Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!umnd-cs!umn-cs!hyper!dean
From: dean@hyper.UUCP (Dean Gahlon)
Newsgroups: comp.sys.cbm
Subject: Re: C Power 128 query
Message-ID: <264@hyper.UUCP>
Date: Mon, 18-May-87 11:01:47 EDT
Article-I.D.: hyper.264
Posted: Mon May 18 11:01:47 1987
Date-Received: Tue, 19-May-87 06:18:48 EDT
References: <1861@rutgers.rutgers.edu]
Organization: Network Systems Corp., Mpls. MN
Lines: 35

in article <1861@rutgers.rutgers.edu], prindle@nadc says:
] 
] Has any user of "C Power 128" been able to obtain the fix to the "floating
] point to integer conversion problem"?  Supposedly, it has been fixed by the
] author, but I just can't seem to locate the fix (I believe it would be in
] library function "conv52" or "cnvfi").  Briefly, the problem is:
] 
] 	main()
] 		{
] 		int i;
] 		float f;
] 
] 		f=123.4;
] 		i=(int)f;		/*the (int) is optional in this case*/
] 		printf("i=%d\n",i);	/*should print 123*/
] 		}
] 
] prints garbage for "i"!  I wrote to the company describing this and many other
] bugs in "C Power 128".  All other significant bugs were fixed and the repaired
] files posted to the C Power BBS.  Not so the fix to the bug above.  Any
] clues?
	I've come up with a partial fix for this problem, but there seem to be
other problems with floating point. (The partial fix was to call the correct
BASIC ROM routine -- the routine which the existing libraries call does NOT
leave its result in the A and Y registers (even though its name in Commodore
documentation leads one to believe this), and the code in the library expects
it to do so. )
	On another note, does anybody out there know about what C-power 128's
problem with regard to scanf/fscanf/sscanf of hex numbers is? It doesn't seem
to work right, yet I've looked at the library code, and can't seem to find
anything wrong.


						Dean C. Gahlon
						...ihnp4!umn-cs!hyper!dean