Article View: borland.public.jbuilder.java.language
Article #17583Re: Floating-Point Range of Values...
From: Matt Welsh
Date: Wed, 21 Oct 1998 00:00
Date: Wed, 21 Oct 1998 00:00
17 lines
1201 bytes
1201 bytes
Steve Schafer (TeamB) wrote: >SNIP> > Your code that works with byte variables doesn't "prove" anything. > You've come up with a clever trick that happens to give you the right > answer. There's nothing in the Java language specification that says > that incrementing an integer-type variable that has the maximum value > will give you the corresponding integer-type variable with the minimum > value; Not only is Steve correct, but the JLS *specifically* states that the built in integer types do not "indicate overflow or underflow in any way" (section 4.2.2). -- Matt Welsh Inprise Corporation R&D Engineer, JBuilder
Technical Details
Message-ID:
<362E4E81.FF3414A9@inprise.com>Path:
arc1-ca.pugleaf.net!unknown!not-for-mailReferences:
<362DCA9D.CEFABA69@cbn.net.id> <369d498a.253050490@forums.borland.com>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: f74b3,da07c345f9ebf065 X-Google-Attributes: gidf74b3,public From: Matt Welsh <mwelsh@inprise.com> Subject: Re: Floating-Point Range of Values... Date: 1998/10/21 Message-ID: <362E4E81.FF3414A9@inprise.com> X-Deja-AN: 403693064 Content-Transfer-Encoding: 7bit References: <362DCA9D.CEFABA69@cbn.net.id> <369d498a.253050490@forums.borland.com> Content-Type: text/plain; charset=us-ascii Organization: Inprise Corporation Mime-Version: 1.0 Newsgroups: borland.public.jbuilder.java.language