- 27 Jun, 2012 18 commits
-
-
Felix Paul Kühne authored
-
Erwan Tulou authored
-
Erwan Tulou authored
And better performance since we were still using the rest of a polling mechanism (every 100ms) just to ensure we were alive.
-
Erwan Tulou authored
interaction_Register and interaction_Unregister are weird #define that don't seem used at all nowadays
-
erwan authored
-
Erwan Tulou authored
-
Erwan Tulou authored
Use the mouse-hide-timeout parameter for that purpose, as the other interfaces do in vlc. this fixes #7017
-
Erwan Tulou authored
-
Rémi Denis-Courmont authored
That horrible hack caused all subtitles to be parsed as Unicode if one (but not necessarily all) opened subtitles started with a UTF-8 or UTF-16 Byte Order Mark. If any other subtitle was neither in UTF-8 nor in UTF-16 with a BOM, that hack failed.
-
Rémi Denis-Courmont authored
This should not be needed (BOM is a zero-width character). If any stream reader barks on a BOM, it really needs fixing on its own. The subtitle demuxer was fixed in the previous commit. In the case of transparent conversion from UTF-16 to UTF-8, the BOM will be converted to UTF-8 as any other character. Thus the stream reader can (correctly) detect a UTF-8 text stream.
-
Rémi Denis-Courmont authored
If an UTF-8 BOM is found at the beginning of the text file, skip it (it can confuse some parsers). Also mark the subtitle track explicitly as UTF-8 encoded.
-
Felix Paul Kühne authored
-
Damien Erambert authored
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>
-
Damien Erambert authored
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>
-
Damien Erambert authored
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>
-
Jean-Baptiste Kempf authored
10L
-
Jean-Baptiste Kempf authored
-
Jean-Baptiste Kempf authored
-
- 26 Jun, 2012 5 commits
-
-
Jean-Baptiste Kempf authored
-
Jean-Baptiste Kempf authored
-
David Fuhrmann authored
Now, the sidebar will not will not resized while resizing the window. Also, this commit sets a min width for the the sidebar. But it can still be hidden, just drag the slider to the left.
-
Francois Cartegnie authored
-
Francois Cartegnie authored
When calling a modal dialog, Qt switches to a new event loop. The call may still have some event to process in its own queue, and deleting it when doing the Vout update will have those events refering to a no longer existing widget.
-
- 25 Jun, 2012 9 commits
-
-
Rafaël Carré authored
Since 42a7f46d it is required to re-run gettextize because we run autoreconf. Unfortunately gettextize doesn't run in non interactive mode (this is an advertised "feature"). This gpg-error patch will appear in the next release so it doesn't need maintenance: http://git.gnupg.org/cgi-bin/gitweb.cgi?p=libgpg-error.git;a=commit;h=c35dbfdaa2e30e08185e4a468278a29e2c48b500
-
Rafaël Carré authored
Tested on HP Touchpad up to 1280x720 1920x1080 gives: E/OMX-VDEC-1080P: ERROR: IP Requirements(#1: 2097152) Requested(#2: 4177920) Which probably means that provided buffers are too small for the decoder to operate
-
Martin Storsjö authored
This is required on HTC Sensation. That device uses a special tiled pixel format, which probably contains a bit of padding, too. The size that the omx codec suggests for the output port is bigger than what VLC calculates, and we thus should keep the larger value instead of overwriting it with what we calculated ourselves. Signed-off-by: Rafaël Carré <funman@videolan.org>
-
Felix Paul Kühne authored
-
David Fuhrmann authored
-
David Fuhrmann authored
close #6926
-
Francois Cartegnie authored
As Qt >= 4.6.0 provides AbstractAnimation we no longer need QTimer based animators. PixmapAnimator allows to be used by Q*Painter. Unused still frame support is removed, but can easily be implemented.
-
Jean-Baptiste Kempf authored
-
David Fuhrmann authored
-
- 24 Jun, 2012 8 commits
-
-
Felix Paul Kühne authored
-
Felix Paul Kühne authored
macosx: fixed HiDPI fspanel background artwork, which accidently included a shadow which lead to drawing artifacts since the 2 variants didn't match their sizes
-
Felix Paul Kühne authored
no user-visible change.
-
Jean-Baptiste Kempf authored
-
Jean-Baptiste Kempf authored
-
Jean-Baptiste Kempf authored
-
Felix Paul Kühne authored
This is a broken concept, since according to the HIGL, one point always equals one pixel on non-HiDPI screens regardless of the actual DPIs or the actual screen resolution. Additionally, this interferes with our support for HiDPI
-
Damien Erambert authored
Signed-off-by: Felix Paul Kühne <fkuehne@videolan.org>
-