def upload_blob(self, container: Container, filename: FileLike,
blob_name: str = None, acl: str = None,
meta_data: MetaData = None, content_type: str = None,
content_disposition: str = None, cache_control: str = None,
chunk_size: int = 1024,
extra: ExtraOptions = None) -> Blob:
meta_data = {} if meta_data is None else meta_data
extra = {} if extra is None else extra
blob_name = blob_name or validate_file_or_path(filename)
if not content_type:
if isinstance(filename, str):
content_type = file_content_type(filename)
else:
content_type = file_content_type(blob_name)
if isinstance(filename, str):
self.client.fput_object(container.name,
blob_name,
filename,
content_type=content_type,
metadata=meta_data)
else:
length = extra.pop('length', len(filename.read()))
filename.seek(0)
self.client.put_object(container.name,
blob_name,
filename,
length,
content_type=content_type,
metadata=meta_data)
return self.get_blob(container, blob_name)
def download_blob(self, blob: Blob, destination: FileLike) -> None:
data = self.client.get_object(blob.container.name, blob.name)
if isinstance(destination, str):
with open(destination, "wb") as blob_data:
for d in data.stream(4096):
blob_data.write(d)
else:
for d in data.stream(4096):
destination.write(d)
def download_blob(self, blob: Blob, destination: FileLike) -> None:
try:
data = self.object_store.download_object(
obj=blob.name, container=blob.container.name)
if isinstance(destination, str):
with open(destination, 'wb') as out:
out.write(data)
else:
destination.write(data)
except ResourceNotFound:
raise NotFoundError(messages.BLOB_NOT_FOUND %
(blob.name, blob.container.name))
def file_checksum(filename: FileLike,
hash_type: str = 'md5',
block_size: int = 4096) -> HASH:
"""Returns checksum for file.
.. code-block:: python
from cloudstorage.helpers import file_checksum
picture_path = '/path/picture.png'
file_checksum(picture_path, hash_type='sha256')
# '03ef90ba683795018e541ddfb0ae3e958a359ee70dd4fccc7e747ee29b5df2f8'
Source: `get-md5-hash-of-big-files-in-python <https://stackoverflow.com/
questions/1131220/get-md5-hash-of-big-files-in-python>`_
:param filename: File path or stream.
:type filename: str or FileLike
:param hash_type: Hash algorithm function name.
:type hash_type: str
:param block_size: (optional) Chunk size.
:type block_size: int
:return: Hash of file.
:rtype: :class:`_hashlib.HASH`
:raise RuntimeError: If the hash algorithm is not found in :mod:`hashlib`.
.. versionchanged:: 0.4
Returns :class:`_hashlib.HASH` instead of `HASH.hexdigest()`.
"""
try:
file_hash = getattr(hashlib, hash_type)()
except AttributeError:
raise RuntimeError('Invalid or unsupported hash type: %s' % hash_type)
if isinstance(filename, str):
with open(filename, 'rb') as file_:
for chunk in read_in_chunks(file_, block_size=block_size):
file_hash.update(chunk)
else:
for chunk in read_in_chunks(filename, block_size=block_size):
file_hash.update(chunk)
# rewind the stream so it can be re-read later
if filename.seekable():
filename.seek(0)
return file_hash
def download_blob(self, blob: Blob, destination: FileLike) -> None:
blob_path = self._get_file_path(blob)
if isinstance(destination, str):
base_name = os.path.basename(destination)
if not base_name and not os.path.exists(destination):
raise CloudStorageError("Path %s does not exist." %
destination)
if not base_name:
file_path = os.path.join(destination, blob.name)
else:
file_path = destination
shutil.copy(blob_path, file_path)
else:
with open(blob_path, "rb") as blob_file:
for data in read_in_chunks(blob_file):
destination.write(data)
def read_in_chunks(file_object: FileLike,
block_size: int = 4096) -> Generator[bytes, None, None]:
"""Return a generator which yields data in chunks.
Source: `read-file-in-chunks-ram-usage-read-strings-from-binary-file
<https://stackoverflow.com/questions/17056382/
read-file-in-chunks-ram-usage-read-strings-from-binary-files>`_
:param file_object: File object to read in chunks.
:type file_object: file object
:param block_size: (optional) Chunk size.
:type block_size: int
:yield: The next chunk in file object.
:yield type: `bytes`
"""
for chunk in iter(lambda: file_object.read(block_size), b''):
yield chunk
