[Storage] Decompression for Binary Response in Download APIs

sdk/storage/azure-storage-blob/azure/storage/blob/_blob_client.py

@@ -732,6 +732,8 @@ def download_blob(
             function(current: int, total: int) where current is the number of bytes transferred
             so far, and total is the total size of the download.
         :paramtype progress_hook: Callable[[int, int], None]
+        :keyword bool decompress: If True, any compressed content, identified by the Content-Type header, will be
+            decompressed automatically before being returned. Default value is True.
         :keyword int timeout:
             Sets the server-side timeout for the operation in seconds. For more details see
             https://learn.microsoft.com/rest/api/storageservices/setting-timeouts-for-blob-service-operations.

sdk/storage/azure-storage-blob/azure/storage/blob/_blob_client.pyi

@@ -206,6 +206,7 @@ class BlobClient(StorageAccountHostsMixin, StorageEncryptionMixin):
         max_concurrency: int = 1,
         encoding: str,
         progress_hook: Optional[Callable[[int, int], None]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> StorageStreamDownloader[str]: ...
@@ -227,6 +228,7 @@ class BlobClient(StorageAccountHostsMixin, StorageEncryptionMixin):
         max_concurrency: int = 1,
         encoding: None = None,
         progress_hook: Optional[Callable[[int, int], None]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> StorageStreamDownloader[bytes]: ...
@@ -248,6 +250,7 @@ class BlobClient(StorageAccountHostsMixin, StorageEncryptionMixin):
         max_concurrency: int = 1,
         encoding: Optional[str] = None,
         progress_hook: Optional[Callable[[int, int], None]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> Union[StorageStreamDownloader[str], StorageStreamDownloader[bytes]]: ...

sdk/storage/azure-storage-blob/azure/storage/blob/_download.py

@@ -422,6 +422,10 @@ def __len__(self):
     def _get_encryption_data_request(self) -> None:
         # Save current request cls
         download_cls = self._request_options.pop('cls', None)
+
+        # Temporarily removing this for the get properties request
+        decompress = self._request_options.pop('decompress', None)
+
         # Adjust cls for get_properties
         self._request_options['cls'] = deserialize_blob_properties
 
@@ -434,6 +438,9 @@ def _get_encryption_data_request(self) -> None:
         # Restore cls for download
         self._request_options['cls'] = download_cls
 
+        if decompress is not None:
+            self._request_options['decompress'] = decompress
+
     @property
     def _download_complete(self):
         if is_encryption_v2(self._encryption_data):

sdk/storage/azure-storage-blob/azure/storage/blob/aio/_blob_client_async.py

@@ -745,6 +745,8 @@ async def download_blob(
             function(current: int, total: int) where current is the number of bytes transferred
             so far, and total is the total size of the download.
         :paramtype progress_hook: Callable[[int, int], Awaitable[None]]
+        :keyword bool decompress: If True, any compressed content, identified by the Content-Type header, will be
+            decompressed automatically before being returned. Default value is True.
         :keyword int timeout:
             Sets the server-side timeout for the operation in seconds. For more details see
             https://learn.microsoft.com/rest/api/storageservices/setting-timeouts-for-blob-service-operations.

sdk/storage/azure-storage-blob/azure/storage/blob/aio/_blob_client_async.pyi

@@ -208,6 +208,7 @@ class BlobClient(  # type: ignore[misc]
         max_concurrency: int = 1,
         encoding: str,
         progress_hook: Optional[Callable[[int, int], Awaitable[None]]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> StorageStreamDownloader[str]: ...
@@ -229,6 +230,7 @@ class BlobClient(  # type: ignore[misc]
         max_concurrency: int = 1,
         encoding: None = None,
         progress_hook: Optional[Callable[[int, int], Awaitable[None]]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> StorageStreamDownloader[bytes]: ...
@@ -250,6 +252,7 @@ class BlobClient(  # type: ignore[misc]
         max_concurrency: int = 1,
         encoding: Optional[str] = None,
         progress_hook: Optional[Callable[[int, int], Awaitable[None]]] = None,
+        decompress: Optional[bool] = None,
         timeout: Optional[int] = None,
         **kwargs: Any
     ) -> Union[StorageStreamDownloader[str], StorageStreamDownloader[bytes]]: ...

sdk/storage/azure-storage-blob/azure/storage/blob/aio/_download_async.py

@@ -292,6 +292,10 @@ def __len__(self):
     async def _get_encryption_data_request(self) -> None:
         # Save current request cls
         download_cls = self._request_options.pop('cls', None)
+
+        # Temporarily removing this for the get properties request
+        decompress = self._request_options.pop('decompress', None)
+
         # Adjust cls for get_properties
         self._request_options['cls'] = deserialize_blob_properties
 
@@ -304,6 +308,9 @@ async def _get_encryption_data_request(self) -> None:
         # Restore cls for download
         self._request_options['cls'] = download_cls
 
+        if decompress is not None:
+            self._request_options['decompress'] = decompress
+
     async def _setup(self) -> None:
         if self._encryption_options.get("key") is not None or self._encryption_options.get("resolver") is not None:
             await self._get_encryption_data_request()

sdk/storage/azure-storage-blob/tests/test_common_blob.py

@@ -3595,4 +3595,27 @@ def test_upload_blob_partial_stream_chunked(self, **kwargs):
         result = blob.download_blob().readall()
         assert result == data[:length]
 
+    @pytest.mark.live_test_only
+    @BlobPreparer()
+    def test_download_blob_decompress(self, **kwargs):
+        storage_account_name = kwargs.pop("storage_account_name")
+        storage_account_key = kwargs.pop("storage_account_key")
+
+        # Arrange
+        self._setup(storage_account_name, storage_account_key)
+        blob_name = self._get_blob_reference()
+        blob = self.bsc.get_blob_client(self.container_name, blob_name)
+        compressed_data = b'\x1f\x8b\x08\x00\x00\x00\x00\x00\x00\xff\xcaH\xcd\xc9\xc9WH+\xca\xcfUH\xaf\xca,\x00\x00\x00\x00\xff\xff\x03\x00d\xaa\x8e\xb5\x0f\x00\x00\x00'
+        decompressed_data = b"hello from gzip"
+        content_settings = ContentSettings(content_encoding='gzip')
+
+        # Act / Assert
+        blob.upload_blob(data=compressed_data, content_settings=content_settings)
+
+        result = blob.download_blob(decompress=True).readall()
+        assert result == decompressed_data
+
+        result = blob.download_blob(decompress=False).readall()
+        assert result == compressed_data
+
     # ------------------------------------------------------------------------------

sdk/storage/azure-storage-blob/tests/test_common_blob_async.py

@@ -3526,4 +3526,29 @@ async def test_upload_blob_partial_stream_chunked(self, **kwargs):
         result = await (await blob.download_blob()).readall()
         assert result == data[:length]
 
+    @pytest.mark.live_test_only
+    @BlobPreparer()
+    async def test_download_blob_decompress(self, **kwargs):
+        storage_account_name = kwargs.pop("storage_account_name")
+        storage_account_key = kwargs.pop("storage_account_key")
+
+        # Arrange
+        await self._setup(storage_account_name, storage_account_key)
+        blob_name = self._get_blob_reference()
+        blob = self.bsc.get_blob_client(self.container_name, blob_name)
+        compressed_data = b'\x1f\x8b\x08\x00\x00\x00\x00\x00\x00\xff\xcaH\xcd\xc9\xc9WH+\xca\xcfUH\xaf\xca,\x00\x00\x00\x00\xff\xff\x03\x00d\xaa\x8e\xb5\x0f\x00\x00\x00'
+        decompressed_data = b"hello from gzip"
+        content_settings = ContentSettings(content_encoding='gzip')
+
+        # Act / Assert
+        await blob.upload_blob(data=compressed_data, content_settings=content_settings)
+
+        downloaded = await blob.download_blob(decompress=True)
+        result = await downloaded.readall()
+        assert result == decompressed_data
+
+        downloaded = await blob.download_blob(decompress=False)
+        result = await downloaded.readall()
+        assert result == compressed_data
+
 # ------------------------------------------------------------------------------