Module transfer_manager (2.14.0)

Concurrent media operations.

Modules Functions

download_chunks_concurrently

download_chunks_concurrently(
    blob,
    filename,
    chunk_size=33554432,
    download_kwargs=None,
    deadline=None,
    worker_type="process",
    max_workers=8,
    *,
    crc32c_checksum=True
)

Download a single file in chunks, concurrently.

In some environments, using this feature with mutiple processes will result in faster downloads of large files.

Using this feature with multiple threads is unlikely to improve download performance under normal circumstances due to Python interpreter threading behavior. The default is therefore to use processes instead of threads.

Parameters
NameDescription
blob Blob

The blob to be downloaded.

filename str

The destination filename or path.

chunk_size int

The size in bytes of each chunk to send. The optimal chunk size for maximum throughput may vary depending on the exact network environment and size of the blob.

download_kwargs dict

A dictionary of keyword arguments to pass to the download method. Refer to the documentation for blob.download_to_file() or blob.download_to_filename() for more information. The dict is directly passed into the download methods and is not validated by this function. Keyword arguments "start" and "end" which are not supported and will cause a ValueError if present. The key "checksum" is also not supported in download_kwargs, but see the argument crc32c_checksum (which does not go in download_kwargs) below.

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

crc32c_checksum bool

Whether to compute a checksum for the resulting object, using the crc32c algorithm. As the checksums for each chunk must be combined using a feature of crc32c that is not available for md5, md5 is not supported.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded. google.resumable_media.common.DataCorruption if the download's checksum doesn't agree with server-computed checksum. The google.resumable_media exception is used here for consistency with other download methods despite the exception originating elsewhere.

download_many

download_many(
    blob_file_pairs,
    download_kwargs=None,
    threads=None,
    deadline=None,
    raise_exception=False,
    worker_type="process",
    max_workers=8,
    *,
    skip_if_exists=False
)

Download many blobs concurrently via a worker pool.

Parameters
NameDescription
blob_file_pairs List(Tuple('google.cloud.storage.blob.Blob', IOBase or str))

A list of tuples of blob and a file or filename. Each blob will be downloaded to the corresponding blob by using APIs identical to blob.download_to_file() or blob.download_to_filename() as appropriate. Note that blob.download_to_filename() does not delete the destination file if the download fails. File handlers are only supported if worker_type is set to THREAD. If worker_type is set to PROCESS, please use filenames only.

download_kwargs dict

A dictionary of keyword arguments to pass to the download method. Refer to the documentation for blob.download_to_file() or blob.download_to_filename() for more information. The dict is directly passed into the download methods and is not validated by this function.

threads int

DEPRECATED Sets worker_type to THREAD and max_workers to the number specified. If worker_type or max_workers are set explicitly, this parameter should be set to None. Please use worker_type and max_workers instead of this parameter.

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

raise_exception bool

If True, instead of adding exceptions to the list of return values, instead they will be raised. Note that encountering an exception on one operation will not prevent other operations from starting. Exceptions are only processed and potentially raised after all operations are complete in success or failure.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations. PROCESS workers do not support writing to file handlers. Please refer to files by filename only when using PROCESS workers.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

skip_if_exists bool

Before downloading each blob, check if the file for the filename exists; if it does, skip that blob.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded.
Returns
TypeDescription
listA list of results corresponding to, in order, each item in the input list. If an exception was received, it will be the result for that operation. Otherwise, the return value from the successful download method is used (which will be None).

download_many_to_path

download_many_to_path(
    bucket,
    blob_names,
    destination_directory="",
    blob_name_prefix="",
    download_kwargs=None,
    threads=None,
    deadline=None,
    create_directories=True,
    raise_exception=False,
    worker_type="process",
    max_workers=8,
    *,
    skip_if_exists=False
)

Download many files concurrently by their blob names.

The destination files are automatically created, with paths based on the source blob_names and the destination_directory.

The destination files are not automatically deleted if their downloads fail, so please check the return value of this function for any exceptions, or enable raise_exception=True, and process the files accordingly.

For example, if the blob_names include "icon.jpg", destination_directory is "/home/myuser/", and blob_name_prefix is "images/", then the blob named "images/icon.jpg" will be downloaded to a file named "/home/myuser/icon.jpg".

Parameters
NameDescription
bucket Bucket

The bucket which contains the blobs to be downloaded

blob_names list(str)

A list of blobs to be downloaded. The blob name in this string will be used to determine the destination file path as well. The full name to the blob must be blob_name_prefix + blob_name. The blob_name is separate from the blob_name_prefix because the blob_name will also determine the name of the destination blob. Any shared part of the blob names that need not be part of the destination path should be included in the blob_name_prefix.

destination_directory str

A string that will be prepended (with os.path.join()) to each blob_name in the input list, in order to determine the destination path for that blob. For instance, if the destination_directory string is "/tmp/img" and a blob_name is "0001.jpg", with an empty blob_name_prefix, then the source blob "0001.jpg" will be downloaded to destination "/tmp/img/0001.jpg" . This parameter can be an empty string. Note that this parameter allows directory traversal (e.g. "/", "../") and is not intended for unsanitized end user input.

blob_name_prefix str

A string that will be prepended to each blob_name in the input list, in order to determine the name of the source blob. Unlike the blob_name itself, the prefix string does not affect the destination path on the local filesystem. For instance, if the destination_directory is "/tmp/img/", the blob_name_prefix is "myuser/mystuff-" and a blob_name is "0001.jpg" then the source blob "myuser/mystuff-0001.jpg" will be downloaded to "/tmp/img/0001.jpg". The blob_name_prefix can be blank (an empty string).

download_kwargs dict

A dictionary of keyword arguments to pass to the download method. Refer to the documentation for blob.download_to_file() or blob.download_to_filename() for more information. The dict is directly passed into the download methods and is not validated by this function.

threads int

DEPRECATED Sets worker_type to THREAD and max_workers to the number specified. If worker_type or max_workers are set explicitly, this parameter should be set to None. Please use worker_type and max_workers instead of this parameter.

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

create_directories bool

If True, recursively create any directories that do not exist. For instance, if downloading object "images/img001.png", create the directory "images" before downloading.

raise_exception bool

If True, instead of adding exceptions to the list of return values, instead they will be raised. Note that encountering an exception on one operation will not prevent other operations from starting. Exceptions are only processed and potentially raised after all operations are complete in success or failure. If skip_if_exists is True, 412 Precondition Failed responses are considered part of normal operation and are not raised as an exception.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

skip_if_exists bool

Before downloading each blob, check if the file for the filename exists; if it does, skip that blob. This only works for filenames.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded.
Returns
TypeDescription
listA list of results corresponding to, in order, each item in the input list. If an exception was received, it will be the result for that operation. Otherwise, the return value from the successful download method is used (which will be None).

upload_chunks_concurrently

upload_chunks_concurrently(filename, blob, content_type=None, chunk_size=33554432, deadline=None, worker_type='process', max_workers=8, *, checksum='md5', timeout=60, retry=<google.api_core.retry.Retry object>)

Upload a single file in chunks, concurrently.

This function uses the XML MPU API to initialize an upload and upload a file in chunks, concurrently with a worker pool.

The XML MPU API is significantly different from other uploads; please review the documentation at https://cloud.google.com/storage/docs/multipart-uploads before using this feature.

The library will attempt to cancel uploads that fail due to an exception. If the upload fails in a way that precludes cancellation, such as a hardware failure, process termination, or power outage, then the incomplete upload may persist indefinitely. To mitigate this, set the AbortIncompleteMultipartUpload with a nonzero Age in bucket lifecycle rules, or refer to the XML API documentation linked above to learn more about how to list and delete individual downloads.

Using this feature with multiple threads is unlikely to improve upload performance under normal circumstances due to Python interpreter threading behavior. The default is therefore to use processes instead of threads.

ACL information cannot be sent with this function and should be set separately with ObjectACL methods.

Parameters
NameDescription
filename str

The path to the file to upload. File-like objects are not supported.

blob Blob

The blob to which to upload.

content_type str

(Optional) Type of content being uploaded.

chunk_size int

The size in bytes of each chunk to send. The optimal chunk size for maximum throughput may vary depending on the exact network environment and size of the blob. The remote API has restrictions on the minimum and maximum size allowable, see: https://cloud.google.com/storage/quotas#requests

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

checksum str

(Optional) The checksum scheme to use: either "md5", "crc32c" or None. Each individual part is checksummed. At present, the selected checksum rule is only applied to parts and a separate checksum of the entire resulting blob is not computed. Please compute and compare the checksum of the file to the resulting blob separately if needed, using the "crc32c" algorithm as per the XML MPU documentation.

timeout float or tuple

(Optional) The amount of time, in seconds, to wait for the server response. See: configuring_timeouts

retry google.api_core.retry.Retry

(Optional) How to retry the RPC. A None value will disable retries. A google.api_core.retry.Retry value will enable retries, and the object will configure backoff and timeout options. Custom predicates (customizable error codes) are not supported for media operations such as this one. This function does not accept ConditionalRetryPolicy values because preconditions are not supported by the underlying API call. See the retry.py source code and docstrings in this package (google.cloud.storage.retry) for information on retry types and how to configure them.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded.

upload_many

upload_many(
    file_blob_pairs,
    skip_if_exists=False,
    upload_kwargs=None,
    threads=None,
    deadline=None,
    raise_exception=False,
    worker_type="process",
    max_workers=8,
)

Upload many files concurrently via a worker pool.

Parameters
NameDescription
file_blob_pairs List(Tuple(IOBase or str, 'google.cloud.storage.blob.Blob'))

A list of tuples of a file or filename and a blob. Each file will be uploaded to the corresponding blob by using APIs identical to blob.upload_from_file() or blob.upload_from_filename() as appropriate. File handlers are only supported if worker_type is set to THREAD. If worker_type is set to PROCESS, please use filenames only.

skip_if_exists bool

If True, blobs that already have a live version will not be overwritten. This is accomplished by setting if_generation_match = 0 on uploads. Uploads so skipped will result in a 412 Precondition Failed response code, which will be included in the return value but not raised as an exception regardless of the value of raise_exception.

upload_kwargs dict

A dictionary of keyword arguments to pass to the upload method. Refer to the documentation for blob.upload_from_file() or blob.upload_from_filename() for more information. The dict is directly passed into the upload methods and is not validated by this function.

threads int

DEPRECATED Sets worker_type to THREAD and max_workers to the number specified. If worker_type or max_workers are set explicitly, this parameter should be set to None. Please use worker_type and max_workers instead of this parameter.

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

raise_exception bool

If True, instead of adding exceptions to the list of return values, instead they will be raised. Note that encountering an exception on one operation will not prevent other operations from starting. Exceptions are only processed and potentially raised after all operations are complete in success or failure. If skip_if_exists is True, 412 Precondition Failed responses are considered part of normal operation and are not raised as an exception.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations. PROCESS workers do not support writing to file handlers. Please refer to files by filename only when using PROCESS workers.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded.
Returns
TypeDescription
listA list of results corresponding to, in order, each item in the input list. If an exception was received, it will be the result for that operation. Otherwise, the return value from the successful upload method is used (which will be None).

upload_many_from_filenames

upload_many_from_filenames(
    bucket,
    filenames,
    source_directory="",
    blob_name_prefix="",
    skip_if_exists=False,
    blob_constructor_kwargs=None,
    upload_kwargs=None,
    threads=None,
    deadline=None,
    raise_exception=False,
    worker_type="process",
    max_workers=8,
    *,
    additional_blob_attributes=None
)

Upload many files concurrently by their filenames.

The destination blobs are automatically created, with blob names based on the source filenames and the blob_name_prefix.

For example, if the filenames include "images/icon.jpg", source_directory is "/home/myuser/", and blob_name_prefix is "myfiles/", then the file at "/home/myuser/images/icon.jpg" will be uploaded to a blob named "myfiles/images/icon.jpg".

Parameters
NameDescription
bucket Bucket

The bucket which will contain the uploaded blobs.

filenames list(str)

A list of filenames to be uploaded. This may include part of the path. The file will be accessed at the full path of source_directory + filename.

source_directory str

A string that will be prepended (with os.path.join()) to each filename in the input list, in order to find the source file for each blob. Unlike the filename itself, the source_directory does not affect the name of the uploaded blob. For instance, if the source_directory is "/tmp/img/" and a filename is "0001.jpg", with an empty blob_name_prefix, then the file uploaded will be "/tmp/img/0001.jpg" and the destination blob will be "0001.jpg". This parameter can be an empty string. Note that this parameter allows directory traversal (e.g. "/", "../") and is not intended for unsanitized end user input.

blob_name_prefix str

A string that will be prepended to each filename in the input list, in order to determine the name of the destination blob. Unlike the filename itself, the prefix string does not affect the location the library will look for the source data on the local filesystem. For instance, if the source_directory is "/tmp/img/", the blob_name_prefix is "myuser/mystuff-" and a filename is "0001.jpg" then the file uploaded will be "/tmp/img/0001.jpg" and the destination blob will be "myuser/mystuff-0001.jpg". The blob_name_prefix can be blank (an empty string).

skip_if_exists bool

If True, blobs that already have a live version will not be overwritten. This is accomplished by setting if_generation_match = 0 on uploads. Uploads so skipped will result in a 412 Precondition Failed response code, which will be included in the return value, but not raised as an exception regardless of the value of raise_exception.

blob_constructor_kwargs dict

A dictionary of keyword arguments to pass to the blob constructor. Refer to the documentation for blob.Blob() for more information. The dict is directly passed into the constructor and is not validated by this function. name and bucket keyword arguments are reserved by this function and will result in an error if passed in here.

upload_kwargs dict

A dictionary of keyword arguments to pass to the upload method. Refer to the documentation for blob.upload_from_file() or blob.upload_from_filename() for more information. The dict is directly passed into the upload methods and is not validated by this function.

threads int

DEPRECATED Sets worker_type to THREAD and max_workers to the number specified. If worker_type or max_workers are set explicitly, this parameter should be set to None. Please use worker_type and max_workers instead of this parameter.

deadline int

The number of seconds to wait for all threads to resolve. If the deadline is reached, all threads will be terminated regardless of their progress and concurrent.futures.TimeoutError will be raised. This can be left as the default of None (no deadline) for most use cases.

raise_exception bool

If True, instead of adding exceptions to the list of return values, instead they will be raised. Note that encountering an exception on one operation will not prevent other operations from starting. Exceptions are only processed and potentially raised after all operations are complete in success or failure. If skip_if_exists is True, 412 Precondition Failed responses are considered part of normal operation and are not raised as an exception.

worker_type str

The worker type to use; one of google.cloud.storage.transfer_manager.PROCESS or google.cloud.storage.transfer_manager.THREAD. Although the exact performance impact depends on the use case, in most situations the PROCESS worker type will use more system resources (both memory and CPU) and result in faster operations than THREAD workers. Because the subprocesses of the PROCESS worker type can't access memory from the main process, Client objects have to be serialized and then recreated in each subprocess. The serialization of the Client object for use in subprocesses is an approximation and may not capture every detail of the Client object, especially if the Client was modified after its initial creation or if Client._http was modified in any way. THREAD worker types are observed to be relatively efficient for operations with many small files, but not for operations with large files. PROCESS workers are recommended for large file operations.

max_workers int

The maximum number of workers to create to handle the workload. With PROCESS workers, a larger number of workers will consume more system resources (memory and CPU) at once. How many workers is optimal depends heavily on the specific use case, and the default is a conservative number that should work okay in most cases without consuming excessive resources.

additional_blob_attributes dict

A dictionary of blob attribute names and values. This allows the configuration of blobs beyond what is possible with blob_constructor_kwargs. For instance, {"cache_control": "no-cache"} would set the cache_control attribute of each blob to "no-cache". As with blob_constructor_kwargs, this affects the creation of every blob identically. To fine-tune each blob individually, use upload_many and create the blobs as desired before passing them in.

Exceptions
TypeDescription
`concurrent.futures.TimeoutErrorif deadline is exceeded.
Returns
TypeDescription
listA list of results corresponding to, in order, each item in the input list. If an exception was received, it will be the result for that operation. Otherwise, the return value from the successful upload method is used (which will be None).