Hi Elke, Eventually downloaded the source, but can't find any tables in vsp81.c!
Do they need to be created? If so, could you send me a copy of the built version? Many thanks, David "Zabach, Elke" <[EMAIL PROTECTED]> on 17/11/2003 10:10:11 AM To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]> cc: Subject: RE: UPPER with DBCS Hi, please see file http://www.unicode.org/Public/UNIDATA/UnicodeData.txt and the description for this http://www.unicode.org/Public/3.2-Update/UnicodeData-3.2.0.html for those character mapped to uppercase resp vsp81.c in our source. For korean / chinese there are only few entries, for greek, latin 1 you will find several mapped to upper Elke SAP Labs Berlin > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Donnerstag, 13. November 2003 00:11 > To: Zabach, Elke > Subject: RE: UPPER with DBCS > > > > Hi Elke, > > still trying to get my head around what is and is not > supported at this > time. > > can you give me some examples of "single ucs2 characters which have a > well-defined Single UCS2 uppercase counterpart"? Do you know > if this would > cover simplified chinese or korean for example? > > Also, is all this the same when it comes to sorting with DBCS? > > Many thanks, > > David > > > > > "Zabach, Elke" <[EMAIL PROTECTED]> on 08/10/2003 02:36:00 AM > > To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>, [EMAIL PROTECTED] > cc: > Subject: RE: UPPER with DBCS > > > [EMAIL PROTECTED] wrote: > > > > Hi, > > > > How does SAP handle the UPPER function with DBCS? Isn't > > uppercase pretty > > language-specific in double-byte character sets? > > > You are right, it is language-specific. It is handled as J�rg > wrote last > year: > lower/uppercase mapping for unicode is implemented using > tables extracted > from unicode.org ftp side. (see vsp81.c created by > genUCS2CaseMaps.pl). > There is a compromise however, between a database and a text > editor for > some characters (like German-Sharp-S, ligatures, some precomposed > characters and all letters with 'iota-subscript' or > 'iota-adjust'. In short > words: all convertions, that would modifiy the length of the > string are not > handled correctly and could not, since the internal interface does not > support different input and output length before and after conversion. > > Only those SINGLE UCS2 characters which have a well-defined > SINGLE UCS2 > uppercase counterpart are therefore handled as expected. > > Elke > SAP Labs Berlin > > > > > -- MaxDB Discussion Mailing List For list archives: http://lists.mysql.com/maxdb To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
