Re: delete based on multipart key

Rudy Fernandes wrote: > > In article <pz67yez52w.fsf@eeyore.ibcinc.com>, I can't figure out an efficient way > >to delete all the rows whose keys appear in a second table. > delete from first_table > where a || b in ( select a || b from second_table); This works, but it is not exactly efficient since it will not take advantage of the index on first_table columns (a,b). Using a cursor in a procedural language will do the trick. Which reminds me, I've mentioned this before and I'll mention it again, there ought to be syntax for this multi-part sort of thing like so: delete from first_table where a,b in (select a,b from second_table) Seems simple and reasonable enough to me. How does one get this into the next set of ANSI standards? Of course, they also ought to have case-insensitive searching. Even mSQL provides a 'CLIKE' operator for such a purpose. Enough griping, back to work, Douglas Wilson