Quammen (5):
      Add to IMAGE_PATH
      Fix retrieval of increments based on data array
      Relax requirements of connectivity filter running in parallel
      Add an IgnoreFileTime option to vtkExodusIIReader
      Stop looking for .xml, .dart, and .dta files alongside Exodus II files

David E. DeMarle (5):
      change commit string to 8.1
      change commit string to 8.1
      address compiler warnings
      Better hide ospray from updated wrapping API
      fix test compilation under msvc

David Gobbi (1):
      Check for hierarchy files before opening output file.

David Thompson (2):
      Provide a method to iterate block visibility info.
      Avoid breaking ParaView's CS wrapping of vtkGDAL.

Expat Upstream (1):
      expat 2017-12-12 (36f8db49)

HDF Group (1):
      hdf5 2014-07-22 (c345afd5)

Hans Johnson (2):
      COMP: Add nullptr instead of NULL or 0
      COMP: Add more C++11 override instrumentation

Jean-Christophe Fillion-Robin (2):
      cmake: re-associate RuntimeLibraries component with python install rules
      cmake: Add hierarchy dependency to python target if it exists

Joachim Pouderoux (3):
      Update numpy data type access for Graph, Molecule and DataObject in general
      Replace std::to_string as it is not supported by some compilers (eg Android)
      Fix GPU based iso surfaces rendering when no contour values are given.

KWSys Upstream (2):
      KWSys 2018-01-08 (f7990fc2)
      KWSys 2018-01-10 (49a21d48)

Ken Martin (11):
      use a smaller exponent for the shadow maps
      remove some code marked legacy in the last release
      use the standard way of freeing/releasing opengl resources
      remove some leftover methods
      remove old 32bit simulator for ios
      maybe fix memory leak in MomentsHelper class
      try fixing two wrapping warnings introduced by recent changes
      fix hardware pick to use the zbuffer not lock it
      fix warning due to not specifying handling of CMP0042
      backport mr 3690 to 8.1 release
      fix compile warning of unused variable in VolumeShaderComposer.h

Mathieu Westphal (3):
      Adding a README explaining why OpenTURNS license can be used in VTK
      Adding a option to generate vertex instead of polyvertex
      Adding a AdaptiveStepReintegration check in the model

Max Smolens (1):
      COMP: WIP: Fix inconsistent-missing-override warning

Menno Deij - van Rijswijk (11):
      New polyhedral contouring, clipping approach
      added a test and solved 2 problems in polyhedral contouring
      Better face triangulation for slightly concave polyhedra, see TestPolyhedron5
      fixed warnings on CDash
      New polyhedral contouring, clipping approach
      added a test and solved 2 problems in polyhedral contouring
      Better face triangulation for slightly concave polyhedra, see TestPolyhedron5
      fixed warnings on CDash
      fix size_t vs vtkIdType vs int warning C4267
      added clamp for acos input. added check for non-manifold triangulation
      replaced all cerr output with vtkGenericWarningMacro

Michael Migliore (3):
      Isosurface blend mode for vtkVolumeMapper
      Improve matrix construction in vtkOpenGLGlyph3DMapper
      vtkCompositePolyDataMapper2 enhancement

Mmanu Chaturvedi (6):
      Fix two leaks in wrapper
      Fix leaks from vtkParsePreprocess.c
      Fix leaks in vtkParseData.c and vtkWrapHierarchy.c
      Fix leak in vtkParse.tab.c
      Fix a leak in vtkParse.tab.c
      Fix float conversion in vtkVolumeMapper.cxx

Robert Maynard (4):
      VTK-m: Consistently use the free standing CastAndCall function
      Bump the VTK-m submodule so that we compile on Intel.
      Propagate down VTK lib install dir and other important information to VTK
      VTK-m used from VTK build directory is now valid

Roxana Bujack (1):
      Add the moment invariants module.

Sankhesh Jhaveri (5):
      Added Address Sanitizer suppresion for libX11
      Add missing overrides for C++11 compatibility
      Add missing override keyword
      Remove unused file
      DynamicAnalysis: Blacklist third party modules and libraries

Sean McBride (3):
      Changed more int to vtkTypeBool
      Introduce typedef unsigned int vtkTypeUBool
      Fixed all compiler errors when vtkTypeBool is actually bool

Steve Pieper (1):
      BUG: WIP: fix vtkPickingManager interaction with widgets

Tiff Upstream (1):
      tiff 2017-12-12 (073877ce)

Todd Martin (1):
      Remove OrientationMarkerWidget from render window on destruction

Unknown (2):
      Uniformity and several typo fixes
      More Misc. typos

Utkarsh Ayachit (14):
      Split `vtk_module_python_module`.
      Add new `vtkmodules` package.
      import vtkCommonCore instead of `vtk` in vtkVariant.py
      Make `vtk` pseudo-package.
      fix invalid `vtk.vtk....` usage in spatialRepAll.
      Fix LagrangeGeometricOperations imports.
      Avoid `import vtk` in `vtk.util.misc`
      Update doc string for vtkmodules.
      Make the trampoline code more robust & clean.
      `vtk_hash_source`: add function to hash a source file.
      vtkParallelCore now uses `vtk_hash_source`.
      remove obsolete vtkHashSource tool.
      Always generate `VTKCompileToolsConfig.cmake`
      Fix warning in FFMPEG 2.* or newer.

Will Schroeder (3):
      New class for extracting surfaces from label masks
      New extrusion filter extrudes to trim surface
      Clean up dashboard warnings

luz.paz (3):
      Misc. typos
      More typo fixes
      misc. typo fixes

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26830 3bd1e089-480b-0410-8dfb-8563597acbee
Thanks: Hans Meine <meine@uni-bremen.de>
Thanks: Isaiah Norton <inorton@bwh.harvard.edu>

$ git shortlog 14580c3..86b5bd6 --no-merges
Hans Meine (10):
      ENH: make test error output more readable
      ENH: output test headers
      ENH: add comment about potential refactoring trap
      BUG: properly format numeric error codes in output
      BUG: add FIXMEs about inconsistent naming
      BUG: fix error handling in qMidasAPI
      ENH: assert that queryId points to a valid result
      BUG: fix success/error reporting of synchonousQuery()
      STYLE: remove excessive spaces
      BUG: correctly forward errors

Isaiah Norton (2):
      ENH: don't assert out when zero results returned
      ENH: safer parse: use QScriptEngine JSON.parse internal function

Jean-Christophe Fillion-Robin (1):
      README: Update build and testing instructions

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26832 3bd1e089-480b-0410-8dfb-8563597acbee
This commit fixes the following error introduced by r26829 (BUG: Fix
Slicer style setting display)

/path/to/Slicer/Base/QTGUI/qSlicerSettingsStylesPanel.cxx: In function ‘QString {anonymous}::toCamelCase(const QString&)’:
/path/to/Slicer/Base/QTGUI/qSlicerSettingsStylesPanel.cxx:163:18: error: no match for ‘operator==’ (operand types are ‘const QChar’ and ‘const char [2]’)
     if (s[i - 1] == " ")
                  ^

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26833 3bd1e089-480b-0410-8dfb-8563597acbee
Wrap hierarchy stamp file removed in VTK9, so use base hierarchy file instead

From: Adam Rankin <adam.rankin@gmail.com>

git-svn-id: http://svn.slicer.org/Slicer4/trunk@26834 3bd1e089-480b-0410-8dfb-8563597acbee
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants