Path: utzoo!attcan!uunet!aplcen!samsung!cs.utexas.edu!tut.cis.ohio-state.edu!snorkelwacker!mit-eddie!rutgers!njin!uupsi!rodan!jfbruno
From: jfbruno@rodan.acs.syr.edu (John Bruno)
Newsgroups: comp.sys.ibm.pc.programmer
Subject: Re: atof() problem in Turbo C 2.0
Summary: RTFM
Message-ID: <3702@rodan.acs.syr.edu>
Date: 12 Jun 90 16:23:55 GMT
References: <1990Jun11.195944.26213@wuphys.wustl.edu>
Reply-To: jfbruno@rodan.acs.syr.edu (John Bruno)
Organization: Integrated Healthcare Technologies, Inc.
Lines: 21

In article <1990Jun11.195944.26213@wuphys.wustl.edu> marty@wuphys.UUCP (Marty Olevitch) writes:
 >Here is something I've just noticed in using the atof() library routine in
 >Turbo C 2.0. When the following test program is compiled with the "-DWEIRD"
 >option, the correct answer is printed. However, when "WEIRD" is not
 >#defined, the error message "scanf : floating point formats not linked" is
                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 >printed at runtime. Does the compilation of the sscanf() load in some other
 >code needed for atof()? If so, how do I get this code loaded without the
 >dummy call to sscanf() (which in the example is never executed).
 >
 >Thanks,
 >
 >Marty Olevitch				Internet: marty@wuphys.wustl.edu
 >Washington University			UUCP:     uunet!wugate!wuphys!marty
 >Physics Department, Campus Box 1105	Bitnet:   marty@wuphys.WUSTL
 >St Louis MO 63130 USA			Tel:      (314) 889-6285
 >
 > [Short program that calls atof() deleted]

Page 380 Turbo-C User's Guide (Using Floating Point Libraries)
---jb