From mihai.bazon at gmail.com Mon Feb 3 20:37:04 2014 From: mihai.bazon at gmail.com (=?UTF-8?Q?Mihai_C=C4=83lin_Bazon?=) Date: Mon, 3 Feb 2014 22:37:04 +0200 Subject: Infinite loop when adjusting a time stamp backwards across a DST boundary. In-Reply-To: <52EFFCAE.7000904@onshored.com> References: <52EFFCAE.7000904@onshored.com> Message-ID: I wonder if it's related to this: http://lists.common-lisp.net/pipermail/local-time-devel/2010-December/000217.html (and if so, sad to see the problem still there after so many years). On Mon, Feb 3, 2014 at 10:31 PM, Lee Ayres wrote: > When adjusting a time-stamp across the DST boundary it is possible to > throw the function into an infinite loop. > > For example, in US Central Time the following never returns: > > (adjust-timestamp @2014-03-09T03:00:00.000000-05:00 (:offset :sec -1)) > > Attached is a patch that bails out before getting caught in the loop. > -- Mihai Bazon, http://lisperator.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From lee at onshored.com Mon Feb 3 20:31:42 2014 From: lee at onshored.com (Lee Ayres) Date: Mon, 03 Feb 2014 14:31:42 -0600 Subject: Infinite loop when adjusting a time stamp backwards across a DST boundary. Message-ID: <52EFFCAE.7000904@onshored.com> When adjusting a time-stamp across the DST boundary it is possible to throw the function into an infinite loop. For example, in US Central Time the following never returns: (adjust-timestamp @2014-03-09T03:00:00.000000-05:00 (:offset :sec -1)) Attached is a patch that bails out before getting caught in the loop. -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Add-a-bailout-condition-when-adjusting-time-accross-.patch Type: text/x-patch Size: 982 bytes Desc: not available URL: