From: Warren Young
Date: February 9 2008 1:05am
Subject: Re: possible bug in rc3 ?
List-Archive: http://lists.mysql.com/plusplus/7444
Message-Id: <47ACFC66.6030300@etr-usa.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit

Thomas Werth wrote:
>> You have exceptions turned off and don't show that you're checking for 
>> error codes.  How do you know that the option was set successfully?
>>
> All code is surronded by try block ending with
> catch ( std::exception &e )
>     {
>         cerr << e.what();
>     }

Doesn't matter.  You passed 'false' to the Connection ctor, which 
disables all optional exceptions in it and any children it creates that 
can throw optional exceptions.  Thus, the only way MySQL++ can signal 
most errors -- including option setting errors -- is with return codes.

> if ( !conn.set_option( 

Okay, this tells us that set_option() isn't encountering an error, which 
means the MySQL C API has accepted the SSL parameters without complaint. 
  This means it probably isn't MySQL++'s fault, and that changing to 
v2.3.2 will show the same problem.

It also means enabling exceptions won't help.  I included the previous 
comments for your information only, because you seem a little fuzzy on 
this.  You can't catch exceptions that don't get thrown. :)

> For additional info i used ssl key setup from 
> http://dev.mysql.com/doc/refman/5.0/en/secure-create-certs.html
> and mysql from xamp.

Thanks for the pointer.  I haven't actually tested this feature of 
MySQL++ -- I added it, but it scratches someone else's itch -- so this 
will help if I decide I need to set this up for testing.