Closed gstrauss closed 1 month ago
@gene1wood was the repo config changed recently?
Review required New changes require approval from someone other than the last pusher.
And yet, under the Reviewers sidebar on the right:
At least 0 approving reviews are required to merge this pull request.
This merge fixes the issue in #227 where only the Modern config was displayed.
@gene1wood: it looks like the recent changes you made to the repo config have had the desired effect to remove the requirement for Approvals before merge. Thanks!
I would still prefer testing and approval from someone else than the author, but this was so broken it's better to publish now, and refine later. (I will be looking into formatting and also going through the version logic as described by the OP in their contribution — to follow what should be in the hbs and what's already taken care of in js — on a first glance this looks right, but will give it more than a first glance, however the priority is lower than the other things. I'd still generally expect a say ~14day window for reviews before self-approvals unless something's a hotfix though. Not a problem here, great job getting this sorted out so promptly, thanks!)
@gene1wood This deploy at least confirms the publishing source settings are A–OK and we're good to publish just the built artifacts now, and not compete with the static deploy;) Thanks!
@gene1wood This deploy at least confirms the publishing source settings are A–OK and we're good to publish just the built artifacts now, and not compete with the static deploy;) Thanks!
@gene1wood would you like a separate issue opened and assigned to you to track these requests? Please disable the static deploy (and document what was done in case it needs to be re-enabled).
This deploy at least confirms the publishing source settings are A–OK and we're good to publish just the built artifacts
Please disable the static deploy
These were set correctly, and already working correctly, as can be confirmed from recent action runs. Thanks Gene, no action needed.
(2/2 is correct, instead of the 5/5 before…)
Stunnel refactor to correctly use the output data instead of hardcoding version conditionals, that never had the intended effect in the first place (where the template logic would only display
modern
config for all options, unless deliberately lowering the server versions).Fixes #227