Re: [Firebird-net-provider] UNICODE_FSS Character set question

[mailmur <ma...@ya...>]

[warning, long post]
Ive understood same that Interbase(Firebird) has used
UTF-8 as an internal UNICODE_FSS charset. Dont know
why they named it so weird.

About ISO8859-1 charset:
This is what we use here in up north if unicode
charset is not available. It has scandinavian letters
as an extended ascii codes. ALT+0228 gives me "ä"
character in Notepad.

BUT it's not UTF-8 equalent where "ä" letter is
represented with two bytes. Here is Win2k Notepad UTF8
file from hexeditor: 
EB BB BF C3 A4 where EBBBBF is a unicode BOM-marker.
So, C3A4 represents "ä" character in UTF-8 format.

I can run tomorrow few tests with greek and cyrillic
letters. They definitely should not work with
ISO8859-1 encoding.
Greek: ΨΩΪ
Cyrillic: ЗИЙ
Scandinavian: åäö

I will download Jaybird jdbc driver and test it
against the same testdb as well.

What happens if database is UNICODE_FSS and
connectionStr does not provide a charset parameter. Is
it _server_ that does all the conversion and most
likely will fall to default ISO8859-1 if nothing is
provided?

Native Interbase/Firebird drivers, jaybird and Delphi
drivers use "lc_ctype" parameter name for charset. I
think NETProviders "charset" is more general name for
such parameter. 

That probably does not mean anything here as it's just
a logical name for client driver reading value from
connection string.

--- Carlos Guzman Alvarez <car...@te...> >
Hello:
> mailmur has reported a problem with the UNICODE_FSS
> character set in the .NET provider, i have this
character set mapped to
> UTF-8 in the .NET Provider ( is this really correct
?? i think it's ok
> :) ), the problem gives with a field ( varchar field
) with a value of
> ( database was created using UNICODE_FSS as
character set and i
> have made the test using Firebird 1.5 RC4 ) :
> 
> Känkkäränkkä
> 
> but in the .net provider when i read the field value
> i'm receiving 12 bytes of data:
> 
> 	[0]	75
> 	[1]	228
> 	[2]	110
> 	[3]	107
> 	[4]	107
> 	[5]	228
> 	[6]	114
> 	[7]	228
> 	[8]	110
> 	[9]	107
> 	[10]	107
> 	[11]	228
> 
> When i transfor this buffer into a string using the
> UTF-8 encoding of .NET the result is:
> 
> Knkkrnkk
> 
> Instead if i map the UNICODE_FSS character set to
> ISO8859_1 all runs fine, any idea about what can be
the problem ??

__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com