GitLab

The current dependency relationship in vtk is unclear and misleading.
This MR tries to rewrite them based on header files inclusion of headers
and source files in each module. Corner cases are considered and
modules are sorted in alphabetical order to facilitate future reference.
See details in my gitlab python based script project. In future we can continue
using this script to clean the VTK Dependency easily from
time to time.

It doesn't seem to have got everything, but it's a start.

Some resulting whitespace is suboptimal, but one day
we can fix that with clang-format. :)

Just ran this command:
run-clang-tidy.py -checks=-*,modernize-use-override -fix

(I actually had to build my own clang to make clang-tidy
output "VTK_OVERRIDE" instead of "override".)

vtk(.*)\(const vtk\1&\);\s*//\s*Not implemented[\.]*
to
vtk\1(const vtk\1\&) VTK_DELETE_FUNCTION;

vtk(.*)\(const vtk\1 &\);\s*//\s*Not implemented[\.]*
to
vtk\1(const vtk\1 \&) VTK_DELETE_FUNCTION;

vtk(.*)\( const vtk\1 & \);\s*//\s*Not implemented[\.]*
to
vtk\1( const vtk\1 \& ) VTK_DELETE_FUNCTION;

vtk(.*)\( const vtk\1& \);\s*//\s*Not implemented[\.]*
to
vtk\1( const vtk\1\& ) VTK_DELETE_FUNCTION;

vtk(.*) \(const vtk\1&\);\s*//\s*Not implemented[\.]*
to
vtk\1 (const vtk\1\&) VTK_DELETE_FUNCTION;

(operator\s*=.*);\s*//\s*Not\s*implemented[\.]*
to
\1 VTK_DELETE_FUNCTION;

After that, this finds basically nothing:

operator.*implemented

then manually reverted changed files in VPIC and KWSys folders.

Code extracted from:

    https://gitlab.kitware.com/third-party/proj.git

at commit 41bdf05361007f8c5186f3df9944d86bb273ed13 (for/vtk).

It’s apparently undefined for a macro expansion to produce ‘defined’.
Tried to simplify logic too.

Mostly of the form:

A || (!A && B)

to simply:

A || B

This tool warns where macros should be using parens.
Fixed many issues it pointed out by either:
 - adding parens (which it can do automatically)
 - removed unused (or rarely used) macros
 - replacing macros with functions, notably vtkMath stuff

This makes the python wrappers ignore WRAP_EXCLUDE, and instead use
the new property WRAP_EXCLUDE_PYTHON that excludes fewer classes.
The WRAP_SPECIAL flag, which used to act as a whitelist for python,
has been removed.

Because this change causes classes to be wrapped in python by default,
some third-party VTK packages might break until they modify their own
CMakeLists.txt files to add WRAP_EXCLUDE_PYTHON where necessary.

This fix addresses bug reports 0014698 and 0014700. Header files that were
named identically to source files were installed. The added flag allows for
a source file to be flagged with SKIP_HEADER_INSTALL, which prevents its
associated header file from being installed (thanks Ben Boeckel!).

Since NUMFLUIDS is never going to be greater than MAXINT, we simply add
a static_cast<int> to avoid this build issue.

When "NUMFLUID" is present, solid element properties include fluid
density and volume fractions as history variables. Added support to read
those.

aka MSVC++ 8.0 / Visual Studio 2005.

VTK now requires MSVC++ 9.0 / Visual Studio 2008.

Fixed all clang -Wimplicit-fallthrough warnings:
- Some fixed by adding missing breaks
- Other fixed by adding VTK_FALLTHROUGH statement

Under (modern) clang with C++11:
 #define VTK_FALLTHROUGH [[clang::fallthrough]]

Otherwise:
 #define VTK_FALLTHROUGH ((void)0)

VTK_FALLTHROUGH serves two purposes:
 - under clang, suppresses -Wimplicit-fallthrough warnings
 - makes code more self-documenting

Co-authored-by: Yuanxin Liu <leo.liu@kitware.com>
Co-authored-by: Berk Geveci <berk.geveci@kitware.com>

 -The semantics of each unsigned char in the ghost arrays changes:
  Instead of storing a numeric value representing how far a cell is
  from the boundary, it is now a bit field specified by
  vtkDataSetAttributes::CellGhostTypes and
  vtkDataSetAttributes::PointGhostTypes.  The bit field is consistent
  with VisIt specs.

- Previously, filters strip all ghost cells they request from upstream
  before finalizing the output. This is no longer done.

- vtkUniform grids previously supported blanking through member arrays
  vtkUniformGrid::CellVisibility and
  vtkUniformGrid::PointVisibility. These arrays are removed and the
  blanking functionality are supported through the new ghost arrays
  instead.

- the "vtkGhostLevel" arrays for cell and point data are renamed to
  vtkDataSetAttributes::GhostArrayName() ("vtkGhostType").

- the version for VTK Legacy files is increased to 4.0 and the version for
  VTK XML files is increased to 2.0. When reading older files we
  convert vtkGhostLevels array to vtkGhostType.

Fixes cppcheck warning, should be a little faster

Change-Id: I24ebe2085e2f1c771d2a83fba6ebfb8761cfb9ab

Used the following command:

find . \( -iname *.h -and -type f \) -exec perl -pi -w -e 's/__vtk(.*)_h/vtk$1_h/g;' {} \;

Fixes many, but not all, clang -Wreserved-id-macro warnings.

Change-Id: I694ff053cb204c70e37b261296f496064d91fc07

Some gcc-4.2 builds complain when a type is defined within an
anonymous namespace, but subsequently used to define members of
any class declared outside of the anonymous namespace.  This is
probably a spurious warning, but it is easy to fix by making the
type itself a member of the class that is using it.

Change-Id: I4cf5cf2e9538f6a0fd611046193e9f0bd53b76d1

Change-Id: I76fea205f42b796bcfc5d47b6f171d39eadd44a4

Did this for cxx files where another vtk include was already
using the "" style.  I find this is more correct, especially
for implementation files.

Change-Id: I632178a9af13991ccba96bfaabf5ccc4bfea3c18

This resolves PGI ~ 600 warnings "#111-D: statement is unreachable"

Change-Id: I9937c4f22eec4e9f56b6b770e35a6a1ca7390974

Change-Id: Ic3c5e9ffa8cc4890c3e4ae731a8b60e459876a57

The main kits are:

    vtkCommon
    vtkFilters
    vtkIO
    vtkImaging
    vtkInteraction
    vtkOpenGL (required due to a dependency cycle from
               vtkRenderingOpenGL on vtkImagingHybrid)
    vtkParallel
    vtkRendering
    vtkViews
    vtkWrapping

Modules which deal with a specialization of another module belong to the
kit related to the specialization (i.e., FiltersParallel and
IOParallelXML both belong to the vtkParallel kit) to avoid circular
dependencies that arise when put into the same kit as the "lesser"
module.

Change-Id: Icae0baa78d62cc0dcce84546541df600d4f311dd

Change-Id: Ieaac07aa6ad89fd964f2215be790ee2f05978c5e

An LS-Dyna simulation may have time-varying nodal coordinates.
In this case, the initial, static coordinates are ignored since
they are overwritten by the deflected coordinates as each timestep
is read. However, it is apparently possible to configure LS-Dyna
not to write the deflected coordinates. In which case, we *must*
read the static coordinate data since that is all that's present.
Do so.

Change-Id: I4db3af4bbd0948212f179a620bde3bc0fefe23b3

These tests should work with either backend, and were inadvertently
merged with a hard dependency on OpenGL2 backend.

Change-Id: I25356b499d2317143db37e3fa59b04c5b1c53710

Change-Id: Id76d3bc545a64675a0820cb1d9b45c61a51efc6a

Refactoring the way VTK goes between piece and structured
extents. Before, extent translators were used when the pipeline
moved from structured to unstructured data converting piece
request to extent request. This caused many problems with filters
that altered extents, mainly a lot of redundant IO due to
repartitioning of different extents. This became extremely
cumbersome to manage when running distributed. The new behavior
pushes the extent translation all to way to the readers and
only when readers are able to read a subset. This works much
better. The only downside is that filters need to be able to
handle data extents different than update extents. Most filters
can do this but many imaging filters cannot. Those that are
needed in parallel will have to be updated.

As part of this work, I also removed MAXIMUM_NUMBER_OF_PIECES
and added CAN_HANDLE_PIECE_REQUEST. MAXIMUM_NUMBER_OF_PIECES had
reduced to being a boolean. 1 for serial sources, -1 for parallel
sources. I removed it and added a CAN_HANDLE_PIECE_REQUEST instead.
This key, produced by a source, tells the executive that a source is
able to handle piece request. It is a source only key produced in
RequestInformation and is not propagated downstream. If this key is
not present, the executive will only execute the source for piece 0
to produce the entire data. It is then up to the user to add a filter
that splits the data for other piece requests. The only exception to
this is when CAN_PRODUCE_SUB_EXTENT is present, in which case the
executive will split using an extent translator AT THE source - not
downstream.

Conflicts:
	Filters/FlowPaths/vtkParticleTracerBase.cxx

Change-Id: I8db4040289ff87331adeecded4a738313d9b52df

Change-Id: I10a54acbd076e055d8071414f2fff1ebafe1653d

Change-Id: I92102c8ac0609f80308f6954f6bc89aabf42e963

Change-Id: I461204856d7cccdc7f7edf41773391edd69daecd

In other words, changed from the C to the C++ header name.
Did this only in .cxx files, and not in ThirdParty.

Change-Id: I81e2b2c7f2db00e98b37d507232fbc84022e8d8f

A simple implementation of private dependencies, with some of the most
obvious private dependencies moved over. This should be used for all
dependencies that are not exposed in the public API of the classes in
the module.

The target_link_libraries call defaults to private due to the way CMake
implements the link logic. The private dependency logic can help reduce
link line and include directory length for private links where the
dependency's API is not exposed in the module. Also added code to ensure
include directories are not duplicated for modules in the build tree.

Change-Id: If4535338e8d5e9957d7bcce7d3ab0eadec21d2c4

Change-Id: Ief1911b72184635ea73581de1e5a66cf7d54b4c3

Add to the Testing/Data directory ExternalData content links mirroring
the content and layout we use from the VTKData/Data directory.

Add a CMake/vtkLegacyData.cmake module, included from CMakeLists.txt and
Testing/External/CMakeLists.txt, to call ExternalData_Expand_Arguments
for all data directories added from VTKData/Data.  This will bulk-fetch
all test input data currently used by tests of any module, enabled or
not.  We can make the DATA{} references more granular later.

Add a VTK_TEST_DATA_DIR variable to refer to the directory we tell the
ExternalData module to populate the real data files.  Replace references
to VTK_DATA_ROOT in test command lines (after -D) with references to
VTK_TEST_DATA_DIR.  Drop the VTK_DATA_ROOT CMake option and stop using
the VTKData repository.

Remove the TESTING_DATA option from the vtk_add_test_cxx API and remove
the VTK_DATA_ROOT option from the vtk_add_test_mpi API since all test
input data are now handled through ExternalData and Testing/Data content
links.

Change-Id: Id02490b76ea2e161b9038188264a4830485039d8

Add a Testing/Data directory with ExternalData content links mirroring
the content and layout we use from the VTKLargeData/Data directory.
Add a VTK_USE_LARGE_DATA option to enable tests using such data.

Call ExternalData_Expand_Arguments in */*/Testing/Cxx/CMakeLists.txt for
tests that reference these data files in order to tell the ExternalData
module to populate a ExternalData/Testing/Data directory in the build
tree.  Pass the ExternalData/Testing directory in the build tree as the
"-D" data directory option to tests using these data so that they can
find the files where ExternalData puts them at build time.

In the vtk_add_test_* API replace the LARGE_DATA options with a new
TESTING_DATA option to indicate that a test uses the directory
ExternalData/Testing/Data populated by ExternalData from the
Testing/Data source directory.

Drop VTK_LARGE_DATA_ROOT and stop using the VTKLargeData repository.
Drop references to VTKLargeData from test source comments.

Change-Id: Iaaf36293a840e65d6c171ae69b5c99db4c61443e

Baseline images are now referenced by ExternalData content links stored
directly in each module source tree.  Drop the old VTKData baseline
subdirectory name from the vtk_add_test_* signatures.

Change-Id: I5bdc7a5bf87fc7c8fa77dcd50b79f2f54d6eba31