noamcheck

import pandas as pd
from datetime import datetime
from pytz import timezone


def est_to_utc(date_str):
    datetime_obj = datetime.strptime(date_str, "%m/%d/%y %H:%M")
    datetime_obj_utc = datetime_obj.replace(tzinfo=timezone('UTC'))
    return datetime_obj_utc.strftime("%Y-%m-%d %H:%M:%S %Z%z")


def convert_utc():
    cost_df = pd.read_csv("cost_1.csv")
    revenue_df = pd.read_csv("revenue_1.csv")
    cost_df["utc"] = cost_df["data_date"].apply(lambda x: est_to_utc(x))
    cost_df.to_csv("cost_utc.csv")
    revenue_df["utc"] = revenue_df["data_date"].apply(lambda x: est_to_utc(x))
    revenue_df.to_csv("revenue_utc.csv")


if __name__ == '__main__':
    convert_utc()
    revenue_df = pd.read_csv("revenue_utc.csv")
    cost_df = pd.read_csv("cost_utc.csv")
    # x = revenue_df[:10].set_index('campaign_id').join(cost_df[:10]).set_index('campaign_id'))
    small = revenue_df[:10].merge(cost_df[:10], left_on=['campaign_id'], right_on='campaign_id').groupby({'uv'})
    small['uv'] = small["revenue"] / small["clicks"]
    small["cpc"] = small["costs"] / small["clicks"]
    small["roi"] = small["uv"] / small["cpc"]
    small["profit"] = small["revenue"] - small["cost"]

    pass
;