For Blog: Anonymize.py @ MinsaPay

anonymize.py

import pandas as pd
Dataframe = pd.read_csv('raw.csv')

def anonymize(df, targetColumn):
    anon = {}
    id = 0
    for x in range(len(df)):
        user = df.loc[x, targetColumn]
        if user in anon:
            df.loc[x, targetColumn] = anon[user]
        else:
            if id < 10:
                unknown = "#00" + str(id)
            elif id < 100:
                unknown = "#0" + str(id)
            else:
                unknown = "#" + str(id)
            anon[user] = targetColumn + str(unknown)
            id += 1
            df.loc[x, targetColumn] = anon[user]

anonymize(Dataframe, 'user')
anonymize(Dataframe, 'booth')

Dataframe.to_csv("anonymized.csv", mode='w')