From:	"A. Sinan Unur" <asu1 AT cornell DOT edu>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: Q: detecting valid floating point numbers?
Date:	9 Dec 1996 16:02:06 GMT
Organization:	Cornell University
Lines:	25
Sender:	asu1 AT cornell DOT edu (Verified)
Message-ID:	<58hd5u$t5t@newsstand.cit.cornell.edu>
References:	<58gvqc$26p AT newsfeed DOT cs DOT auc DOT dk>
NNTP-Posting-Host:	128
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

ff AT kom DOT auc DOT dk (Frank Frederiksen) wrote:
>
>When we symify the program the first two lines are:
>__doprnt+1432
>_fprintf+105
>
>So, is there a (simple)way of detecting if a floating point number
>is legal, before printing it, or do we have to investigate the
>numbers individually to see if they are valid numbers?
>
i would assume that if trying to print a floating point number gives an
error, the comparison you would make would also be unreliable.

i would concentrate my effort on making sure the receiving program
accepts only uncorrupted data. this does not mean examining individual numbers
but using some sort of mechanism (like checksums) to ensure the integrity
of the data your receive, and do a re-transmit if necessary.

i am no expert on the subject. i recall that the author of kermit had a
very good article which might prove to be useful. it was in dr dobbs 
journal (http://www.ddj..com) (if you need the reference, i can try to 
find the magazine.)

sinan.

- Raw text -