On Tue, Feb 01, 2000 at 11:43:37PM +0100, Torsten Landschoff wrote: > On Tue, Feb 01, 2000 at 08:54:45PM +0100, Thierry Laronde wrote: > > > And Torsten found this : > > > > [EMAIL PROTECTED]:~ $ date -d 'Sun May 2 16:20:49 CEST 1999' > > Sun May 2 17:20:49 CEST 1999 > > [EMAIL PROTECTED]:~ $ date -d 'Sun May 2 16:20:49 CET 1999' > > Sun May 2 17:20:49 CEST 1999 > > > > It seems that date ignores the daylight saving flag of the timezone. > > I don't know why hwclock and date share the same behaviour but I don't > > want to check now as I have to go to bed..." > > > > I had a look to the info pages of date. In the node `Timezone item', one > > can't find the correction for `CEST' : this isn't defined. > > > > Don't you think that the problem is simply here ? If this is the case, > > isn't it a documentation issue ? > > No. The program is wrong no matter if CEST is known. If it is unable to > parse CEST then it should fail instead of running with false data. But I > don't think this is the case. How about forwarding this report? > > Should be a simple fix for the persons actually working on the code.
You're right, but what I was looking for is the "principle" responsible of this (the initial bug was sent against `hwclock', and you found the same thing for `date'). If you use the following command : $ date -d 'Sun May 2 16:20:49 CET DST 1999' dim mai 2 16:20:49 CEST 1999 BTW, aren't we, in the occidental Europe, in MET ? And MEST is implemented. I will give a look to the source before forwarding the bug, just in order to try to identify the function used. Cheers, -- Thierry LARONDE [EMAIL PROTECTED] website : http://www.polynum.com
