Subaxis setup #1056
Subaxis setup #1056
Conversation
…o for the axis, adds subroutines that dumps the contents of each of the objects
| endif | ||
| diag_axis => this | ||
| type is (fmsDiagSubAxis_type) | ||
| sub_axis_name = this%subaxis_name//"_sub01" |
There was a problem hiding this comment.
Why is this "01"? Should these be numbered?
There was a problem hiding this comment.
Why isn't this just added to the subaxis name? It gets used in the write too.
There was a problem hiding this comment.
The "_01" is the convention from the previous diag manager, and PP tools expect it to have it.
but I am going to just add it to this%subaxis_name, not sure what i was thinking
| lat(2) = maxval(corners(:,2)) | ||
| end select | ||
|
|
||
| if (is_cube_sphere) then |
There was a problem hiding this comment.
Can you label the ifs and dos in this routine? There's a lot of them and they are large.
| @@ -13,7 +13,7 @@ module fms_diag_field_object_mod | |||
| use diag_data_mod, only: max_field_attributes, fmsDiagAttribute_type | |||
| use diag_data_mod, only: diag_null, diag_not_found, diag_not_registered, diag_registered_id, & | |||
| &DIAG_FIELD_NOT_FOUND | |||
| use mpp_mod, only: fatal, note, warning, mpp_error, mpp_pe, mpp_root_pe | |||
| use mpp_mod, only: fatal, note, warning, mpp_error, stdout, mpp_pe, mpp_root_pe | |||
There was a problem hiding this comment.
You added stdout but I don't think it's used.
| use mpp_mod, only: mpp_get_current_pelist, mpp_npes, mpp_root_pe, mpp_pe, mpp_error, FATAL, stdout | ||
| use time_manager_mod, only: time_type, operator(/=), operator(==), date_to_string |
There was a problem hiding this comment.
This use statement is very similar to one above. Can you either combine it or move it with the other use time_manager_mod
| @@ -613,9 +642,11 @@ subroutine dump_file_obj(this, unit_num) | |||
| end subroutine | |||
|
|
|||
| !< @brief Opens the diag_file if it is time to do so | |||
| subroutine open_diag_file(this, time_step) | |||
| subroutine open_diag_file(this, time_step, file_is_opened) | |||
There was a problem hiding this comment.
They're called history files, not diag files.
There was a problem hiding this comment.
lol we don't call them history files anywhere in the code ...
| end subroutine open_diag_file | ||
|
|
||
| !< @brief Writes the axis metadata for the file | ||
| subroutine write_metadata(this, diag_axis) |
There was a problem hiding this comment.
Is this specific for an axis? If it is, it should be called write_axis_metadata
| !TODO: closing the file here for now, just to see if it works | ||
| call close_file(fileobj) |
There was a problem hiding this comment.
This gets cleaned up in the next PR, once all of the time stuff is set up "correctly"
| @@ -211,15 +211,15 @@ integer function fms_register_diag_field_obj & | |||
| fileptr => this%FMS_diag_files(file_ids(i))%FMS_diag_file | |||
| call fileptr%add_field_id(fieldptr%get_id()) | |||
| call fileptr%set_file_domain(fieldptr%get_domain(), fieldptr%get_type_of_domain()) | |||
| call fileptr%add_axes(axes) | |||
| call fileptr%add_axes(axes, this%diag_axis, this%registered_axis) | |||
There was a problem hiding this comment.
Is axes the id of the axis?
There was a problem hiding this comment.
axes is an array of ids of the axis
diag_manager/fms_diag_object.F90
Outdated
| @@ -426,9 +427,16 @@ subroutine fms_diag_send_complete(this, time_step) | |||
| #else | |||
| class(fmsDiagFileContainer_type), pointer :: diag_file !< Pointer to this%FMS_diag_files(i) (for convenience) | |||
|
|
|||
| logical :: file_is_opened !< True if the file was opened in this time_step | |||
There was a problem hiding this comment.
This variable should be named something like file_opened_this_time_step so it's more descriptive.
|
@thomas-robinson I think this is ready for another round of review.. |
| @@ -29,14 +29,14 @@ module fms_diag_file_object_mod | |||
| get_instance_filename, open_file, close_file, get_mosaic_tile_file | |||
| use diag_data_mod, only: DIAG_NULL, NO_DOMAIN, max_axes, SUB_REGIONAL, get_base_time, DIAG_NOT_REGISTERED, & | |||
| TWO_D_DOMAIN, UG_DOMAIN, prepend_date, DIAG_DAYS, VERY_LARGE_FILE_FREQ | |||
| use time_manager_mod, only: time_type, operator(>), operator(/=), operator(==), get_date | |||
| use time_manager_mod, only: time_type, operator(>), operator(/=), operator(==), get_date, & | |||
| operator(/=), operator(==), date_to_string | |||
There was a problem hiding this comment.
you have the operator(/=), operator(==) twice in this use statement
* Adds fms_diag_time_utils.F90 and copies get_time_string and diag_time_inc * Use diag_time_inc in the fileobj * finish setting up the io to open the file * Adds documentation, non-domain decomposed files now only get written by the root PE * set up the time for the next open and some bug fixes * Adds code to setups the subaxis * Fix bugs in the code that get the starting and ending index of the subregion * fixes bugs in the code that sets up the subaxis object, updates the io for the axis, adds subroutines that dumps the contents of each of the objects Co-authored-by: Uriel Ramirez <uriel.ramirez@noaa.gov>
* Adds fms_diag_time_utils.F90 and copies get_time_string and diag_time_inc * Use diag_time_inc in the fileobj * finish setting up the io to open the file * Adds documentation, non-domain decomposed files now only get written by the root PE * set up the time for the next open and some bug fixes * Adds code to setups the subaxis * Fix bugs in the code that get the starting and ending index of the subregion * fixes bugs in the code that sets up the subaxis object, updates the io for the axis, adds subroutines that dumps the contents of each of the objects Co-authored-by: Uriel Ramirez <uriel.ramirez@noaa.gov>
Description
Fixes # (issue)
How Has This Been Tested?
CI
Checklist:
make distcheckpasses