Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!unido!mikros!mwtech!martin
From: martin@mwtech.UUCP (Martin Weitzel)
Newsgroups: comp.lang.c
Subject: Re: Reading floating point numbers
Message-ID: <933@mwtech.UUCP>
Date: 24 Sep 90 12:16:39 GMT
References: <1990Sep21.160343.18870@eagle.lerc.nasa.gov>
Reply-To: martin@mwtech.UUCP (Martin Weitzel)
Organization: MIKROS Systemware, Darmstadt/W-Germany
Lines: 16

In article <1990Sep21.160343.18870@eagle.lerc.nasa.gov> vvcns@mars.lerc.nasa.gov writes:
[tries to read floating point numbers with fscanf(pointer,"%f",&arg);
but gets only random numbers]

May I suppose "arg" isn't declared as float, but as double?
In this case "%lf" were the proper format specifier.
(Because float-s are propagated to double-s if used as argument
*values*, this problem doesn't appear in printf and hence is often
overlooked with scanf.)

Another possible source of problems are the "locales" (new in ANSI-C).
The character which separates the decimal part may depend on the current
locale and hence he locale should be the same for writing and reading
floats.
-- 
Martin Weitzel, email: martin@mwtech.UUCP, voice: 49-(0)6151-6 56 83