Message-Id: <10693.122970@fatcity.com> From: "Andrey Bronfin" Date: Mon, 27 Nov 2000 19:14:09 +0200 Subject: Re: National characters This is a multi-part message in MIME format. ------=_NextPart_000_008D_01C058A6.383A1580 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Walt ! Glad to see U not driving a bus yet :)=20 Seriously , many thanks for your reply . Just another question : how do U insert the foreign language stuff into the DB ?=20 Is it typed in by users that have National characters on their keyboards = or is it loaded from files ? How do U see that all the data is OK ? TIA . P.S : I'm using 8.1.6=20 ----- Original Message -----=20 From: Weaver, Walt=20 To: 'Andrey Bronfin' ; ORACLE-L@fatcity.com=20 Sent: Monday, November 27, 2000 6:58 PM Subject: RE: National characters=20 Hi Andrey, =20 We're running UTF8 on an 8.1.6 database, using CLOB's and varchar2 to = store English, Portuguese, Spanish, German, and (in test, anyway) = Japanese stuff. =20 The only problem I had was when I migrated the data initially from an = 8.0.5 database running WE8ISO8859P1 to the UTF8 database. I pre-created = all of the tables and indexes, and during the data migration got a = number of "inserted value too large for column" messages. Didn't have = any problems with the LOB columns, but I had to create the varchar2 = columns twice the size they used to be, since UTF8 uses double the space = for accented characters and such.=20 =20 Other than that, I really haven't had any problems or done anything = special. Just remember that LOB support wasn't available until Oracle = 8i, so if you're going to use UTF8, you won't be able to use any earlier = versions of Oracle... =20 HTH, --Walt Weaver Bozeman, Montana, USA =20 -----Original Message----- From: Andrey Bronfin [mailto:bronfin@VisualTop.com] Sent: Monday, November 27, 2000 8:46 AM To: ORACLE-L@lists.sunysb.edu; ORACLE-L@fatcity.com; = oracledba@quickdoc.co.uk Subject: National characters=20 Dear Gurus ! It appears that i need to design , build and support a database that = will support DATA in different languages with different character sets ( = including such "exotic" languages as Chinese , Hindi , Arabic , Russian = , Hungarian and all the European languages ) . The texts in those languages should be stored in Varchar and LOB = columns . I need to support all those various character sets in ONE DB . I = don't need to manipulate it , just to be able to retrieve the text as it = was stored ( typed in ) . I have no experience with DB character sets other than English , = therefore i'd like to ask those of U who has such experience to help me = . Please tell me whether this can be done and how . Where can i read about it ( docs , URLs , etc ... ) ? What should i be aware of ? Are there any limitations ? Thanks a lot in advance . Andrey Bronfin =20 ------=_NextPart_000_008D_01C058A6.383A1580 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Walt !
Glad to see U not driving a bus yet :)=20
 
Seriously , many thanks for your reply=20 .
Just another question :
how do U insert the foreign language = stuff into the=20 DB ?
Is it typed in by users that have = National=20 characters on their keyboards or is it loaded from files ?
How do U see that all the data is OK = ?
 
TIA .
 
P.S : I'm using 8.1.6
 
----- Original Message -----
From:=20 Weaver,=20 Walt
To: 'Andrey=20 Bronfin' ; ORACLE-L@fatcity.com
Sent: Monday, November 27, 2000 = 6:58=20 PM
Subject: RE: National = characters

Hi=20 Andrey,
 
We're running UTF8 on an 8.1.6 database, = using CLOB's=20 and varchar2 to store English, Portuguese, Spanish, German, and (in = test,=20 anyway) Japanese stuff.
 
The=20 only problem I had was when I migrated the data initially from an = 8.0.5=20 database running WE8ISO8859P1 to the UTF8 database. I pre-created all = of the=20 tables and indexes, and during the data migration got a number of = "inserted=20 value too large for column" messages. Didn't have any problems with = the LOB=20 columns, but I had to create the varchar2 columns twice the size they = used to=20 be, since UTF8 uses double the space for accented characters and such. =
 
Other than that, I really haven't had any = problems or=20 done anything special. Just remember that LOB support wasn't available = until=20 Oracle 8i, so if you're going to use UTF8, you won't be able to use = any=20 earlier versions of Oracle...
 
HTH,
--Walt=20 Weaver
  = Bozeman,=20 Montana, USA
 
 -----Original Message-----
From: Andrey Bronfin=20 [mailto:bronfin@VisualTop.com]
Sent: Monday, November 27, = 2000 8:46=20 AM
To: ORACLE-L@lists.sunysb.edu; ORACLE-L@fatcity.com;=20 oracledba@quickdoc.co.uk
Subject: National characters=20

Dear Gurus !
 
It appears that i need to design , = build and=20 support a database that will support DATA in different languages = with=20 different character sets ( including such "exotic" languages as = Chinese ,=20 Hindi , Arabic , Russian , Hungarian and all the European languages = )=20 .
The texts in those languages should = be stored=20 in Varchar and LOB columns .
I need to support all those=20 various character sets in ONE DB . I don't need to = manipulate it ,=20 just to be able to retrieve the text as it was stored ( typed in )=20 .
 
I have no experience with DB = character sets=20 other than English , therefore i'd like to ask those of U who has = such=20 experience to help me .
Please tell me whether this can be = done and how=20 .
Where can i read about it ( docs , = URLs , etc=20 ... )  ?
What should i be aware of = ?
Are there any limitations = ?
 
Thanks a lot in advance = .

Andrey Bronfin