After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 743496 - libcheese: port to clutter-gst 3.0
libcheese: port to clutter-gst 3.0
Status: RESOLVED FIXED
Product: cheese
Classification: Applications
Component: general
git master
Other All
: Normal normal
: ---
Assigned To: Cheese Maintainer(s)
Cheese Maintainer(s)
Depends on:
Blocks: 751185
 
 
Reported: 2015-01-25 17:20 UTC by Lionel Landwerlin
Modified: 2015-11-30 15:09 UTC
See Also:
GNOME target: ---
GNOME version: ---

Attachments
libcheese: port to clutter-gst 3.0 (8.20 KB, patch)
2015-01-25 17:20 UTC, Lionel Landwerlin 		none Details | Review
libcheese: port to clutter-gst 3.0 (13.20 KB, patch)
2015-01-27 12:25 UTC, Lionel Landwerlin 		committed Details | Review

Description Lionel Landwerlin 2015-01-25 17:20:03 UTC 
Here is a patch to port to clutter gst 3.0. This new version adds
support for color correction using shaders. It would even be possible
to write shaders to effects too.

Sadly this code with a tiny api break for libcheese. Let me know if I
can help to port the apps using it.
Comment 1 Lionel Landwerlin 2015-01-25 17:20:05 UTC 
Created attachment 295386 [details] [review]
libcheese: port to clutter-gst 3.0
Comment 2 Lionel Landwerlin 2015-01-27 12:25:55 UTC 
Created attachment 295527 [details] [review]
libcheese: port to clutter-gst 3.0
Comment 3 Lionel Landwerlin 2015-01-27 12:27:07 UTC 
Sorry, I couldn't test this at home, because my webcam doesn't have a linux driver.
Fixed the remaining issues.
Comment 4 Emmanuele Bassi (:ebassi) 2015-03-27 12:45:42 UTC 
Review of attachment 295527 [details] [review]:

::: libcheese/cheese-camera.c
@@ +1043,3 @@
+
+  g_signal_connect (G_OBJECT (clutter_actor_get_content (texture)),
+                                         "sink", display_element,

If the content has a preferred size, then you can use CLUTTER_REQUEST_CONTEXT_SIZE from Clutter 1.22, instead of using a signal and a fixed size.

@@ +1621,3 @@
+                                         NULL),
+                NULL);
+  g_signal_connect (G_OBJECT (clutter_actor_get_content (priv->video_texture)),

Same as above.
Comment 5 Lionel Landwerlin 2015-03-27 15:33:46 UTC 
That won't work, because the size of the content is only known much later, when the first images are extracted from the webcam.
ClutterContent cannot emit a queue_layout on their actor.
Comment 6 Lionel Landwerlin 2015-08-03 22:54:30 UTC 
Ping?
Comment 7 David King 2015-08-17 16:41:11 UTC 
Review of attachment 295527 [details] [review]:

Pushed to master as commit 9df993253d58c2442ab92df613f5754ead5b5e20, with minor fixes for the .pc file.