Wolfgang Bauer, September 18, 2008 - 3:15 pm UTC
Thanks for your answer Tom.
Is there any documentation available how merge works internally? E.g. why it doews not work with text indexes (at least if you don't use 11g).
Regards
Wolfgang
September 18, 2008 - 9:51 pm UTC
I don't know how knowing the "internals" of merge would be useful here?
It was not supported in 10g and before to merge into a table with a domain index, in 11g - that functionality is now supported.
Determinism
Duke Ganote, September 18, 2008 - 3:22 pm UTC
Wolfgang Bauer, September 20, 2008 - 11:44 am UTC
Tom,
my question about "internal" documentation has nothing to do with the original question - after your answer it is clear to me why the ORA-00001 happends.
It was just asking for vested interest.
Regards
Wolfgang
you should use distinct
Yonglei, September 21, 2008 - 5:39 pm UTC
Merge can't handle duplicate rows, you need to add distinct on source table:
(select distinct id,col1 from src) src
September 22, 2008 - 4:58 pm UTC
adding distinct would be an obvious approach (remove the duplicates) - but would not have "answered" the question.
the question was "why"
This was very useful - Thank you.
suzanne weiss, October 11, 2012 - 6:53 pm UTC
We have a third party file that we've processed with a MERGE in the past. The most recent file had duplicate entries. The very first question in this thread asked the question I was trying to answer. Thank you. The answer that you gave very clearly answered all of my questions. We'll now go request a cleared up file.