[Bug 725221] Add GstUri object for URI handling

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=725221
  GStreamer | gstreamer (core) | git

--- Comment #24 from Sebastian Dröge (slomo) <slomo at coaxion.net> 2014-09-22 20:32:41 UTC ---
Thanks for the patch update :)

(In reply to comment #23)

> > @@ +929,3 @@
> > + * fragment NULL.
> > + *
> > + * Returns: A new GstUri object. [transfer full]
> > 
> > No need to add gtk-doc documentation for private functions
> > 
> Removed those although did leave in an explanation of one function as a simple
> comment.

That's ok (and good :) )

> > @@ +2433,3 @@
> > + * gst_uri_set_query_table:
> > + * @uri: (transfer none)(nullable): The #GstUri to modify.
> > + * @query_table: (transfer none)(nullable): The new query table to use.
> > 
> > Not sure if a GHashTable is ideal for this. Are the GI annotations to tell
> > about the key and value types?
> > 
> Yes, those detail the key and value types.

Ah yes, there's apparently "(element-type KTYPE VTYPE)". But you didn't add
those. Can you add them? Otherwise I'll add that when merging :)

> > Maybe we need a GstUriQueryElement struct, and then here use a GList of
> > GPtrArray of them?
> > 
> I've left this as a GHashTable for now as I didn't want to reinvent a key/value
> pair type. I thought the hash table would be the best fit here as most of the
> time you'd want to lookup a query value string by it's key name, or insert
> key/value pairs. I couldn't see a situation where this structure wouldn't fit,
> so it seemed best. It maybe worth revisiting later.

Conceptionally it makes sense, yes. I was just worried about the bindings
friendliness here :)

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.