Close all those PRs that won't ever get merged

[Coolthulhu]

We have quite a bit of ancient PRs that are not mergeable and no longer maintained. They only serve to confuse the build bot and make the PR list bigger than it needs to be.

I'd close all unmergeable PRs with more than a month of no updates with a note that this closing is not a rejection, but maintenance of the PR hopper and that I'll re-open them if the author replies.

Anything wrong with my idea?

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[Rivet-the-Zombie]

Yeah, we've really needed to do this for a long time.

Good idea!

[drbig]

I fully support the idea.

I also feel similar treatment could be applied to issues.

[i2amroy]

I'm cool with PR's, and what issues we can filter out (notably older bug ones); but a large portion of the older issues are either feature enhancement plans that can't be closed so easily or are ancient NPC-type bugs that are still around in the present day. If someone wants to filter through the issues list and take off duplicates and solved ones that would be great, but it's not going to be anywhere near the ease of "cut off everything unmergeable from before X date" that the PR's are going to be.

[Zireael07]

Let's close the ancient unmergeable PRs first before getting into closing issues.

[kevingranade]

My current project is reimplementing the oldest unmergeable PR on the list, and I plan to reimplement several others as I have time. If nothing else they serve as reminders of the issue, if they're confusing the build bot, consider making or finding a matching issue, linking to the PR, THEN close the PR, but please don't simply close them unless they're abandoned or the idea itself is bad or obsolete (as in implemented in a better way, not just unmergeable).

[Coolthulhu]

don't simply close them unless they're abandoned

I went by last update time, mergeability and user's presence. Most of the closed PRs were ones by users who are unlikely to return.
Some were truly obsolete (AI updates that are already in, in better form).

I'll list:

• [Needs Testing]Added isometric support for tiles #3798 (ancient, author no longer present, someone on forums is doing the same thing)
• Makes MTRIG_FRIEND_HURT/DIED effect only monsters that are friendly towards hurt/dead monster #4029 (old "omnifaction" AI system)
• [WIP] Logic for generic tile categories #4283 (technical, but no longer mergeable and maintained)
• New game: Add "Make new world for this character" option to worldselet menu. #4366 Abandoned, however this one fits your "work on last unimplemented" idea. May be already implemented depending on how we look at it.
• [CR] [WIP] Ice content #4975 (abandoned, highly sci-fi, the contributor "salvaged" the useable parts for a newer PR)
• [WIP] [CR] More dynamic food nutrition/enjoyability values #6564 (abandoned, littered with debug stuff)
• Basic railroad track map tiles #6925 Another possibly re-openable. Abandoned for a while, but could be brought back to service with moderate amount of work. That said, writing it from scratch would also be moderate amount of work.
• [WIP] Tool quality rehash #6954 (regards old tool quality system, giant and unsalvageable)
• [WIP][CR] Atmospheric and Environmental Modifications to Cities #7892 Possibly re-openable. Abandoned and rather heavily unfinished, but a solid idea.
• Rough Morale trap Framework. #8159 (abandoned, a bit hacky and not really a good idea)
• Light-emitting and toggleable furniture #9992 (abandoned, not really finished, makes furniture that does some of the jobs vehicle system does)
• Weapon FANCY kits #10505 Abandoned. IMO this is a bad idea (modding weapons to be fancy? Should be inherent item quality, not a mod), but could be reopened if you don't mind the silliness.
• Facility items 004 gunpowders & stuff [WIP][CR] #10767 Possibly reopenable, but would take the author to come back and finish the job and the job is far more than just "making it mergeable".
• [WIP] Add physics calculations for vehicles. #10666 (abandoned, very "heavy", finishing it would take more work than rewriting from scratch. Idea itself is good, but is quite obvious and recurrent)

I think that's it.

[Frost-wood]

I'd argue:
#11313 (user inactive, doesn't seem to have the desire to make mergable).
#7749 inactive, really easy to redo).

[Coolthulhu]

#7749 could possibly be merged even right now, I think.

[BevapDin]

#7749 doesn't do anything. It adds terrain types that are used nowhere, if someone wants to use them, they can always merge that branch into their own work.