def resolve(cls, uri: str) -> Iterable[str]:
"""Resolve the dataset URI, and returns a list of parquet files."""
parsed = urlparse(uri)
if parsed.scheme in cls._RESOLVERS:
logger.debug("Use extended resolver for scheme: %s", parsed.scheme)
return cls._RESOLVERS[parsed.scheme].resolve(uri)
return cls._RESOLVERS[cls._DEFAULT_SCHEME].resolve(uri)
def validate(self):
"""Validate model spec
Raises
------
SpecError
If the spec is not well-formatted.
"""
logger.debug("Validating spec: %s", self._spec)
try:
validate(instance=self._spec, schema=MODEL_SPEC_SCHEMA)
except ValidationError as e:
raise SpecError(e.message) from e
def get_schema(cls, uri: str):
"""Get the schema of the dataset
Parameters
----------
uri : str
URI of the dataset
"""
parsed = urlparse(uri)
if parsed.scheme in cls._RESOLVERS:
logger.debug("Use extended resolver for scheme: %s", parsed.scheme)
return cls._RESOLVERS[parsed.scheme].get_schema(uri)
return cls._RESOLVERS[cls._DEFAULT_SCHEME].get_schema(uri)
def copy(source: str, dest: str) -> str:
"""Copy a file from source to destination, and return the URI of
the copied file.
Parameters
----------
source : str
The source URI to copy from
dest : str
The destination uri or the destionation directory. If ``dest`` is
a URI ends with a "/", it represents a directory.
Return
------
str
Return the URI of destination.
"""
source = _normalize_uri(source)
dest = _normalize_uri(dest)
parsed_source = urlparse(source)
if dest and dest.endswith("/"):
dest = join(dest, basename(parsed_source.path))
parsed_dest = urlparse(dest)
logger.debug("Copying %s to %s", source, dest)
if parsed_dest.scheme == parsed_source.scheme:
# Direct copy with the same file system
if parsed_dest.scheme == "s3":
s3fs, source_path = fs.FileSystem.from_uri(source)
_, dest_path = fs.FileSystem.from_uri(dest)
s3fs.copy(source_path, dest_path)
return dest
# TODO: find better i/o utilis to copy between filesystems
filesystem, dest_path = fs.FileSystem.from_uri(dest)
with filesystem.open_output_stream(dest_path) as out_stream:
src_fs, src_path = fs.FileSystem.from_uri(source)
with src_fs.open_input_stream(src_path) as in_stream:
while True:
buf = in_stream.read(_BUFSIZE)
if not buf:
break
out_stream.write(buf)
return dest
def copy(source: str, dest: str) -> str:
"""Copy a file from source to destination, and return the URI of
the copied file.
Parameters
----------
source : str
The source URI to copy from
dest : str
The destination uri or the destination directory. If ``dest`` is
a URI ends with a "/", it represents a directory.
Return
------
str
Return the URI of destination.
"""
source = _normalize_uri(source)
dest = _normalize_uri(dest)
parsed_source = urlparse(source)
if dest and dest.endswith("/"):
dest = join(dest, basename(parsed_source.path))
parsed_dest = urlparse(dest)
logger.debug("Copying %s to %s", source, dest)
if parsed_dest.scheme == parsed_source.scheme:
# Direct copy with the same file system
scheme = parsed_dest.scheme
if scheme == "s3":
s3fs, source_path = fs.FileSystem.from_uri(source)
_, dest_path = fs.FileSystem.from_uri(dest)
s3fs.copy(source_path, dest_path)
return dest
elif scheme == "gs":
_gcsfs().copy(source, dest)
return dest
with _open_output_stream(dest) as out_stream, open_input_stream(
source
) as in_stream:
shutil.copyfileobj(in_stream, out_stream)
return dest
