zmap / zlint

X.509 Certificate Linter focused on Web PKI standards and requirements.
https://zmap.io
Apache License 2.0
358 stars 107 forks source link

Cutting zlint 3.6.0 #776

Closed robplee closed 9 months ago

robplee commented 10 months ago

By my count, there've been over 30 PRs merged including quite a number of new lints since we last cut a release so should we make zlint 3.6.0?

I've tested a zlint integration with importing the latest changes and don't see any issues necessitating a major release. Particular as #775 was scoped to checking new zlint code rather than existing code that may exist around zlint as it is imported.

I feel that after #775 and #767, the logical next step is to remove the deprecated interfaces but that'll definitely have to be called 4.0.0. Therefore, I suggest we cut a minor release now to get the new lints around SMIME and CRLs out into the wild and then 4.0.0 can be about removing LintInterface and we can try to keep release notes for that short.

zakird commented 10 months ago

Makes sense to me! In general, I'd love to get in the habit of doing smaller point releases more often. I do think we will likely do a major release for the changed Lint interface.

cardonator commented 10 months ago

Agreed, was going to comment the same thing today and forgot. Regarding 4.0.0 release, should we wait a bit until the SMIME lints are more robust before making another major bump? I know we are anxious for more SMIME but will need to do some refactoring of other code when we do remove the old interfaces.

christopher-henderson commented 9 months ago

I was literally thinking on this the other day. To several points...

Minor Bump, not Major

I agree that now is a good time to do a minor bump. I reckon that we will eserve the major for when we have

  1. Completed deleted the the Lint interface
  2. Taken inventory on other items that are on peoples' wishlist that would necessitate a major release (just so that we get everything in one go).

SMIME

Releasing the extant lints in a minor release is a good canary in the coal mine, I think. It's tempting to wait for a release to have more of them, however I admit that I don't see anything inherently wrong with trickling them out.


Given no objections, I can certainly cut a release before the holiday storm is upon us.

zakird commented 9 months ago

I agree, let's get this done sooner rather than later. Postponing releases for future work almost always leads to releases not happening. Makes sense re Lint deprecation.

christopher-henderson commented 9 months ago

Release candidate up at https://github.com/zmap/zlint/releases/tag/v3.6.0-rc1

christopher-henderson commented 9 months ago

Cut in https://github.com/zmap/zlint/releases/tag/v3.6.0 :tada: