The problem is solved. The coalesce() function was the solution. Thank you very much for your input.
With best regards Horst Düster --------------------------------- Dr. Horst Düster GIS-Koordinator /Amtschef Stv. Kanton Solothurn Amt für Geoinformation Abteilung SO!GIS Koordination Werkhofstr. 65 CH-4509 Solothurn Tel.: ++41 (0)32 627 25 32 Fax: ++41 (0)32 627 22 14 [EMAIL PROTECTED] www.sogis.so.ch -----Ursprüngliche Nachricht----- Von: Dawid Kuroczko [mailto:[EMAIL PROTECTED] Gesendet am: Freitag, 8. Juli 2005 11:08 An: Düster Horst Cc: pgsql-admin@postgresql.org Betreff: Re: [ADMIN] where with NULL values are not selected On 7/8/05, Düster Horst <[EMAIL PROTECTED]> wrote: > I try to join two tables whereas some column values do have NULL values with > the following query: > > select table1.column from table1, table2 where table1.column=table2.column > > table1.column and table2.column may have NULL values. The problem is that > these columns where not selected. Does there exists any solution to > select/join the NULL value colums also. A hint: qnex=# SELECT 'ok' WHERE NULL=NULL; ?column? (0 rows) In other words -- NULL is not equal to NULL. NULL is not a value, NULL is a state. If you want to join those columns, you cannot use NULL as a joining key. It's the way SQL works. Anyway, a quick-and-dirty solution might be: select table1.column from table1, table2 where coalesce(table1.column,-1)=coalesce(table2.column, -1); ...assuming column doesn't have '-1' value. A better solution would be to rethink your design (don't use NULLs). Incientally, instead of using this syntax: select table1.column from table1, table2 where table1.column=table2.column; ...try using explicit inner joins: select table1.column from table1 JOIN table2 USING(column); Regards, Dawid ---------------------------(end of broadcast)--------------------------- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
