Commit Graph

3909 Commits

Author SHA1 Message Date
8268d8aaba BUG: redistributePar: avoid par comms. Fixes #1953 2020-12-10 15:43:28 +00:00
89f2cda3ab ENH: mpi: use per-application communicator. 2020-12-09 15:17:44 +00:00
be783632f2 ENH: isoSurfaceParams
- bundles selection and control parameters used when creating
  iso-surfaces. This simplifies selection and specification

- drop old compatibility handling of "cell" as a bool

- harmonize filter/regularisation flags for iso-surface

- for dictionary input, accept "isoMethod" and "isoAlgorithm" as being
  synonymous. Using "isoMethod" is less subject to typing errors.
2020-12-08 13:31:22 +01:00
df74e8448c ENH: robuster fileOperations splitProcessorPath
- robuster matching behaviour when encountering paths that themselves
  contain the word "processor" in them. For example,

    "/path/processor0generation2/case1/processor10/system"
    will now correctly match on processor10 instead of failing.

- use procRangeType for encapsulating the processor ranges

- provision for information of distributed vs non-distributed roots.
  The information is currently available from the initial setup, but
  can useful to access directly within fileOperation.

STYLE: modernize list iteration
2020-12-08 11:58:28 +01:00
b966b7cd4b ENH: static test methods for matching simple cell shapes
- (tet, pyr, hex) can be identified from their number of faces
  and vertices. For these common shapes can use static `test()`
  method instead of the virtual isA() method.

  This is much cheaper for calling on an individual basis since
  it avoids the overhead of constructing an object.

ENH: tetCell edge/reverseEdge (already had tetEdge)
2020-12-08 11:58:27 +01:00
b4eb214d56 BUG: transformPoints flagged zero or negative scaling as an error
- additional check was introduced by 51b2490258, but scaling with
  zero is useful for flattening a mesh
2020-11-26 01:04:55 +01:00
52c6642a4c Merge remote-tracking branch 'origin/master' into develop 2020-11-25 21:30:49 +01:00
7349b97ecc STYLE: use labelRange for identity 2020-11-25 19:53:02 +01:00
ae49b51182 ENH: improve offsets handling in extrudeToRegionMesh (#1933)
- non-uniform offsets are generated due to truncation errors,
  which can lead to problems later on (e.g. redistributePar).

  Detect if the offsets are close to being uniform.
2020-11-25 19:53:02 +01:00
07bbae0c55 STYLE: use brace constructors for Foam::zero and Foam::one
- makes it clearer that these are constructors and not function calls
2020-11-19 16:57:45 +01:00
6dd2cc808e CONFIG: provisional cuda rule. Update naming for CGAL rules 2020-11-17 21:29:07 +01:00
bed1305d96 BUG: extraneous brackets after Fluent cell types (fixes #1872)
ENH: limit output to 25 cell types per line for readability

- makes it easier to trace potential format errors etc

STYLE: downgrade warning about polyhedrals to a simple info message

- can assume that polyhedral support is widespread enough to not
  warrant a warning.
2020-10-29 11:21:39 +01:00
3b7100a9a1 BUG: PDRsetFields fails with non-orthogonal outer region (fixes #1907)
- first sort the cells into their ijk bins, and restrict testing for
  face orientation to those faces with an owner or neighbour that has
  an ijk bin.

ENH: ensure polyMesh from PDRblockMesh is marked as AUTO_WRITE

- the particular polyMesh constructor inherits the writeOpt, which
  makes is dependent on the caller and thus somewhat fragile for the
  top level caller.
2020-10-28 20:56:32 +01:00
f959c8eb36 ENH: generate local orientation vectors for blockMesh VTK output (#1903)
- can help when designing/debugging blockMesh layouts

- propagate low-level cellModel methods face() and edge() to cellShape

STYLE: relocate blockMesh OBJ output to application only

- remove blockTopology files in cleanCase function

- improve code consistency in top-level blockMesh, PDRblockMesh
  generation.
2020-10-28 16:04:12 +01:00
9eb73a92e2 TUT: fix use of "constant/triSurface" etc (#1887)
- use bracketed syntax (eg, "<constant>/triSurface") instead for
  implicit case resolution.
2020-10-28 16:04:12 +01:00
51b2490258 ENH: additional options for transformPoints and surfaceRefineRedGreen
- add -recentre option to shift points according to the bounding box
  centre before performing other operations

- add -auto-origin to use the bounding box centre as the origin for
  rotations. Has lower priority than the -origin option.

- surfaceTransformPoints now has distinct -read-scale, -write-scale
  options (as per surfaceMeshConvert). Silently accepts -scale as
  equivalent to -write-scale, for backwards compatibility and
  similarity with transformPoints

- add -steps option for surfaceRefineRedGreen for successive
  refinement
2020-10-22 14:59:12 +02:00
132ae101f6 COMP: avoid autoPtr copy assignment and implicit casting 2020-10-15 17:21:33 +02:00
f83bc8c22b Merge remote-tracking branch 'origin/master' into develop 2020-10-12 12:52:17 +02:00
9c7aa9727a ENH: add UPstream::parRun(const bool)
- slightly neater code when disabling/restoring parallel state

  New
      const bool oldParRun = Pstream::parRun(false);
      ...
      Pstream::parRun(oldParRun);

  Old
      const bool oldParRun = Pstream::parRun();
      Pstream::parRun() = false;
      ...
      Pstream::parRun() = oldParRun;
2020-10-12 12:51:22 +02:00
c1ad429564 BUG: parallel foamToEnsight fails depending on lagrangian order (fixes #1873)
- the problem arises when output fields are missing on some
  processors.

  When the information is combined, the resulting HashTables can have
  different insertion orders. This poses an issue when there are hash
  key collisions and thus different chaining.

- Use sorted order.
2020-10-08 18:23:54 +02:00
8939a55653 ENH: relocate blockMesh polyMesh generation into library (for code reuse)
STYLE: adjust blockMesh advanced/non-advanced options

- make -merge-points "non-advanced" (for better exposure)
- make -write-obj "advanced" (-write-vtk is preferred)
2020-10-05 12:36:04 +02:00
5dc5ea928a ENH: add UPstream::subProcs() static method
- returns a range of `int` values that can be iterated across.
  For example,

      for (const int proci : Pstream::subProcs()) { ... }

  instead of

      for
      (
          int proci = Pstream::firstSlave();
          proci <= Pstream::lastSlave();
          ++proci
      )
      {
          ...
      }
2020-09-28 14:26:07 +02:00
e18ff114a6 ENH: add UPstream::allProcs() method
- returns a range of `int` values that can be iterated across.
  For example,

      for (const int proci : Pstream::allProcs()) { ... }

  instead of

      for (label proci = 0; proci < Pstream::nProcs(); ++proci) { ... }
2020-09-28 14:25:59 +02:00
1e95af4d57 STYLE: use more generic terms allow/deny for selections 2020-09-09 13:54:34 +02:00
9423d2bd83 CONFIG: improve support for compiler/link options (#1830)
- 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)
2020-09-07 09:45:51 +02:00
1c71c7cb23 Merge remote-tracking branch 'origin/master' into develop 2020-09-03 12:04:45 +02:00
73057447a2 COMP: pass-through targetType for a few Allwmake files (#1824) 2020-09-03 11:35:18 +02:00
719d3f77be COMP: Intel compiler bug with inheriting constructors (fixes #1821)
- 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.
2020-09-01 11:49:27 +02:00
b2bded48c9 STYLE: use Time::printExecutionTime() method
- makes format of ExecutionTime = ... output configurable (#788)
  and reduces code clutter.

STYLE: more consistent line-breaks after "End" tag
2020-08-07 09:24:56 +02:00
42ce617b43 ENH: prefer 'model' keyword instead of 'RASModel',... (#149)
- since the context (laminar/RAS/LES) is already given by the
  sub-dictionary, it is redundant to use as prefix as well.

- silently support the longer names as compat methods
2020-08-05 15:29:03 +02:00
74a062d1d5 Merge remote-tracking branch 'origin/master' into develop 2020-07-28 08:42:12 +02:00
d5884c8d85 STYLE: use HashSet xor instead of two operations (faceZoneSet)
STYLE: use global operator instead of HashSet -= operator
2020-07-27 09:47:33 +02:00
e8b06ac222 STYLE: remove incorrect branch condition in setExprFields 2020-07-23 16:52:15 +02:00
bcda061f54 ENH: checkMesh: improved checking on writeFields. Fixes #1786. 2020-07-23 17:26:09 +01:00
12c91b9472 STYLE: check autoPtr as plain bool instead of valid()
- cleaner code, more similarity with unique_ptr

  Now
      if (ptr)
      if (!ptr)

  instead
      if (ptr.valid())
      if (!ptr.valid())
2020-07-16 11:39:24 +02:00
9af3f85cf9 STYLE: simplify short-circuit involving autoPtr (#1775)
- with '&&' conditions, often better to check for non-null autoPtr
  first (it is cheap)

- check as bool instead of valid() method for cleaner code, especially
  when the wrapped item itself has a valid/empty or good.
  Also when handling multiple checks.

  Now
      if (ptr && ptr->valid())
      if (ptr1 || ptr2)

  instead
      if (ptr.valid() && ptr->valid())
      if (ptr1.valid() || ptr2.valid())
2020-07-16 10:17:25 +02:00
3baebcb101 STYLE: replace uses of autoPtr::empty() with bool check (#1775)
- less clutter using plain tests with the bool operator:

      (!ptr)  vs  (ptr.empty())
      (ptr)   vs  (!ptr.empty())
2020-07-16 08:58:22 +02:00
5dc04530d9 COMP: additional linkage libraries for AMD compiler (#1627) 2020-07-14 12:17:09 +02:00
a86860430a BUG: checkMesh: index into surfaceScalarField. Fixes #1771. 2020-07-13 13:02:07 +01:00
331e86cf17 BUG: checkMesh: index into surfaceScalarField. Fixes #1771. 2020-07-13 13:00:13 +01:00
6d965d699b BUG: mergeOrSplitBaffles -dict option ignored (fixes #1759)
- regression from f721b5344f
2020-07-01 18:41:31 +02:00
538d749220 REL: Updated headers to version v2006 2020-06-29 17:27:54 +01:00
df7044845e GIT: remove whitespace, backslash escape angle brackets in comments
- fix older '> >' template parameters as '>>'
2020-06-26 12:28:49 +02:00
7c1c9fee0c ENH: Updated AMI dependent code 2020-06-24 22:10:20 +01:00
676a904c99 BUG: flipped cellZones compat setting for foamToEnsight (#1579)
STYLE: reword text for -excludePatches
2020-06-22 11:19:34 +02:00
8526e44884 ENH: dlLibraryTable InfoProxy output (#1735) 2020-06-22 10:22:45 +02:00
2dcbc09b3e ENH: add foamHasLibrary to test for availability of libraries (#1735)
- For some cases it can be helpful to test if additional libraries can
  be properly resolved.

  This can be useful in scripts to test for additional capability:

      if foamHasLibrary geometricVoF
      then ...
      fi

  But also directly from the command-line to help resolve configuration
  issues:

      foamHasLibrary -verbose petscFoam

  Could not load "petscFoam"
  libpetsc.so.3.13: cannot open shared object file: No such file or directory
2020-06-19 12:52:00 +02:00
b0136d835e ENH: lumped point motion using local linear basic functions (#1341)
- the earlier implementation of externally controlled lumped point
  motion (see merge request !120 and OpenFOAM-v1706 release notes) was
  conceived for the motion of simple structures such as buildings or
  simple beams. The motion controller was simply defined in terms of
  an orientation axis and divisions along that axis.

  To include complex structures, multiple motion controllers are
  defined in terms of support points and connectivity.

  The points can have additional node Ids associated with them, which
  makes it easier to map to/from FEA models.

  OLD system/lumpedPointMovement specification
  --------------------------------------------

      //- Reference axis for the locations
      axis            (0 0 1);

      //- Locations of the lumped points
      locations       (0 0.05 .. 0.5);

  NEW system/lumpedPointMovement specification
  --------------------------------------------

      // Locations of the lumped points
      points
      (
          (0  0  0.00)
          (0  0  0.05)
          ...
          (0  0  0.50)
      );

      //- Connectivity for motion controllers
      controllers
      {
          vertical
          {
              pointLabels (0 1 2 3 4 5 6 7 8 9 10);
          }
      }

  And the controller(s) must be associated with the given
  pointDisplacement patch. Eg,

     somePatch
     {
         type            lumpedPointDisplacement;
         value           uniform (0 0 0);
         controllers     ( vertical );   // <-- NEW
     }

TUT: adjust building motion tutorial

- use new controllor definitions
- replace building response file with executable
- add updateControl in dynamicMeshDict for slowly moving structure
2020-06-17 15:16:27 +02:00
01ec92fd35 GIT: remove leading/trailing blank lines, trailing whitespace 2020-06-17 10:46:26 +02:00
c1c995d3fe COMP: fix some linkage issues for mingw
COMP: automatically add FlexLexer.h to MSwindows OSspecific

- useful, frequently forgotten step for cross-compiling
2020-06-16 23:56:41 +02:00