http://d.puremagic.com/issues/show_bug.cgi?id=2917
--- Comment #3 from Glenn Haecker <ghaec...@idworld.net> 2009-05-14 17:06:15 PDT --- This bit of code shows the bug. There are larger effects as well. ------------------------------------------- string[] testdt = [ "1969-12-31 23:59:58 UTC", "1969-12-31 23:59:59 UTC", "1970-01-01 00:00:00 UTC", // epoch "1970-01-01 00:00:01 UTC", "1970-01-01 00:00:02 UTC"]; foreach(int i, string s; testdt) writefln("%s = %5d", s, parse(s)); d_time[] times = [-2000,-1000,0,1000,2000]; foreach(tm; times) writefln("%5d = %s", tm, toUTCString(tm)); ------------------------------------------- Unpatched results: 1969-12-31 23:59:58 UTC = -9223372036854775808 (d_time_nan) 1969-12-31 23:59:59 UTC = -9223372036854775808 (d_time_nan) 1970-01-01 00:00:00 UTC = 0 1970-01-01 00:00:01 UTC = 1000 1970-01-01 00:00:02 UTC = 2000 -2000 = Wed, 31 Dec 1969 23:59:57 UTC (off by 1 sec) -1000 = Wed, 31 Dec 1969 23:59:58 UTC (off by 1 sec) 0 = Thu, 01 Jan 1970 00:00:00 UTC 1000 = Thu, 01 Jan 1970 00:00:01 UTC 2000 = Thu, 01 Jan 1970 00:00:02 UTC Patched results: 1969-12-31 23:59:58 UTC = -2000 1969-12-31 23:59:59 UTC = -1000 1970-01-01 00:00:00 UTC = 0 1970-01-01 00:00:01 UTC = 1000 1970-01-01 00:00:02 UTC = 2000 -2000 = Wed, 31 Dec 1969 23:59:58 UTC -1000 = Wed, 31 Dec 1969 23:59:59 UTC 0 = Thu, 01 Jan 1970 00:00:00 UTC 1000 = Thu, 01 Jan 1970 00:00:01 UTC 2000 = Thu, 01 Jan 1970 00:00:02 UTC -- Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email ------- You are receiving this mail because: -------