Age | Commit message (Collapse) | Author |
|
temporal position reporting
CVS patchset: 2766
CVS date: 2002/10/01 03:52:53
|
|
libraries:
lib_LTLIBRARIES = \
libA.la \
libC.la \
libD.la
# libB.la
Do NOT do:
lib_LTLIBRARIES = \
libA.la \
# libB.la \
libC.la \
libD.la
Commenting out libraries in this way automake appears to get confused. It
does not look at this like a compiler does. It will notice that this is
a comment but it seems to process the '\'. So it will complain about the
Makefile.am variables for libC.la and libD.la (e.g. libC_la_SOURCES) as
being unused.
CVS patchset: 2765
CVS date: 2002/10/01 03:21:19
|
|
CVS patchset: 2764
CVS date: 2002/09/30 14:48:32
|
|
CVS patchset: 2763
CVS date: 2002/09/30 14:46:09
|
|
The fixes involved getting blend_rgb16 to correctly handle different colours in the
highlight area and general overlay area.
Another fix involved fixing the overlay scaling function so that it scales in the same
proportion as the yuv2rgb code.
Currently, blend_yuv has not been updated.
blend_rgb16 has been improved.
blend_rgb24 and blend_rgb32 have not been changed and will probably not work now.
The performance improvements are due to two things: -
1) Special detection of overlays without any buttons/clip areas.
This is detected by ensuring overl->clip_top = -1 when no button is present.
2) Reorganise the blending code so that for each time round the loop, the fewest
possible IF statements are executed.
So, for overlays without any buttons, not checks for the clip area is done.
For overlay with buttons, state is kept as to whether we are above, to the left, inside
or to the right or below the clip area, thus saving having to do IF statements each time
round the loop to find out where we are.
CVS patchset: 2762
CVS date: 2002/09/30 05:16:45
|
|
reporting, and modified demuxer to be able to seek on start
CVS patchset: 2761
CVS date: 2002/09/30 01:16:15
|
|
Removed creation of acinclude.m4 from cvscompile.sh. aclocal allows for multiple declarations
of the -I tag. So far for me this has worked. If there are problems we can back out this
change.
Added "-I m4" to autogen.sh. This allows us to include the m4 files listed in the m4
directory. This is a more elegant choice than concatenating all the m4 files into
an acinclude.m4
CVS patchset: 2760
CVS date: 2002/09/30 01:00:27
|
|
CVS patchset: 2759
CVS date: 2002/09/29 01:01:58
|
|
"unpredictable segfaults"
CVS patchset: 2758
CVS date: 2002/09/28 23:00:22
|
|
cleaner and also clears up some audio pops at the start of a file
CVS patchset: 2757
CVS date: 2002/09/28 22:47:01
|
|
file format
CVS patchset: 2756
CVS date: 2002/09/28 22:06:31
|
|
certain files from crashing the demuxer
CVS patchset: 2755
CVS date: 2002/09/28 17:13:16
|
|
CVS patchset: 2754
CVS date: 2002/09/28 11:10:04
|
|
CVS patchset: 2753
CVS date: 2002/09/27 13:08:59
|
|
device
CVS patchset: 2752
CVS date: 2002/09/27 13:07:43
|
|
CVS patchset: 2751
CVS date: 2002/09/26 16:54:33
|
|
defined in
configure.ac". The same for CCASFLAGS.
CVS patchset: 2750
CVS date: 2002/09/26 14:04:46
|
|
CVS patchset: 2749
CVS date: 2002/09/26 13:35:26
|
|
CVS patchset: 2748
CVS date: 2002/09/25 13:48:32
|
|
CVS patchset: 2747
CVS date: 2002/09/25 12:47:49
|
|
by adding -I $(prefix)/share/aclocal to the XINE_ACFLAGS.
CVS patchset: 2746
CVS date: 2002/09/25 01:54:26
|
|
discontinuity
CVS patchset: 2745
CVS date: 2002/09/24 13:10:25
|
|
playback (which is infeasible due to the audio encoding)
CVS patchset: 2744
CVS date: 2002/09/24 03:09:09
|
|
handling
CVS patchset: 2743
CVS date: 2002/09/23 03:34:03
|
|
CVS patchset: 2742
CVS date: 2002/09/22 17:07:52
|
|
CVS patchset: 2741
CVS date: 2002/09/22 16:01:06
|
|
- bring our beloved xine_t * back (no more const there)
- remove const on some input plugin functions
where the data changes with media (dvd, ...) changes
and is therefore not const
CVS patchset: 2740
CVS date: 2002/09/22 14:29:40
|
|
CVS patchset: 2739
CVS date: 2002/09/22 14:25:11
|
|
CVS patchset: 2738
CVS date: 2002/09/22 14:24:30
|
|
CVS patchset: 2737
CVS date: 2002/09/22 04:27:00
|
|
CVS patchset: 2736
CVS date: 2002/09/22 04:23:24
|
|
CVS patchset: 2735
CVS date: 2002/09/21 20:27:02
|
|
CVS patchset: 2734
CVS date: 2002/09/21 19:47:11
|
|
CVS patchset: 2733
CVS date: 2002/09/21 19:18:37
|
|
CVS patchset: 2732
CVS date: 2002/09/21 18:48:25
|
|
disqualify MPEG files wrapped in QT-like headers
CVS patchset: 2731
CVS date: 2002/09/21 18:35:10
|
|
spec
CVS patchset: 2730
CVS date: 2002/09/21 18:18:45
|
|
CVS patchset: 2729
CVS date: 2002/09/21 17:22:21
|
|
CVS patchset: 2728
CVS date: 2002/09/21 12:20:47
|
|
CVS patchset: 2727
CVS date: 2002/09/21 11:58:40
|
|
CVS patchset: 2726
CVS date: 2002/09/21 11:39:45
|
|
CVS patchset: 2725
CVS date: 2002/09/21 07:46:45
|
|
CVS patchset: 2724
CVS date: 2002/09/20 22:34:26
|
|
CVS patchset: 2723
CVS date: 2002/09/20 15:53:24
|
|
- first attempt to solve the skin problem
CVS patchset: 2722
CVS date: 2002/09/20 15:52:21
|
|
called AS. Don't use standard AM_PROG_AS, but our modified version (!x86
builder, report failure please).
CVS patchset: 2721
CVS date: 2002/09/20 15:05:59
|
|
CVS patchset: 2720
CVS date: 2002/09/20 13:24:53
|
|
CVS patchset: 2719
CVS date: 2002/09/20 12:57:02
|
|
CVS patchset: 2718
CVS date: 2002/09/20 12:53:53
|
|
CVS patchset: 2717
CVS date: 2002/09/20 12:51:58
|