1. 30 Nov, 2017 4 commits
  2. 29 Nov, 2017 1 commit
  3. 17 Nov, 2017 1 commit
  4. 15 Nov, 2017 3 commits
  5. 14 Nov, 2017 2 commits
    • Sean McBride's avatar
      Fixed, again, a slightly incorrect assert · e1e413fa
      Sean McBride authored
      Fixes issue #17171.
    • Utkarsh Ayachit's avatar
      Python interpreter fixes. · 78dcea63
      Utkarsh Ayachit authored
      This commit includes various cleanups to Python interpreter
      initialization in VTK.
      vtkpython/pvtkpython executables now simply use vtkPythonInterpreter.
      Previously, the code (which was added before vtkPythonInterpreter was
      created) made direct Python API calls. Now, we simply defer to
      vtkPythonInterpreter. That makes it possible to have just one
      location to setup Python initialization logic in VTK.
      Of HPC installations, Python default logic to locate prefix/exec_prefix
      relative to the executable name had the potential to fail when using
      custom Python builds. The new code tries to setup executable name in the
      directory with the Python libraries, if possible.
      The code to locate VTK's python modules (both platform dependent and
      independent components) has been updated to look for them relative to
      the VTK libraries, rather than the executable.
      vtkpython now respects `-v` and `-vv` command line arguments to print
      debugging information about paths looked up when setting up module
      search paths.
  6. 07 Nov, 2017 1 commit
  7. 06 Nov, 2017 1 commit
  8. 01 Nov, 2017 1 commit
  9. 31 Oct, 2017 1 commit
    • Sean McBride's avatar
      General review and cleanup of vtkLookupTable and friends · 63366102
      Sean McBride authored
      - made vtkScalarsToColors::SetRange(double[2]) virtual, since the other SetRange() is.  Likewise vtkLookupTable::SetTableRange(double[2]).
      - changed most uses of dangerous GetTableValue() (that returns inner buffer) in favour of version that copies the values out.
      - use fixed-size memcpy in a few places for a potential speedup
      - improved some comments
      - added a few consts, especially with buffers
      - removed const on plain old ints
      - moved a nan check earlier, saving work when true
      - moved some variable declarations closer to initialization
      - removed some unneeded semi colons
      - removed some useless casts, added others for clarity
  10. 25 Oct, 2017 1 commit
  11. 20 Oct, 2017 1 commit
    • Ken Martin's avatar
      update openvr support to better handle events · eb6bf099
      Ken Martin authored
      Includes support for trackpad event data
      Fix for new steam vr version that adds many events causing
      delays in butten response.
      Fixes picking when using multisamples.
      Better hardware picking by starting at the center pick point
      and working out.
      New menthods in the menuwidget and rep to rename/remove menu items
      Remap next pose control to left trigger so as to not overload
      the grip buttons.
      Handle higlighting of picked lines and polylines better.
      Fix issues with the panel representation when it is called
      before the controllers have been loaded.
      Fix an issue when hardware line width was being forced by the
      render window to a different value.
      Better handling of pose for unloaded or untracked devices.
  12. 18 Oct, 2017 1 commit
  13. 16 Oct, 2017 1 commit
  14. 12 Oct, 2017 1 commit
    • David Gobbi's avatar
      Fix tabs and utf-8 quotes in source code · d636213c
      David Gobbi authored
      Some VTK source code still had tabs (even some of the .py files!).
      Also, several comments used utf-8 quotes, which are hard to tell
      from regular quotes by eye and can lead to confusing mistakes.
  15. 10 Oct, 2017 1 commit
  16. 05 Oct, 2017 1 commit
    • Sean McBride's avatar
      Improved performance of vtkLookupTable a bit · 30acb904
      Sean McBride authored
      Did a few things to speed things up:
      - eliminated some double<->integer conversions.
      - repeated the last colour in the lut as a 4th "special" colour, eliminating need for a test in very hot function. Improved shift & scale calculations against pathological cases.
      - use memcpy() (with fixed size) instead of manual copying byte by byte. Modern compilers optimize this well, as verified with godbolt.
      Did some other cleanup while investigating my performance issues: added more const safety, fixed comments, etc.
  17. 04 Oct, 2017 2 commits
  18. 03 Oct, 2017 1 commit
  19. 02 Oct, 2017 2 commits
    • Kitware Robot's avatar
      Replace VTK_OVERRIDE with override. · 657d021e
      Kitware Robot authored
      Now that VTK requires a C++11 compiler we can ditch using custom typedefs
    • Bastien Jacquet's avatar
      Predictable output for vtkTensorGlyph on non-symmetric tensors · 1e579cd7
      Bastien Jacquet authored
      Previous version implicitly assumed that the tensor was symmetric
      when using eigenvalue decomposition. General tensors have
      non-real eigenvalues and eigenvectors.
      It did not crash on general input because vtkMath::Jacobi expects
      real symmetric matrix, and only use the upper right triangular
      part of the matrix. So computations were as if the lower
      triangular part of the tensor was symmetric of its the upper
      triangular part.
      It make more sense to apply the eigenvalue decomposition to the
      symmetrical part of the tensor.
  20. 29 Sep, 2017 1 commit
    • Andrew Bauer's avatar
      Improvements for vtkNew for stl containers and comparison operations · fd04b442
      Andrew Bauer authored
      Adding in functionality to allow inserting vtkNew objects into
      stl containers of vtkSmartPointer objects as well as comparisons
      between vtkNew objects and raw pointers. Also removing Get() and
      GetPointer() calls from vtkNew objects since in most instances we
      can just pass in the vtkNew<> object instead of having to use the
      Get() or GetPointer() methods to get the raw pointer.
  21. 25 Sep, 2017 2 commits
  22. 20 Sep, 2017 2 commits
  23. 15 Sep, 2017 1 commit
  24. 14 Sep, 2017 1 commit
  25. 09 Sep, 2017 1 commit
    • Sean McBride's avatar
      Fixed undefined ptr overflow · f76e167a
      Sean McBride authored
      Fixed the same UB in 3 tests:
      Where GetBuffer() returned 0 and MaxId was -1, resulting in a pointer to 0xFFFFFFFF.  Even though 1 is added right after, it's apparently UB to even have this temporary overflowed pointer. Crazy. Anyway, calculating the offset first seems no worse for readability and correctness.
      Found by clang 5 UBSan.
  26. 08 Sep, 2017 1 commit
  27. 31 Aug, 2017 2 commits
  28. 30 Aug, 2017 1 commit
  29. 28 Aug, 2017 1 commit