Re: Are all commands mandatory

This is, as Andreas points out, the reason why we have capabilites that are
returned. From the point of view of httpd I guess it would need to
implement everything and then handle errors from the UA if it can't handle
the call properly.

David


On 8 March 2015 at 20:37, Andreas Tolfsen <ato@mozilla.com> wrote:

> On 13 Feb 2015, at 00:26, James Graham <james@hoppipolla.co.uk> wrote:
> > Are all implementations expected to support all commands? It seems like
> > there are some for which this is not possible e.g. on a mobile device
> > changing the window size might not be possible. On the other hand it
> > feels like supporting navigation ought to be mandatory. This sort of
> > suggests that the commands can be divided up into ones that a browser
> > SHOULD supprt, if it can, and ones that it MUST support. Does that seem
> > sensible? If so, which are in which category?
>
> Not all commands are mandatory to implement.
>
> It is perhaps poorly described in the spec text as it is, but the idea is
> for the remote end to state which capabilities it supports when a new
> session is created.
>
> It's obvious that window manipulation (resizing, maximising) isn't
> possible across all devices.  The spec should be clear that a
> supportsWindowHandling capability needs to be set if the driver/UA supports
> this.
>
>
>