Download dataset fail via SDK

[lixjohn]

step -

1. Open dataset 美国法律案例摘要 download page https://opencsg.com/datasets/MagicAI/BVA_Cases
2. Click downlaod button and copy SDK code
3. Update 'token' and 'cache_dir' field. (Token is from my 'Settings/User Access Token')
4. Excute code, the error reported, please see attachment error.txt
   error.txt
   Seems like txt files downloaded successfully but fail when downloading py file https://hub.opencsg.com/hf/datasets/MagicAI/BVA_Cases/resolve/main/attribution/__init__.py, please see snapshot.

Please notice Use git download command 'git clone https://opencsg.com/datasets/MagicAI/BVA_Cases.git', everything goes well.

BTW - I tried dataset Dataset 历代中医文献数据库 via SDK but failed as well with different error, the error attached here
error_cmbook.txt

[lixjohn]

just wondering when will be avaliable on production env? Thanks!

[SeanHH86]

It has been fixed. Thanks!

[lixjohn]

Thanks, it works now. Just quick question - SDK download will create folder like this /MagicAI/BVA_Cases, however, https download just create BVA_Cases folder, is it design?

[SeanHH86]

Which function do you use in your code? file_download for http_get ?
From what I saw file_download will create fold <repo_id> under cache_dir, and http_get downloads file under cache_dir.

Please log a new issue if you have any trouble in using our SDK.

[lixjohn]

Thanks, file a new issue for this. #56