On Fri, Apr 18, 2008 at 11:11 AM, Felix Meschberger <[EMAIL PROTECTED]> wrote:
> Am Freitag, den 18.04.2008, 10:44 +0200 schrieb Bertrand Delacretaz:
> >... I'd use my suggested rules as the only options here:
> >
> > rule A)
> > {requestMethod}.{requestExtension} == "GET.html" -> required to use an
> > empty string
>
> At first sight, I could agree. But thinking again about how Servlets
> would be merged into the Resource Tree view with their registration
> properties for virtual path building, we should allow GET and/or html.
> Otherwise a servlet could not be registered for, say, GET and POST
> explicitly....
Ok, I didn't consider the servlets selection - I agree with your
suggested optional parts then, except resourceTypeLabel which is
required, as discussed.
> ... We should not treat html special in that we prevent it from being used
> as part of the script name. Rather we make it optional; always. It is
> then the responsibility of the script programmer to cope with the
> situation...
Ok, if that makes the overall resolution mechanism better.
> ...It is also easy to describe: The requestExtension is optional (all
> the time actually, see above) while the requestMethod is required for
> non-GET (and non-HEAD) only and optional for GET/HEAD requests. This
> still reflects the separation of safe vs. unsafe methods...
Ok.
-Bertrand
