Re: Bug: yyleng declared as type size_t

[Hans-Bernhard Broeker]

On Tue, 18 Sep 2001, W. L. Estes wrote:

> > Yep. OTOH, there *are* some existing definitions (POSIX, Single Unix
> > Spec.) of what 'lex' (and thus, by compatibility, flex) should behave. One
> > of them is that 'yyleng' is an int, not a size_t.  That definition may
> > look silly by today's standards, but that's what we have.
> > 
> > So, at least in lex compatibility mode, yyleng has to remain 'int', I
> > think.
> 
> If that's the case, then the skeleton needs changing. Silly
> standards. Also, are these standards available online somehow for us
> mere  mortals to get and consult?

At least the SUS is (it's by the Open Group). Just google for "single unix
specification" and you'll see it.  POSIX is an ISO standard, I think,
which would mean it's only available for $$$ (even for the PDF edition!).

-- 
Hans-Bernhard Broeker (address@hidden)
Even if all the snow were burnt, ashes would remain.