This should also allow query selections on field data to
be plotted over time (though ParaView does not yet support them).

Change-Id: Ia7e19723c9c2fca46bbdb08b9befe3f447fb2959

Instead of breaking a selection into a separate time-history
table for each (block,ID)-tuple, you may call
ReportStatisticsOnly(). Then a single table per
block of the input dataset will report the minimum, maximum,
quartiles, and (for numerical arrays) the average and standard
deviation of the selection over time.

This adds dependencies on the FiltersGeneral and
FiltersStatistics modules to the FiltersExtraction module.

Change-Id: I42c2ef3314a3a4497bd50e1c0dd6cc9fde8ce5fb

The main change is to remove the use of vtkTemporalDataSet and move
the support of multiple temporal data sets from the execution
pipeline to filters. To be specific,

- Before, a filter can request objects from multiple time steps by
  setting the key UPDATE_TIME_STEPS to a vector of doubles; the
  resulting objects get wrapped by the pipeline into a single
  vtkTemporalDataSet object.

- After, a filter can only ask for a single time step from the
  pipeline by setting the key UPDATE_TIME_STEP to a single double.
  The "wrapping" no longer happens. If a filter want to request
  multiple time steps, it needs to either inherit from
  vtkMultiTimeStepAlgorithm or use the CONTINUE_EXECUTION to loop the
  upstream pipeline and store the data from each iteration.

The following key changes/constants are backward incompatible:

vtkStreamingDemandDrivenPipeline::UPDATE_TIME_STEPS(() ->
   vtkStreamingDemandDrivenPipeline::UPDATE_TIME_STEP((),
   vtkMultiStepAlgorithm::UPATE_TIME_STEPS()

vtkStreamingDemandDrivenPipeline::PREVIOUS_UPDATE_TIME_STEPS() ->
  vtkStreamingDemandDrivenPipeline::PREVIOUS_UPDATE_TIME_STEP()

deleted: vtkCompositeDataPipeline::REQUIRES_TIME_DOWNSTREAM()

depreicated: VTK_TIME_EXTENT

Change-Id: I635b6401ae4f0a7ea7c4b5c466ced40ee75963c7

Fixed: SetUpdateExtent() and its ilk were setting the update
extent on the input. Fixed it to be on the output. This is
more consistent with the executive API. Also updated other
code accordingly.

Change-Id: Iac92df21b8fa8dbeb58e4e27c5ffc796fadc4eaa
Added: API to get update extent from the output.

Exclude ThirdParty, Utilities/MetaIO, and Utilities/KWSys as these
are maintained outside VTK.

Co-Author: Marcus D. Hanwell <marcus.hanwell@kitware.com>
Co-Author: Chris Harris <chris.harris@kitware.com>
Co-Author: Brad King <brad.king@kitware.com>

Move source files from their former monolithic VTK location to their new
location in modular VTK without modification.  This preserves enough
information for "git blame -M" and "git log --follow" to connect
modularized VTK files to their original location and history.

Co-Author: Marcus D. Hanwell <marcus.hanwell@kitware.com>
Co-Author: Chris Harris <chris.harris@kitware.com>
Co-Author: Brad King <brad.king@kitware.com>
Co-Author: Nikhil Shetty <nikhil.shetty@kitware.com>

The vtkstd forwarding headers are deprecated, move to using the STL
headers directly.

Change-Id: Iaeb5ea0c10d40b8e934b6a8ca18f28c3b022b924

Since the behaviour of SetInput changed - it no
longer connects the pipeline but sets up the data
object as the input -, it was decided to rename
it to SetInputData. This includes similar methods
such as AddInputData. This commit makes on
sweeping change that compiles.

The CVS $Revision$ keyword replacement will no longer maintain these
macros automatically.  They were used only to implement CollectRevisions
and vtkObjectBase::PrintRevisions, an API that was never used.

Automated as follows:

  pass 0:
  catch templates
  ---------------
  $ git grep 'vtk\(Cxx\|Type\)RevisionMacro' | grep '<'

  pass 1:
  main changes
  ---------------
  $ git grep -l '^vtkCxxRevisionMacro' | while read file; do
    echo "$file" 1>&2
    perl -i -ne 'print unless (/^vtkCxxRevisionMacro/ and /\)/)' $file
  done
  $ git grep -l -e 'vtkTypeRevisionMacro(' | while read file; do
    echo "$file" 1>&2
    perl -i -pe 's/vtkTypeRevisionMacro/vtkTypeMacro/g' $file
  done

  pass 2:
  verify
  ---------------
  $ git grep 'vtk\(Cxx\|Type\)RevisionMacro'

Fixed multi-line vtkCxxRevisionMacro and templates by hand.

The CVS $Revision$ keyword replacement will no longer maintain these
macros automatically.  They were used only to implement CollectRevisions
and vtkObjectBase::PrintRevisions, an API that was never used.

Automated as follows:

  pass 0:
  catch templates
  ---------------
  $ git grep 'vtk\(Cxx\|Type\)RevisionMacro' | grep '<'

  pass 1:
  main changes
  ---------------
  $ git grep -l '^vtkCxxRevisionMacro' | while read file; do
    echo "$file" 1>&2
    perl -i -ne 'print unless (/^vtkCxxRevisionMacro/ and /\)/)' $file
  done
  $ git grep -l -e 'vtkTypeRevisionMacro(' | while read file; do
    echo "$file" 1>&2
    perl -i -pe 's/vtkTypeRevisionMacro/vtkTypeMacro/g' $file
  done

  pass 2:
  verify
  ---------------
  $ git grep 'vtk\(Cxx\|Type\)RevisionMacro'

Fixed multi-line vtkCxxRevisionMacro and templates by hand.

* Moving vtkExtractSelectedRows from InfoVis to Graphics.
* Adding support to vtkExtractSelectedRows to optionally produce
  vtkOriginalRowIds array.
* Extended vtkExtractSelection to use vtkExtractSelectedRows to extract rows
  from vtkTable.
* Fixed vtkExtractArraysOverTime to work with vtkTables.

don't add it to the output.

get added by vtkExtractSelection and don't provide any extra information in the
output of vtkExtractArraysOverTime filter anyways.

of vtkRectilinearGrid.

* All parameters such as content type, field type, selection list are on
  vtkSelectionNode.
* All constants are on vtkSelectionNode instead of vtkSelection.
* The SELECTIONS content type is deprecated.
* To create a simple single-node selection, you must create an
  instance of vtkSelectionNode and an instance of vtkSelection, and add
  the node to the selection. You may also use vtkSelectionSource.
* vtkConvertSelection has a few new useful static functions
  To obtain the list of selected items in a data object.

called before the RequestUpdateExtent pass. This method will evetually remove
all update-related keys from the input information. Currently it only removes
the fast-path related keys.

exodus reader switched to multiblocks and vtkExtractArraysOverTime started
support extracting of multiple cells/points.

In the new framework, fast path only works if the selection being extracted is a
global id based selection. For multiple cells/points, we execute the pipeline
once for each cell/point, thus avoiding any need to change the fast path-based
keys in the executive.

vtkCompositeDataPipeline::NeedToExecuteData() now takes into consideration
changes in fast path keys (similar to vtkStreamingDemandDrivenPipeline).

ENH: vtkExtractArraysOverTime names the blocks based on which cell/pt they represent.
BUG: Fixes BUG #6262. Fixed all extract selection filters to pass global ids.
BUG: If composite dataset has no names, use default values

ENH: Fix vtkExtractArraysOverTime to work with composite datasets. Also changing the implementation to use vtkExtractSelection filter to extract the selection. Also, it now produces a multi-block output. Thus, making is possible to extract multiple cells/points over time.

ENH :Added a new vtkProbeSelectedLocations which uses vtkProbe to probe different locations indicates by the vtkSelection.

BUG: Update the proxy tab information when the information is gathered by the source proxy.

BUG: Don't bother doing any globalId-id conversions if the number of ids==0

     readable time values.
BUG: Fix bug where the x-axis of time plots displayed
     time steps instead of time values when the fast-path was used.
     Fix plotting cell variables over time.

     (From Eric Stanton)

be consistent with probe. Removed "Cell's Point Ids" from the list of available
arrays.

BUG: Work around a hard to fix bug by not using vtkDataSetAttribute::InterpolateAllocate/InterpolatePoint. The bug shows up, in parallel, when the point data has a globalIDS attribute, on the first probe. In that case the arrays at allocate lack the globalid array and later when the interpolate happens, the arrays become misaligned. To 'fix' I am doing the interpolation manually, array by array.

BUG: correct and simplify probe location over time. Now, it produces a validity array and sets values to 0 at invalid samples instead of deleting the samples. Deleting was not OK because the extent from RequestInformation must match that produced by RequestData. Also when probing a location interpolate into the point data and copy the cell data (as long as their are no name collisions).

ENH: Add probe a location over time (deja vu), and make extract cell (by id or location) over time record point ids. See bug #4849

ENH: remove invalid points from the plot. This can happen in extract by id over time when the cell or point being plotted is destroyed. It also happens when plotting a location over time when that location doesn't happen to be inside any cell.

BUG: Fixed wrong return value. It was always returning the id it received. In case of global ids, it should return the id of the point/cell that has the given global id.

STYLE: Improved documentation and comments.

ENH: Added new filter to extract a point/cell over time as a rectilinear grid. This class will probably merge with vtkExtractDataOverTime at one point.