To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=92164
------- Additional comments from [EMAIL PROTECTED] Mon Jul 28 12:04:30 +0000 2008 ------- fs->oj: there's two problems here: First, when a query is asked for its columns, its first approach is to use a SingleSelectQueryComposer. Only if this fails, a fallback to a prepared statement (and its result set meta data) is used. I suppose the proper solution here is to *not* use a query composer if the query's EscapeProcessing is set to FALSE. Second, the SingleSelectQueryComposer builds an SQL statement from the SELECT part of the underlying statement, a WHERE 0 = 1 condition, and the original GROUP BY clause. In particular, it completely drops the original where condition. Now the query in the example selects a cartesian product of three tables, but limits the results with a WHERE condition. The SingleSelectQueryComposer drops the WHERE condition, so what's left is a mere cartesian product over 3 very large tables - and this is *extremely* expensive. (Yes, in theory the Jet Engine could recognize and value the "WHERE 0=1" condition, but obviously it doesn't.) So, I think we should first fix the Query implementation, which will make the bug here vanish. However, we should also fix the SingleSelectQueryComposer, to retain the original WHERE clause, else the bug might re-appear in other circumstances. --------------------------------------------------------------------- Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
