Re: [g-a-devel] Re: OO.o / new a11y bridge ...

[Oliver Braun <Oliver Braun Sun COM>]

Willie Walker wrote:

> > > If/when the code is incorporated, I'm curious about the following:
> > >
> > >   o Will there be regression tests to determine if the old (Java)
> > >     and new (direct) methods are the same?
> >
> > are you talking about manual or automatic regression tests here ?
>
>
> Well...that's up to you. ;-)  The main concern I have is that the
> event model, focus model, etc. (i.e., all the unspecified stuff in
> the AT-SPI) remains somewhat consistent.  If it changes, then one
> might as well write a new script.

Sounds like such scripts are a perfect candidates for regression test ;-).

> > >     an assistive technology to determine versioning information so
> > >     a scripting screen reader can adapt appropriately?
> >
> > The toolkit name should be sufficient here.
>
>   o If the old/new cannot be the same, will there be some way for
>
> Sorry - I should have dug into the AT-SPI IDL more.  One can actually
> use toolkitName and version for these purposes.

No problem.

Regards,
Oliver