sergio authored 4 years ago and Andrew Heather committed 4 years ago

1) Adding LiquidEvapFuchsKnudsen model for lagrangian evaporation.
   This models is based on a diffusion type of evaporation/
   condensation on particles composed of solution (liquid + solid).

2) Adding modes of calculating the particle rho and volume change.
   The new keyword in constantProperties is 'volumeUpdateMethod'
   which three options:
        a) constantRho
        b) constantVolume
        c) updateRhoAndVol

   The old keyword 'constantVolume' true/face is still valid

3) The entry rho0 is now optional for multicomponent parcels.
   If defined , it is used, but if it is not the actual mixture
   provided is used to calculate rho0 of the particle.
   T0 is still used as initial T and Cp0 is over-written in the
   multicomponent cloud but still required.

4) Adding tutorial for evaporation/condensation model

BUG: faMatrix::residual changes source vector (fixes #1835)

ENH: improve code alignment between faMatrix and fvMatrix

- support setValues() with a single value

- refactor UniformField accordingly

- can use ListOps::appendEqOp as the more general form.
  Note that this uses a different template parameter.
  Eg,
      `globalMeshData::ListPlusEqOp<labelList>()`
  vs. `ListOps::appendEqOp<label>()`

- select default nastran PLOAD2 or PLOAD4 based on field type.
  Default to PLOAD2 for scalar types and PLOAD4 for vectors etc.

- relocate nastran SHELL/MAT cards.
  Previously wrote at the end of the file, now emit when writing the
  geometry itself. This improves modularity (of code and files)

- initial support for common geometry file for nastran

- abstracted out from ensight surface writer for potential reuse by
  other surface writers.

- for CAE formats such as abaqus, nastran, starcd, etc, the element id
  is already part of the output format itself. For these cases, there
  is no use in generating an additional "Ids" field.

ENH: add code to ignore negative face ids

- these will arise from very special cases, such as when a
  solid element and side are encoded into a single integer.

BUG: starcd surface values output did not use original face ids

When truncating layers in one go (nBufferCellsNoExtrude -1) it
should set the nCells per face to be the min of the number of points
per edge.

- slight speed gain for recompilation and provisions for future
  refactoring

- introduce WM_COMPILE_CONTROL variable to convey control information
  into the build rules.

  The convention (as per spack):
      - '+' to select a feature
      - '~' to deselect a feature

  Eg, to select the gold linker, and disable openmp
  (spaces are not required):

      WM_COMPILE_CONTROL="+gold ~openmp"

CONFIG: accept FOAM_EXTRA_LDFLAGS for AMD, gold, Mingw linkers

CONFIG: generalize PROJECT_LIBS (-ldl used almost universally)

- simplifies cases where Tuple2 is used as a Pair replacement
  (for output format reasons)

- originally (incorrectly) was a Field(0.0), which was generalized
  to Field(Zero), but Field() is the correct form

STYLE: rename 'mustGet' to more standard 'mandatory' variable

- read header info once and reuse
- short-circuit logic to avoid unneeded checks
- additional debug information if field cannot be found

STYLE: remove unused old code remnants from #1206

- the compiler reports

      Internal error loop: assertion failed:
      find_seq_in_lookup_table: seq_number not found
      (shared/cfe/edgcpfe/il.c, line 4118)

  Seems to be the same as the bug report
  https://community.intel.com/t5/Intel-C-Compiler/Internal-error-loop-assertion-failed-find-seq-in-lookup-table/td-p/1087603

This _should_ be fixed in icc 17.0 update 1, but appears to have
struck here as well.

- workaround: explicitly construct member elements.

- slipped in with changes to csvTableReader (commit 59ed3ba1) so
  only affects the 2006 version.

- adjust constructor to expect "componentColumns", but also accept
  "valueColumns" as 1912 and earlier-compatibility. This not only
  fixes the reported bug, but also ensure proper compatibility with
  older files.

ENH: use "refColumn" instead of "timeColumn" for csvTableReader

- consistent with the CSV Function1.
  Support 'timeColumn' as 1912 and earlier-compatibility.

TUT: remove unused table-reader entry

- experienced while reusing src/Pstream/Allwmake-mpi to create
  additional mpi-layers after installation. Since the copied sources
  are not located within the OpenFOAM source-tree (and/or the
  source-tree is non-writable), it should not and does not use the
  central build/WM_OPTIONS directory.

  However, when exploring for the appropriate local Make directory, it
  searched for the current '.' directory instead of checking for the
  resolved directory.

  This fails, since there is no src/Pstream/Make directory.
  Must check for src/Pstream/mpi/Make directory first!

- Adjust wclean to always remove a local build directory
  (Make/WM_OPTIONS) for additional safety.
  After which, attempt to remove central build/WM_OPTIONS version too.

  Script to copy/create mingw run-time installation from the Linux
  cross-compilation. Packs everything into a tar or a zip file.

  To accommodate Windows, all .dll files are also placed in the
  platforms bin/ directory where they are easily found via the PATH.

  Futhermore, ThirdParty dll files (including those from mingw itself)
  are also placed in the same directory.

---
Steps
  - bundles common files and directories (bin, etc, META-INFO, ...)

  - copies .exe files from FOAM_APPBIN and .dll files from FOAM_LIBBIN
    to the new target platforms/win64MingwDPInt32Opt/bin.

  - copies mingw sys-root .dll files to the new target
    platforms/win64MingwDPInt32Opt/bin

  - copies other ThirdParty dll files (scotch, fftw, etc) to
    platforms/win64MingwDPInt32Opt/bin

  - copies tutorials (can be deactivated)

---
Note
    Can only be called when the linux64Mingw environment is active.

- handle relative directory names for FOAM_CONFIG_ETC

- makes it easier to use in combination with various 'New' selectors,
  which mostly return an autoPtr.

ENH: add very simple FFT test

- basic sanity test that the library links properly

- previously hidden as Detail::[IO]FstreamAllocator, now exposed
  directly as [io]fstreamPointer, which allows reuse for
  std::ifstream, std::ofstream wrapping, without the additional
  ISstream, OSstream layers.

  These stream pointers have some characteristics similar to a
  unique_ptr.

- restrict direct gzstream usage to two files (fstreamPointers.C,
  gzstream.C) which improves localization and makes it simpler to
  enable/disable with the `HAVE_LIBZ` define.

  The HAVE_LIBZ define is currently simply hard-coded in the
  Make/options.

  If compiled WITHOUT libz support:
    - reading gz files : FatalError
    - writing gz files : emit warning and downgrade to uncompressed
    - warn if compression is specified in the case controlDict
      and downgrade to uncompressed

ENH: minor updates to gzstream interface for C++11

- support construct/open with std::string for the file names.

CONFIG: provisioning for have_libz detection as wmake/script

- makes format of ExecutionTime = ... output configurable (#788)
  and reduces code clutter.

STYLE: more consistent line-breaks after "End" tag

- WM_ARCH=win64 to indicate the runtime environment

ENH: cross-compile wmake toolchain for the target-side as well

- silence failed restoration of controlDict from controlDict.orig
  (idempotent?)

- the various information queries MUST be executed with
  the '--no-print-directory' or risk polluting values
  in the information queries.

  This is mostly seen with the 'canCompile' test for tutorials running
  in parallel.

ENH: finer granularity for handling functionObject failure (#1779)

See merge request !380

- the various information queries MUST be executed with
  the '--no-print-directory' or risk polluting values
  in the information queries.

  This is mostly seen with the 'canCompile' test for tutorials running
  in parallel.

- additional "errors" entry with enumerated values
  (default|warn|ignore|strict) for defining warning or error at
  construct or runtime stage

- default : construct = warn, runtime = fatal
- warn    : construct = warn, runtime = warn
- ignore  : construct = silent, runtime = silent
- strict  : construct = fatal, runtime = fatal

The errors control can be added at the top-level and/or for individual
function objects.