Add some of the io for the field's metadata #1102

uramirez8707 · 2023-01-04T18:36:38Z

Description

Writes the field's metadata to the corresponding netcdf file. There is still more work needed for the diurnal variables and the uncompressed land files, but that will be done in a seperate PR

Fixes # (issue)

How Has This Been Tested?
CI

Checklist:

My code follows the style guidelines of this project
I have performed a self-review of my own code
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings
Any dependent changes have been merged and published in downstream modules
New check tests, if applicable, are included
make distcheck passes

uramirez8707 · 2023-01-05T18:23:34Z

diag_manager/fms_diag_object.F90

+
+!> @brief Loops through all the files, open the file, writes out axis and
+!! variable metadata and data when necessary.
+subroutine fms_diag_do_io(this, time_step, is_end_of_run)


@thomas-robinson
The behavior of this function will need to be updated #391 (comment)

But we should leave it the way it is now because it makes it easier to test (to check if all metadata is there). I will update this in the next, next PR

OK, we should also reach out to the scientists and make sure that they are ok with this behavior change

thomas-robinson · 2023-01-25T14:37:08Z

diag_manager/fms_diag_field_object.F90

@@ -220,9 +229,14 @@ subroutine fms_register_diag_field_obj &
 !> get the optional arguments if included and the diagnostic is in the diag table
  if (present(longname))      this%longname      = trim(longname)
  if (present(standname))     this%standname     = trim(standname)
-  if (present(units))         this%units         = trim(units)
+
+  !> Ignore the units if the file they are set to "none". This it to reproduce previous diag_manager


you have a typo. Should be "This is to reproduce previous diag_manager"

thomas-robinson · 2023-01-25T16:42:55Z

diag_manager/fms_diag_field_object.F90

+  character(len=120), intent(out), allocatable :: rslt(:)
+
+  integer :: i     !< For do loops
+  integer :: j     !< For do loops


This is not for do loops, it's the axis ID.

thomas-robinson · 2023-01-25T16:44:16Z

diag_manager/fms_diag_field_object.F90

+  allocate(rslt(naxis))
+
+  do i = 1, size(this%axis_ids)
+    j = this%axis_ids(i)


I suggest renaming this variable to axis_id or something more descriptive. I also think you can make it a "pointer for convenience"

This was replaced with a pointer 6d73416

thomas-robinson · 2023-01-25T17:41:40Z

diag_manager/fms_diag_file_object.F90

-      trim(time_var_name)//"_bounds", str_len=len_trim(time_var_name//"_bounds"))
+      trim(time_var_name)//"_bnds", str_len=len_trim(time_var_name//"_bnds"))


Is this to get it in like with the current?

yes, it should be _bnds

diag_manager/fms_diag_file_object.F90

thomas-robinson · 2023-01-25T17:46:42Z

diag_manager/fms_diag_file_object.F90

+  class(fmsDiagFile_type), pointer     :: diag_file      !< Diag_file object to open
+
+  integer :: i           !< For do loops
+  integer :: j           !< For do loops


This is a field ID not an index for looping. It should probably be a pointer.

This was replaced with a pointer 6d73416

thomas-robinson · 2023-01-25T17:47:04Z

diag_manager/fms_diag_file_object.F90

+    !TODO I think if the area and the volume field are no in the same file, a global attribute containing the
+    !the file that the fields are in needs to be added


Do we need to discuss this?

thomas-robinson · 2023-01-25T17:49:00Z

diag_manager/fms_diag_file_object.F90

+
+  do i = 1, size(diag_file%field_ids)
+    j = diag_file%field_ids(i)
+    if (.not. diag_file%field_registered(i)) cycle !TODO do something else here


Maybe instead of a do this could be a where

The where won't work because you are writing.

thomas-robinson · 2023-01-31T11:48:06Z

diag_manager/fms_diag_field_object.F90

  integer :: naxis !< Number of axis for the field
+  class(fmsDiagAxisContainer_type), pointer :: axis_ptr !diag_axis(this%axis_ids(i), for convenience


This is so much cleaner now.

thomas-robinson · 2023-01-31T11:50:13Z

diag_manager/fms_diag_field_object.F90

@@ -1120,6 +1121,51 @@ subroutine write_field_metadata(this, fileobj, file_id, yaml_id, diag_axis, unli
      trim(adjustl(cell_measures)), str_len=len_trim(adjustl(cell_measures)))

 end subroutine write_field_metadata
+!> @brief Gets a fields data buffer
+!! @return a pointer to the data buffer


We'll have to see if there are any issues with returning a pointer down the road.

thomas-robinson · 2023-01-31T12:10:34Z

diag_manager/fms_diag_file_object.F90

+
+  do i = 1, size(diag_file%field_ids)
+    j = diag_file%field_ids(i)
+    if (.not. diag_file%field_registered(i)) cycle !TODO do something else here


The where won't work because you are writing.

Add some of the io for the field's metadata

…FDL#1102)

uramirez8707 added 2 commits January 4, 2023 13:39

Add some of the io for the field's metadata

dde1ce4

Fix typos and add some missing documentation

bf61ff1

uramirez8707 marked this pull request as ready for review January 4, 2023 20:00

uramirez8707 requested a review from thomas-robinson as a code owner January 4, 2023 20:00

uramirez8707 commented Jan 5, 2023

View reviewed changes

uramirez8707 mentioned this pull request Jan 23, 2023

Accept data #1088

Merged

8 tasks

thomas-robinson requested changes Jan 25, 2023

View reviewed changes

uramirez8707 added 2 commits January 30, 2023 11:23

Merge branch 'dmUpdate' of github.com:NOAA-GFDL/FMS into most_field_io

2423838

fix some typos, add some pointers for convenience

6d73416

thomas-robinson approved these changes Jan 31, 2023

View reviewed changes

thomas-robinson merged commit af0b3ca into NOAA-GFDL:dmUpdate Jan 31, 2023

rem1776 pushed a commit to rem1776/FMS that referenced this pull request May 1, 2024

Add some of the io for the field's metadata (NOAA-GFDL#1102)

662d817

Add some of the io for the field's metadata

rem1776 pushed a commit to rem1776/FMS that referenced this pull request May 1, 2024

Add some of the io for the field's metadata (NOAA-GFDL#1102)

27263bb

Add some of the io for the field's metadata

rem1776 pushed a commit to rem1776/FMS that referenced this pull request May 1, 2024

feat: modern diag Add some of the io for the field's metadata (NOAA-G…

8a9c47f

…FDL#1102)

rem1776 pushed a commit to rem1776/FMS that referenced this pull request May 1, 2024

feat: modern diag Add some of the io for the field's metadata (NOAA-G…

9c62d12

…FDL#1102)

rem1776 pushed a commit to rem1776/FMS that referenced this pull request May 1, 2024

feat: modern diag Add some of the io for the field's metadata (NOAA-G…

4c736d4

…FDL#1102)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add some of the io for the field's metadata #1102

Add some of the io for the field's metadata #1102

uramirez8707 commented Jan 4, 2023 •

edited

Loading

uramirez8707 Jan 5, 2023

thomas-robinson Jan 25, 2023

thomas-robinson Jan 25, 2023

thomas-robinson Jan 25, 2023

thomas-robinson Jan 25, 2023

uramirez8707 Jan 30, 2023

thomas-robinson Jan 25, 2023

uramirez8707 Jan 30, 2023

thomas-robinson Jan 25, 2023

uramirez8707 Jan 30, 2023

thomas-robinson Jan 25, 2023

thomas-robinson Jan 25, 2023

thomas-robinson Jan 31, 2023

thomas-robinson Jan 31, 2023

thomas-robinson Jan 31, 2023

thomas-robinson Jan 31, 2023

		trim(time_var_name)//"_bounds", str_len=len_trim(time_var_name//"_bounds"))
		trim(time_var_name)//"_bnds", str_len=len_trim(time_var_name//"_bnds"))

		!TODO I think if the area and the volume field are no in the same file, a global attribute containing the
		!the file that the fields are in needs to be added

		integer :: naxis !< Number of axis for the field
		class(fmsDiagAxisContainer_type), pointer :: axis_ptr !diag_axis(this%axis_ids(i), for convenience

Add some of the io for the field's metadata #1102

Add some of the io for the field's metadata #1102

Conversation

uramirez8707 commented Jan 4, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

uramirez8707 commented Jan 4, 2023 •

edited

Loading