Why do I get an error type with an argument that is used twice for the same purpose

Question

First of all, I have these types :

type position = float * float

type node = position

type path = position list

Here are the two pieces of code causing the error :

let build_path map source target =
  let rec build_aux acc map source x initial_target =
    if (((DistMap.find_opt x map) = None) || x = source) then acc@[initial_target]
    else build_aux ((DistMap.find x map)::acc) map source (DistMap.find x map) initial_target
  in build_aux [] map source target target

let shortest_path graph source target : path =
  build_path (snd (dijkstra graph source target)) source target

path has type position list for clarity.

Here's the error :

361 |   build_path (snd (dijkstra graph source target)) source target
                                                               ^^^^^^
Error: This expression has type position list
       but an expression was expected of type position = float * float

I just don't get it. I've tried the build_path function in Utop, by having a Map filled like this :

DistMap.bindings prevMap;;
- : (node * (float * float)) list =
[((1., 1.), (7., 7.)); ((2., 2.), (1., 1.)); ((3., 3.), (2., 2.));
 ((4., 4.), (3., 3.)); ((5., 5.), (4., 4.))]

let l = build_list prevMap (1.,1.) (5.,5.);;
val l : node list = [(1., 1.); (2., 2.); (3., 3.); (4., 4.); (5., 5.)]

shortest_path has to, with 100% certainty, receive target with type node . The thing is, no error is raised when target is used as an argument for the dijsktra function, which requires a graph and two nodes source and target .

So I'm really confused why target suddenly has the wrong type for build_path and not dijkstra .

Anyway to resolve this issue?

Answer 1

感谢@Pierre G. 的帮助，我们确定target类型被dijkstra函数约束到一个position list ，因为我正在将dijkstra的列表与目标进行比较，一旦错误被修复并且target与dijkstra另一个node进行比较，问题就解决了。