def read_ad_data(account_id, campaign_id):
path_manager = EnvManager()
insight_dir = path_manager.get_insight_dir(mconstant.NODE_TYPE_AD)
account_dir = os.path.join(insight_dir, account_id)
config_path = path_manager.get_conf_dir()
output_path = os.path.join(path_manager.get_output_dir(),
mconstant.NODE_TYPE_AD)
fhelper.make_dir(output_path)
handler_factory = HandlerFactory(config_path)
insight_handler = handler_factory.get_insight_handler(
mconstant.NODE_TYPE_AD)
in_act_files = fhelper.get_file_list(account_dir)
data_per_day = read_ad_insight_day(in_act_files, insight_handler,
campaign_id)
ad_name_key = ''
group_data = group_by_name(data_per_day, ad_name_key)
if ad_name_key.strip():
file_name = account_id + '_' + ad_name_key + '_' + str(
time.time()) + '.xlsx'
act_file = os.path.join(output_path, file_name)
ExcelExporter.export_data_excel(group_data, act_file)
else:
file_name = account_id + '_' + str(time.time()) + '.xlsx'
act_file = os.path.join(output_path, file_name)
ExcelExporter.export_excel_by_key(iconstant.NEW_FIELD_GROUP_AD_NAME,
iconstant.NEW_FIELD_GROUP_AD_NAME,
group_data, act_file)
def merger_all_account_data(node_type):
path_manager = EnvManager()
insight_dir = path_manager.get_insight_dir(node_type)
retention_dir = path_manager.get_retention_dir(node_type)
account_list = fhelper.get_subdir_name_list(retention_dir)
config_path = path_manager.get_conf_dir()
merger_handler = MergerDataHandler(config_path)
output_path = os.path.join(path_manager.get_output_dir(), node_type)
fhelper.make_dir(output_path)
for account_id in account_list:
in_act_dir = os.path.join(insight_dir, account_id)
re_act_dir = os.path.join(retention_dir, account_id)
in_act_files = fhelper.get_file_list(in_act_dir)
re_act_files = fhelper.get_file_list(re_act_dir)
merger_data = merger_handler.merger_in_re_per_day(
in_act_files, re_act_files, node_type)
output_act_path = os.path.join(output_path, account_id)
fhelper.make_dir(output_act_path)
ExcelExporter.export_merger_retention(dhelper.get_key_id(node_type),
dhelper.get_key_id(node_type),
merger_data, output_act_path)
def read_all_account_data(node_type):
path_manager = EnvManager()
insight_dir = path_manager.get_insight_dir(node_type)
account_list = fhelper.get_subdir_name_list(insight_dir)
config_path = path_manager.get_conf_dir()
output_path = os.path.join(path_manager.get_output_dir(), node_type)
fhelper.make_dir(output_path)
handler_factory = HandlerFactory(config_path)
insight_handler = handler_factory.get_insight_handler(node_type)
for account_id in account_list:
in_act_dir = os.path.join(insight_dir, account_id)
in_act_files = fhelper.get_file_list(in_act_dir)
file_name = account_id + '_' + str(time.time()) + '.xlsx'
act_file = os.path.join(output_path, file_name)
read_insight_per_day(in_act_files, insight_handler, act_file, node_type, account_id)
class InsightReader:
def __init__(self,
account_id,
handler_factory,
start_date='',
end_date='',
node_types=None,
insight_path=''):
self._path_manager = EnvManager()
self._handler_factory = handler_factory
self._current_act_id = account_id
self._all_data = {}
self._all_handler = {}
self._start_date = start_date
self._end_date = end_date
if not end_date:
self._end_date = mhelper.get_now_date()
if node_types:
self._node_type = node_types
else:
self._node_type = [
mconstant.NODE_TYPE_CAMPAIGN, mconstant.NODE_TYPE_ADSET,
mconstant.NODE_TYPE_AD
]
self._insight_path = insight_path
self._read_all_data()
def read_daily_data(self, node_type, start_date='', end_date=''):
if node_type not in self._all_handler:
return None
if node_type not in self._all_data:
return None
handler = self._all_handler[node_type]
insight_data = self._all_data[node_type]
data_per_day = handler.get_insight_data_per_day(insight_data)
if not end_date:
end_date = self._end_date
index_mask = np.logical_and(data_per_day.index >= start_date,
data_per_day.index <= end_date)
filter_data = data_per_day[index_mask]
return filter_data
def read_hourly_data(self, node_type, start_date='', end_date=''):
if node_type not in self._all_handler:
return None
if node_type not in self._all_data:
return None
handler = self._all_handler[node_type]
insight_data = self._all_data[node_type]
increment_data = handler.read_increment_by_data(insight_data)
hour_data = handler.get_increment_per_hour(increment_data)
if not end_date:
end_date = self._end_date
start_object = mhelper.convert_str_date(start_date, '%Y-%m-%d').date()
end_object = mhelper.convert_str_date(end_date, '%Y-%m-%d').date()
index_mask = np.logical_and(hour_data.index.date >= start_object,
hour_data.index.date <= end_object)
filter_data = hour_data[index_mask]
return filter_data
def _read_all_data(self):
for ntype in self._node_type:
data_handler = self._handler_factory.get_insight_handler(ntype)
file_list = self._get_file_list(ntype)
insight_datas = [
data_handler.read_insight_data(file_path, self._current_act_id)
for file_path in file_list
]
if not insight_datas:
continue
all_insight_data = pd.concat(insight_datas)
self._all_data[ntype] = all_insight_data
self._all_handler[ntype] = data_handler
def _get_file_list(self, node_type):
node_dir = self._path_manager.get_insight_dir(node_type,
self._insight_path)
account_dir = ospath.join(node_dir, self._current_act_id)
node_files = fhelper.get_file_list(account_dir)
filtered_files = [
path for path in node_files if self._filter_file(path)
]
return filtered_files
def _filter_file(self, file_path):
date_dirname = ospath.basename(file_path)
pattern = '\d+-\d+-\d+'
search_date = re.search(pattern, date_dirname)
if not search_date:
return True
date_dir = search_date.group()
if self._start_date <= date_dir <= self._end_date:
return True
else:
return False