From: "Juan Manuel Guerrero" <ST001906 AT HRZ1 DOT HRZ DOT TU-Darmstadt DOT De>
Organization: Darmstadt University of Technology
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Date: Thu, 19 Sep 2002 11:30:07 +0100
Subject: Re: conversion specifiers and _doprnt
CC: djgpp-workers AT delorie DOT com
References: <AB6DAB53745 AT HRZ1 DOT hrz DOT tu-darmstadt DOT de>
In-reply-to: <Pine.SUN.3.91.1020919071649.27969G-100000@is>
X-mailer: Pegasus Mail for Windows (v2.54DE)
Message-ID: <C1E45A8735D@HRZ1.hrz.tu-darmstadt.de>
X-MailScanner: Found to be clean
Reply-To: djgpp-workers AT delorie DOT com

On Thu, 19 Sep 2002, Eli Zaretskii wrote:

>> 2: F (uppercase f) Not implemented                                Defined.
>
> What does F do?  (Sorry don't have the references to look it up where I 
> type this.)

The difference between f and F is that:
  f prints: inf or infinity and nan.
  F prints: INF or INFINITY and NAN.

This applies also to e,E and g,G.
IIRC, libc implementation always prints NaN and Inf. (Sorry don't have the
sources to look it up where I type this.)
Anyway to get as much as possible C99 compatibility, I have modified f,F, e,E
and g,G as explained above.
As Charles Sandman suggested some time ago, I have commented out (#if 0)
the O conversion specifier from the sources and retained unchanged everything
else to keep BORLAND/MSC compatibility.

Regards,
Juan M. Guerrero