def from_json(filepath: Union[str, Path],
encoding='utf-8',
logger: Logger = silent_logger(),
**kwargs):
"""Load dictionary from json file"""
filepath = plain_path(filepath)
logger.info(f'Load data from {filepath} ...')
with filepath.open(mode='r', encoding=encoding) as f:
return json.load(f, **kwargs)
def read_lines(filepath: Union[str, Path],
logger: Logger = silent_logger(),
**kwargs) -> Iterator[str]:
"""Read lines from text file"""
filepath = plain_path(filepath)
kwargs.setdefault('mode', 'r')
kwargs.setdefault('encoding', 'utf-8')
logger.info(f'Read lines from {filepath} ...')
with filepath.open(**kwargs) as f:
for line in f:
yield line.rstrip()
def load_pickle(filepath: Union[str, Path],
use_dill: bool = False,
logger: Logger = silent_logger()):
"""Deserialize object with pickle"""
filepath = plain_path(filepath)
logger.info(f'Load data from {filepath} ...')
serializer = dill if use_dill else pickle
compress = str(filepath).endswith('.gz')
file = gzip.open(filepath, mode='rb') if compress else filepath.open(
mode='rb')
with file as f:
return serializer.load(f)
def write_lines(filepath: Union[str, Path],
lines: List[str],
logger: Logger = silent_logger(),
**kwargs):
"""Write lines into text file"""
filepath = ensure_filedir(filepath)
kwargs.setdefault('mode', 'w')
kwargs.setdefault('encoding', 'utf-8')
logger.info(f'Write {len(lines)} lines into {filepath} ...')
with filepath.open(**kwargs) as f:
for line in lines:
f.write(line + '\n')
def to_json(filepath: Union[str, Path],
data: dict,
encoding='utf-8',
logger: Logger = silent_logger(),
**kwargs):
"""Dump dictionary into json file"""
filepath = ensure_filedir(filepath)
logger.info(f'Dump data to {filepath} ...')
kwargs.setdefault('indent', 4)
kwargs.setdefault('ensure_ascii', False)
with filepath.open(mode='w', encoding=encoding) as f:
json.dump(data, f, **kwargs)
def from_tsv(filepath: Union[str, Path],
logger: Logger = silent_logger(),
**kwargs) -> pd.DataFrame:
"""Load table from tsv file"""
filepath = plain_path(filepath)
kwargs.setdefault('sep', '\t')
kwargs.setdefault('na_values', NAN_IDENTIFIER)
kwargs.setdefault('keep_default_na', False)
kwargs.setdefault('dtype', object)
if str(filepath).endswith('.gz'):
kwargs.setdefault('compression', 'gzip')
logger.info(f'Load data from {filepath} ...')
return pd.read_csv(filepath, **kwargs)
def to_tsv(filepath: Union[str, Path],
data: pd.DataFrame,
logger: Logger = silent_logger(),
**kwargs):
"""Write table into tsv file"""
filepath = ensure_filedir(filepath)
kwargs.setdefault('sep', '\t')
kwargs.setdefault('na_rep', NAN_IDENTIFIER)
kwargs.setdefault('encoding', 'utf-8')
kwargs.setdefault('index', False)
if str(filepath).endswith('.gz'):
kwargs.setdefault('compression', 'gzip')
logger.info(f'Dump {len(data)} rows into {filepath} ...')
data.to_csv(filepath, **kwargs)
def dump_pickle(filepath: Union[str, Path],
obj: "serializable object",
use_dill: bool = False,
skip_fields: List[str] = None,
logger: Logger = silent_logger()):
"""Serialize object with pickle"""
filepath = ensure_filedir(filepath)
logger.info(f'Dump data into {filepath} ...')
skip_fields = [] if not skip_fields else skip_fields
protocol = 2 if sys.version_info[0] == 2 else 4
serializer = dill if use_dill else pickle
compress = str(filepath).endswith('.gz')
with ExitStack() as stack, gzip.open(
filepath, mode='wb') if compress else filepath.open(
mode='wb') as file:
for field in skip_fields:
stack.enter_context(CachedObject.parse_from(obj=obj, field=field))
try:
serializer.dump(obj, file, protocol=protocol)
except RecursionError:
logger.warning('Set recursion limit to 10000!')
sys.setrecursionlimit(10000)
serializer.dump(obj, file, protocol=protocol)
