CHANGELOG.md

Release History

12.3.2 (2020-6-12)

Fixes

• Fixed issue where batch requests could not be combined with SAS (#9534)
• Batch requests now support applying parameters to individual blobs within the request via passing in a dictionary.
• Metadata cannot have leading space (#11457)
• Improve the performance of upload when using max_concurrency

Notes

• Updated dependency from azure-core<2.0.0,>=1.2.2 to azure-core<2.0.0,>=1.6.0

12.3.1 (2020-04-29)

Fixes

• Fixed issue where batch requests could not be combined with token credentials (#9534)
• Skip '/' in url encoding.

12.3.0 (2020-03-10)

New features

• stage_block now propagates the response from the service.

Fixes

• Fixed a bug where a new transport is being passed in the get_blob_client method instead of using the existing one in the ContainerClient.

Notes

• The StorageUserAgentPolicy is now replaced with the UserAgentPolicy from azure-core. With this, the custom user agents are now added as a prefix instead of being appended.

12.2.0

New features

• Added support for the 2019-07-07 service version, and added api_version parameter to clients.
• Added support for encryption scopes that that could be used to encrypt blob content.
• Added get_page_range_diff_for_managed_disk API which returns the list of valid page ranges diff between a snapshot and managed disk or another snapshot.

Fixes

• Responses are always decoded as UTF8

12.1.0 (2019-12-04)

New features

• Added download_blob method to the container_client.
• All the clients now have a close() method to close the sockets opened by the client when using without a context manager.

Fixes and improvements

• Fixes a bug where determining length breaks while uploading a blob when provided with an invalid fileno.
• Fix metadata not being included in commit_block_list operation.

12.0.0 (2019-10-31)

Breaking changes

• set_container_access_policy has required parameter signed_identifiers.
• NoRetry policy has been removed. Use keyword argument retry_total=0 for no retries.
• StorageStreamDownloader is no longer iterable. To iterate over the blob data stream, use StorageStreamDownloader.chunks.
• The public attributes of StorageStreamDownloader have been limited to:
  • name (str): The name of the blob.
  • container (str): The container the blob is being downloaded from.
  • properties (BlobProperties): The properties of the blob.
  • size (int): The size of the download. Either the total blob size, or the length of a subsection if sepcified. Previously called download_size.
• StorageStreamDownloader now has new functions:
  • readall(): Reads the complete download stream, returning bytes. This replaces the functions content_as_bytes and content_as_text which have been deprecated.
  • readinto(stream): Download the complete stream into the supplied writable stream, returning the number of bytes written. This replaces the function download_to_stream which has been deprecated.
• Module level functions upload_blob_to_url and download_blob_from_url functions options are now keyword only:
  • overwrite
  • max_concurrency
  • encoding
• Removed types that were accidentally exposed from two modules. Only BlobServiceClient, ContainerClient, BlobClient and BlobLeaseClient should be imported from azure.storage.blob.aio
• Logging has been renamed to BlobAnalyticsLogging.
• Client and model files have been made internal. Users should import from the top level modules azure.storage.blob and azure.storage.blob.aio only.
• All operations that take Etag conditional parameters (if_match and if_none_match) now take explicit etag and match_condition parameters, where etag is the Etag value, and match_condition is an instance of azure.core.MatchConditions.
• The generate_shared_access_signature methods on each of BlobServiceClient, ContainerClient and BlobClient have been replaced by module level functions generate_account_sas, generate_container_sas and generate_blob_sas.
• The batch APIs now have an additional keyword only argument raise_on_any_failure which defaults to True. This will raise an error even if there's a partial batch failure.
• LeaseClient has been renamed to BlobLeaseClient.
• get_service_stats now returns a dict
• get_service_properties now returns a dict with keys consistent to set_service_properties

New features

• Added async module-level upload_blob_to_url and download_blob_from_url functions.
• ResourceTypes, and Services now have method from_string which takes parameters as a string.

12.0.0b4 (2019-10-08)

Breaking changes

• Permission models.
  • AccountPermissions, BlobPermissions and ContainerPermissions have been renamed to AccountSasPermissions, BlobSasPermissions and ContainerSasPermissions respectively.
  • enum-like list parameters have been removed from all three of them.
  • __add__ and __or__ methods are removed.
• max_connections is now renamed to max_concurrency.
• ContainerClient now accepts only account_url with a mandatory string param container_name. To use a container_url, the method from_container_url must be used.
• BlobClient now accepts only account_url with mandatory string params container_name and blob_name. To use a blob_url, the method from_blob_url must be used.
• Some parameters have become keyword only, rather than positional. Some examples include:
  • loop
  • max_concurrency
  • validate_content
  • timeout etc.
• APIs now take in offset and length instead of range_start and range_end consistently. length is the number of bytes to take in starting from the offset. The APIs that have been changed include:
  • get_page_ranges
  • upload_page
  • upload_pages_from_url
  • clear_page
  • append_block_from_url
• block_id is not optional in BlobBlock model.

New features

• Add support for delete_blobs API to ContainerClient (Python 3 only)
• Add support for set_standard_blob_tier_blobs to ContainerClient (Python 3 only)
• Add support for set_premium_page_blob_tier_blobs to ContainerClient (Python 3 only)
• Added support to set rehydrate blob priority for Block Blob, including Set Standard Blob Tier/Copy Blob APIs
• Added blob tier support for Block Blob, including Upload Blob/Commit Block List/Copy Blob APIs.
• AccountSasPermissions, BlobSasPermissions, ContainerSasPermissions now have method from_string which takes parameters as a string.

Fixes and improvements

• Downloading page blobs now take advantage of their sparseness.
• The length param in download_blob now takes the number of bytes to take in starting from the offset instead of a harde set end value.

Dependency updates

• Adopted azure-core 1.0.0b4

  • If you later want to revert to previous versions of azure-storage-blob, or another Azure SDK library requiring azure-core 1.0.0b1 or azure-core 1.0.0b2, you must explicitly install the specific version of azure-core as well. For example:

  pip install azure-core==1.0.0b2 azure-storage-blob==12.0.0b2

12.0.0b3 (2019-09-10)

New features

• Added SAS support for snapshot and identity.
• Distributed tracing framework OpenCensus is now supported.
• Added support for append_block_from_url API for append blobs.
• Added support for upload_pages_from_url API for page blobs.
• Added support for client provided encryption key to numerous APIs.

Dependency updates

• Adopted azure-core 1.0.0b3

  • If you later want to revert to previous versions of azure-storage-blob, or another Azure SDK library requiring azure-core 1.0.0b1 or azure-core 1.0.0b2, you must explicitly install the specific version of azure-core as well. For example:

  pip install azure-core==1.0.0b2 azure-storage-blob==12.0.0b2

Fixes and improvements

• Fix where content-type was being added in the request when not mentioned explicitly.

12.0.0b2 (2019-08-06)

Breaking changes

• Renamed copy_blob_from_url to start_copy_from_url and changed behaviour to return a dictionary of copy properties rather than a polling object. Status of the copy operation can be retrieved with the get_blob_properties operation.
• Added abort_copy operation to the BlobClient class. This replaces the previous abort operation on the copy status polling operation.
• The behavior of listing operations has been modified:
  • The previous marker parameter has been removed.
  • The iterable response object now supports a by_page function that will return a secondary iterator of batches of results. This function supports a continuation_token parameter to replace the previous marker parameter.
• Some parameters have become keyword only, rather than positional. Some examples include:
  • timeout
  • lease
  • encoding
  • Modification conditions, e.g. if_modified_since, if_match , maxsize_condition, etc

New features

• Added async APIs to subnamespace azure.storage.blob.aio.
• Distributed tracing framework OpenCensus is now supported.

Dependency updates

• Adopted azure-core 1.0.0b2

  • If you later want to revert to azure-storage-blob 12.0.0b1, or another Azure SDK library requiring azure-core 1.0.0b1, you must explicitly install azure-core 1.0.0b1 as well. For example:

  pip install azure-core==1.0.0b1 azure-storage-blob==12.0.0b1

Fixes and improvements

• Fix for SAS URL encoding (#6500)
• General refactor of duplicate and shared code.

12.0.0b1 (2019-07-02)

Version 12.0.0b1 is the first preview of our efforts to create a user-friendly and Pythonic client library for Azure Storage Blobs. For more information about this, and preview releases of other Azure SDK libraries, please visit https://aka.ms/azure-sdk-preview1-python.

Breaking changes: New API design

• Operations are now scoped to a particular client:

  • BlobServiceClient: This client handles account-level operations. This includes managing service properties and listing the containers within an account.
  • ContainerClient: The client handles operations for a particular container. This includes creating or deleting that container, as well as listing the blobs within that container and managing properties and metadata.
  • BlobClient: The client handles operations for a particular blob. This includes creating or deleting that blob, as well as upload and download data and managing properties. This BlobClient handles all blob types (block, page and append). Where operations can behave differently according to type (i.e. upload_blob) the default behaviour will be block blobs unless otherwise specified.
  • LeaseClient: Handles all lease operations for both containers and blobs.

  These clients can be accessed by navigating down the client hierarchy, or instantiated directly using URLs to the resource (account, container or blob). For full details on the new API, please see the reference documentation.

• Copy blob operations now return a polling object that can be used to check the status of the operation, as well as abort the operation.

• New module level operations for simple upload and download using a blob URL.

• Download operations now return a streaming object that can download data in multiple ways:

  • Iteration: The streamer is an iterable object that will download and yield the content in chunks. Only supports single threaded download.
  • content_as_bytes: Return the entire blob content as bytes. Blocking operation that supports multi-threaded download.
  • content_as_text: Return the entire blob content as decoded text. Blocking operation that supports multi-threaded download.
  • download_to_stream: Download the entire content to an open stream handle (e.g. an open file). Supports multi-threaded download.

• New underlying REST pipeline implementation, based on the new azure-core library.

• Client and pipeline configuration is now available via keyword arguments at both the client level, and per-operation. See reference documentation for a full list of optional configuration arguments.

• Authentication using azure-identity credentials

  • see the Azure Identity documentation for more information

• New error hierarchy:

  • All service errors will now use the base type: azure.core.exceptions.HttpResponseError
  • The are a couple of specific exception types derived from this base type for common error scenarios:
    • ResourceNotFoundError: The resource (e.g. queue, message) could not be found. Commonly a 404 status code.
    • ResourceExistsError: A resource conflict - commonly caused when attempting to create a resource that already exists.
    • ResourceModifiedError: The resource has been modified (e.g. overwritten) and therefore the current operation is in conflict. Alternatively this may be raised if a condition on the operation is not met.
    • ClientAuthenticationError: Authentication failed.

• Operation set_blob_properties has been renamed to set_http_headers.

• Operations get_blob_to_<output> have been replaced with download_blob. See above for download output options.

• Operations create_blob_from_<input> have been replace with upload_blob.

• Operation create_blob has been renamed to separate create_page_blob and create_append_blob.

• Operations get_container_acl and set_container_acl have been renamed to get_container_access_policy and set_container_access_policy.

• Operation snapshot_blob has been renamed to create_snapshot.

• Operation copy_blob has been renamed to copy_blob_from_url.

• Operations put_block and put_block_from_url have been renamed to stage_block and stage_block_from_url.

• Operation put_block_list has been renamed to commit_block_list.

• No longer have specific operations for get_metadata - use get_properties instead.

• No longer have specific operations for exists - use get_properties instead.

• Operation incremental_copy_blob has been replaced by an optional boolean flag in the copy_blob_from_url operation.

• Operation update_page has been renamed to upload_page.

• Operation get_page_ranges_diff has been replaced by an optional str flag in the get_page_ranges operation.

2.0.1

• Updated dependency on azure-storage-common.

2.0.0

• Support for 2018-11-09 REST version. Please see our REST API documentation and blog for information about the related added features.
• Added support for append block from URL(synchronously) for append blobs.
• Added support for update page from URL(synchronously) for page blobs.
• Added support for generating and using blob snapshot SAS tokens.
• Added support for generating user delegation SAS tokens.

1.5.0

• Added new method list_blob_names to efficiently list only blob names in an efficient way.

1.4.0

• azure-storage-nspkg is not installed anymore on Python 3 (PEP420-based namespace package)
• copy_blob method added to BlockBlobService to enable support for deep sync copy.

1.3.1

• Fixed design flaw where get_blob_to_* methods buffer entire blob when max_connections is set to 1.
• Added support for access conditions on append_blob_from_* methods.

1.3.0

• Support for 2018-03-28 REST version. Please see our REST API documentation and blog for information about the related added features.
• Added support for setting static website service properties.
• Added support for getting account information, such as SKU name and account kind.
• Added support for put block from URL(synchronously).

1.2.0rc1

• Support for 2017-11-09 REST version. Please see our REST API documentation and blog for information about the related added features.
• Support for write-once read-many containers.
• Added support for OAuth authentication for HTTPS requests(Please note that this feature is available in preview).

1.1.0

• Support for 2017-07-29 REST version. Please see our REST API documentation and blogs for information about the related added features.
• Added support for soft delete feature. If a delete retention policy is enabled through the set service properties API, then blobs or snapshots could be deleted softly and retained for a specified number of days, before being permanently removed by garbage collection.
• Error message now contains the ErrorCode from the x-ms-error-code header value.

1.0.0

• The package has switched from Apache 2.0 to the MIT license.
• Fixed bug where get_blob_to_* cannot get a single byte when start_range and end_range are both equal to 0.
• Optimized page blob upload for create_blob_from_* methods, by skipping the empty chunks.
• Added convenient method to generate container url (make_container_url).
• Metadata keys are now case-preserving when fetched from the service. Previously they were made lower-case by the library.

0.37.1

• Enabling MD5 validation no longer uses the memory-efficient algorithm for large block blobs, since computing the MD5 hash requires reading the entire block into memory.
• Fixed a bug in the _SubStream class which was at risk of causing data corruption when using the memory-efficient algorithm for large block blobs.
• Support for AccessTierChangeTime to get the last time a tier was modified on an individual blob.