GitLab

- what was previously termed 'setLargeCellSubset()' is now simply
  'setCellSubset()' and supports memory efficient interfaces.

  The new parameter ordering avoids ambiguities caused by default
  parameters.

  Old parameter order:

      setLargeCellSubset
      (
          const labelList& region,
          const label currentRegion,
          const label patchID = -1,
          const bool syncCouples = true
      );

  New parameter order:

      setCellSubset
      (
          const label regioni,
          const labelUList& regions,
          const label patchID = -1,
          const bool syncCouples = true
      );

   And without ambiguity:

      setCellSubset
      (
          const labelUList& selectedCells,
          const label patchID = -1,
          const bool syncCouples = true
      );

- support bitSet directly for specifying the selectedCells for
  memory efficiency and ease of use.

- Additional constructors to perform setCellSubset() immediately,
  which simplifies coding.

  For example,

      meshParts.set
      (
          zonei,
          new fvMeshSubset(mesh, selectedCells)
      );

  Or even

      return autoPtr<fvMeshSubset>::New(mesh, selectedCells);

- get<label>, get<scalar> instead of readLabel, readScalar, etc.

- regularise referencing of OpenFOAM env variables, sources etc.

CONFIG: cmake targets are globally unique

- unify CMakeLists naming and align content

- the build is now much more strongly oriented on cmake.

  The wmake wrapper only adds some minor versioning on the build
  and uses a standard CMAKE_INSTALL_PREFIX to define its installation
  location.

  These changes can make it easier for some packagers (eg, spack)
  to manage.

 - LIBRARY_OUTPUT_PATH is removed
 - CMAKE_LIBRARY_OUTPUT_DIRECTORY is used instead of LIBRARY_OUTPUT_PATH
 - CMAKE_INSTALL_PREFIX is used as install destination

- allows restriction of the output patches to specific names and hiding
  others. Allow groups when matching patch names

- forces release of memory for the associated backends

- log which channels are being emitted.
  The printChannels pipeline was redundant anyhow.

- removed fvMesh sub-channels if favour of selector flags.
  To enable/disable portions of the geometry:

      internal    true/false
      boundary    true/false

  This replaces the previous method of sub-channels in which restrictions
  were propagated from the pipeline through.

  Before
  ======
      Function Object:

          input
          {
              fields    (U p);
          }

      Pipeline:
          input = coprocessor.CreateProducer(datadescription, 'input/patches')

  Now
  ===
      Function Object:

          input
          {
              internal  false;
              fields    (U p);
          }

      Pipeline:
          input = coprocessor.CreateProducer(datadescription, 'input')

- allows cmake override of the default output library directory,
  which is normally $FOAM_LIBBIN

ENH: additional debug verbosity

- extra information (in parallel) when debug level > 1, which can help
  when isolating problems.

- if sigFpe is enabled within OpenFOAM, any errors in the Catalyst
  pipeline will potentially kill the entire simulation.

- use the new sigFpe::ignore helper class to manage a local disabling
  of SIGFPE handling.

- can remove directly from the temporary objectRegistry prior to
  converting by field type. Cannot, however, do the same for volFields
  or areaFields since these reside on their respective mesh objects.

- LINK_LIBRARIES are required for proper try_compile

STYLE: report output directory

- ParaView Catalyst with MPI is advisable (especially for larger
  simulations), but not a requirement

BUG: general patch movement was inconsistent

- previously had separate function objects for each category of source
  (finite volume, finite area, lagrangian). This lead to isolation of
  the individual input sources, which highly restricted the flexibility.

- now have a single catalyst function object with an arbitrary number
  of inputs (sources). The catalyst channel name is that of the
  corresponds dictionary name. For volume mesh sources, it is possible
  to address "mesh" and "patches" sub-channels.

  Eg,
      mesh = coprocessor.CreateProducer(datadescription, 'myregion/mesh')

  This restructuring also aims at integration of other OpenFOAM content
  (eg, surfMesh, sampled probes, sampled surfaces) as a future
  development.

- the output directory (by default "<case>/insitu") is passed onto
  catalyst as its working directory, using the upstream patches that
  now close issue #4

- reflects current change being made to the vtkCPProcessor API to
  include chdir capabilities.

- prevents accidental overwrite of runtime conversions.

- uses the cellCellStencil information to blank holes and the
  interpolated cells.

- needs more detailed attention for handling cases when three or more
  meshes are overlapping.

- Note that stopping the catalyst function object via a `timeEnd` value
  does not actually call the corresponding end() method.

  See: Development/OpenFOAM-plus#791

ENH: support mkdir as dictionary keyword

- this is a convenient means of creating an output directory from within
  the function object without corresponding python or shell script.

  Some pipelines (image generation) balk if the output directory does
  not exist. Others (vtm writer etc) will create their own.

This integration represents a collective work
  - CINECA (Simone.Bna@cineca.it)
  - OpenCFD
  - with additional input from KitWare (Andy.Bauer@kitware.com)

NB: requires ParaView or ParaView Catalyst with MPI and Python

- do not need the stringOps version if there is no dictionary context

- support VTP input in functionObjectCloud scene element

- additional fallback lookup of cloud information from state properties
  instead of cloud OutputProperties

- make the purpose more explicit, and reduces some work for the
  compiler as well.

- disallow insert() of raw pointers, since a failed insertion
  (ie, entry already existed) results in an unmanaged pointer.

  Either insert using an autoPtr, or set() with raw pointers or autoPtr.

- IOobjectList::add() now takes an autoPtr instead of an object reference

- IOobjectList::remove() now returns an autoPtr instead of a raw pointer

- generalize some of the library extensions (.so vs .dylib).
  Provide as wmake 'sysFunctions'

- added note about unsupported/incomplete system support

- centralize detection of ThirdParty packages into wmake/ subdirectory
  by providing a series of scripts in the spirit of GNU autoconfig.
  For example,

      have_boost, have_readline, have_scotch, ...

  Each of the `have_<package>` scripts will generally provide the
  following type of functions:

      have_<package>          # detection
      no_<package>            # reset
      echo_<package>          # echoing

  and the following type of variables:

      HAVE_<package>          # unset or 'true'
      <package>_ARCH_PATH     # root for <package>
      <package>_INC_DIR       # include directory for <package>
      <package>_LIB_DIR       # library directory for <package>

  This simplifies the calling scripts:

      if have_metis
      then
          wmake metisDecomp
      fi

  As well as reducing clutter in the corresponding Make/options:

      EXE_INC = \
          -I$(METIS_INC_DIR) \
          -I../decompositionMethods/lnInclude

      LIB_LIBS = \
          -L$(METIS_LIB_DIR) -lmetis

  Any additional modifications (platform-specific or for an external build
  system) can now be made centrally.

- in many cases can just use lookupOrDefault("key", bool) instead of
  lookupOrDefault<bool> or lookupOrDefault<Switch> since reading a
  bool from an Istream uses the Switch(Istream&) anyhow

STYLE: relocated Switch string names into file-local scope

- sentinel was not working properly when building user-space routines

old "positions" file form

The change to barycentric-based tracking changed the contents of the
cloud "positions" file to a new format comprising the barycentric
co-ordinates and other cell position-based info.  This broke
backwards compatibility, providing no option to restart old cases
(v1706 and earlier), and caused difficulties for dependent code, e.g.
for post-processing utilities that could only infer the contents only
after reading.

The barycentric position info is now written to a file called
"coordinates" with provision to restart old cases for which only the
"positions" file is available. Related utilities, e.g. for parallel
running and data conversion have been updated to be able to support both
file types.

To write the "positions" file by default, use set the following option
in the InfoSwitches section of the controlDict:

    writeLagrangianPositions 1;