Python argparse error

Question

Hi I am trying to use this python script (running with bash script) to do my analysis. I am getting the following error:

File "my_python_script.py", line 33 logging.info("Reading input2 file") ^ SyntaxError: invalid syntax

My Python script is as folllows and albeit I omitted the indentation and my full apologies therein, please keep in mind I'm not the coder in this instance.

import itertools
import argparse
import logging
import sys
import numpy as np
import scipy
import pandas as pd

def parse_commandline():
"""Parse command line parameters.
Return type: Object
"""
parser = argparse.ArgumentParser(description="Association of variants")
parser.add_argument("--input1", metavar="FILE1.tsv", help="gene data file", required=True)
parser.add_argument("--input2", metavar="FILE2.tsv", help="mutation data file", required=True)
parser.add_argument("--output", metavar="FILE.tsv", help="Output file name", required=True)
if len(sys.argv) == 1:
    parser.print_help()
    sys.exit(0)
return parser.parse_args()

def main():
params = parse_commandline()

#load genedata
logging.info("Reading input1 file")
try:
gene_data = pd.read_csv(params.input1, sep="\t", index_col ="Model")
sys.exit(0)
    
#load mutation data
logging.info("Reading input2 file")
try:
mutation_data = pd.read_csv(params.input2, sep="\t", 
index_col="Mutation")
sys.exit(0)

##All mutations and gene in data files
all_mutations = list(mutation_data.index)
all_genes = list(gene_data.columns)
cell line Models in mutation_data file
models_mutations = list(mutation_data.columns)

## models in gene_data file
models_genes = list(gene_data.index)
intersection_set = set.intersection(set(models_genes), 
set(models_mutations))
models_common = list(intersection_set)
unique_series_gene = gene_data.index.isin(models_common)
end_gene_table = gene_data[unique_series_gene]
transposed_mutation_data = mutation_data.transpose()
unique_series_mutation = 
transposed_mutation_data.index.isin(models_common)
end_mutation_table = transposed_mutation_data[unique_series_gene]

##calcuate p-values
all_p_values = []
genes = []
mutations = []
for gene, mutation in itertools.product(all_genes, all_mutations):
gene_table_to_test = np.array(end_gene_table[gene])
mutation_table_to_test = np.array(end_mutation_table[mutation])    
p_value = scipy.stats.wilcoxon(gene_table_to_test, 
mutation_table_to_test)[1]
genes.append(gene)
mutations.append(mutation)
all_p_values.append(p_value)

## calculating adjusted_pvalues
adjusted_pvalues = p_adjust_bh(all_p_values)
d = {'Gene': genes, 'Mutation': mutations, 'pvalue': 
all_p_values,'p_adj':adjusted_pvalues}
df = pd.DataFrame(data=d)

# Print output file
logging.info("Writing into file")
df.to_csv(params.output, sep="\t")

if __name__ == "__main__":
main()

My Bash script

#!/bin/sh

### This script will run python script

python3 my_python_script.py --input1 Gene.tsv --input2 
Mutation.tsv --output tmp
diff tmp adjusted_pvalues.tsv
rm tmp

Answer 1

It is certainly possible the 'error' is the input file 2.

However, both try: routines are weird and if this is the real code it would fall over. Something isn't quite right however, which I'll touch on at the end.

What the code is supposed to say is:

try:
    load stuff
except:
    raise Exception ('Input file 2 is has deviation from csv format and will not load')

raise Exception isn't needed, the following would be a more harmonised approach:

try:
    load stuff
except:
    print ("Input file 2 is has deviation from csv format and will not load')
finally:
    sys.exit('Abandon ship!')

The idea is that if the CSV file didn't load, it would stop the code and tell you approximately where to start looking. You might then ask, can it actually sees the file and then peer into the loaded CSV. If it loads the code continues.

You could just stay:

try:
    load stuff
except:
    sys.exit('Oh the file did not to load')

This isn't cool, but who am I (who sticks Exceptions in except statements), but it will work and it could be that is the actual code.

What the code actually says is

try:
    load stuff
    sys.exit()

or

try: 
    load stuff
sys.exit()

Same outcome the code stop because sys.exit stops everything ... nothing is left. What the code should really say is not just perform a try: except: but perform a 'getter' and 'setter' on the input file to check the CSV format, this is because your the user of the code not the developer and the coder has to put safe guards in because stuff happens.

sys.exit is really used in development, for example in debugging.

However, somehow you got past the first try:, sys.exit and I don't know how you did it, because that should not have been possible. I would carefully check the original code.

Just to point out as @ChrisR mentioned, if this was your actual python code it would fall over because you've omitted all the identation, which in python is essential. Its not like shell where indentitation matters not a jot.

Finally, as @supertech has mentioned, there are numerous line breaks in the code and whether you are in shell or python this can cause problems. I think the line break(s) are within an extra layer of identation, which Python can read through (I never understood how).

Personally, I'd go back to the coder and point out the try: sys.exit() and just point out any beginners webpage on try: except:, this is not advanced coding. If anything sys.exit() risks weird behaviour for a user particularly as no distinguishing information is provided.