Re: FYI: fixed yycheck signed/unsigned comparison warnings

[Paul Eggert]

"Joel E. Denny" <address@hidden> writes:

> Should b4_int_type trigger an error when things are beyond the size 
> expected?

Sorry, I don't know what you mean by that.  If you mean that the
values are less than 2 Gi or greater than 2 Gi, that's just part of
the overall problem that Bison has with values like that; this is just
one of many places.  I should warn you that if/when we get around to
fixing that, we'll start to run into M4 bugs once we start trying to
support 64-bit arithmetic better.

>> save space; the values are intended to be 'int' values.
>
> Does this mean unsigned int should never be returned by b4_int_type?

No, because in places it might be helpful to do that (e.g., line
numbers).  When I wrote 'int' I was thinking more of state numbers and
things like that.