Add recoding possibility to create_blob_from_bytes

[romanovacca]

Which service(blob, file, queue) does this issue concern?

Blob

What problem was encountered?

During the creation of a blob, it would be useful if you could add the encoding of the file ,together with the desired encoding. So if you have a file stored that is in big5 encoding, but you want it to be in utf-8, you would simple add two parameters and that could be used.

Currently, one has to re-encode the file themselves, which is sometimes time consuming with large files.

Why should this be implemented?

To be able to create a table in azure sql dwh using polybase, the only allowed encodings are utf-8 and utf-16. This means you have to re-encode all the files yourself first before creation tables for these. But enabling this feature would make the process way easier, since then you would have them converted during uploading.

[zezha-msft]

Hi @romanovacca, thanks for reaching out!

I'm a bit confused by your feature request.

So if you have a file stored that is in big5 encoding, but you want it to be in utf-8, you would simple add two parameters and that could be used.

Do you mean that you want the service to automatically decode your data and re-encode it? If the SDK did it, the process would still be time consuming since the computing is happening on the client side.

Perhaps you could explore using create_blob_from_stream instead, then wrap your source file stream to decode and re-encode the content on the go.