Regexp_replace help

Question

Hi,
I am writing a regexp_replace and I am a newbie to this concept. I am looking for help to convert the following pattern as follows:
If I see a pattern of two lowercase followed by a period and a space, then I need to convert the pattern to two lowercase followed by a period and a new line character.
"lowercaselowercaseperiodspace" --> "lowercaselowercaseperiodnewline"
Any help would be appreciated.
Ajay

Answer 1

Hi,
[[:lower:]]or (depending on your NLS settings)
[a-z]means a lower-case letter, so if newline is CHR (10) on your system, you can say:
REGEXP_REPLACE ( str
            , '([[:lower:]]{2}\.) '
            , '\1' || CHR (10)
            )When writing or debugging regular expressions, you may want to be excruciatingly clear about what you're doing, like this:
REGEXP_REPLACE ( str
            , '('               || -- \1 is ...
                 '[[:lower:]]{2}'     || --       any 2 lower-case letters
              '\.               || --         followed by a dot
           ')'               || -- end \1
           ' '                  -- followed by a space
            , '\1'                || -- \1 is copied without changes
              CHR (10)             -- but the space is replaced by newline
            )Edited by: Frank Kulash on Mar 7, 2011 2:56 PM