Error reading block blob from Python API

Using v3 blob service (running from source in docker), the following code throws an error:

from azure.storage.blob import BlockBlobService
blob_conn = BlockBlobService(is_emulated=True, custom_domain='localhost:10000')
blob_conn.create_container('azurite-test')
blob_conn.create_blob_from_text('azurite-test', 'test-file', 'Test content')
print(blob_conn.get_blob_to_text('azurite-test', 'test-file').content)

TypeError: '<' not supported between instances of 'int' and 'NoneType'

This code works with v2.7.0

[XiaoningLiu]

Thanks for debugging! That’s very interesting. I get the root cause.

For V3, when GET BLOB request requires to download the full blob, it doesn’t return content-range header, which leads to Python SDK error. While other SDKs like Node.js or .Net doesn’t depend on this response header.

We can fix this by always return content-range header when request includes a ranges or x-ms-ranges headers which fits REST API spec. https://docs.microsoft.com/en-us/rest/api/storageservices/get-blob

[efiop]

@XiaoningLiu Looking forward to it! 🎉 Thank you!