From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7374 invoked from network); 31 Oct 2002 18:55:05 -0000 Received: from sunsite.dk (130.225.247.90) by ns1.primenet.com.au with SMTP; 31 Oct 2002 18:55:05 -0000 Received: (qmail 6100 invoked by alias); 31 Oct 2002 18:54:49 -0000 Mailing-List: contact zsh-workers-help@sunsite.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 17889 Received: (qmail 5930 invoked from network); 31 Oct 2002 18:54:42 -0000 Date: Thu, 31 Oct 2002 10:53:40 -0800 From: Wayne Davison To: Zsh hackers list Subject: Re: PATCH: my "SECONDS can be floating point" tweaks Message-ID: <20021031185339.GE8814@scuzzy.blorf.net> References: <20021030211750.GA6296@scuzzy.blorf.net> <24071.1036060485@csr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <24071.1036060485@csr.com> User-Agent: Mutt/1.3.28i On Thu, Oct 31, 2002 at 10:34:45AM +0000, Peter Stephenson wrote: > Looks OK, although I still vaguely think a struct timeval is more > appropriate for the raw values (which doesn't fit inside struct param). That is certainly an option, and it would be easy to tweak my code to implement this if it is desired (after adding a "struct timeval" to the param.u union). In my limited testing the sub-second value was restored exactly, but I can imagine that certain fractional representations would cause the sub-second value to get tweaked by a small amount the first time it was saved and restored (and remain constant after that). I haven't checked yet to know for sure how big this tweak could get, but I'll look at it more closely soon. (My gut says that the values will be close enough that the change will not be noticed.) I've gone ahead and committed my previous patch. ..wayne..