[Bug 746818] v4l2: add uri handler interface

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=746818

Nicolas Dufresne (stormer) <nicolas at ndufresne.ca> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #347626|none                        |needs-work
             status|                            |

--- Comment #6 from Nicolas Dufresne (stormer) <nicolas at ndufresne.ca> ---
Review of attachment 347626:
 --> (https://bugzilla.gnome.org/review?bug=746818&attachment=347626)

::: sys/v4l2/gstv4l2object.c
@@ +588,3 @@
+          gst_uri_unref (v4l2object->uri);
+
+        v4l2object->uri = gst_uri_from_string (g_value_get_string (value));

This may fail.

@@ +590,3 @@
+        v4l2object->uri = gst_uri_from_string (g_value_get_string (value));
+
+        gst_object_set_properties_from_uri_query_parameters (G_OBJECT
(v4l2object->element), v4l2object->uri);

It's nice to have a generic interface here, though there is some stuff I really
don't like. I'll bring that to the respective bug. In this specific case, we
MUST filter certain properties (name, parent, device, device-fd, io-mode, are
the one I would deliberatly prevent from accessing from the URI). We should
also find a way to validate the values in a way that it won't assert.

@@ +591,3 @@
+
+        gst_object_set_properties_from_uri_query_parameters (G_OBJECT
(v4l2object->element), v4l2object->uri);
+        path = gst_uri_get_path(v4l2object->uri);

Coding style, need space before (.

@@ +710,3 @@
+        gchar *string = gst_uri_to_string (v4l2object->uri);
+
+        g_value_set_string (value, string);

g_value_take_string() please, you can probably make it a one liner.

::: sys/v4l2/gstv4l2sink.c
@@ +102,3 @@
+        gst_v4l2sink_video_orientation_interface_init);
+    G_IMPLEMENT_INTERFACE (GST_TYPE_URI_HANDLER,
+        gst_v4l2sink_uri_handler_init));

This is a non-sense, URI handler are strictly for sources.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.