• via Digitalmars-d-bugs (60/60) Oct 27 2016 https://issues.dlang.org/show_bug.cgi?id=16642

via Digitalmars-d-bugs <digitalmars-d-bugs puremagic.com> writes:

https://issues.dlang.org/show_bug.cgi?id=16642

          Issue ID: 16642
           Summary: byCodeUnit doesn't work AutodecodableStrings unless
                    they're actually strings or alias a variable that's a
                    string
           Product: D
           Version: D2
          Hardware: All
                OS: All
            Status: NEW
          Severity: normal
          Priority: P1
         Component: phobos
          Assignee: nobody puremagic.com
          Reporter: issues.dlang jmdavisProg.com

This code fails to compile

import std.utf;

struct S
{
    string s;
    string str() { return s; }
    alias str this;
}

void main()
{
    auto s = S("hello");
    auto range = byCodeUnit(s);
}


Similarly, this fails to compile

import std.utf;

enum E { a = "hello" }

void main()
{
    auto s = E.a;
    auto range = byCodeUnit(s);
}

Also, this compiles but uses the alias rather than the range API and thus fails
the assertion:

import std.range.primitives;
import std.utf;

static struct RangeAndStringish
{
    string data;

    string s;
    alias s this;

    bool empty() { return data.empty; }
    char front() { return data[0]; }
    void popFront() { data = data[1 .. $]; }
}

void main()
{
    auto fn = RangeAndStringish("test.d", "other");
    auto x = fn.byCodeUnit();
    assert(x.front == 't');
}

I suppose that it could be argued that this last case is doing the right thing
- or at least that it's ambiguous, but normally, an implicit conversion is only
supposed to be used if the type itself doesn't work directly, and in this case,
the type itself would work without the implicit conversion.

--