vyas-group-scripts 0.3.0__py3-none-any.whl

vyas_group_scripts/__init__.py

@@ -0,0 +1,4 @@
+from . import gen_batch
+from .templates import g16_batch_inputs
+from .templates import orca_batch_inputs
+from . import run_batch

vyas_group_scripts/gen_batch.py

@@ -0,0 +1,161 @@
+from glob import glob
+from .templates.g16_batch_inputs import simple_g16_template, complex_g16_template
+from .templates.orca_batch_inputs import orca_template, orca_restart_template
+import os
+import argparse
+
+
+
+def process_g16_files(file_pattern="*.gjf",complex = False ,time=os.environ.get("DEFAULT_DFT_TIME") ,num_nodes=os.environ.get("NUM_NODES") ,benchmarking=False):
+
+    benchmarking_line = ""
+    batch_file = simple_g16_template
+    if complex:
+        batch_file = complex_g16_template
+    if benchmarking:
+        benchmarking_line = "\n#SBATCH --exclusive"
+
+    for gjf_file in glob(file_pattern):
+        if ".gjf" !=  gjf_file[-4:]:
+            continue
+        
+        bash_file_name = gjf_file[:-4]+".s"
+        print(bash_file_name)
+        if  os.path.exists(bash_file_name):
+            print(f"{bash_file_name} already exists, skipping.")
+            continue
+
+        this_file_string = batch_file.format(TIME=time,
+                                             ACTNUM=os.environ.get("ACTNUM"),
+                                             JOB_NAME=gjf_file[:-4],
+                                             TASKS_PER_NODE=1,
+                                             N_TASKS=num_nodes,
+                                             INPUT_FILE_NAME=gjf_file.split("/")[-1][:-4],
+                                             NUM_NODES=num_nodes,
+                                             BENCHMARKING_LINE = benchmarking_line)
+        with open(bash_file_name,"w") as bash_file:
+            bash_file.write(this_file_string)
+        
+        apply_parallelization(gjf_file,num_nodes)
+
+
+
+def process_inp_files(file_pattern="*.inp",complex = False ,time=os.environ.get("DEFAULT_DFT_TIME") ,num_nodes=os.environ.get("NUM_NODES") ,benchmarking=False,restart=False):
+
+    # Select the bash template
+    batch_file = orca_template
+
+    # Set Benchmarking line
+    benchmarking_line = ""
+    if benchmarking:
+        benchmarking_line = "\n#SBATCH --exclusive"
+
+    # loop through input files
+    for inp_file in glob(file_pattern):
+        if ".inp" != inp_file[-4:]:
+            continue
+        bash_file_name = inp_file[:-4]+".s"
+
+        if  os.path.exists(bash_file_name):
+            print(f"{bash_file_name} already exists, skipping.")
+            continue
+
+
+        # Modify bash script 
+        this_file_string = batch_file.format(TIME=time,
+                                             ACTNUM=os.environ.get("ACTNUM"),
+                                             JOB_NAME=inp_file[:-4],
+                                             TASKS_PER_NODE=1,
+                                             N_TASKS=num_nodes,
+                                             INPUT_FILE_NAME=inp_file.split("/")[-1][:-4],
+                                             NUM_NODES=num_nodes,
+                                             BENCHMARKING_LINE = benchmarking_line)
+        with open(bash_file_name,"w") as bash_file:
+            bash_file.write(this_file_string)
+        
+        # Add parallelization to input file
+        apply_parallelization(inp_file,num_nodes)
+
+
+def apply_parallelization(input_file,nprocs = 36):
+
+    if ".gjf" == input_file[-4:]:
+        if nprocs != 0:
+            print(f"Adding parallelization to {input_file}.")
+            with open(input_file,"r") as file:
+                file_string = file.read()
+            file_string = f"%mem=5GB\n%nprocshared={nprocs}\n"+file_string
+
+            with open(input_file,"w") as file:
+                file.write(file_string)
+        
+    if ".inp" ==  input_file[-4:]:
+        if nprocs != 0:
+            if not nprocs in [6,10,30,36]:
+                print("invalid nprocs selected.")
+                return None
+            memory = int(144000/nprocs)
+            file_string = f"%maxcore {memory}\n%pal nprocs {nprocs} end\n"
+            with open(input_file,"r") as file:
+                file_list = file.readlines()
+            lastinput_line = 1
+            for index,line in enumerate(file_list): 
+                if len(line.strip())>0 and line.strip()[0] == "!":
+                    lastinput_line = index
+            file_list.insert(lastinput_line+1,file_string)
+            with open(input_file,"w") as file:
+                file.writelines(file_list)
+
+
+
+        
+
+
+
+
+
+
+
+def gen_all_batch():
+
+    parser = argparse.ArgumentParser(prog="gab")
+    parser.add_argument("-t",default=os.environ.get("DEFAULT_DFT_TIME"),help="Set the time limit for the batch script (format: HH:MM:SS)." \
+    " If not specified, the default from the DFT_DEFAULT_TIME environment variable is used.")
+    parser.add_argument("-b",action="store_true",help="Enable benchmarking mode. This sets the node allocation" \
+    " to exclusive to ensure consistent performance for benchmarking runs.")
+    parser.add_argument("-k",default="*",help="Provide a comma seperated regex pattern to match input files. " \
+    "Only batch files for matching input files will be generated.\nExample: -k 'xyz,*mol' (quotes reccomended)")
+    parser.add_argument("-p",default=-1,type=int,choices=[0,6,10,30,36],help="Specify the number of processors to use"\
+    " Acceptable values 36, 30, 10, 6, or 0." \
+    " Memory sttings will be adjusted based on this value. Use 0 to skip adding parallelization directives entirely." \
+    " When the -p flag is not called the default options in DFT_DEFAULT_MEM and DFT_DEFAULT_PROCESSORS environment variables" \
+    " are used for all calculations.\n\tOrca jobs:\n\t\tInsert content after first '!' in input file.\n\tGaussian Jobs:\n\t\tAppends" \
+    " parralleliszation options to the top of input file\n\t\tOnly '36' and '0' options are available for gaussian jobs.")
+    parser.add_argument("-r",default="*",help="Restart ORCA jobs from one or more specified folders. Provide a comma-seperated list of" \
+    " folder names.\n\tExample: -r 'job1_TIMEOUT,job2_TIMEOUT")
+    parser.add_argument("-v",default=None,help="Specify the quantum chemistry software version to use." \
+    " Currently only downgrading to ORCA 5 is supported.")
+    parser.add_argument("-c",action="store_false",help="Uses a more simple .s file for gaussian jobs that does not create a bunch of subdirectories.")
+    args = parser.parse_args()
+
+    patterns_to_search = args.k.split(",")
+    complex_choice = args.c
+    time_choice = args.t
+    processors_choice = args.p
+    benchmarking_choice = args.b
+    orca_version = args.v
+    restarts = args.r
+    if processors_choice == -1:
+        processors_choice = os.environ.get("DEFAULT_DFT_NUM_NODES")
+
+    for file_pattern in patterns_to_search:
+
+        for file in glob(file_pattern):
+
+            process_g16_files(file,complex_choice,time_choice,processors_choice,benchmarking_choice)
+            process_inp_files(file,complex_choice,time_choice,processors_choice,benchmarking_choice)
+
+
+
+if __name__ == "__main__":
+    gen_all_batch()

vyas_group_scripts/run_batch.py

@@ -0,0 +1,35 @@
+import os
+from glob import glob
+from rich import print
+import argparse
+
+
+def check_file_has_pair(s_file:str):
+    print(f"Checking {s_file}")
+    if os.path.exists(s_file[:-1]+"inp"):
+        return True
+    if os.path.exists(s_file[:-1]+"gjf"):
+        return True
+    print(f"[red]Matching file not found for {s_file}")
+    return False
+
+def submit_job_pattern(sfile_pattern:str="*.s"):
+    for s_file in glob(sfile_pattern):
+        if check_file_has_pair(s_file):
+            os.system(f"sbatch {s_file}")
+
+
+def run_all_batch():
+
+    parser = argparse.ArgumentParser(prog="rab")
+    parser.add_argument("-k",default="*.s",help="provide a regesx pattern for .s files to submit")
+    
+    args = parser.parse_args()
+
+    patterns_to_submit = args.k
+    
+    submit_job_pattern(patterns_to_submit)
+
+
+if __name__ == "__main__":
+    run_all_batch()

vyas_group_scripts/templates/g16_batch_inputs.py

@@ -0,0 +1,180 @@
+
+##################################################
+# This a simple g16 batch file template python
+# Is responsible for feeding in the following 
+# information:
+# JOB_NAME
+# ACTNUM
+# NUM_NODES
+# TASKS_PER_NODE
+# N_TASKS
+# INPUT_FILE_NAME
+# TIME
+##################################################
+simple_g16_template = """#!/bin/bash -x
+
+#SBATCH --job-name={JOB_NAME}
+#SBATCH --account={ACTNUM} # you can find the account number by running $ sacctmgr show Account
+#SBATCH --nodes={NUM_NODES}
+#SBATCH --ntasks-per-node={TASKS_PER_NODE}
+#SBATCH --ntasks={N_TASKS}
+#SBATCH --export=ALL
+#SBATCH --time={TIME} # time when job will automatically terminate- want the smallest possible overestimate{BENCHMARKING_LINE}
+
+export KMP_AFFINITY=respect,verbose
+
+module load apps/gaussian16/c01
+
+cd $SLURM_SUBMIT_DIR
+
+eval "$VGS_LOAD_G16"
+INPUT_FILE="{INPUT_FILE_NAME}"
+
+OUTPUT_FILE="${{INPUT_FILE: 0:-4}}.log"
+
+JOBID=`echo $SLURM_JOBID`
+
+#export OMP_NUM_THREADS=1
+#export GAUSS_EXEDIR=/opt/g09/g09
+
+echo "Running Job"
+
+
+g16 <"${{INPUT_FILE}}">"${{OUTPUT_FILE}}"
+
+
+
+echo "job has finished" """
+
+
+##################################################
+# This a more complex g16 batch file template python
+# Is responsible for feeding in the following 
+# information:
+# JOB_NAME
+# ACTNUM
+# NUM_NODES
+# TASKS_PER_NODE
+# N_TASKS
+# INPUT_FILE_NAME
+# TIME
+##################################################
+
+complex_g16_template = """#!/bin/bash -x
+
+# --------------------------------------------------------------------------------------------------------------
+#
+# This script will submit the input file specified to Gaussian 16 via the slurm scheduler. The script will 
+# automatically create a new output directory with the same title as the input file and send all files there. 
+# Script timing will be recorded in the slurm.out file which will be copied to the output directory only when 
+# the job has completed. Email status updates will be sent if configured in the config.toml.
+#
+# --------------------------------------------------------------------------------------------------------------
+
+# must set the account number as an env variable manually - gab will automatically use this
+#SBATCH --account={ACTNUM} # you can find the account number by running $ sacctmgr show Account
+#SBATCH --nodes={NUM_NODES}
+#SBATCH --ntasks-per-node={TASKS_PER_NODE}
+#SBATCH --ntasks={N_TASKS}
+#SBATCH --export=ALL
+#SBATCH --time={TIME} # time when job will automatically terminate- want the smallest possible overestimate{BENCHMARKING_LINE}
+maxtime=3590 # buffer time to allow cleanup, should be ~10 seconds less than auto termination time
+
+# --------------------------------------------------------------------------------------------------------------
+
+INPUT_FILE="{INPUT_FILE_NAME}"
+
+# --------------------------------------------------------------------------------------------------------------
+
+# record submission time
+mystart=$(date +%Y-%m-%d %H:%M:%S)
+
+# make a new output folder with the job name as the title and direct the output there
+OUTPUT_DIR="${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME: 0:-2}}"
+mkdir -p "${{OUTPUT_DIR}}_IN-PROGRESS"
+
+# copy the input file to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}" "${{OUTPUT_DIR}}_IN-PROGRESS/${{INPUT_FILE}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}"
+
+# copy the input script to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}" "${{OUTPUT_DIR}}_IN-PROGRESS/${{SLURM_JOB_NAME}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}"
+
+
+# configure KMP_AFFINITY to communicate hardware threads to OMP parallelizer
+export KMP_AFFINITY=respect,verbose
+
+# load the gaussian module and print JOBID to slurm out for debugging
+eval "$VGS_LOAD_G16"
+JOBID=`echo $SLURM_JOBID`
+
+# configure and pass OMP parameters
+#export OMP_NUM_THREADS=1
+#export GAUSS_EXEDIR=/opt/g09/g09
+
+# go the output folder
+cd "${{OUTPUT_DIR}}_IN-PROGRESS"
+
+# run the input file and generate an output file of the same name
+# if the timeout is reached it will return exit 124, otherwise it returns the calc exit status
+start=`date +%s.%N`
+OUTPUT_FILE="${{INPUT_FILE: 0:-4}}.log"
+timeout -s SIGTERM $maxtime g16 <"${{OUTPUT_DIR}}_IN-PROGRESS/${{INPUT_FILE}}">"${{OUTPUT_FILE}}"
+CALC_STATUS=$?
+end=`date +%s.%N`
+
+# get the job status
+if [[ $CALC_STATUS == 124 ]]; then 
+	status="TIMEOUT"
+elif [[ $CALC_STATUS != 0 ]]; then
+	status="ERROR"
+elif [[ $CALC_STATUS == 0 ]]; then
+	status="NORMAL"
+fi
+
+# log the time for benchmarking in the outputfile
+runtime=$( echo "$end - $start" | bc -l )
+echo $runtime
+exec 3>>"${{OUTPUT_FILE}}"
+echo "">&3
+echo "slurmID:    ${{SLURM_JOBID}}">&3
+echo "totalRuntime[s]:    ${{runtime}}">&3
+exec 3>&-
+
+# get the number of basis functions used in the first calculation
+myBasis=$(grep -o -m 1 '[0-9]\+ basis functions' myFile | cut -d ' ' -f 1)
+myBasis=($myBasis)
+
+# record the completion time
+myend=$(date +%Y-%m-%d %H:%M:%S)
+
+# write the total job timing to the job_timings file in the submit directory as a CSV
+cd "${{SLURM_SUBMIT_DIR}}"
+if [ ! -f job_timings.csv ]; then
+	echo "filename,slurmID,nbasisfuncs,start,end,runtime[s],jobstatus" > job_timings.csv
+fi
+exec 3>>job_timings.csv
+echo "${{INPUT_FILE}},${{SLURM_JOBID}},${{myBasis}},${{mystart}},${{myend}},${{runtime}},${{status}}">&3
+exec 3>&-
+
+# copy the slurm output to the output folder and delete from the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out" "${{OUTPUT_DIR}}_IN-PROGRESS/slurm-${{SLURM_JOBID}}-${{INPUT_FILE: 0:-4}}.out"
+rm "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out"
+
+# rename the output directory appropriately
+if [[ $CALC_STATUS == 124 ]]; then 
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}_TIMEOUT"
+	exit 124
+
+elif [[ $CALC_STATUS != 0 ]]; then
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}_ERROR"
+	exit 2
+
+elif [[ $CALC_STATUS == 0 ]]; then
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}"
+	exit 0
+fi"""

vyas_group_scripts/templates/orca_batch_inputs.py

@@ -0,0 +1,315 @@
+orca_template = """#!/bin/bash -x
+
+# --------------------------------------------------------------------------------------------------------------
+#
+# This script will submit the input file specified to ORCA via the slurm scheduler. The script will automatically
+# create a new output directory with the same title as the input file and send all files there. Script timing 
+# will be recorded in the slurm.out file which will be copied to the output directory only when the job has 
+# completed. Email status updates will be sent if configured in the config.toml.
+#
+# --------------------------------------------------------------------------------------------------------------
+
+# must set the account number as an env variable manually - gab will automatically use this
+#SBATCH --account={ACTNUM} # you can find the account number by running $ sacctmgr show Account
+#SBATCH --nodes={NUM_NODES}
+#SBATCH --ntasks-per-node={TASKS_PER_NODE}
+#SBATCH --ntasks={N_TASKS}
+#SBATCH --export=ALL
+#SBATCH --time={TIME} # time when job will automatically terminate- want the smallest possible overestimate{BENCHMARKING_LINE}
+maxtime=3590 # buffer time to allow cleanup, should be ~10 seconds less than auto termination time
+
+# --------------------------------------------------------------------------------------------------------------
+
+INPUT_FILE="{INPUT_FILE_NAME}"
+
+# --------------------------------------------------------------------------------------------------------------
+
+ml purge 
+
+# record submission time
+mystart=$(date +%Y-%m-%d %H:%M:%S)
+
+# make a new output folder with the job name as the title and direct the output there
+OUTPUT_DIR="${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME: 0:-2}}"
+mkdir -p "${{OUTPUT_DIR}}_IN-PROGRESS"
+
+# copy the input orca file to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}" "${{OUTPUT_DIR}}_IN-PROGRESS/${{INPUT_FILE}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}"
+
+# copy the input xyz file to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME: 0:-2}}.xyz" "${{OUTPUT_DIR}}_IN-PROGRESS/${{SLURM_JOB_NAME: 0:-2}}.xyz" || echo "NO XYZ INPUT FOR THIS JOB"
+rm "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME: 0:-2}}.xyz" || echo "NO XYZ FILE TO REMOVE"
+
+# copy the input script to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}" "${{OUTPUT_DIR}}_IN-PROGRESS/${{SLURM_JOB_NAME}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}"
+
+# load and export NBO path
+eval "$VGS_LOAD_NBO"
+
+# load ORCA path
+eval "$VGS_LOAD_ORCA"
+
+# load the mpi module and print JOBID to slurm out for debugging
+eval "$VGS_LOAD_MPI"
+JOBID=`echo $SLURM_JOBID`
+
+# configure and pass OMP parameters
+#export OMP_NUM_THREADS=1
+
+# go the output folder
+cd "${{OUTPUT_DIR}}_IN-PROGRESS"
+
+
+#################################################################################################
+# run the input file and generate an output file of the same name
+# if the timeout is reached it will return exit 124, otherwise it returns the calc exit status
+start=`date +%s.%N`
+OUTPUT_FILE="${{INPUT_FILE: 0:-4}}.out"
+timeout -s SIGTERM $maxtime $VGS_ORCA_PATH "${{OUTPUT_DIR}}_IN-PROGRESS/${{INPUT_FILE}}">"${{OUTPUT_FILE}}"
+CALC_STATUS=$?
+end=`date +%s.%N`
+#################################################################################################
+
+
+# try to generate the orbital *.cube files and *.html files if the ORCA job was successful
+# also try to extract the NBO output (not always called, will pass if no NBO)
+if [[ $CALC_STATUS == 0 ]]; then 
+	cubegen="${{VGSPATH}}/orca_orbital_cubegen.s"
+	chmod +x "${{cubegen}}"
+	"${{cubegen}}" -f "${{INPUT_FILE: 0:-4}}.gbw" || echo "MO VISUALIZATION FAILED"
+
+	sed -n '/Now starting NBO\.\.\./,/returned from  NBO  program/p' "${{OUTPUT_FILE}}" | tail -n +2 | head -n -1 > "${{INPUT_FILE: 0:-4}}.nbout" || echo "NBO SCRAPING FAILED"
+fi
+
+
+# get the job status
+# get the job status
+if [[ $CALC_STATUS == 124 ]]; then 
+    status="TIMEOUT"
+elif [[ $CALC_STATUS != 0 ]]; then
+    status="ERROR"
+elif [[ $CALC_STATUS == 0 ]]; then
+    # check if the words "ORCA TERMINATED NORMALLY" appear in the last 10 lines of the output file
+    if tail -n 10 "$OUTPUT_FILE" | grep -q "****ORCA TERMINATED NORMALLY****"; then
+        status="NORMAL"
+    else
+        CALC_STATUS=2
+        status="INCOMPLETE"
+    fi
+fi
+
+
+# log the time for benchmarking in the outputfile
+runtime=$( echo "$end - $start" | bc -l )
+echo $runtime
+exec 3>>"${{OUTPUT_FILE}}"
+echo "">&3
+echo "slurmID:    ${{SLURM_JOBID}}">&3
+echo "totalRuntime[s]:    ${{runtime}}">&3
+exec 3>&-
+
+# get the number of basis functions used in the first calculation by the SHARK package
+myBasis=$(grep -m 1 "Number of basis functions" "${{OUTPUT_FILE}}" | awk '{{print $NF}}' | tr -d '[:space:]')
+myBasis=($myBasis)
+
+# record the completion time
+myend=$(date +%Y-%m-%d %H:%M:%S)
+
+# write the total job timing to the job_timings file in the submit directory as a CSV
+cd "${{SLURM_SUBMIT_DIR}}"
+if [ ! -f job_timings.csv ]; then
+	echo "filename,slurmID,nbasisfuncs,start,end,runtime[s],jobstatus" > job_timings.csv
+fi
+exec 3>>job_timings.csv
+echo "${{INPUT_FILE}},${{SLURM_JOBID}},${{myBasis}},${{mystart}},${{myend}},${{runtime}},${{status}}">&3
+exec 3>&-
+
+
+
+# copy the slurm output to the output folder and delete from the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out" "${{OUTPUT_DIR}}_IN-PROGRESS/slurm-${{SLURM_JOBID}}-${{INPUT_FILE: 0:-4}}.out"
+rm "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out"
+
+
+# rename the output directory appropriately
+if [[ $status == "TIMEOUT" ]]; then 
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}_TIMEOUT"
+	exit 124
+
+elif [[ $status == "ERROR" ]]; then
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}_ERROR"
+	exit 2
+
+elif [[ $status == "INCOMPLETE" ]]; then
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}_INCOMPLETE"
+	exit 2
+
+elif [[ $status == "NORMAL" ]]; then
+
+	mv "${{OUTPUT_DIR}}_IN-PROGRESS" "${{OUTPUT_DIR}}"
+	cd "${{OUTPUT_DIR}}"
+	# cleanup empty files
+	if [[ -e "2" && ! -s "2" ]]; then
+		rm "2"
+	fi
+	for nbout_file in *.nbout; do
+		if [[ -e "$nbout_file" && ! -s "$nbout_file" ]]; then
+			rm "${{nbout_file}}"
+		fi
+	done
+	cd ..
+
+	exit 0
+fi"""
+
+orca_restart_template = """#!/bin/bash -x
+
+# --------------------------------------------------------------------------------------------------------------
+#
+# This script will re-submit the input file specified to ORCA via the slurm scheduler. This script automatically  
+# edits the last input file to ensure that the previous geometry and wavefunction is read. All output will be 
+# generated in the same directory as the original calculation that timed out. Script timing will be recorded in 
+# the slurm.out file which will be copied to the original output directory only when the job has completed. Email
+# status updates will be sent if configured in the config.toml.
+#
+# --------------------------------------------------------------------------------------------------------------
+
+# must set the account number as an env variable manually - gab will automatically use this
+#SBATCH --account={ACTNUM} # you can find the account number by running $ sacctmgr show Account
+#SBATCH --nodes={NUM_NODES}
+#SBATCH --ntasks-per-node={TASKS_PER_NODE}
+#SBATCH --ntasks={N_TASKS}
+#SBATCH --export=ALL
+#SBATCH --time={TIME} # time when job will automatically terminate- want the smallest possible overestimate{BENCHMARKING_LINE}
+maxtime=3590 # buffer time to allow cleanup, should be ~10 seconds less than auto termination time
+
+# --------------------------------------------------------------------------------------------------------------
+
+INPUT_FILE="{INPUT_FILE_NAME}"
+OLD_JOB_DIR="my_old_job"
+
+# --------------------------------------------------------------------------------------------------------------
+
+# record submission time
+mystart=$(date +%Y-%m-%d %H:%M:%S)
+
+# make a new output folder with the job name as the title and direct the output there
+OUTPUT_DIR="${{SLURM_SUBMIT_DIR}}/${{OLD_JOB_DIR}}"
+mv "${{OUTPUT_DIR}}" "${{OUTPUT_DIR}}_RESTARTED"
+
+# copy the input script to the output folder and delete the copy in the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}" "${{OUTPUT_DIR}}_RESTARTED/${{SLURM_JOB_NAME}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{SLURM_JOB_NAME}}"
+cp "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}" "${{OUTPUT_DIR}}_RESTARTED/${{INPUT_FILE}}"
+rm "${{SLURM_SUBMIT_DIR}}/${{INPUT_FILE}}"
+
+# load and export NBO path
+eval "$VGS_LOAD_NBO"
+
+# load the mpi module and print JOBID to slurm out for debugging
+eval "$VGS_LOAD_MPI"
+JOBID=`echo $SLURM_JOBID`
+
+# go the output folder
+cd "${{OUTPUT_DIR}}_RESTARTED"
+
+
+#################################################################################################
+# run the input file and generate an output file of the same name
+# if the timeout is reached it will return exit 124, otherwise it returns the calc exit status
+start=`date +%s.%N`
+OUTPUT_FILE="${{INPUT_FILE: 0:-4}}.out"
+timeout -s SIGTERM $maxtime $SS_ORCA_PATH "${{OUTPUT_DIR}}_RESTARTED/${{INPUT_FILE}}">"${{OUTPUT_FILE}}"
+CALC_STATUS=$?
+end=`date +%s.%N`
+#################################################################################################
+
+
+# try to generate the orbital *.cube files and *.html files if the ORCA job was successful
+# also try to extract the NBO output (not always called, will pass if no NBO)
+if [[ $CALC_STATUS == 0 ]]; then 
+	cubegen="${{VGSPATH}}/orca_orbital_cubegen.s"
+	chmod +x "${{cubegen}}"
+	"${{cubegen}}" -f "${{INPUT_FILE: 0:-4}}.gbw" || echo "MO VISUALIZATION FAILED"
+
+	sed -n '/Now starting NBO\.\.\./,/returned from  NBO  program/p' "${{OUTPUT_FILE}}" | tail -n +2 | head -n -1 > "${{INPUT_FILE: 0:-4}}.nbout" || echo "NBO SCRAPING FAILED"
+fi
+
+
+# get the job status
+if [[ $CALC_STATUS == 124 ]]; then 
+	status="TIMEOUT"
+elif [[ $CALC_STATUS != 0 ]]; then
+	status="ERROR"
+elif [[ $CALC_STATUS == 0 ]]; then
+	status="NORMAL"
+fi
+
+
+# log the time for benchmarking in the outputfile
+runtime=$( echo "$end - $start" | bc -l )
+echo $runtime
+exec 3>>"${{OUTPUT_FILE}}"
+echo "">&3
+echo "slurmID:    ${{SLURM_JOBID}}">&3
+echo "totalRuntime[s]:    ${{runtime}}">&3
+exec 3>&-
+
+# get the number of basis functions used in the first calculation by the SHARK package
+myBasis=$(grep -m 1 "Number of basis functions" "${{OUTPUT_FILE}}" | awk '{{print $NF}}' | tr -d '[:space:]')
+myBasis=($myBasis)
+
+# record the completion time
+myend=$(date +%Y-%m-%d %H:%M:%S)
+
+# write the total job timing to the job_timings file in the submit directory as a CSV
+cd "${{SLURM_SUBMIT_DIR}}"
+if [ ! -f job_timings.csv ]; then
+	echo "filename,slurmID,nbasisfuncs,start,end,runtime[s],jobstatus" > job_timings.csv
+fi
+exec 3>>job_timings.csv
+echo "${{INPUT_FILE}},${{SLURM_JOBID}},${{myBasis}},${{mystart}},${{myend}},${{runtime}},${{status}}">&3
+exec 3>&-
+
+
+
+# copy the slurm output to the output folder and delete from the submit directory
+cp "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out" "${{OUTPUT_DIR}}_RESTARTED/slurm-${{SLURM_JOBID}}-${{INPUT_FILE: 0:-4}}.out"
+rm "${{SLURM_SUBMIT_DIR}}/slurm-${{SLURM_JOBID}}.out"
+
+
+# rename the output directory appropriately
+if [[ $CALC_STATUS == 124 ]]; then 
+
+	mv "${{OUTPUT_DIR}}_RESTARTED" "${{OUTPUT_DIR}}_TIMEOUT"
+	exit 124
+
+elif [[ $CALC_STATUS != 0 ]]; then
+
+	mv "${{OUTPUT_DIR}}_RESTARTED" "${{OUTPUT_DIR}}_ERROR"
+	exit 2
+
+elif [[ $CALC_STATUS == 0 ]]; then
+
+	FINAL_OUTPUT="${{OUTPUT_DIR%_TIMEOUT}}"
+	mv "${{OUTPUT_DIR}}_RESTARTED" "${{FINAL_OUTPUT}}"
+	cd "${{OUTPUT_DIR}}"
+	# cleanup empty files
+	if [[ -e "2" && ! -s "2" ]]; then
+		rm "2"
+	fi
+	for nbout_file in *.nbout; do
+		if [[ -e "$nbout_file" && ! -s "$nbout_file" ]]; then
+			rm "${{nbout_file}}"
+		fi
+	done
+	cd ..
+
+	exit 0
+fi"""
+

vyas_group_scripts-0.3.0.dist-info/METADATA

@@ -0,0 +1,11 @@
+Metadata-Version: 2.3
+Name: vyas-group-scripts
+Version: 0.3.0
+Summary: Add your description here
+Author: Ben Payton
+Author-email: Ben Payton <bennyp2494@gmail.com>
+Requires-Dist: argparse>=1.4.0
+Requires-Dist: rich>=14.2.0
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+

vyas_group_scripts-0.3.0.dist-info/RECORD

@@ -0,0 +1,9 @@
+vyas_group_scripts/__init__.py,sha256=PZJwQzDvN8PbfBDMCXOb0ciJKMir9J35nU1EXRqgARg,131
+vyas_group_scripts/gen_batch.py,sha256=WHaw2DsC0gq9FSrkOpnCvwX8LqC-3bHJ-9x1OaGGhOc,7025
+vyas_group_scripts/run_batch.py,sha256=RUqEM2UATNq1KOorfszeCfNcdABhwiSlQf6MHnXdENU,881
+vyas_group_scripts/templates/g16_batch_inputs.py,sha256=3BbLCZaPOeiivJJlLNc5lL-8bRDcOkAyGIuChqsdUEY,6138
+vyas_group_scripts/templates/orca_batch_inputs.py,sha256=moYHIJmhX8NRH5J6NK_TcGLSh_nzWoQVQ1HiDA25yfg,12087
+vyas_group_scripts-0.3.0.dist-info/WHEEL,sha256=eh7sammvW2TypMMMGKgsM83HyA_3qQ5Lgg3ynoecH3M,79
+vyas_group_scripts-0.3.0.dist-info/entry_points.txt,sha256=7cL-_aZU96ATik2TKSKoUHO0Ar4b7Rhoa3NqoTG8Ki0,121
+vyas_group_scripts-0.3.0.dist-info/METADATA,sha256=1qBK7c-hiHuDf2yQdH6-dzDP_Ts_1kIj6niDoX8TTcY,287
+vyas_group_scripts-0.3.0.dist-info/RECORD,,

vyas_group_scripts-0.3.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: uv 0.8.24
+Root-Is-Purelib: true
+Tag: py3-none-any

vyas_group_scripts-0.3.0.dist-info/entry_points.txt

@@ -0,0 +1,4 @@
+[console_scripts]
+pygab = vyas_group_scripts.gen_batch:gen_all_batch
+pyrab = vyas_group_scripts.run_batch:run_all_batch
+