From: Peter J. <pet...@gm...> - 2003-11-14 10:32:53
|
Hi Carlos, All, > > Please apologize, both for disturbing the thread structure by > > replying to the digest, and for giving my half-informed view. > > You are not disturbing!!, all comments are wellcome :) Trying to find some backing information on the WWW, now I feel only quarter-informed. But you invited me to comment... > For char and varchar fields Firebird sends the field character set in > the sqlsubtype, i have readed this in a message of Nickolay Samofatov in > the Firebird-Java list ( getParameterMetaData - 06/10/2003 13:04), Perhaps, I was wrong and the 'on the wire' data is really traveling with different character sets. But this only leads to a more radical proposal: (a) The connection character set is the character set seen by the user (i.e. the programmer in this case) (b) .NET works internally always in UTF-16 (a) AND (b) => (c) The whole character set issue is void for .NET, the connection should always be Unicode, the programmer should always only see the UTF-16 encoded data and all necessary translation is done by data provider (there may be a backdoor needed for getting the actual byte sequence from the server, for very special cases). It follows, that a column charset declaration only affects the possible character repertoire for this column and sorting behaviour, but is otherwise unnoticed by the C# program. From the question of the original poster and from vague memories of earlier I18n questions on the list, that seems to be the behaviour that is expected. It would be most helpful to compare the behaviour of other DBMS's data provider implementations, notable PostgreSQL. Best Reagrds, Peter Jacobi -- NEU FÜR ALLE - GMX MediaCenter - für Fotos, Musik, Dateien... Fotoalbum, File Sharing, MMS, Multimedia-Gruß, GMX FotoService Jetzt kostenlos anmelden unter http://www.gmx.net +++ GMX - die erste Adresse für Mail, Message, More! +++ |