On 2011/09/27 02:02 PM, Yury Tarasievich wrote:
On 09/27/2011 01:46 PM, Robert Elz wrote: ...
What we should really be doing is making it clear to one and all that these abbreviations are meaningless, rather than FET or EEFT or BYT or anything else, call all the Russian (and related) zones "EST" (they're Eastern, with respect to most of Europe anyway, and they have a standard time), for no better reason than to make it clear to everyone that the abbreviations have no practical purpose, and are better never used, anywhere.
In fact, rather than the abbreviations, I'd rather see designations based somehow on the corresponding international timezones (/chasovyye poyasa/). That's the long established practice here, after all. Something like UTC+3 or Z02+1 (w/r to the 0..23 numbering) would make do nicely both for Belarus and Ukraine, at least.
If you want to include digits, '+' and '-' characters in the abbreviations, they'd also have to start with '<' and end with '>'. See the description of the TZ environment variable in the X/Open Single UNIX Specification. Some experiments on my Linux system: $ LANG=C TZ='UTC' date Tue Sep 27 14:05:51 UTC 2011 $ LANG=C TZ='UTC+3' date Tue Sep 27 11:05:56 UTC 2011 $ LANG=C TZ='<UTC+3>' date Tue Sep 27 14:06:02 UTC+3 2011 $ LANG=C TZ='<UTC+3>3' date Tue Sep 27 11:06:05 UTC+3 2011 You can see the effect of the '<' '>' quoting characters in the above commands. In the TZ='UTC+3' case the 'UTC+3' was broken down and parsed (hence the 3 hour shift in the printed time relative to the 'UTC' case) instead of being treated as a literal abbreviation. -- -=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@mev.co.uk> )=- -=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=-