-
Notifications
You must be signed in to change notification settings - Fork 50
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[doc] Add README section on semver #431
Comments
YES! If we are OK with this I'd love to enhance the package API: #422 @behnazh-w curious if you have feedback here. I've been holding off on library updates because of that, but if we only want to guarantee CLI for now, would be very happy to do more cleanup. |
I don't think there are many users of the verifier API. The only public one seems to be https://github.com/ddworken/hishtory by @ddworken. @suzuki-shunsuke also created #373 to fix @ddworken @suzuki-shunsuke Do you have any thoughts? To be fair, it's also not a huge burden to just bump the semver but we'd want to minimize the number of times we have to bump so we'd want to do all the changes to the API in one go. Probably as part of the API updates for the "slsa-verifier as a service". |
aqua, which is a declarative CLI Version Manager I maintain, uses slsa-verifer as Go library. ref. https://aquaproj.github.io/docs/reference/cosign-slsa |
@suzuki-shunsuke Got it thanks! Hmm, Strange. If you have it listed in your go.mod, I wonder why aqua doesn't show up in the dependents for slsa-verifier... |
aqua depends on not |
Aha, I didn't realize they split it up by versioned package name. |
@asraa I think the main question would be whether you would like the slsa-verifier to be used with its CLI interface or as a library. I would prefer slsa-verifier to be mainly available as a library because it gives a lot of flexibility to end users. In this case the library would be versioned and the CLI would be just a wrapper that is always compatible with the library APIs. |
I think that we would like to go in the direction of supporting it as an API. Whether we should support it as an API now though is perhaps the question. I'm not sure we ever intended to provide the current API as stable API and I think we may like to do some more iteration before it becomes more stable but I think we can stick to semver without it being too burdensome. |
+1 to what @ianlewis said. We want to revamp the API and have the CLI just "wrap" it #46 There are only a handful of users who use the API, so it should be OK to make the API experimental and subject to changes until we have the issue above resolved. As @ianlewis said, we did not intend for the API to be stable atm. Our mistake for not making this more explicitly in the README. @suzuki-shunsuke , we can work with you when we update it to help you migrate. Would that be OK? Let's create a new milestone for it as well. Would that work for everyone? re: semver Section in README. Let's also call out that CLI arguments marked as |
Hmm. CLI users wants CLI to follow semver. |
We do follow semver for CLI. There are sometimes some new arguments we label as |
Oh, I see. I misunderstood. Thank you for your explanation. |
Sorry for the late reply, but I'll chime in to say that I am using |
Let's add a section in the README to call out which parts of the code follow semver: the CLI arguments and behavior. Currently no guarantees on CLI output and API.
The text was updated successfully, but these errors were encountered: