[Spice-devel] [PATCH 11/18] worker: use spice_return_if_fail() instead of spice_assert() in release_item

Frediano Ziglio fziglio at redhat.com
Tue Nov 24 02:17:57 PST 2015


> 
> On Mon, 2015-11-23 at 17:01 +0000, Frediano Ziglio wrote:
> > From: Marc-André Lureau <marcandre.lureau at gmail.com>
> > 
> > ---
> >  server/red_worker.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/server/red_worker.c b/server/red_worker.c
> > index 656f9ab..65d5dea 100644
> > --- a/server/red_worker.c
> > +++ b/server/red_worker.c
> > @@ -4453,7 +4453,7 @@ static void release_item(RedChannelClient *rcc,
> > PipeItem
> > *item, int item_pushed)
> >  {
> >      DisplayChannelClient *dcc = RCC_TO_DCC(rcc);
> >  
> > -    spice_assert(item);
> > +    spice_return_if_fail(item != NULL);
> >      dcc_release_item(dcc, item, item_pushed);
> >  }
> >  
> 
> Early return seems fine here, but in order to actually return and not abort,
> we
> should use g_return_if_fail()
> 

I like spice_assert here.
However changing to spice_return_if_fail would be a no-op and is not
useful to mark a future conversion to g_return_if_fail.

I could have an easy suggestion if we decide we want to change the
current spice_assert with a future g_return_if_fail:

before:
  spice_assert(whatever);

after:
  /* change to g_return_if_fail */
  spice_return_if_fail(whatever);

This is the second proposal I do... I'm starting to be tired of
these chit-chat.

Frediano


More information about the Spice-devel mailing list