[PATCH wayland v2 1/2] client: update obsolete comments

[Pekka Paalanen]

On Wed, 3 Dec 2014 13:05:16 +0100
Marek Chalupa <mchqwerty at gmail.com> wrote:

> On 28 November 2014 at 14:30, Pekka Paalanen <ppaalanen at gmail.com> wrote:
> 
> > On Fri, 28 Nov 2014 12:18:55 +0100
> > Marek Chalupa <mchqwerty at gmail.com> wrote:
> >
> > > 1) there is nothing like main thread since
> > >    3c7e8bfbb4745315b7bcbf69fa746c3d6718c305 anymore, so remove
> > >    it from documentation and update the doc accordingly.
> > >
> > > 2) use calling 'default queue' instead of 'main queue'. In the code
> > >    we use display->default_queue, so it'll be easier the understand.
> > >
> > > 3) update some obsolete or unprecise pieces of documentation
> > >
> > > Signed-off-by: Marek Chalupa <mchqwerty at gmail.com>
> > > ---
> > >  src/wayland-client.c | 80
> > ++++++++++++++++++++++++++++++++--------------------
> > >  src/wayland-client.h | 26 ++++++++---------
> > >  2 files changed, 61 insertions(+), 45 deletions(-)
> > >
> > > diff --git a/src/wayland-client.c b/src/wayland-client.c
> > > index 36380fe..0ab94e9 100644
> > > --- a/src/wayland-client.c
> > > +++ b/src/wayland-client.c

...

> > > +
> > > + * \note Display queue (for wl_display events like delete_id or so)
> > > + * is dispatched always.
> >
> > I wonder if referring to the display queue is just confusing, because I
> > don't see us really explaining what it is anywhere. And it's not the
> > default queue. In fact, I don't think the users of libwayland would even
> > care about it.
> >
> 
> True.
> 
> Hmm, just thinking about it - I should rephrase the thing about checking
> return value of wl_display_dispatch_queue(). Because when
> it dispatched some events in display_queue, then it won't return 0 even
> when it dispatched none of the queue-assigned events, so what I wrote in
> the comment to wl_display_dispatch_queue() is wrong.
> 
> But to clarify that it can return non-0 even when it dispatched no events,
> I have to say the thing about display queue, don't I?

Huh. I wonder if that should be fixed in code instead?
Could anything outside libwayland-client care about when the hidden
internal display queue dispatches? Errors maybe?

Dispatching the display queue cannot create requests one might need to
flush, right?

Hmm, but the correction you did to the doc is good, accurate to the
code AFAIK, and not doing a non-obvious change in the code is probably
for the better. Yeah, nevermind, I think. :-)


Thanks,
pq