is presently its sole maintainer,
You can support him:
Sticking problem recap
2 ways to handle problem: search in T, search in P
Problems with search in T
Solving the problem The Witness way
Problems with search in position
How to implement search in P in Handmade Hero
Why you need a flood fill
The problem with players with volume
Implementing search in P
Breaking apart code
Casey on messy code (rant)
Simplifying the movement code
Problem with unsigned integers on the zero boundary
Finding the closest tile for each loop
Check for empty tiles
Defining the valid player region inside the tile
How to perform the search
Getting the length of a vector
Cleaning up to finish on day 46
Why different versions of GetTileValue()?
You start variables uppercase and structs lowercase. Why?
When npc's are introduced, will they reuse some of the code shown here?
Are you going to implement push/squeeze through doorways?
It seems like the possible movement is more like a pizza-slice?
How would we, using this system, handle when the player would be killed by collision?
Will we have to push boulders out of the way of doors?
Will you ever fall for some of the dogmatic traps when coding?
Does the search method here give the sense of rounded corners?
I don't really understand this. Why not just change the players vector with normals?
Are you going to worry about how many objects the collision detection system can handle?
For search in p, don't you have a problem with passing through walls?
Will the search code work if he hits a small notch that would make him bounce back?
Will you implement secrets when none of us are watching?
Will the characters have skeletal animations?
Will the search in p work with moving objects?
A great deal of the problems you've faced seems to be around tradeoffs..