Sender: rich AT phekda DOT freeserve DOT co DOT uk Message-ID: <3E007233.784010EE@phekda.freeserve.co.uk> Date: Wed, 18 Dec 2002 13:03:47 +0000 From: Richard Dawe X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.23 i586) X-Accept-Language: de,fr MIME-Version: 1.0 To: djgpp-workers AT delorie DOT com Subject: Re: strtof (C99) [PATCH] References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Hello. Eli Zaretskii wrote: > > On Wed, 18 Dec 2002, Richard Dawe wrote: > > > > > Test 6: Inf -> 0 > > > > > > Why doesn't this return an Inf (the float representation of it)? > > > > Because strtof does not parse Inf. strtod and strtold do not either. > > Does that mean our printf can generate Inf and NaN, but our scanf cannot > read them? Oops! Yes. Looking at the code, it seems that sscanf(sprintf(x)) != x for x == INFINITY or x == NAN. > > C99 requires strto* to understand Inf and NaN (and other capitalisations). > > I think we can fix those problems in another round of C99-compliance work. > > I think we should indeed fix that. How about adding this to TODO? Done. I've created a page for C99 compliance issues: http://www.phekda.freeserve.co.uk/richdawe/djgpp/2.04/c99-204.html I thought my main DJGPP 2.04 page would be too full, if I added them to it. This C99 compliance page referenced from my DJGPP 2.04 page: http://www.phekda.freeserve.co.uk/richdawe/djgpp/2.04/status204.html > > > Also, what about a NaN? > > > > I can add a test for NaNs, if that's what you are asking. > > Yes. If we don't support converting Inf, we probably don't support NaN > either, so this is also something to be fixed, IMHO. (There's also the > new C99 complication with NaN-0xFFFFFF.... thingie.) OK, I will add a test for a NaN, but it won't work for now... Bye, Rich =] -- Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]