ndarray: type convertion of _div_scalar (#357)

Ref: #353

- and handle the case of integer NDArray divided by zero

src/ndarray.jl

@@ -669,13 +669,21 @@ Matrix (2D NDArray) multiplication.
 
 Elementwise divide a scalar or an `NDArray` of the same shape from `dst`. Inplace updating.
 """
-function div_from!(dst::NDArray{T}, arg::NDArrayOrReal) where {T}
+function div_from!(dst::NDArray, arg::NDArrayOrReal)
   @assert dst.writable
   if isa(arg, Real)
-    _div_scalar(dst, scalar = convert(T, arg), out = dst)
+    _div_scalar(dst, scalar = arg, out = dst)
   else
     _div(dst, arg, out = dst)
   end
+  dst
+end
+
+function div_from!(dst::NDArray{T}, arg::Real) where {T<:Integer}
+  @assert dst.writable
+  @assert(round(T, arg) != zero(T), "Integer divided by zero")
+  _div_scalar(dst, scalar = arg, out = dst)
+  dst
 end
 
 """

test/unittest/ndarray.jl

@@ -396,6 +396,19 @@ function test_div()
   t6, a6 = rand_tensors(Float16, dims)
   scalar_large = 1e4
   @test t6 ./ scalar_large ≈ copy(a6 ./ scalar_large)
+
+  info("NDArray::div::scalar::type convert")
+  let x = mx.NDArray([1, 2, 3])
+    y = x ./ 1.1
+    @test eltype(y) == Int
+    @test copy(y) == [1, 2, 3]
+
+    y = x ./ 2
+    @test eltype(y) == Int  # this differs from julia
+    @test copy(y) == [0, 1, 1]
+
+    @test_throws AssertionError x ./ 0.5
+  end
 end