1

There is different letters for 'KAF' (ک , ك) and 'YA' (ی , ي) in Persian with different unicode codes and same spell.

I am using these commands to set Persian sort and comparing in my project

ALTER SESSION SET nls_sort=persian;

ALTER SESSION SET nls_comp =linguistic;

With setting nls_comp and nls_sort parameter I receieve these output

select first_name from customer 
where first_name like '%ك'; -- with result

select first_name from customer 
where first_name like '%ک';--with no result

How can receive same output for these queries? Can I customize nls_comp ?

Nik Kashi
  • 4,447
  • 3
  • 40
  • 63
  • Have you investigated using regexp instead of like? Just reading around the topic it looks like there are character equivalences in Posix that might not be available in LIKE. http://docs.oracle.com/cd/B28359_01/server.111/b28298/ch5lingsort.htm#i1008619 – David Aldridge Sep 15 '13 at 08:12
  • I need a systemic solution. There is more than 200 tables and hundreds of queries in our project! – Nik Kashi Sep 15 '13 at 08:51

0 Answers0