def validate_speechFile(cls, speech_file: UploadedFile) -> None: # pylint: disable=invalid-name
"""Проверить mimetype - принимать только FLAC/WAVE файлы"""
if magic.from_buffer(speech_file.read(
cls.MAXIMAL_MAGIC_BYTE_LENGTH_FOR_ACCEPTED_FILETYPES),
mime=True) not in [
# pylint: disable=bad-continuation
'audio/flac',
'audio/x-wav',
]:
# pylint: enable=bad-continuation
raise exceptions.UnsupportedMediaTypeError('Invalid speech file')
speech_file.open()
def __csv_to_dataframe(self, file_data: UploadedFile,
is_no_header: bool) -> pd.DataFrame:
with file_data.open(mode='rt') as f:
lines = f.readlines()
encode = self.__get_upload_csv_encode(lines)
lines_data = [str(line, encoding=encode) for line in lines]
reader = csv.reader(lines_data)
# self.__logger.debug(f"reader = {reader}")
# self.__logger.debug(f"reader.dialect = {reader.dialect}")
# self.__logger.debug(f"reader.line_num = {reader.line_num}")
header = []
header_format = "列{header}"
if is_no_header:
row = next(reader)
for i in range(len(row)):
header.append(header_format.format(header=(i + 1)))
reader = csv.reader(lines_data)
else:
header = next(reader)
# self.__logger.debug(f"header = {header}")
# self.__logger.debug(f"reader = {reader}")
# self.__logger.debug(f"reader.line_num = {reader.line_num}")
result = pd.DataFrame(data=reader, index=None, columns=header)
self.__logger.debug(f"result = {result}")
return result
