def __init__(self, schema_name, report_path, compression, provider_uuid):
"""Initialize the report processor.
Args:
schema_name (str): The name of the customer schema to process into
report_path (str): Where the report file lives in the file system
compression (CONST): How the report file is compressed.
Accepted values: UNCOMPRESSED, GZIP_COMPRESSED
"""
self._processor = None
_, self.report_type = utils.detect_type(report_path)
if self.report_type == utils.OCPReportTypes.CPU_MEM_USAGE:
self._processor = OCPCpuMemReportProcessor(schema_name,
report_path,
compression,
provider_uuid)
elif self.report_type == utils.OCPReportTypes.STORAGE:
self._processor = OCPStorageProcessor(schema_name, report_path,
compression, provider_uuid)
elif self.report_type == utils.OCPReportTypes.NODE_LABELS:
self._processor = OCPNodeLabelProcessor(schema_name, report_path,
compression, provider_uuid)
elif self.report_type == utils.OCPReportTypes.NAMESPACE_LABELS:
self._processor = OCPNamespaceLabelProcessor(
schema_name, report_path, compression, provider_uuid)
elif self.report_type == utils.OCPReportTypes.UNKNOWN:
raise OCPReportProcessorError("Unknown OCP report type.")
def divide_csv_daily(file_path, filename):
"""
Split local file into daily content.
"""
daily_files = []
directory = os.path.dirname(file_path)
data_frame = pd.read_csv(file_path)
report_type, _ = utils.detect_type(file_path)
unique_times = data_frame.interval_start.unique()
days = list({cur_dt[:10] for cur_dt in unique_times})
daily_data_frames = [{
"data_frame":
data_frame[data_frame.interval_start.str.contains(cur_day)],
"date":
cur_day
} for cur_day in days]
for daily_data in daily_data_frames:
day = daily_data.get("date")
df = daily_data.get("data_frame")
day_file = f"{report_type}.{day}.csv"
day_filepath = f"{directory}/{day_file}"
df.to_csv(day_filepath, index=False, header=True)
daily_files.append({"filename": day_file, "filepath": day_filepath})
return daily_files
def report_type(self):
"""Report type for OpenShift, else None."""
if self.provider_type == Provider.PROVIDER_OCP:
for file_name in self.file_list:
report_type, _ = detect_type(file_name)
if report_type:
return report_type
return None
