On Saturday, 1 October 2016 at 02:31:23 UTC, Andrei Alexandrescu
wrote:
https://github.com/dlang/phobos/pull/4827 still allows that but
specifies that phobos algorithms are not required to. -- Andrei
I don't have much experience, but IMHO the length should be
restricted to be any of the built-in unsigned integral types
(ubyte, ushort, uint, size_t, ulong). This way, all comparisons
for equality or inequality are defined and correct, and the
common type exists and is the one you'd expect. Also, every other
operation on them works as expected, because smaller types are
widened and the absence of signed types guarantees no strange
results.
The only problem that arises with this definition of length is
that functions cannot take a length parameter as size_t, because
that could truncate it. Possible solutions, ordered from the best
to the worst:
1) functions should try not to take lengths; they should take
ranges or slices, which can be queried for their length;
2) functions could take lengths as the greatest possible type
(ulong);
3) functions could be parameterized on the length type;
4) a minority of functions could, if deemed really necessary (for
performance or whatever), use size_t, and it should be documented
that they expect the length to be limited by the addressable
length of the machine, and for which reason they do.
