Aloha Bernard,
i'm very sorry that this ticket is stuck. Since i will be away for some time, i took a short, but not too deep, look at it. It will need an rebase, but i can't see any showstopper or obvious issues. So: please go ahead and big thank you for your work and patience!
Greetings,
Torsten

Okay, i'm fine with this :)

Why not downgrading rubygem-html-pipeline27 to 2.7.1 since it is exactly only for GitLab to run?

The update itself looks fine so far. Good that you catched this error! Now there are two possibilities:

• Do the upgrade even with the bug in mind and warn the users
• Only fix the security issues by porting the fixes as patches (and leave the broken fix)

Thanks swills! :)

In D15990#338873, @mfechner wrote:

There is already an entry existing (20180611).
If we start to add for each php application an entry that will cause the UPDATING file to explode or?

While the patch is fine, UPDATING needs an entry for it!

No? Maybe we should lift the embargo for approval of flavors? :D

In this case approved :)

You could also include the "Reported by:" header, if some reports the issue to you. Its not needed for me, just want to point it out.

Please update after requested changes ;)

Why did you remove the patches? If you do so you should mention the reason in the commit log. Most times upstream incorporated them, but not always.

Just another ping. @mat maybe? ;)

Just a short ping to portmgr. Since adding flavors to a port currently needs approval by portmgr.