Using REGEXP_REPLACE to remove XML data from between XML tags

Question

Hello there,
I am trying to use the REGEXP_REPLACE SQL function on a piece of XML, to remove the actual XML data from the elements. My end goal is to end up with a comma-delimited list of XML element names. As part of the first step, I just want to rip out all the actual data.
I tested the following query, and it initially appeared to work:
SELECT REGEXP_REPLACE('>THIS IS A TEST<',
                      '>([[:alnum:]]\s|\S)*<',
                      '><' ) AS test_result
  FROM dual;Unfortunately, when I applied this to a full XML string, it didn't work. Here is the test query I used:
SELECT REGEXP_REPLACE('<ROW><TEST_ELEMENT1>123</TEST_ELEMENT1><TEST_ELEMENT2>THIS IS A TEST!</TEST_ELEMENT2></ROW>',
                      '>([[:alnum:]]\s|\S)*<',
                      '><') AS test_result
  FROM dual;I ended up with the following output:
*<ROW></ROW>*
What I was trying for was:
*<ROW><TEST_ELEMENT1></TEST_ELEMENT1><TEST_ELEMENT2></TEST_ELEMENT2></ROW>*
If you're reading this and you're a Posix RegExp Guru, please could you let me know exactly where I'm going wrong? RegExps are not my strong point, but I'd like to get better at them.

Answer 1

jmcnaug2 wrote:
Just out of interest, how would you go about writing the XQuery function so that it returns the actual XML Element values in comma-delimited form (This is Oracle Canonical XML, so has a very basic layout)? ;)Do you mean like this ?
SQL> SELECT xmlquery(
  2    'for $i in $d/ROWSET/ROW
  3     return string-join($i/*, ",")'
  4    passing dbms_xmlgen.getXMLType('SELECT employee_id, last_name, first_name FROM employees WHERE rownum < 10') as "d"
  5    returning content
  6  ).getClobVal() as csv
  7  FROM dual;
CSV
100,King,Steven
101,Kochhar,Neena
102,De Haan,Lex
103,Hunold,Alexander
104,Ernst,Bruce
105,Austin,David
106,Pataballa,Valli
107,Lorentz,Diana
108,Greenberg,Nancy
SQL>