Generate Soda CL based on a csv

py

import csv
import pandas as pd
import yaml

# Provide a link to the csv file with the rules.
file_name = 'rules.csv'

df = pd.read_csv(file_name)

df = df.reset_index()  # make sure indexes pair with number of rows
df.fillna("nan",inplace=True)
checks = []
table_name = ''
agreement = []

for index, row in df.iterrows():
    if(row['script'] != 'nan'):
        rule = {
                    'failed rows': {
                        'name': row['pk_control_id'],
                        'attributes': {
                                'criticality': row['criticality'],
                                'fk_dim': row['fk_dim'],
                                'environment': row['environment'],
                            },
                        'fail query': row['script']
                    }
                }
        checks.append(rule)
        if (table_name != row['table_name']):
            table_name = row['table_name']

            checks_dict = {
                'checks for ' + row['table_name']: checks
            }
            print(yaml.dump(checks_dict))