When I fixed threadtest2.c yesterday, it quickly uncovered a bug (not thread related but a bug all the same) that was introduced by the fix to the database corruption problem of earlier this week. The new bug does not cause database corruption, but it can cause a segfault if you have a lot of sqlite3_open() calls (100s per second) to the same database from many threads or processes. See
http://www.sqlite.org/cvstrac/chngview?cn=3563 The question is: should I rush out 3.3.10 to cover this important bug fix, wait a week to see if any other bugs surface, or do the usual 1-2 month release schedule and let people effected by this bug apply the patch above. What is the opinion of the user community? -- D. Richard Hipp <[EMAIL PROTECTED]> ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
