Will and I have refactored Capabilities extensively. Review would be appreciated.

(In reply to comment #1)
> Will and I have refactored Capabilities extensively. Review would be
> appreciated.

I have only a few comments after such a big branch :-)


src/ft-channel.c
+      /* FIXME: should we check for SI, bytestreams and/or IBB too? */
--> Reply to the question or open a bug for that.

src/capabilities.c
gsize feature_handles_refcount = 0;
--> What isn't it static?

src/presence.c
  if (priv->cap_set == NULL)
    { 
      tmp = gabble_capability_set_dump (priv->cap_set, "  ");
--> do you mean '!=' instead of '==' ?  (2 occurences of this pattern in this file)

(In reply to comment #2)
> src/ft-channel.c
> +      /* FIXME: should we check for SI, bytestreams and/or IBB too? */
> --> Reply to the question or open a bug for that.

I'll open a bug, unless Guillaume can insta-answer it...

> src/capabilities.c
> gsize feature_handles_refcount = 0;
> --> What isn't it static?

A very good question. It should be.

> src/presence.c
>   if (priv->cap_set == NULL)
>     { 
>       tmp = gabble_capability_set_dump (priv->cap_set, "  ");
> --> do you mean '!=' instead of '==' ?  (2 occurences of this pattern in this
> file)

Yes, you're right.

(In reply to comment #3)
> (In reply to comment #2)
> > src/ft-channel.c
> > +      /* FIXME: should we check for SI, bytestreams and/or IBB too? */
> > --> Reply to the question or open a bug for that.
> 
> I'll open a bug, unless Guillaume can insta-answer it...

Bug #23777.

> > src/capabilities.c
> > gsize feature_handles_refcount = 0;
> > --> What isn't it static?
> A very good question. It should be.

Fixed in the branch; make check is re-running now.

(en_FR check: "Why" should have been "What" in your review comment :-)

> > src/presence.c
> [...]
> Yes, you're right.

Fixed in the branch; make check is re-running now.

The branch is ok for me

Merged to git master, will be in 0.9.0.