fix the output file name and diag time #404

junwang-noaa · 2021-09-29T03:22:07Z

Description

The output file name does not have correct time stamp when the output time is not a integer. E.g. output_fh=0.25, the file name is currently atmf00.nc and sfcf00.nc. Also for random output time, the diag_time may not be updated when the output time is not multiple of fhzero. This PR contains fixes for the two issues.

Issue(s) addressed

Link the issues to be closed with this PR, whether in this repository, or in another repository.
(Remember, issues should always be created before starting work on a PR branch!)

fv3atm issue #403

Testing

How were these changes tested? Tested on hera.
What compilers / HPCs was it tested with?
Are the changes covered by regression tests? (If not, why? Do new tests need to be added?)
Have the ufs-weather-model regression test been run? On what platform?
Yes on hera.

Will the code updates change regression test baseline? If yes, why? Please show the baseline directory below.
No

Dependencies

If testing this branch requires non-default branches in other repositories, list them.
Those branches should have matching names (ideally)

ufs-weather-model PR
fv3atm PR

merge to latest develop

The HAFS related developments for the write_grid_component (NOAA-EMC#10)

change ifmin to zero at fh00 for inline POST (NOAA-EMC#14)

merge to latest develop

Update EMC dev/emc from NCAR dtc/develop 2019/12/17 (NOAA-EMC#30)

GFSv16 netcdf post ficein cpl (NOAA-EMC#48)

Regain bit-for-bit identical results between IPD and CCPP for coupled…

Merge NCAR:dtc/develop into develop 2020/04/14 (NOAA-EMC#98)

Port to Orion (NOAA-EMC#102)

Remove dynamic CCPP build (NOAA-EMC#103)

…ame_nonint

climbfuji

I don't understand the logic, especially in lines 763 to 772. Is it possible to document that code? Just a brief one liner what each of those blocks does would be great.

DusanJovic-NOAA · 2021-09-30T17:05:08Z

Why do need all this logic with lflname_fulltime, why can we just switch to using the filename format that always include HH-MM-SS, which will simplify the code substantially. Scripting will be also simplified by always using the same file naming scheme, that does not change based on the value output_fh.

junwang-noaa · 2021-09-30T17:08:38Z

Why do need all this logic with lflname_fulltime, why can we just switch to using the filename format that always include HH-MM-SS, which will simplify the code substantially. Scripting will be also simplified by always using the same file naming scheme, that does not change based on the value output_fh.

This is for backward compatible. Otherwise all the downstream jobs will be broken.

DusanJovic-NOAA · 2021-09-30T17:16:15Z

Why do need all this logic with lflname_fulltime, why can we just switch to using the filename format that always include HH-MM-SS, which will simplify the code substantially. Scripting will be also simplified by always using the same file naming scheme, that does not change based on the value output_fh.

This is for backward compatible. Otherwise all the downstream jobs will be broken.

Can't we announce the change and downstream jobs can update scripts. This is after all develop branch, we should be able to make breaking changes. Are we going to keep backward compatibility forever. Every change we make in any of the config file structure or output is potential breaking change for somebody.

climbfuji

Thanks for adding those comments. A few suggestions/corrections of typos, otherwise looks good.

climbfuji · 2021-09-30T17:34:55Z

fv3_cap.F90

@@ -746,6 +746,9 @@ subroutine InitializeAdvertise(gcomp, rc)
              endif
              do i=2,nfh
                output_fh(i) = (i-1)*outputfh2(1) + output_startfh
+                ! Except fh000, which is the first time output, if any other of the 
+                ! output time is not integer hour, set lflname_fulltime to be ture, so the 


Suggested change

! output time is not integer hour, set lflname_fulltime to be ture, so the

! output time is not integer hour, set lflname_fulltime to be true, so the

climbfuji · 2021-09-30T17:35:20Z

fv3_cap.F90

@@ -760,6 +763,10 @@ subroutine InitializeAdvertise(gcomp, rc)
             count=noutput_fh, rc=rc)
          if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return
          if( output_startfh == 0) then
+            ! if the output time in output_fh array contains first time stamp output,


Suggested change

! if the output time in output_fh array contains first time stamp output,

! If the output time in output_fh array contains first time stamp output,

climbfuji · 2021-09-30T17:36:17Z

fv3_cap.F90

@@ -760,6 +763,10 @@ subroutine InitializeAdvertise(gcomp, rc)
             count=noutput_fh, rc=rc)
          if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return
          if( output_startfh == 0) then
+            ! if the output time in output_fh array contains first time stamp output,
+            ! check the rest of output time, otherwise, check all the output time. 
+            ! if any of them is not integer hour, the history file names will


Suggested change

! if any of them is not integer hour, the history file names will

! If any of them is not integer hour, the history file names will

climbfuji · 2021-09-30T17:37:01Z

fv3_cap.F90

@@ -773,6 +780,9 @@ subroutine InitializeAdvertise(gcomp, rc)
          else
            do i=1,noutput_fh
              output_fh(i) = output_startfh + output_fh(i)
+              ! when output_startfh >0, check all the output time, if any of


Suggested change

! when output_startfh >0, check all the output time, if any of

! When output_startfh >0, check all the output time, if any of

climbfuji · 2021-09-30T17:37:11Z

fv3_cap.F90

@@ -773,6 +780,9 @@ subroutine InitializeAdvertise(gcomp, rc)
          else
            do i=1,noutput_fh
              output_fh(i) = output_startfh + output_fh(i)
+              ! when output_startfh >0, check all the output time, if any of
+              ! them is not integer hour, set lflname_fulltime to be ture. The 


Suggested change

! them is not integer hour, set lflname_fulltime to be ture. The

! them is not integer hour, set lflname_fulltime to be true. The

junwang-noaa · 2021-09-30T17:40:04Z

@climbfuji I added some comments to explain, please check. Basically for frequency output, we will don't count the the first time step output as it is labeled as fh000.

climbfuji · 2021-09-30T17:41:59Z

@climbfuji I added some comments to explain, please check. Basically for frequency output, we will don't count the the first time step output as it is labeled as fh000.

I did, please look at/accept my suggestions.

junwang-noaa · 2021-09-30T17:44:46Z

@DusanJovic-NOAA Well, this will impact future operational implementations. As we don't know if we will have higher frequency output, I think we may need to get the downstream involved before we make the hard code change. Right now if the output time is non-integer (needs min/sec info), the output file will have full time stamp, otherwise if it's all integer hours, file names will only contain hours.

DeniseWorthen and others added 30 commits November 18, 2019 11:11

Merge pull request #1 from NOAA-EMC/develop

9328c87

merge to latest develop

Merge pull request #2 from NOAA-EMC/develop

cb908e9

The HAFS related developments for the write_grid_component (NOAA-EMC#10)

Merge pull request #3 from NOAA-EMC/develop

c96eaef

change ifmin to zero at fh00 for inline POST (NOAA-EMC#14)

Merge pull request #4 from NOAA-EMC/develop

1067e95

merge to latest develop

Merge pull request NOAA-EMC#9 from NOAA-EMC/develop

c8abdc6

Update EMC dev/emc from NCAR dtc/develop 2019/12/17 (NOAA-EMC#30)

Merge pull request NOAA-EMC#10 from NOAA-EMC/develop

3e0e847

GFSv16 netcdf post ficein cpl (NOAA-EMC#48)

Merge pull request NOAA-EMC#12 from NOAA-EMC/develop

5b2831e

Merge pull request NOAA-EMC#13 from NOAA-EMC/develop

69b7a01

Merge pull request NOAA-EMC#15 from NOAA-EMC/develop

2a2b89a

Regain bit-for-bit identical results between IPD and CCPP for coupled…

Merge pull request NOAA-EMC#16 from NOAA-EMC/develop

0671b18

Merge pull request NOAA-EMC#17 from NOAA-EMC/develop

1e25e61

Merge NCAR:dtc/develop into develop 2020/04/14 (NOAA-EMC#98)

Merge pull request NOAA-EMC#18 from NOAA-EMC/develop

fce0b06

Port to Orion (NOAA-EMC#102)

Merge pull request NOAA-EMC#19 from NOAA-EMC/develop

e655629

Remove dynamic CCPP build (NOAA-EMC#103)

Merge remote-tracking branch 'upstream/develop' into develop

3df3898

Merge remote-tracking branch 'upstream/develop' into develop

7f89d1f

Merge remote-tracking branch 'upstream/develop' into develop

d3a9665

Merge remote-tracking branch 'upstream/develop' into develop

c2aae4e

Merge remote-tracking branch 'upstream/develop' into develop

a1e05e0

Merge remote-tracking branch 'upstream/develop' into develop

24d4fe6

Merge remote-tracking branch 'upstream/develop' into develop

ebfcf47

Merge remote-tracking branch 'upstream/develop' into develop

ff8c950

Merge remote-tracking branch 'upstream/develop' into develop

7e1489d

Merge remote-tracking branch 'upstream/develop' into develop

9619fe7

Merge remote-tracking branch 'upstream/develop' into develop

a87558f

Merge remote-tracking branch 'upstream/develop' into develop

d8e5772

Merge remote-tracking branch 'upstream/develop' into develop

484782a

Merge remote-tracking branch 'upstream/develop' into develop

5514ecc

Merge remote-tracking branch 'upstream/develop' into develop

7beaa31

Merge remote-tracking branch 'upstream/develop' into develop

624b4e2

Merge remote-tracking branch 'upstream/develop' into develop

e174f2a

junwang-noaa and others added 5 commits September 29, 2021 02:41

fix the diag time issue with output_fh

d2c8b71

Merge branch 'NOAA-EMC:develop' into feature/removesuites

0230f2b

fix lflname_fulltime

c69f676

fix lflname_fulltime for output time starting with 0

036e270

Merge remote-tracking branch 'upstream/develop' into filename_nonint

c3b08ce

junwang-noaa mentioned this pull request Sep 30, 2021

Global testfix DeniseWorthen/ufs-weather-model#18

Merged

14 tasks

DeniseWorthen and others added 3 commits September 30, 2021 10:45

Merge branch 'NOAA-EMC:develop' into feature/removesuites

bfb50fb

Merge remote-tracking branch 'upstream/develop' into filename_nonint

090b893

Merge remote-tracking branch 'denise/feature/removesuites' into filen…

d86c57a

…ame_nonint

junwang-noaa requested review from climbfuji, DusanJovic-NOAA and DeniseWorthen September 30, 2021 16:28

DeniseWorthen mentioned this pull request Sep 30, 2021

Update coupled tests to use P7 configuration, add standalone P7 test suite ufs-community/ufs-weather-model#765

Merged

27 tasks

climbfuji reviewed Sep 30, 2021

View reviewed changes

add doc for lflname_fulltime

439f485

climbfuji requested changes Sep 30, 2021

View reviewed changes

DusanJovic-NOAA approved these changes Sep 30, 2021

View reviewed changes

fix typo

8fe3d7a

climbfuji approved these changes Sep 30, 2021

View reviewed changes

DeniseWorthen approved these changes Oct 1, 2021

View reviewed changes

DeniseWorthen mentioned this pull request Oct 1, 2021

Remove unused coupled suites #394

Closed

junwang-noaa merged commit 799b157 into NOAA-EMC:develop Oct 1, 2021

AnningCheng-NOAA mentioned this pull request Oct 7, 2021

Update merra2_thompson from Dom AnningCheng-NOAA/fv3atm#1

Merged

junwang-noaa deleted the filename_nonint branch October 20, 2021 01:56

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix the output file name and diag time #404

fix the output file name and diag time #404

junwang-noaa commented Sep 29, 2021

climbfuji left a comment

DusanJovic-NOAA commented Sep 30, 2021

junwang-noaa commented Sep 30, 2021 •

edited

Loading

DusanJovic-NOAA commented Sep 30, 2021

climbfuji left a comment

climbfuji Sep 30, 2021

climbfuji Sep 30, 2021

climbfuji Sep 30, 2021

climbfuji Sep 30, 2021

climbfuji Sep 30, 2021

junwang-noaa commented Sep 30, 2021

climbfuji commented Sep 30, 2021

junwang-noaa commented Sep 30, 2021 •

edited

Loading

	! output time is not integer hour, set lflname_fulltime to be ture, so the
	! output time is not integer hour, set lflname_fulltime to be true, so the

	! if the output time in output_fh array contains first time stamp output,
	! If the output time in output_fh array contains first time stamp output,

	! if any of them is not integer hour, the history file names will
	! If any of them is not integer hour, the history file names will

	! when output_startfh >0, check all the output time, if any of
	! When output_startfh >0, check all the output time, if any of

	! them is not integer hour, set lflname_fulltime to be ture. The
	! them is not integer hour, set lflname_fulltime to be true. The

fix the output file name and diag time #404

fix the output file name and diag time #404

Conversation

junwang-noaa commented Sep 29, 2021

Description

Issue(s) addressed

Testing

Dependencies

climbfuji left a comment

Choose a reason for hiding this comment

DusanJovic-NOAA commented Sep 30, 2021

junwang-noaa commented Sep 30, 2021 • edited Loading

DusanJovic-NOAA commented Sep 30, 2021

climbfuji left a comment

Choose a reason for hiding this comment

climbfuji Sep 30, 2021

Choose a reason for hiding this comment

climbfuji Sep 30, 2021

Choose a reason for hiding this comment

climbfuji Sep 30, 2021

Choose a reason for hiding this comment

climbfuji Sep 30, 2021

Choose a reason for hiding this comment

climbfuji Sep 30, 2021

Choose a reason for hiding this comment

junwang-noaa commented Sep 30, 2021

climbfuji commented Sep 30, 2021

junwang-noaa commented Sep 30, 2021 • edited Loading

junwang-noaa commented Sep 30, 2021 •

edited

Loading

junwang-noaa commented Sep 30, 2021 •

edited

Loading