I am not familiar with how charactersets work because I have never
needed anything other than simple ASCII, however if I recall
correctly, Firebird 1.5 has some issues with certain character sets.
You could perhaps check Firebird 2 (beta) to see if you issue has been
resolved there. I can not give you the details nor any work arounds
because it has never been an issue for me, but the first thing I would
check is whether your problem is the way the data is stored or in the
data access components used by IBO, and the easiest way to do that is
to check FB 2.

Adam

--- In firebird-support@yahoogroups.com, "mailmur" <mailmur@y...> wrote:
>
> I made a simple testdb without utf-8 charset then iboconsole.exe
> worked fine. But its not what I need. If my db has UNICODE_FSS charset
> it was able to list data, but most string values were garbage if
> contained two-byte unicode (>127 ascii) letter.
>
> My db applications _must_ support unicode from point to point. I could
> use ISO-8859-1 to support finnish letters but then greek would be
> impossible. To make letter problems go away unicode is only solution
> so I must use UTF-8 charset in databases.
>
> We currently use MySQL4.1 in a production use as it has a smooth
> unicode support. phpMysqladmin have no problems with unicode 'cause
> webbrowser does it well. But still some tasks are easier to do with
> MySqlQueryBrowser or MySqlAdministrator windows applications.
>
> Java and dotNet client applications have no problems rendering unicode
> letters. But Javabased admin tools I've tried lack too much features.
> Client applications are mostly java and dotnet what we use.
>
> I havent tried FB2 which is said to have a unicode support. FB1.x
> versions are practically problem when it comes to unicode (can store
> even 6 chars to varchar(3) columns, string utf functions does not
> work well, sorting and/or collation problems)
>
> And hard to understand how to setup a proper utf-8 support, and then
> clients dont break everything if connectionstrings miss a small charst
> attribute. MSSQL server does it well without too much hassle, so do
> current MySQL version and all functions and sortings work well.
>
> I am planning on installing FB2 version at home and test how much
> progress utf-8 support have.
>
> > I am using the ISO8859_1 character set for a database in Norway. Using
> > IBOConsole, I can update the tables with Scandinavian characters
> > without problem.
> >
> > I think the error you are seeing is coming from the database. Since
> > Unicode has multi byte characters, are you trying to enter more
> > characters than the column width (in bytes) permits?
>
>
> > > I downloaded IBOConsole and tested it with my current FB1.5.0.4290
> > > installation. I have a database with UNICODE_FSS default charset.
> > >
> > > Registered localhost server, registered TESTUTF database and then
> > > tried to insert non-USascii letters (a with two dots, a with ring
> etc scandinavian letters).
> > >
> > > "Trying to post a new record it just gives me the following error.
> > > ISC ERROR CODE: 335544321, aritmetic exception, numeric overflow,
> or string truncation. Cannot transliterate character between character
> sets."
> > >
> > > I have tried few Java based dbManagers but they lack too much
> > > features. But Java and/or dotNet should give a smooth full unicode
> > > support from point to point. Maybe native UIs, like IBOConsole,
> dont cope with unicode charsets well enough.
> > >
> > > Table SQL
> > > ============
> > > CREATE TABLE "Customer" (
> > > "id" INTEGER NOT NULL,
> > > "Name" VARCHAR(50) CHARACTER SET UNICODE_FSS,
> > > "Code" VARCHAR(3) CHARACTER SET UNICODE_FSS NOT NULL,
> > > CONSTRAINT "PK_Customer" PRIMARY KEY ("id")
> > > );
>