GitLab

The normal of a vector is the vector divided by the vector's magnitude.
However, if the vector is the 0 vector, you get 0/0, which results in
non-finite numbers.

There might be legitimate situations where you (unintentionally) try to
normalize a 0 vector. For example, you might be using the derivative of
a field to compute the normal to a contour of the field. If the field is
in a static region or critical point, then the gradient goes to 0.

In these cases, we want the `Normal` and `Normalize` functions to return
something reasonable. These functions now return the 0 vector back. This
is no more "correct" than producing NaN values, but it is less likely to
cause problems later.

Matt Larsen authored Jun 08, 2020 and Kitware Robot committed Jun 08, 2020

ce3a64b4 fix comparison
ab6bb14d Merge branch 'respect_max_distance' of gitlab.kitware.com:mclarsen/vtk-m into respect_max_distance
a6ed4102 use bounds based epsilon for tiny meshes
59775c4e

 respect ray max distance
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Sudhanshu Sane <ssane@cs.uoregon.edu>
Merge-request: !2129

Robert Maynard authored Jun 08, 2020 and Kitware Robot committed Jun 08, 2020

748f8f8c

 Correct flying edge cell-id mapping
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Matt Larsen <larsen30@llnl.gov>
Merge-request: !2124

Kenneth Moreland authored Jun 05, 2020 and Kitware Robot committed Jun 05, 2020

7be44c84

 Enable setting invalid value in probe filter
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Sujin Philip <sujin.philip@kitware.com>
Merge-request: !2122

Nick Thompson authored Jun 03, 2020 and Kitware Robot committed Jun 03, 2020

3265460e

 Fix C++17 build.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2121

Initially, the probe filter would simply not set a value if a sample was
outside the input `DataSet`. This is not great as the memory could be
left uninitalized and lead to unpredictable results. The testing
compared these invalid results to 0, which seemed to work but is
probably unstable.

This was partially fixed by a previous change that consolidated to
mapping of cell data with a general routine that permuted data. However,
the fix did not extend to point data in the input, and it was not
possible to specify a particular invalid value.

This change specifically updates the probe filter so that invalid values
are set to a user-specified value.

Kenneth Moreland authored Jun 03, 2020 and Kitware Robot committed Jun 03, 2020

7ded85f0 Remove reference to Canvas in ImageReaderBase
7fa2c9f7

 Update doxygen comments in Tuple.h.in
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2119

It is no longer necessary for the image reader to access a Canvas, and
this does not work anyway if rendering is turned off.

The `Tuple.h` file was updated with some comments that were not updated
in `Tuple.h.in`. This caused the pyexpander script to fail with
mismatched files.

Robert Maynard authored Jun 03, 2020 and Kitware Robot committed Jun 03, 2020

930c3d2f reproduce_ci_env better detects which stage a job is for
71d9d431 Doxygen is now generated by gitlab and done on each master merge
a652f7fe Update vtk-m doxygen to generate less warnings
97a2408f

 Correct warnings for using TBB 2020 deprecated functions
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2117

Previously stages like `doxygen` wouldn't be detected by
reproduce_ci_env as they didn't start with `build:` or `test:`.
Now we properly look for the `stage` key.

Kenneth Moreland authored Jun 03, 2020 and Kitware Robot committed Jun 03, 2020

8239eb46

 Fix an issue with structured (non-uniform) gradients
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Robert Maynard <robert.maynard@kitware.com>
Merge-request: !2118

Kenneth Moreland authored Jun 02, 2020 and Kitware Robot committed Jun 02, 2020

1949cad4 Add get/set filename methods for image reader/writer
e8c5ee5e Remove pixel type selection from PNG reader
6c97054b Update image writes to style and library
1e91b32a

 Update image readers to style and library
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Nickolas Davis <nadavi@sandia.gov>
Merge-request: !2109

This fixes a bug introduced in commit
93d87e06

Part of the commit moved fromusing a `BoundaryState` object to keeping
boundary flags in a `vtkm::Vec<bool,3>`. However, the meaning of the
flag got inverted. The `BoundaryState::IsRadiusIn?Boundary` methods
returned true if the current location was inside the bounds (i.e. _not_
on the boundary), but the new flags are true if the current location is
on the boundary (the inverse).

The updating of the code for the `StructuredPointGradient::Jacobian`
method got the flag somewhat wrong and scaled the wrong case.

Abhishek Yenpure authored Jun 01, 2020 and Kitware Robot committed Jun 01, 2020

e9fd545b Merge branch 'master' of gitlab.kitware.com:vtk/vtk-m into euler_integrator_fix
23126bbe Integrator status fix for consistant behavior
3e18f105 Fixing Integrator Status Behavior
ff112e42 Integrator status behavior check
343f2d57 Merge branch 'master' of gitlab.kitware.com:vtk/vtk-m into euler_integrator_fix
82cfba27 Rectify testing code
8ddecbb4 Fixing test issues
6274b165

 Updating Euler Integrator to work
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2023

Robert Maynard authored Jun 01, 2020 and Kitware Robot committed Jun 01, 2020

f7846f90 Add openmpi library path to LD_LIBRARY_PATH
e7ca8be0

 Update docker images to use cmake 3.17.3
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Sujin Philip <sujin.philip@kitware.com>
Merge-request: !2114

Selecting the pixel type just means that lodepng casts it to that type
for it then to be cast to the type VTK-m uses. Since VTK-m always uses
the same type, there is little value for changing the intermediate cast.

Updated the image writer classes to better follow VTK-m naming
convention (i.e. use `ImageWriterPNG` instead of `PNGWriter`). The
structure of the image writer class now also better matches the
interface for the VTK writer.

Also put the implementation of the image writers into the `vtkm_io`
library so that it only has to be compiled once.

Updated the image reader classes to better follow VTK-m naming
convention (i.e. use `ImageReaderPNG` instead of `PNGReader`).
The structure of the image reader class now also better matches
the interface for the VTK reader.

Also put the implementation of the image readers into the `vtkm_io`
library so that it only has to be compiled once.

The 3.17.3 release fixes a regression in ctest where NOT_RUN
tests didn't get reported to ctest or cdash

Robert Maynard authored Jun 01, 2020 and Kitware Robot committed Jun 01, 2020

565d5a3a

 Remove CMake workarounds for version < 3.12
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2110

Kenneth Moreland authored May 28, 2020 and Kitware Robot committed May 28, 2020

3273f1c4 Update to AddPointField of dataset
eef30c23 Merge branch 'master' into add/hyperstructstats
1014dbf0 Fix BOV reader support in ContourTreeApp
e908826c Temporary disabled BOV reader in ContourTreeApp pending #517
8a27ff51 Merge branch 'master' into add/hyperstructstats
43e470de Fix g++ conversion warning in ContourTreeApp
a9a4736c Add BOV reader support and fix TBB settings in contour tree app
38da8dc4

 Fix TBB check bug in ContourTreeApp
...
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2002

Robert Maynard authored May 28, 2020 and Kitware Robot committed May 28, 2020

0f867c05

 vtkm::io::reader::BOVDataSet fix 'constructor delegates to itself'
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2107

Nick Thompson authored May 28, 2020 and Kitware Robot committed May 28, 2020

f33578d6 Add newline to docs file.
8fe17e0a Remove DataSetFieldAdd from examples as well.
fe4f71ab

 Deprecate DataSetFieldAdd.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2106

Fixes #517