def read_campaign_retention():
path_manager = EnvManager()
resource_path = path_manager.get_resource_dir()
source_dir = ospath.join(resource_path, 'campaign_retention')
source_file = ospath.join(source_dir, 'campaign_retention.csv')
source_data = pd.read_csv(
source_file, usecols=['Date', 'Campaign', 'Install Day', 'Day 1'])
sa_mask = source_data['Campaign'].apply(isContain, sub_string='_SA_')
sa_data = source_data[sa_mask]
sa_sum = sa_data.groupby('Date').sum()
row_num = len(sa_sum.index)
if row_num > 0:
country_list = pd.Series(['SA'] * row_num, index=sa_sum.index)
sa_sum['country'] = country_list
ph_mask = source_data['Campaign'].apply(isContain, sub_string='_ph_')
ph_data = source_data[ph_mask]
ph_sum = ph_data.groupby('Date').sum()
row_num = len(ph_sum.index)
if row_num > 0:
country_list = pd.Series(['PH'] * row_num, index=ph_sum.index)
ph_sum['country'] = country_list
result_data = pd.concat([sa_sum, ph_sum])
writer = pd.ExcelWriter('sum_retention.xlsx')
result_data.to_excel(writer, 'retention')
writer.save()
def analysis_typany_retention():
path_manager = EnvManager()
resource_path = path_manager.get_resource_dir()
typany_path = ospath.join(resource_path, 'typany')
retention_file = ospath.join(typany_path, 'retention_typany.xlsx')
retention_data = ReadExcelDataHandler.read_typany_retention(retention_file)
# pivot_data = pd.pivot_table(retention_data, index=['Date', 'Country'], values=['Install Day', 'Day 1'],
# aggfunc=np.sum, fill_value=0)
pivot_data = retention_data.groupby(['Date', 'Country'],
as_index=False).sum()
output_dir = path_manager.get_output_dir()
output_file = ospath.join(output_dir, 'typany.xlsx')
writer = pd.ExcelWriter(output_file)
pivot_data.to_excel(writer, 'retention')
writer.save()
conversion_data['All conv.'] = conversion_data['All conv.'].apply(str.replace, args=(',', '')).apply(float)
conversion_data['Cost'] = conversion_data['Cost'].apply(str.replace, args=(',', '')).apply(float)
install_data = pd.read_excel(install_file, parse_cols="B, C")
install_data['campaign'] = install_data['campaign'].apply(str)
install_data.rename(columns={'campaign': 'Campaign ID', 'install': 'AF Install'}, inplace=True)
merger_data = pd.merge(conversion_data, install_data, on='Campaign ID', how='outer')
merger_data.fillna(0, inplace=True)
merger_data['Other Conversion'] = merger_data['All conv.'] - merger_data['Conversions']
merger_data['Cost/AF'] = map(comhelper.division_operation, merger_data['Cost'], merger_data['AF Install'])
merger_data['Cost/Coversion'] = map(comhelper.division_operation, merger_data['Cost'], merger_data['Conversions'])
merger_data['Cost/All'] = map(comhelper.division_operation, merger_data['Cost'], merger_data['All conv.'])
merger_data['Cost/Other'] = map(comhelper.division_operation, merger_data['Cost'], merger_data['Other Conversion'])
export_data = merger_data[['Campaign ID', 'Campaign', 'AF Install', 'Conversions', 'All conv.', 'Other Conversion',
'Cost', 'Cost/AF', 'Cost/Coversion', 'Cost/All', 'Cost/Other']]
date_str = comhelper.get_delta_date(-1)
file_name = 'adword_conversion_' + date_str + '.xlsx'
export_file = ospath.join(output_root_path, file_name)
ExcelExporter.export_data_excel(export_data, export_file)
if __name__ == '__main__':
path_manager = EnvManager()
resource_path = path_manager.get_resource_dir()
conversion_path = ospath.join(resource_path, 'aw_conversion')
output_path = path_manager.get_output_dir()
export_conversion(conversion_path, output_path)