Merge branch 'master' of https://github.com/NCAR/CESM_postprocessing

Machines/cheyenne_modules

@@ -6,7 +6,7 @@ module load python/2.7.14
 module load intel/17.0.1
 module load ncarenv
 module load ncarcompilers
-module load mpt/2.15f
+module load mpt/2.19
 module load netcdf/4.6.1
 module load nco/4.7.4
 module load ncl/6.4.0

Machines/machine_postprocess.xml

@@ -18,7 +18,6 @@
       <module>module purge</module>
     </reset_modules>
     <modules>
-      <module>module load python/2.7.14</module>
       <module>module load intel/17.0.1</module>
       <module>module load ncarenv</module>
       <module>module load ncarcompilers</module>
@@ -140,11 +139,10 @@
       <module>module purge</module>
     </reset_modules>
     <modules>
-      <module>module load python/2.7.14</module>
       <module>module load intel/17.0.1</module>
       <module>module load ncarenv</module>
       <module>module load ncarcompilers</module>
-      <module>module load mpt/2.15f</module>
+      <module>module load mpt/2.19</module>
       <module>module load netcdf/4.6.1</module>
       <module>module load nco/4.7.4</module>
       <module>module load ncl/6.4.0</module>

Tools/ration_script

@@ -1,28 +1,35 @@
-#! /usr/bin/env bash
+#!/usr/bin/env bash
 #
 # template to activate the virtualenv, call post process program, deactivate virtualenv
 #
 
-#BSUB -n 4
-#BSUB -R "span[ptile=2]"
-#BSUB -q geyser
-#BSUB -N
-#BSUB -a poe
-#BSUB -J ration_script
-#BSUB -W 00:02
-#BSUB -P P93300606 
+#PBS -N ration
+#PBS -q regular
+#PBS -l select=1:ncpus=8:mpiprocs=8
+#PBS -l walltime=00:00:02
+#PBS -A P93300606
 
-. /glade/apps/opt/lmod/lmod/init/bash
+source /etc/profile.d/modules.sh
 
-export MP_LABELIO=yes
+export MPI_UNBUFFERED_STDIO=true
+export TMPDIR=$TMPDIR
 
-module load python/2.7.7
+module purge
 
-. /glade/p/work/aliceb/sandboxes/dev/postprocessing/cesm-env2/bin/activate
+## activate the virtualenv that contains all the non-bootstrapped dependencies
+cd /glade/work/aliceb/sandboxes/dev/postprocessing_new/cesm-env2/bin
+echo "Running from virtualenv directory:"
+pwd
+. activate
 
-module load mpi4py/2.0.0
+## load the boot-strap modules 
+##module load python/2.7.14
+module load intel/17.0.1
+module load ncarenv
+module load ncarcompilers
+module load mpt/2.19
 
-mpirun.lsf ./ration_example.py >> ./ration.log
+mpiexec_mpt dplace -s 1 /glade/work/aliceb/sandboxes/dev/postprocessing_new/Tools/ration_test.py >> /glade/work/aliceb/sandboxes/dev/postprocessing_new/Tools/ration.log
 
 status=$?
 echo $status

Tools/ration_script_geyser → Tools/ration_script_dav

@@ -1,24 +1,27 @@
 #!/bin/bash -l
 
-## test the mpi4py and ASAPPyTools utility on geyser with ncar_pylib virtualenv 
+## test the mpi4py and ASAPPyTools utility on DAV with ncar_pylib virtualenv 
 
-#SBATCH -t 00:05:00
 #SBATCH -n 4
 #SBATCH -N 2
 #SBATCH --ntasks-per-node=2
+#SBATCH -t 00:05:00
 #SBATCH -p dav
 #SBATCH -J ration_test
 #SBATCH -A P93300606
-#SBATCH -C geyser
 #SBATCH --mem 1G
 #SBATCH -e ration_test.err.%J
 #SBATCH -o ration_test.out.%J
 
-export MP_LABELIO=yes
-
+module purge
 module load python/2.7.14
+module load intel/17.0.1
+module load ncarenv
+module load ncarcompilers
+module load impi
+
 
-. /glade2/work/aliceb/sandboxes/dev/postprocessing_geyser/cesm-env2/bin/activate
+. /gpfs/fs1/work/aliceb/sandboxes/dev/postprocessing_dav/cesm-env2/bin/activate
 
 srun ./ration_test.py >> ./ration.log
 

Tools/ration_test.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python2
 import sys
 
 try:

diagnostics/diagnostics/ocn/ocn_avg_generator.py

@@ -358,7 +358,7 @@ def createClimFiles(start_year, stop_year, in_dir, htype, tavgdir, case, tseries
         if len(averageListMoc) > 0:
             # call the pyAverager with the inVarList
             if 'MOC' in inVarList:
-                tmpInVarList = ['MOC']
+                tmpInVarList = ['MOC','SALT', 'TEMP']
             else:
                 tmpInVarList = ['SALT', 'TEMP']
             if main_comm.is_manager():

timeseries/timeseries/cesm_tseries_generator.py

@@ -184,7 +184,7 @@ def readArchiveXML(caseroot, input_rootdir, output_rootdir, casename, standalone
                             log[comp+stream] = {'slices':[],'index':0}
                         ts_log_dates = log[comp+stream]['slices']
                         index = log[comp+stream]['index']
-                        files,dates,index = chunking.get_chunks(tper, index, size_n, stream_dates, ts_log_dates, cal, units, completechunk)
+                        files,dates,index = chunking.get_chunks(tper, index, size_n, stream_dates, ts_log_dates, cal, units, completechunk, tseries_tper)
                         for d in dates:
                             log[comp+stream]['slices'].append(float(d))
                         log[comp+stream]['index']=index

timeseries/timeseries/chunking.py

@@ -13,7 +13,6 @@ def num2date(time_value, unit, calendar):
         time_value = int(round(time_value))
     if ('common_year' in unit):
         my_unit = unit.replace('common_year', 'day')
-##        my_time_value = time_value * 365
         my_time_value = int(round(time_value)) * 365
     else:
         my_unit = unit
@@ -130,7 +129,7 @@ def get_input_dates(glob_str, comm, rank, size):
     comm.sync()
     return g_stream_dates,g_file_slices,calendar.lower(),units,time_period_freq
 
-def get_cesm_date(fn,t=None):
+def get_cesm_date(fn,tseries_tper,t=None):
 
     '''
     Open a netcdf file and return its datestamp
@@ -156,28 +155,42 @@ def get_cesm_date(fn,t=None):
            # for the first lnd and rof file
            if ( -1.0 < d < 0.0):
                d = 0
-        if t == 'bb':
+        elif t == 'bb':
            d = f.variables[att['bounds']][0][0]
            # for the first lnd and rof file
            if ( -1.0 < d < 0.0):
                d = 0
            elif(d > 1):
-               d = d = f.variables[att['bounds']][0][1]
+               d = f.variables[att['bounds']][0][1]
         elif t == 'e':
            l = len(f.variables[att['bounds']])
            d = (f.variables[att['bounds']][l-1][1])-1
         elif t == 'ee':
            l = len(f.variables[att['bounds']])
            d = (f.variables[att['bounds']][l-1][1])
+
+        # problem if global attr time_period_freq does not exist in the nc file
+        if 'time_period_freq' in f.ncattrs():
+            if 'month' in f.time_period_freq:
+                if t=='bb' or t=='b':
+                    d = (f.variables[att['bounds']][0][0] + f.variables[att['bounds']][0][1]) / 2
+                if t=='ee' or t=='e':
+                    l = len(f.variables[att['bounds']])
+                    d = (f.variables[att['bounds']][l-1][0]+f.variables[att['bounds']][l-1][1])/2
+        elif 'month' in tseries_tper:
+            if t=='bb' or t=='b':
+                d = (f.variables[att['bounds']][0][0] + f.variables[att['bounds']][0][1]) / 2
+            if t=='ee' or t=='e':
+                l = len(f.variables[att['bounds']])
+                d = (f.variables[att['bounds']][l-1][0]+f.variables[att['bounds']][l-1][1])/2
+
     else:
         # problem if time has only one value when units are common_year
         try:
             d = f.variables['time'][1]    
         except:
             d = f.variables['time'][0]    
 
-
-##    d1 = cf_units.num2date(d,att['units'],att['calendar'].lower())
     d1 = num2date(d,att['units'],att['calendar'].lower())
     f.close()
 
@@ -202,7 +215,6 @@ def get_chunk_range(tper, size, start, cal, units):
     '''
 
     # Get the first date
-##    d1 = cf_units.num2date(start, units, cal)
     d1 = num2date(start, units, cal)
 
     # Figure out how many days each chunk should be
@@ -222,17 +234,15 @@ def get_chunk_range(tper, size, start, cal, units):
             y2 = y2 + 1
             m2 = m2 - 12
         d2 = datetime.datetime(y2, m2, d1.day, d1.hour, d1.minute)
-##        end = cf_units.date2num(d2, units, cal)
         end = date2num(d2, units, cal)
 
     elif 'year' in tper: #year
         d2 = datetime.datetime(int(size)+d1.year, d1.month, d1.day, d1.hour, d1.minute)
-##        end = cf_units.date2num(d2, units, cal)
         end = date2num(d2, units, cal)
 
     return start, end 
 
-def get_chunks(tper, index, size, stream_dates, ts_log_dates, cal, units, s):
+def get_chunks(tper, index, size, stream_dates, ts_log_dates, cal, units, s, tseries_tper):
 
     '''
     Figure out what chunks there are to do for a particular CESM output stream
@@ -249,6 +259,7 @@ def get_chunks(tper, index, size, stream_dates, ts_log_dates, cal, units, s):
     units(string) - the units to use to figure out chunk size
     s(string) - flag to determine if we need to wait until we have all data before we create a chunk or
                 if it's okay to do an incomplete chunk
+    tseries_tper - time_period_freq read from XML rather than nc file
 
     Output:
     files(dictionary) - keys->chunk, values->a list of all files needed for this chunk and the start and end dates 
@@ -307,10 +318,10 @@ def get_chunks(tper, index, size, stream_dates, ts_log_dates, cal, units, s):
                         files[chunk_n] = {}
                         files[chunk_n]['fn'] = sorted(cfiles)
                         if chunk_n > 0:
-                            files[chunk_n]['start'] = get_cesm_date(cfiles[0],t='bb')
+                            files[chunk_n]['start'] = get_cesm_date(cfiles[0],tseries_tper,t='bb')
                         else:
-                            files[chunk_n]['start'] = get_cesm_date(cfiles[0],t='b')
-                        files[chunk_n]['end'] = get_cesm_date(cfiles[-1],t='e')
+                            files[chunk_n]['start'] = get_cesm_date(cfiles[0],tseries_tper,t='b')
+                        files[chunk_n]['end'] = get_cesm_date(cfiles[-1],tseries_tper,t='e')
                         for cd in sorted(cdates):
                             dates.append(cd)
                     e = True
@@ -319,8 +330,11 @@ def get_chunks(tper, index, size, stream_dates, ts_log_dates, cal, units, s):
                 files[chunk_n] = {}
                 s_cdates = sorted(cdates)
                 files[chunk_n]['fn'] = sorted(cfiles)
-                files[chunk_n]['start'] = get_cesm_date(cfiles[0],t='bb')  
-                files[chunk_n]['end'] = get_cesm_date(cfiles[-1],t='ee') 
+                if chunk_n > 0:
+                    files[chunk_n]['start'] = get_cesm_date(cfiles[0],tseries_tper,t='bb')  
+                else:
+                    files[chunk_n]['start'] = get_cesm_date(cfiles[0],tseries_tper,t='b') 
+                files[chunk_n]['end'] = get_cesm_date(cfiles[-1],tseries_tper,t='ee') 
                 for cd in sorted(cdates):
                     dates.append(cd)
             chunk_n = chunk_n+1