Michael Katz wrote:
> 1) A lack of connection timers seems like a massive omission.
MySQL++ doesn't do any network communication at all. None. All of that
is handled by libmysqlclient. Therefore, you should take this up on the
main MySQL mailing list. Any fixes to that underlying layer will also
If there is any change needed to MySQL++, it will simply be to push
through any C-level API for changing the timeouts. If/when such APIs
are implemented, the necessary change to MySQL++ will probably be obvious.
> If we want to contribute
> connection timer patches for other OS's for the ++client how can we go
> about doing this?
Before submitting MySQL++ patches, read the HACKERS file that comes with
the distribution. For MySQL itself, I have no idea.