Overall, I'm not a fan of the template file approach of assuming vars in the global scope. It makes tracing the lifetime of a variable much more difficult. -1

Overall, I'm not a fan of the template file approach of assuming vars in the global scope.

Vars are not in the global scope though. They are specifically passed into the template. Are you perhaps confusing it with the old way where everything was indeed in the global scope? Or am I missing what you are trying to say?

ping @sgolemon

IMO, this is much cleaner than the old code (and yes, there are no global variables).

What's the status here? Maybe we should seriously consider to retire bugsnet in favor of something else?

Well, PHP's web-bugs requires a lot of fixes and improvements.
But I think there can be a lot of it done in small steps. It requires some motivated PHP people to regularly review, discuss and merge proposed improvements.

Switching to something else may solve a good portion of current 'not fun' things, but may also add new problems like:

• Can it handle 80k+ tasks at same/better speed without increasing cost and/or complexity?
• Adds it dependencies that may hurt in the long run?
• Who does the migration work and how much 'detail loss' is acceptable?

I saw several tries in the open source world where migrations got stuck and still runs the old ugly but at least somehow working solution.

disclosure 1: I am not part of PHP Group, just a dude who uses PHP.
disclosure 2: I am the current maintainer by of https://github.com/Flyspray/flyspray , also a PHP based bug tracker.

bugs.php.net is no longer really used, and hence, we're sunsetting it. Due to that, I don't think it makes sense to merge this patch, which also has conflicts.

Oh, right, further improvements to the bug tracker (especially refactorings) won't make sense anymore.