derived_variables_generator.py

# -*- coding: utf-8 -*-
"""
Created on Wed Nov 10 12:41:12 2021

@author: ctb
"""

import datetime
import pandas as pd
import sys
import numeric_converter
import zipfile
import csv

datafile_path = sys.argv[1]
csv_separator = ","

if len(sys.argv) == 3:
    csv_separator = sys.argv[2]

datafile = pd.read_csv(datafile_path, csv_separator)


datafile = numeric_converter.numeric_conversion(datafile)


age_ranges = list(range(30, 100, 10))


if "DMRAGEYR" in datafile.columns:
    
    derived_age_range = []
    for x in datafile["DMRAGEYR"]:
        age_range = -1
        for i, mark in enumerate(age_ranges):
            if x > mark:
                age_range= str(mark) + "-" + str(age_ranges[i+1] - 1)
                break
        
        if age_range == -1:
            age_range = "18-29"
            
        derived_age_range.append(age_range)
    
    datafile["AGE_RANGE"] = derived_age_range
        
    

if "DATDS" in datafile.columns:
    
    derived_outcome_month = []
    for x in datafile["DATDS"]:
        if str(x) == 'nan': #math.isnan(x):
            derived_outcome_month.append(None)
        else:
            month_num = x.split("/")
            if len(month_num) == 3:
                month_num = month_num[1]
                datetime_object = datetime.datetime.strptime(month_num, "%m")
                full_month_name = datetime_object.strftime("%B")

                derived_outcome_month.append(full_month_name)
            else:
                derived_outcome_month.append(None)
    
    datafile["DISCHARGE_DATE"] = derived_outcome_month



if "DATAD" in datafile.columns:
    
    derived_outcome_month = []
    for x in datafile["DATAD"]:
        if str(x) == 'nan':  # math.isnan(x):
            derived_outcome_month.append(None)
        else:
            month_num = x.split("/")
            if len(month_num) == 3:
                month_num = month_num[1]
                datetime_object = datetime.datetime.strptime(month_num, "%m")
                full_month_name = datetime_object.strftime("%B")

                derived_outcome_month.append(full_month_name)
            else:
                derived_outcome_month.append(None)
    
    datafile["ADMISSION_DATE"] = derived_outcome_month

    
    
new_datafile_path = datafile_path.replace(".csv", "_numeric_derived")
datafile.to_csv(new_datafile_path+".csv", index = False, quoting=csv.QUOTE_NONNUMERIC)

with zipfile.ZipFile(new_datafile_path+".zip", 'w') as myzip:
    myzip.write(new_datafile_path+".csv")