The legacy reader and writers only used dimensions. This did
not support extents that do not start at (0, 0, 0). Added initial
support.

Also bumped the version. The format version is now 4.2.

This commit fox a warning in vtkDataWriter
for a incorrect usage of sprintf with vtkIdType

The source was searched with "grep -l [^ -~]" to find characters outside
of the printable ASCII range 0x20 to 0x7e.  The worst offender is the
en-dash character, which is visually indistinguishable from the minus
character in most code editors.  Several tabs were also found, which
is surprising since they should be blocked by the commit hooks.

This commits corrects usage of int into vtkIdType
to allow dataset to read and write to have more points/cells/rows
than INT_MAX

This commit replace a bunch of stat() usage by Stat
which is more robust especially on windows.

This commit uses uint64 instead of vtkIdType/int for
the ensight FileSize, and ensure the stat version used is always 64bit.

Paired several vtkErrorMacro's with missing return calls to bail on the
function to prevent applications from freezing. A few vtkErrorMacro's
were changed to vtkWarningMacro's since it appears the intention is
for the process to continue.

Kunda authored 8 years ago and Cory Quammen committed 8 years ago

doxy and misc. typos pt2

Kunda authored 8 years ago and Cory Quammen committed 8 years ago

fixes some doxy typos
fixes misc. comment typos

- some post to pre increment changes
- removed unneeded check that unsigned >= 0
- fixed seeming copy-paste where kMax was assigned different values back-to-back
- remvoed unused variable
- marked operator= method with VTK_DELETE_FUNCTION to hopefully silence complaint about return type

This avoids code from ending up tripped by '\r's in Windows ASCII files.
Fixes TestLegacyArrayMetaData on Windows file i/o.

Also updated TestLegacyArrayMetaData to test saving out and reading from
disk files in addition to in-memory strings.

Addresses paraview/paraview#17242.

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

Text in VTK (and PNG) is rendered at the bottom left corner of bounding box,
however text in PDF, postcript and probably all other vector formats
is rendered at baseline left position.

Alignment used to work only for TextAsPath true. Now it works
for TextAsPath false as well.

fix memory leak of materials

Apparently eeloo has a slight difference in the way the text is aligned
relative to the scalar bar.  Everything looks right it is just shifted
slightly.

With the text rendering changes, new baselines are needed.  Removing all
old baselines for these tests and adding new ones as needed.

header should include string

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.

When reading an attribute, vtkNetCDFCFReader would allocate an
std::string of the appropriate length, then grab the C pointer from it
and call the netCDF library to fill it with the value for the attribute.
The problem was that if the attribute was zero-length, std::string would
throw an exception if you tried to grab the pointer. Fix the problem by
not loading any data if the length is 0.

Also added some defensive code for when reading cell bounds when there
are no cells.

The code for reading in indexed colors is a bit
wonkey and can segfault. This topic prevents the
segfault.

This patch removes the upper limit of 4096 time steps in annotated
time XML files. Now an arbitrary number of time steps can be read.

Repurposed vtkXMLDataElementVectorAttributeParse() for the purposes of
counting tokens of a particular type in an attribute by changing the
meaning of a NULL 'data' parameter.

The existing test has been added to try out the new feature, as well
as a small test data set.

Hat tip to David Benn for the bug report and a reference solution.

Expected errors generated when vtkTRUCHASReader::CanReadFile() is
invoked on non-TRUCHAS files were written to stdout, causing false
negative test results caused by the word "Error" being written
to stdout. This changes silences those error messages.

The vtkPExodusIIReader now will go through multiple files when
examining for meta-data during the RequestInformation() pass.
Previously, if the first file didn't have any data then the
information would be wrong but with this change it will be correct.

The reader now supports reading files have vertices but no
faces. These fies are used to store point clouds.

A sample data file and test are provided.

- Made a few APIs' params const
- Changed many usages of the evil GetBounds() to expect a const
  return value, even though it does not (yet?) return const
- Slightly changed return values of vtkTextActor3D::GetBounds()
  to return its own ivar instead of its ImageActor's

The way the ExodusII writer worked had issues when trying
to be used in Catalyst (i.e. if the last time step wasn't
known). Changes include getting the actual time directly
from the input VTK data object and ignoring file name mangling
for the time step if WriteAllTimeSteps isn't set. Note that
file name mangling will still happen in this case if there
is more than a single process doing the writing.

Reported in JVN#37655328