Code review comment for lp:~vanvugt/mir/move

> 220 - typedef uint32_t ValueType;
> 221 + typedef int ValueType;
>
> (and related uint_32 -> int changes)
>
> While these changes fix the problem mentioned here, I imagine that there was a
> reason for choosing uint32_t, and that this change could have unintended
> consequences. Has this been checked?

Alan,

Yes it's been checked to some extent:
1. No test case failures.
2. No visible regressions when running mir binaries.

Furthermore, you have theoretical peace of mind:
1. All the +,-,==,!= operators are unaffected by sign. They produce the same result regardless of the signedness of the operands.
2. Only the >,<,>=,<= might be affected in theory. But you'd have to be unlucky to use them in a way that they are affected and all testing so far has shown no change in behaviour.