Quoted string not properly terminated

Hello,

I am runnig Oracle 8.1.7 on NT4. When I created database I set charachter set to UTF8.

On NT4 the value of NLS_LANG is AMERICAN_AMERICA.UTF8.

Ok lets move to client machines.
Let just say that I have table TEST which has one field TEXT VARCHAR2(5).
I am going to use some of our nacional charachters (Slovenia), hope you will be able to see them.

On clients the NLS_LANG parameter is set to SLOVENIAN_SLOVENIA.UTF8

I was executing this command :

insert into test values (sometext);

And results where :

If sometext was string without any our nacional charachters there is no problems.

Problem occurs if I use in sometext string charachter è. For example
sometext = 'kaèa'. Then I get error ORA-01756: quoted string not properly terminated.

What is interesting if after charachter è there is an a,e,i,o,u and after that again one of our nacional charachter which is not è, then insert statment exectues ok.

For example sometext = 'kèaš' --> 1 row created.

What I tried next is setting NLS_LANG (on client) to SLOVENIAN_SLOVENIA.EEMSWIN1250.

Then I could use all of our charachters, but (there is always but)!

It looks like that every nacional charachters now requreis two places in string. If field is VARCHAR2(10) and I try to insert string with five our charachters plus one additional (sometext = 'šðèæžh') then I get message : ORA-01401: inserted value too large for column.

P.S. On server NLS_LANG must be set to UFT8, becouse clients can be in very diffent languages. (slovenian,bulgarian,english,turkish,croatian,...)

I think your problem is in you Windows NT, naimely: you have to change default location and character set of your PC to Slovenian.
When i do that i can type even in Japanese and save any data i want, but if i dont do that i am getting the same error as you.

I've changed regional setting on NT4 to slovenian, but I keep getting the same errors.

Did you do it on station you are connecting from?
Or did you do it on the Server site?

I've changed it on server side.

Then I started SqlPlus on serves side with nls_lang values american_america.utf8 and to
slovenian_slovenia.utf8

still there was quoted string...

Then I changed it to
american_america.ee8mswin1250 and to
slovenian_slovenia.ee8mswin1250

inserted value to long

P.S.
My machine from where I am connection the regional setting are
Location : Slovenian
Language (checked) : Central Europe (default), Cyrilic, Western Europe and United States.

Is the problem with the language selection?

P.S. Slovenian language is part of CE

On my computer I have W2K.

Yes, you have to change default language as well.

Make default language SLOVENIAN.

Quote:

---

Originally posted by Aljaz
My machine from where I am connection the regional setting are
Location : Slovenian
Language (checked) : Central Europe (default), Cyrilic, Western Europe and United States.

Is the problem with the language selection?

P.S. Slovenian language is part of CE

---

From what I can see I have then sistem regional settings set to slovenian, plus I have the dafault sistem locale set to slovenian and also the default OS locale. I have everything that is possible set to slovenian. My windows are in english language if this plays a role.

If helps this what I get if I guery nls_session_parameters and nls_instance_parametres tables

select * from nls_session_parameters;
PARAMETER VALUE
----------------------------- -----
NLS_LANGUAGE SLOVENIAN
NLS_TERRITORY SLOVENIA
NLS_CURRENCY SIT
NLS_ISO_CURRENCY SLOVENIA
NLS_NUMERIC_CHARACTERS ,.
NLS_CALENDAR GREGORIAN
NLS_DATE_FORMAT DD MM YYYY HH24:MI:SS
NLS_DATE_LANGUAGE SLOVENIAN
NLS_CHARACTERSET UTF8
NLS_SORT SLOVENIAN
NLS_TIME_FORMAT HH24:MI:SSXFF
NLS_TIMESTAMP_FORMAT DD.MM.RR HH24:MI:SSXFF
NLS_TIME_TZ_FORMAT HH24:MI:SSXFF TZH:TZM
NLS_TIMESTAMP_TZ_FORMAT DD.MM.RR HH24:MI:SSXFF TZH:TZM
NLS_DUAL_CURRENCY SIT
NLS_NCHAR_CHARACTERSET UTF8
NLS_COMP BINARY

select * from nls_instance_parameters

PARAMETER VALUE
----------------------------- ----------
NLS_LANGUAGE AMERICAN
NLS_TERRITORY AMERICA
NLS_SORT
NLS_DATE_LANGUAGE
NLS_DATE_FORMAT
NLS_CURRENCY
NLS_NUMERIC_CHARACTERS
NLS_ISO_CURRENCY
NLS_CALENDAR
NLS_TIME_FORMAT
NLS_TIMESTAMP_FORMAT
NLS_TIME_TZ_FORMAT
NLS_TIMESTAMP_TZ_FORMAT
NLS_DUAL_CURRENCY
NLS_COMP

Slovenian charset is ISO8859-2,
NLS_LANG=slovenian_slovenia.EE8ISO8859P2
I do not know already what is wrong, when I changed English Windows to be Japanese, I could use UTF8 charset as client charset.

Slovenian is a language of Easten Europe BUT NOT Central Europe.

You wrote:

Language (checked) : Central Europe (default), Cyrilic, Western Europe and United States.
Is the problem with the language selection?
P.S. Slovenian language is part of CE

I,ve set NLS_LANG = slovenian_slovenia.EE8ISO8859P2 and it is working!

There is no sign of "quoted string not properly terminated" error.

But now I have the other problem. My nacional charachters requires two places in a string. So if field's length is 10 and I insert string which containts 5 our charachters plus one more that I get message "inserted value too large for column"

What I would like to kwow is that normal. Becouse if that is normal I will just increase lenghts of varchar2 fields in database.

That is normal, your Slovenian character takes 2 bytes in UTF8 database, so in VARCHAR2(10) you can place only 5 characters.

To know the size of string:

select lengthb('?') from dual;

Quote:

---

Originally posted by Aljaz
What I would like to kwow is that normal. Becouse if that is normal I will just increase lenghts of varchar2 fields in database.

---

Thank you!