Case Sensitivity Issue with Latin Characters

Question

Hi,
The Latin Character that i am talking abt is "i",
I am trying to provide in my application the support for Azeri-Latin(Turkish family) Characters,
That is the display and the orientation of characters is perfectly fine throughout the flow of application.
But now there is a problem when selecting a data from DB using like constraint in where clause... The issue has been elaborated below...
From the set of characters there are two character in which i am facing Case Sensitivity issue...
They are as,
i ---> the capital of same is ---> İ
ı ---> the capital of same is ---> I
now in english character mapping,
for i ---> the capital is ---> I
So, this is creating an issue...
Now the test i have done is as follows,
I created a table as,
Create Table CS
CS1 VARCHAR2(20)
Following is the data that is inserted in the table,
Insert into CS values('İ');
Insert into CS values('i');
Insert into CS values('I');
Insert into CS values('ı');
now i am running the following query,
Select * from CS where lower(CS1) like lower('%&a%');
for characters "i","İ","I" it returns 3 rows, that are,
i
İ
I
and for character ı it only returns,
ı
ideally for characters "i" & "İ" should be mapped with each other and "ı" & "I" should be mapped with each other respectively.
But that is not the case.
I am using Oracle 10g
my NLS_database_parameters are as follows,
PARAMETER     VALUE
NLS_LANGUAGE     AMERICAN
NLS_TERRITORY     AMERICA
NLS_CURRENCY     $
NLS_ISO_CURRENCY     AMERICA
NLS_NUMERIC_CHARACTERS     .,
NLS_CHARACTERSET     AL32UTF8
NLS_CALENDAR     GREGORIAN
NLS_DATE_FORMAT     DD-MON-RR
NLS_DATE_LANGUAGE     AMERICAN
NLS_SORT     BINARY
NLS_TIME_FORMAT     HH.MI.SSXFF AM
NLS_TIMESTAMP_FORMAT     DD-MON-RR HH.MI.SSXFF AM
NLS_TIME_TZ_FORMAT     HH.MI.SSXFF AM TZR
NLS_TIMESTAMP_TZ_FORMAT     DD-MON-RR HH.MI.SSXFF AM TZR
NLS_DUAL_CURRENCY     $
NLS_COMP     BINARY
NLS_LENGTH_SEMANTICS     BYTE
NLS_NCHAR_CONV_EXCP     FALSE
NLS_NCHAR_CHARACTERSET     AL16UTF16
NLS_RDBMS_VERSION     10.2.0.1.0
and at client i m using,
NLS_LANG = .AL32UTF8
any help on this issue will be appreciated,
Thanks & Regards,
Pratik

Answer 1

I'm not fully understand your question but I guess you may need to set up NLS_LANGUAGE, NLS_TERRITORY db parameters and client NLS_LANG.
eg client NLS_LANG=TURKISH_AZERBAIJAN.AL32UTF8 and db parameters NLS_LANGUAGE=TURKISH, NLS_TERRITORY=AZERBAIJAN
Full list of supported languages, locales see [Oracle® Database Globalization Support Guide local data subpage|http://download.oracle.com/docs/cd/B19306_01/server.102/b14225/applocaledata.htm]
or tip2: check this: ALTER SESSION SET NLS_COMP=LINGUISTIC;
Edited by: Kecskemethy on Mar 24, 2009 3:14 AM
or you need something like this:
Example 5-16 Matching with the Base Letter Operator [==]
Expression: r[[=e=]]sum[[=e=]]
Matches:
resume<<
résumé<<
résume<<
resumé<<Oracle SQL syntax: SQL> SELECT col FROM test WHERE REGEXP_LIKE(col,'r[[=e=]]sum[[=e=]]');Edited by: Kecskemethy on Mar 24, 2009 3:20 AM