dlang.org/type.html gives this example

    ubyte  u1 = cast(byte)-1;  // error, -1 cannot be represented in a ubyte
    ushort u2 = cast(short)-1; // error, -1 cannot be represented in a ushort
    uint   u3 = cast(int)-1;   // ok, -1 can be represented in a uint
    ulong  u4 = cast(long)-1; // ok, -1 can be represented in a ulong

However, all 4 compile, which makes sense, because casts are blunt and don't care about what does and doesn't fit. This won't compile

    ubyte  u1 = -1;
    ushort u2 = -1;

which is a bit interesting given that this compiles just fine

    byte i1 = -1;
    short i2 = -1;
    ubyte  u1 = i1;
    ushort u2 = i2;

But regardless of whether

    ubyte  u1 = -1;
    ushort u2 = -1;

should compile, the example in documentation definitely should, and the documentation claims that it shouldn't, so the documentation is wrong and needs to be updated.

And honestly, I don't understand the logic of how -1 can be represented in a uint but not a ushort like the docs claim, since it can't be represented in either, since they're unsigned, so I don't understand how the example in the docs ever would have made sense.