Re: A Real Abstract BaseDatabaseWrapper Class

2008-09-12 Thread Kevin Teague
On Sep 11, 8:15 am, mtrier <[EMAIL PROTECTED]> wrote: > I'd like to propose that we fix up the abstract BaseDatabaseWrapper > class so that it actually has all the hooks it needs. To be slightly pedantic, if all the methods in a class contain an implementation, it's not an abstract base class,

Re: A Real Abstract BaseDatabaseWrapper Class

2008-09-11 Thread Malcolm Tredinnick
On Thu, 2008-09-11 at 13:17 -0700, mtrier wrote: > Malcolm, > > Thanks for the nice reply. > > On Sep 11, 2:08 pm, Malcolm Tredinnick <[EMAIL PROTECTED]> > wrote: > > Why? If your reason is self-documenting code for somebody implementing > > an external backend, we can probably work something o

Re: A Real Abstract BaseDatabaseWrapper Class

2008-09-11 Thread mtrier
Malcolm, Thanks for the nice reply. On Sep 11, 2:08 pm, Malcolm Tredinnick <[EMAIL PROTECTED]> wrote: > Why? If your reason is self-documenting code for somebody implementing > an external backend, we can probably work something out. Let me be more concrete then. I went to implement a backend.

Re: A Real Abstract BaseDatabaseWrapper Class

2008-09-11 Thread Malcolm Tredinnick
On Thu, 2008-09-11 at 08:15 -0700, mtrier wrote: > I'd like to propose that we fix up the abstract BaseDatabaseWrapper > class so that it actually has all the hooks it needs. Right now the > two missing items I'm seeing are the absence of validation and > introspection. There is the expectation

A Real Abstract BaseDatabaseWrapper Class

2008-09-11 Thread mtrier
I'd like to propose that we fix up the abstract BaseDatabaseWrapper class so that it actually has all the hooks it needs. Right now the two missing items I'm seeing are the absence of validation and introspection. There is the expectation that the implementations will have these and implement th