I haven't tested the patch yet, I'm not sure if this is really the right thing to do. From my understanding, the intention is that towns only build houses adjacent to roads. The only case a house can appears two tiles from a road is when a large house (e.g. 2x2 stadium) north-east/north-west of a road is replaced with a small house, which always appears in the north corner of the 2x2 building's footprint. I think that case might actually be the bug.

If this is actually considered a bug that should be fixed, I don't think it should have a new setting.

On technicalities, you need to change your commit message to conform to the standard, with a keyword in front. Do this by making a amend-commit on your local branch, then force-pushing that to GitHub.
My suggestion would be: Fix: Allow towns to fill the center square of 3x3 blocks
(You are removing code, but the purpose of the removal is to fix a bug, so the appropriate keyword is Fix.)

Well, I'm not sure if this is the right thing to do either. But basically this just makes the game behave the same with regard to house placement regardless if you have the "towns are allowed to build roads" setting on or off.

There is an inconsistency at the moment: unless you have that setting setting turned on, that whole code block never executes (except when generating a map). Since the map generation also runs this code, generated towns at the start of a game will also have houses in those places, but then if you play with that setting disabled they will slowly disappear over time.

Since the setting is called "towns are allowed to build roads", I think it's weird that it would change where houses are chosen to be built, so that's why I think this is probably a bug. Removing these conditions makes house placement identical regardless of the value of that setting. So if you play with road building enabled you won't notice any difference.

There are a few other uses of the _settings_game.economy.allow_town_roads and _generating_world bools but to me it seems like they all properly have to do with road building. Just this one doesn't, but then I'm only looking into the code for the first time so I can't be sure

Edit: modified commit message

Someone in the thread said I should test this to make sure this change doesn't let towns build roads all of a sudden. I didn't take screenshots (will do tomorrow) but I did test it, perhaps a more robust test should be done though.

Basically it seems to work as expected: with allow_town_roads on, everything works as it always has, towns refuse to build houses on tiles where roads will go, and eventually roads show up. With allow_town_roads off, towns freely builds houses on tiles where the roads should go, doesn't make roads, and it also fills in the "gaps" in 3x3 blocks as with the setting on.

Town growth algorithm isn't a good place to start patching tbh. It's quite complex and there're lots of subtleties. Especially you don't wan't to mess up allow_town_roads=false behavior or there will be a mob of citybuilder players coming after you xD
As for your patch how did you even check it? Because for me it's completely broken. Towns build roads when not allowed and reserve spots for the grid. And that's exactly how it should be as you just allowed that code to run with disabled setting.

As for your patch how did you even check it? Because for me it's completely broken. Towns build roads when not allowed and reserve spots for the grid. And that's exactly how it should be as you just allowed that code to run with disabled setting.

That is weird, when I tested it I didn't see new roads being made. All I saw was new houses being made on road tiles, and the gaps being filled up. I clearly saw different behavior based on the roads setting. Are you sure your test was with just this patch and nothing else? I don't even see a code path that leads to roads being built with this change. Just GrowTownWithExtraHouse(), and if allow_house gets turned on it runs LevelTownLand() and BuildTownHouse(). Maybe I'm missing something.

I'm new to the code so I understand that I'm not the most suitable for making a change like this, but I'm determined to try harder and figure out if I can get this working somehow. I will also do another test to see if I can reproduce that behavior, because it seems incompatible with what I saw.

It may not be building all the roads when disabled but it still builds some roadbits.
allow_house there does grid reservation and rcmd is passed to GrowTownWithRoad later in this function. So you'll need to separate GrowTownWithExtraHouse from those.

Thanks for clarifying, my test wasn't good enough. Going back to the code.

We recently switched from Jenkins as CI to Azure Pipelines as CI. This means you need to rebase before this Pull Request will pass its checks. Sorry for the troubles!

Sorry, I've been really busy the past few weeks, but I'll get back to this as soon as I'm able to. I made one last change to my local code, I'll rebase and push that when I've got some time this week. I'll then post a test that displays clearly what the before/after consequences are, to ensure there aren't any weird unintended changes.

This pull request has been automatically marked as stale because it has not had any activity in the last month.
Please feel free to give a status update now, ping for review, or re-open when it's ready.
It will be closed if no further activity occurs within 7 days.
Thank you for your contributions.