Fix pulling from nvcr.io #285
Conversation
|
Nikita, The latest updates added an option to use some external tools from the broader OCI community to handle pulling and unpacking images. This may be a more robust way to deal with these other repositories. But I'll take a look at the PR. |
| # The json content type can include a extra bit defining the character set | ||
| # so we just check the beginning. Worst case is the json.loads will throw | ||
| # an error just after this. | ||
| if not resp.getheader('content-type').startswith('application/json'): |
There was a problem hiding this comment.
Why was this dropped? Does the nvcr case not return json?
There was a problem hiding this comment.
This part is from your original PR, but I can check if it's still needed.
scanon
left a comment
There was a problem hiding this comment.
This looks good as far as I can tell. I would like to run it and make sure there aren't any edge cases that fail.
Sounds good! Is there something I should do to enable this new pulling method? |
|
Here is the documentation on the external mode option.... |
Based on #238 with my suggestions applied.