On Wed, 11 Nov 2009, vatzct wrote:
Hi,
> This is the result of cpinfo with UA codepage:
> 11/11/09 15:52:00 Windows XP 5.1.2600 Service Pack 1 Harbour
> 2.0.0beta3 (Rev. 12818)
> Character encoding: UA1125
> ==================================================
> letter 'џ' (159:AU ) is wrongly sorted
> letter 'Ї' (175:A L ) is wrongly sorted
[...]
Yes, this test nicely shows bugs in old CP implementation.
For many languages like Ukrainian old CP implementation was completely
broken. The new CP code fixed this and few other problems and it's
the reason I left this note in the ChangeLog:
2009-11-08 02:43 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
[...]
Warning !!! some very serious bugs in national sorting have been
fixed. It may cause that old indexes using some national
sorting will not be compatible. Please rebuild existing
indexes.
so do not forget to recreate indexes (reindex) in all your programs
when you will update your Harbour version (Rev. 12818 still uses the
old buggy CP code). Please also remember that xHarbour still uses
broken CP code so it's not possible to share data between Harbour
and xHarbour as long as it will not be fixed. Otherwise the indexes
will be corrupted.
BTW I'm surprised that Ukrainian, Turkish, Serbian, Russian, Greek and
some other users haven't reported the problems for such long time.
best regards,
Przemek
_______________________________________________
Harbour mailing list (attachment size limit: 40KB)
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour
