Aaron Stone wrote:
The bug appears to be very subtle. When I was backing out Mikhail's patches one at a time, there was a one-line hunk of the dbmysql_v2.patch that didn't reverse because it had not been applied.
This probably means that the first version of the dbmysql patch was applied. It did contain a subtle bug, which I fixed in the second version (duly warning the list about it).
Yours, Mikhail Ramendik
