You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Description
While using cosign as a library, I found that the cli.SignCmd does all the logic that I would require for signing myself, however, it is designed for CLI usage with particular assumptions. One such assumption is that it uses the default keychain. I would like to be able to change the keychain used without having to duplicate the whole function in my own codebase with that one line change and having to maintain it whenever the function changes in the upstream.
I was wondering if there was a design decision made regarding this current implementation or would it be valid to extract this logic out of the CLI portion to be more generic and/or configurable as a library function?
The text was updated successfully, but these errors were encountered:
I think the next level of stuff is going to be pushing some of the signing (and verification) stuff into different packages, which compose nicely with the pkg/oci stuff.
Description
While using cosign as a library, I found that the
cli.SignCmd
does all the logic that I would require for signing myself, however, it is designed for CLI usage with particular assumptions. One such assumption is that it uses the default keychain. I would like to be able to change the keychain used without having to duplicate the whole function in my own codebase with that one line change and having to maintain it whenever the function changes in the upstream.I was wondering if there was a design decision made regarding this current implementation or would it be valid to extract this logic out of the CLI portion to be more generic and/or configurable as a library function?
The text was updated successfully, but these errors were encountered: