Replacing the text

Question

Hello All,
I need your help regarding replacing the text with some other text. I am trying to replace the text '\\' with '\'. But i am not able to get the correct result.
I am having string value like   '\\Computername\Path\comp_name\\domain\\location' i want to replace the text '\\' with '\' only for 2nd and 3rd occurance.
Replace \\domain\\ with \domain\ I have tried with below query...
SELECT REGEXP_REPLACE('\\Computername\Path\comp_name\\domain\\location' ,
       '(\\){2,}', '\')
       AS Expression
  FROM dual;i am getting the below result for above query.
  \Computername\Path\comp_name\domain\location But i want the output like below..
\\Computername\Path\comp_name\domain\location I am using Oracke 11g vesion.
Please help me on this .
Thanks a lot for your help.

Answer 1

Hi,
sunitha2010 wrote:
... small doubt--Is there any difference between these two queries?Let's call this Query 1:
SELECT  REGEXP_REPLACE ( str
                 , '\\\\'    -- 2 backslashes, escaped.  See note below
                 , 1 + INSTR ( str
                 )     AS expression
FROM    table_x
;and Let's call the query below Query 2:
select
REGEXP_REPLACE
('\\Computername\Path\comp_name\\domain\\location', '\\\\', '\\',2 )
as Expression
from
dual
;Yes, there is a difference.
Query 1 leaves the first occurrence of '\\' intact, regardless of where that is in the string.
Query 2 ignores the 1st character, but otherwise changes all occurrences of '\\' into a single '\'.
You'll see the difference if you add anything to the very beginning of the test string, for example
'X\\Computername\Path\comp_name\\domain\\location'I mentioned this in my first message:
Frank Kulash wrote:
... I assume the 1st occurrence won't always be at the beginning of the string...If the first occrence always is at the beginning of the string, then you don't need regualr expressions at all. You can say:
SELECT  SUBSTR (str, 1, 1)
     || REPLACE ( SUBSTR (str, 2)
          )     AS expression
FROM    table_x;This will be more efficient than regular expressions.