def load_file_content(fil):
"""
Returns content of files. Designed for very generic file formats:
so far supported contents are `json`, `npy`, `csv`, `tsv`, `ssv`, `jsonable`
:param fil: file to read
:return:array/json/pandas dataframe depending on format
"""
if not fil:
return
fil = Path(fil)
if fil.stat().st_size == 0:
return
if fil.suffix == '.csv':
return pd.read_csv(fil)
if fil.suffix == '.json':
try:
with open(fil) as _fil:
return json.loads(_fil.read())
except Exception as e:
_logger.error(e)
return None
if fil.suffix == '.jsonable':
return jsonable.read(fil)
if fil.suffix == '.npy':
return np.load(file=fil)
if fil.suffix == '.pqt':
return parquet.load(fil)
if fil.suffix == '.ssv':
return pd.read_csv(fil, delimiter=' ')
if fil.suffix == '.tsv':
return pd.read_csv(fil, delimiter='\t')
return Path(fil)
def __init__(self,
username=None,
password=None,
base_url=None,
cache_dir=None,
silent=None):
# get parameters override if inputs provided
self._par = oneibl.params.get(silent=silent)
# can delete those 2 lines from mid January 2021
if self._par.HTTP_DATA_SERVER == 'http://ibl.flatironinstitute.org':
self._par = self._par.set("HTTP_DATA_SERVER",
"https://ibl.flatironinstitute.org")
self._par = self._par.set('ALYX_LOGIN', username
or self._par.ALYX_LOGIN)
self._par = self._par.set('ALYX_URL', base_url or self._par.ALYX_URL)
self._par = self._par.set('ALYX_PWD', password or self._par.ALYX_PWD)
self._par = self._par.set('CACHE_DIR', cache_dir
or self._par.CACHE_DIR)
# init the cache file
self._cache_file = Path(
self._par.CACHE_DIR).joinpath('.one_cache.parquet')
if self._cache_file.exists():
# we need to keep this part fast enough for transient objects
self._cache = parquet.load(self._cache_file)
else:
self._cache = pd.DataFrame()
def __init__(self, username=None, password=None, base_url=None, cache_dir=None, silent=None):
# get parameters override if inputs provided
self._par = oneibl.params.get(silent=silent)
self._par = self._par.set('ALYX_LOGIN', username or self._par.ALYX_LOGIN)
self._par = self._par.set('ALYX_URL', base_url or self._par.ALYX_URL)
self._par = self._par.set('ALYX_PWD', password or self._par.ALYX_PWD)
self._par = self._par.set('CACHE_DIR', cache_dir or self._par.CACHE_DIR)
# init the cache file
self._cache_file = Path(self._par.CACHE_DIR).joinpath('.one_cache.parquet')
if self._cache_file.exists():
# we need to keep this part fast enough for transient objects
self._cache = parquet.load(self._cache_file)
else:
self._cache = pd.DataFrame()
