Add Libgmsh support/ flboundadapt #168
Conversation
tools/Flboundadapt.F90
Outdated
| call calculate_diagnostic_variables_new(states) | ||
|
|
||
| ! Find the external mesh field | ||
| !call find_mesh_field_to_adapt(states(1), old_mesh_field) |
There was a problem hiding this comment.
Commented out code.
Should this perhaps be topology_mesh_name ? The metric doesn't necessarily live on the CoordinateMesh: periodic, extruded, higher order, etc. - I realize not all of these combination may actually work with this tool of course. If Coordinate is the safest choice here, maybe add some comment why you think that is...
configure.in
Outdated
| @@ -1067,7 +1098,7 @@ fi | |||
|
|
|||
| ##### | |||
| # | |||
| LIBS="$LIBS -L./lib" | |||
| LIB="$LIBS -L./lib" | |||
| @@ -1337,4 +1337,25 @@ function extract_prognostic_velocity(state, stat) result(vfield) | |||
|
|
|||
| end function extract_prognostic_velocity | |||
|
|
|||
| subroutine get_surface_ids(bc_path, mesh, surface_ids) | |||
There was a problem hiding this comment.
Could you add a one line comment saying exactly what this routine does (or more if required).
femtools/Fields_Allocates.F90
Outdated
| @@ -1097,6 +1102,10 @@ function make_mesh (model, shape, continuity, name) & | |||
| size(mesh%faces%surface_node_list), name='Surface'//trim(mesh%name)) | |||
| #endif | |||
| end if | |||
| if (associated(model%surface_names)) then | |||
| allocate(mesh%surface_names(size(model%surface_names))) | |||
| mesh%surface_names(:) = model%surface_names(:) | |||
There was a problem hiding this comment.
Pedantic: spurious (:)
| @@ -4199,4 +4200,39 @@ subroutine write_minmax_tensor(tfield, field_expression) | |||
|
|
|||
| end subroutine write_minmax_tensor | |||
|
|
|||
| subroutine set_surface_names(mesh, names, ids) | |||
There was a problem hiding this comment.
Same here: "docstring" please
| end subroutine set_surface_names | ||
|
|
||
| subroutine get_surface_ids_from_names(mesh, names, ids) | ||
| type(mesh_type), intent(in) :: mesh |
femtools/Futils.F90
Outdated
| subroutine copy_c_string_to_fortran(c_string, f_string) | ||
|
|
||
| type(c_ptr), intent(in), target :: c_string | ||
| character(len=*) :: f_string |
femtools/Futils.F90
Outdated
| call c_f_pointer(c_string, arr, [strlen(c_string)]) | ||
| !!! make into a pointer to a Fortran string | ||
| call wrap_to_string(strlen(c_string), arr, f_ptr) | ||
| !!! copy happens here |
There was a problem hiding this comment.
Maybe we should add a check that len(f_string)<=len(c_string)
|
Some more queries on the new functionality to specify a .geo file as the input mesh:
|
|
To address the metaqueries:
|
|
Just a minor comment from me - would it be possible to update the manual to reflect these changes? |
…ing better support for standard libgmsh-dev on 16.04.
This change set gives the option to use functions from the libgmsh library to read and write .msh files and to read a .geo file directly. It also introduces a new tool flboundadapt which generates a .msh file from a .geo file using fluidity metric information taken from a .flml file (eg. a checkpoint file.)