Re: [PATCH spice-gtk] gtk: don't request clipboard targets when there is no owner

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,

On Sat, Jan 12, 2019 at 06:29:00PM +0100, Jakub Janku wrote:
> Hi,
> 
> On Fri, Jan 11, 2019 at 11:40 AM <marcandre.lureau@xxxxxxxxxx> wrote:
> >
> > From: Marc-André Lureau <marcandre.lureau@xxxxxxxxxx>
> >
> > After a clipboard-clear, the owner-changed signal is triggered, with
> > no owner. Let's avoid requesting for targets, that would lead to a
> > failure.
> >
> > Fixes:
> > (lt-spicy:23226): GSpice-DEBUG: 13:36:38.741: spice-gtk-session.c:797 clipboard_clear
> > (lt-spicy:23226): GSpice-DEBUG: 13:36:38.742: spice-gtk-session.c:547 clipboard_get_targets:
> > (lt-spicy:23226): GSpice-DEBUG: 13:36:38.742: spice-gtk-session.c:555 Retrieving the clipboard data has failed
> >
> > Signed-off-by: Marc-André Lureau <marcandre.lureau@xxxxxxxxxx>
> > ---
> >  src/spice-gtk-session.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/spice-gtk-session.c b/src/spice-gtk-session.c
> > index adc72a2..9d7702c 100644
> > --- a/src/spice-gtk-session.c
> > +++ b/src/spice-gtk-session.c
> > @@ -670,7 +670,8 @@ static void clipboard_owner_change(GtkClipboard        *clipboard,
> >
> >      /* This situation happens when clipboard is being cleared by us, when agent
> >       * sends a release-grab for instance */
> > -    if (gtk_clipboard_get_owner(clipboard) == G_OBJECT(self)) {
> > +    GObject *owner = gtk_clipboard_get_owner(clipboard);
> > +    if (owner == NULL || owner == G_OBJECT(self)) {
> >          return;
> >      }
> 
> I'm afraid this commit will have to be reverted as it completely
> disables copy&paste from client to guest:

Yep. My bad in acking that too fast.

> According to the GTK+ docs, gtk_clipboard_get_owner() corresponds to
> the owner object set by gtk_clipboard_set_with_owner().
> So if another application in the client's system grabs the clipboard,
> spice-gtk receives "owner-change" event, gtk_clipboard_get_owner()
> returns NULL and the callback returns without sending a grab to the
> vdagent.

The situation that I was handling, was different indeed, in the
patch you acked before. Oh well.

Many thanks for the clarification above, I'll be sending the
revert shortly.

Victor

Attachment: signature.asc
Description: PGP signature

_______________________________________________
Spice-devel mailing list
Spice-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/spice-devel

[Index of Archives]     [Linux Virtualization]     [Linux Virtualization]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]     [Monitors]