I found the problem and fixed it. There are couple of other small issues
that were found so I plan to put out a 1.0.6 _very soon_.
Thanks for the report.
> -----Original Message-----
> From: Patrick Bennett [mailto:patrick@stripped]
> Sent: Monday, August 29, 2005 9:39 PM
> To: 'Reggie Burnett'
> Cc: mike.griffin@stripped; 'Andrew Preece';
> dotnet@stripped
> Subject: RE: [mysql] Re: Weird character causes statement
> with paramteters to fail
>
> I believe it fixed the decimal parameter problem, but
> introduced a problem (as I described in a message the other
> day) with MySqlParameter's db type changing from what it was
> assigned to originally after setting the Value property.
> IMO, it should try to retain the originally assigned
> parameter type (since that's what the DB expects, right?) and
> convert whatever's assigned to the Value property into the
> appropriate type using System.Convert.ToXxxxx. This would
> then enable user-defined types to implement IConvertible and
> everything would work correctly.
> If a MySqlParameter was created without an assigned dbtype,
> then it would make sense to adopt the type that was assigned
> to the Value property, but once a dbtype is assigned, it
> should strive to convert *to* that type.
>
> I can certainly work around this in my own datatype classes
> by adding a new interface method to have each type set the
> MySqlParameter value property directly, but it's a bit of a
> hack compared to the cleanliness of the built-in IConvertible
> interface. Why this all worked in 1.0.4, I can't say.
>
> -----Original Message-----
> From: Reggie Burnett [mailto:reggie@stripped]
> Sent: Monday, August 29, 2005 4:29 PM
> To: Patrick Bennett
> Cc: mike.griffin@stripped; Andrew Preece;
> dotnet@stripped
> Subject: [mysql] Re: Weird character causes statement with
> paramteters to fail
>
> Patrick & Andew
>
> Please verify that 1.0.5 solves your problem and let me know
> if it doesn't.
>
> -Reggie
>
> Patrick Bennett wrote:
>
> > This is the same problem I've been having as well. It's apparently
> > been fixed in version 1.0.5 of the connector, but they're not
> > releasing it for some reason. :(
> >
>
