GitLab

91ca6e3d vtkOBJImporterInternals: add missing close bracket
Acked-by: Kitware Robot <kwrobot@kitware.com>
Reviewed-by: Brad King <brad.king@kitware.com>
Merge-request: !3347

* objimporter_fixes_6.3:
  fix issues when importing obj files with windows line endings
  Many fixes to OBJImporter

Thanks to Pat Marion for tracking this down and providing
the patch. Handles windows line endings.

This topic fixes a few issues and adds a test executable
that can be used to convert obj files to vtp more easily.

The issues fixed are

1) having a map_Kd without a texture name cause a crash

2) having materials in the mtl file that were not used
resulted in geometry and parts missing from the output

3) having material names with spaces in them caused
the material not to be recognized

4) some packages use map_kd instead of map_Kd causing
textures to not be found.

* void-swig-ptr:
  BUG 16054: Reduce chance of mistaken swig pointer string.

* slice-collection-crash:
  16130: Fix inconsistency in vtkImageSliceCollection.

In the Python wrappers, the vtkDataArray::SetVoidArray() method can take
two kinds of arguments:

1) a string formatted as a swig pointer "_addr_type" where "addr" is
the hexadecimal address and "type" is the type, e.g. "p_void".

2) a Python buffer object (any Python object with buffer protocol)

A Python string (in Python 2) is also a buffer object, therefore there
is some ambiguity.  The use of swig pointers with VTK is exceedingly
rare, so this fix requires that the string exactly matches the format
"_addr_p_void" to be interpreted as a swig pointer.

Note that in Python 3, the ambiguity disappears because a string in Py3K
does not have the buffer protocol.

The Bottom member of vtkCollection was not updated if the added slice
replaced the previous bottom slice of the collection.  This caused an
inconsistency in the data structure which could lead to a crash in
subsequent operations.  I have rewritten the insertion code to reduce
its complexity.

* gl/Xabi/release-6.3-missing-jpeg-mangle:
  Add missing mangle for jpeg symbol used in vtkJPEGReader

* gl/dgobbi/stencildata-fixes:
  Consolidate vtkImageStencilData code, fix bugs.

Merge-request: !824

* gl/dgobbi/15647-PyObject-Wrapping:
  15647 Fix bug with wrapping PyObject args in Python.

Merge-request: !1213

The check for passthrough "PyObject *" args had an erroneous break
statement, which caused all further args to be ignored.

* gl/Xabi/release-6.3-fix-elevation-limit-case:
  Fix camera Elevation and Pitch if angle is 90

* gl/dgobbi/stencil-iterator-overrun-6.3:
  Fix a memory overrun in vtkImageStencilIterator.

If the stencil extent was smaller than the image extent, then the code
incremented the span list pointer upon reaching the first voxel within
the stencil extent, even though the pointer was already at the correct
location for the first voxel in the stencil extent.  The extra increment
could lead to a buffer overrun and unpredicable behavior resulting in
random crashes.

* gl/dgobbi/15934-meta-writer-update:
  Bug 15934: vtkMetaImageWriter incorrect voxel spacing and origin

The vtkMetaImageWriter's Write() method was getting the spacing and origin
from the input's data object before updating the data object.  As a result,
an incorrect spacing and origin was written to the file header.

* 6.3-osmesa-32:
  get a 3.2 context from OS mesa using new funcs

* gl/Xabi/release-6.3-vtkDelaunay2D-fixes:
  Fix a bug with vtkDelaunay2D.

Working with Brian Paul to get a 3.2 osmesa context
requires changes to VTK and 11.2 mesa changes to work

(cherry picked from commit e7e71c4e)

In a previous patch, a mechanism was added to check edges once
constrained edges were applied. However, in some cases, this
check can break some constrained edges.
This patch make sure to not perform this edge check for triangle
edges which contains a constrained edge.

(cherry picked from commit ac7a28f9)

This reverts commit 262fe1ef, reversing
changes made to 1a04e8a1.

* gl/gcasey/fix-macro-redefinition:
  Fix test compilation
  Fix macro redefinition errors

Move the offending macros to class-scope enums

* gl/dgobbi/python-py3k:
  Fixing bug from 5b6e9c49 with test inversion
  PyVTKObject: add .c_str()
  PyVTKObject: shorten the lifetime of the `s` object
  PyVTKObject: own the classname memory

* gl/Xabi/release-6.3-vtkDelaunay2D-fixes:
  Fix occasional failure of testing if a point is in or out of a triangle
  Fix a bad index issue that make constrained delaunay2d crash

(cherry picked from commit 26fc6af6)

(cherry picked from commit b00e8680)

Seems GCC is happy without this?

It isn't needed after this point anymore.

In Python3, we take the actual object and construct a new bytes() object
from it. We then take a pointer to its data and decref the bytes object.
If the garbage collector runs between the decref and the use of the
classname variable further down, it is a use-after-free.

* gl/Xabi/release-6.3-mousewheel-pending-event:
  Return the MouseWheel event as a pending event in Win32 RenderWindow

Merge remote-tracking branch 'gl/Xabi/release-6.3-fix-doc-vtkimagedata-getdimensions' into release-6.3

* gl/Xabi/release-6.3-fix-doc-vtkimagedata-getdimensions:
  Fix the documentation for vtkImageData::GetDimensions

* gl/dgobbi/python-toupper-macro:
  Remove the toupper macro defined by Python.h.