Age | Commit message (Collapse) | Author |
|
I hope removing them does not break any compatibility.
CVS patchset: 3320
CVS date: 2002/11/20 18:38:08
|
|
* new public output interface xine_{audio,video}_port_t instead of
xine_{ao,vo}_driver_t, old names kept as aliases for compatibility
* modified the engine to allow multiple streams per output
* renaming of some internal structures according to public changes
* moving SCR out of per-stream-metronom into a global metronom_clock_t
residing in xine_t and therefore easily available to the output layer
* adapting all available plugins
(note to external projects: the compiler will help you a lot, if a plugin
compiles, it is adapted, because all changes add new parameters to some
functions)
* bump up all interface versions because of xine_t and xine_stream_t changes
CVS patchset: 3312
CVS date: 2002/11/20 11:57:38
|
|
CVS patchset: 3275
CVS date: 2002/11/17 12:14:14
|
|
CVS patchset: 3241
CVS date: 2002/11/11 21:07:54
|
|
CVS patchset: 3215
CVS date: 2002/11/09 23:22:32
|
|
CVS patchset: 3176
CVS date: 2002/11/03 23:11:21
|
|
CVS patchset: 3175
CVS date: 2002/11/03 23:07:14
|
|
Added xine_health_check_s struct.
CVS patchset: 3133
CVS date: 2002/11/02 02:55:46
|
|
basic changes:
- MRLs are a subset of URIs
- important new delimiter is #
- everything before the # will be passed to the input plugin
- after the # you can add several ;-separated stream parameters:
novideo - video will be ignored
noaudio - audio will be ignored
nospu - spus will be ignored
demux:... - specify the demuxer to use (e.g. demux:mpeg_block)
entry:value - assign any config entry a new value
CVS patchset: 3112
CVS date: 2002/10/31 16:58:12
|
|
CVS patchset: 3100
CVS date: 2002/10/30 13:29:31
|
|
- time to cleanup demuxer mess... no more pthread stuff on any demuxer.
of course i haven't tested all demuxers, but they at least compile.
(sorry Mike, we will need to move a few variables around to make most of
your demuxers work again - not big deal i think)
- make api more consistent by providing xine_open/xine_close
and xine_start/xine_stop.
CVS patchset: 3058
CVS date: 2002/10/28 03:24:42
|
|
parameter (not implemented yet), document new plugin loader in changelog
CVS patchset: 3029
CVS date: 2002/10/26 22:50:52
|
|
CVS patchset: 3009
CVS date: 2002/10/26 14:31:36
|
|
Changed xine_health_check to take the cdrom and dvd device links as arguments
CVS patchset: 3003
CVS date: 2002/10/26 03:52:07
|
|
CVS patchset: 3002
CVS date: 2002/10/26 03:50:37
|
|
CVS patchset: 2993
CVS date: 2002/10/24 22:54:27
|
|
CVS patchset: 2991
CVS date: 2002/10/24 19:37:26
|
|
CVS patchset: 2988
CVS date: 2002/10/24 16:41:05
|
|
CVS patchset: 2962
CVS date: 2002/10/23 11:44:30
|
|
Added myself to AUTHORS
configure.ac:
Removed a '-O' flag from DEBUG_CFLAGS when its initialized. The flag
is added later depending on architecture of CPU.
include/xine.h.in:
src/xine-utils/Makefile.am:
src/xine-utils/xine_check.h:
src/xine-utils/xine_check.c:
misc/xine-config.in:
Moved some of the system checks from xine-config.in to xine_check.c. Now
some of the checks are written in C instead of shell script. This is the
first beta/bleeding edge/kick the tires/it might break for you version.
It works for me (Intel P3). It will report to the stdout your kernel version,
MTRR (x86) architecture only (depends on kernel), symbolic link for CDROM is ok,
symbolic link for DVD is ok, and if environment variable "DISPLAY" is
set. It will later support telling if DMA is turned on for IDE based DVD
drives (Need help here - see below) and XVinfo.
DMA: <HELP REQUIRED>: DMA is successfully check but I cannot determine
that from the /dev/dvd symbolic link for those DVD drives that need to
use the ide-scsi module (e.g. My DVD/CDRW). I can verify DMA is set
if I hard code the link into the code (obviously not a prime choice).
So if any solutions are known please pass them along.
misc/Makefile.am:
misc/xine-config.in:
src/xine-engine/Makefile.am:
Removed xine_logo information from xine-lib. Please let me know if I
missed anyting.
CVS patchset: 2869
CVS date: 2002/10/20 02:17:36
|
|
but it was agreed it should be that way), events contain a timestamp now
CVS patchset: 2864
CVS date: 2002/10/19 21:23:52
|
|
CVS patchset: 2828
CVS date: 2002/10/14 15:46:48
|
|
- 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: 2728
CVS date: 2002/09/21 12:20:47
|
|
CVS patchset: 2727
CVS date: 2002/09/21 11:58:40
|
|
Replaced configure.in with configure.ac
Remove acconfig.h (Unnecessary with autoconf 2.52 or above)
Updated configure.ac to remove deprecated macros, upgraded
AC_DEFINE to include comments, merged in acconfig.h.
Combined libxine and libxineutils into one library. libxineutils
is created but not installed. Its statically linked into libxine.
Advanced libxine version from 0.2.0 to 1.0.0
Removed references for skins.
Plugins now installed to <libdir>/xine/plugins/$XINE_MAJOR.$XINE_MORE.
$XINE_SUB
Xine-config updated for new location of plugins
Sorted file lists in Makefile.am to make them more readable.
TODO: remove skins references from xine-config
CVS patchset: 2705
CVS date: 2002/09/19 06:01:18
|
|
CVS patchset: 2686
CVS date: 2002/09/18 11:14:27
|
|
CVS patchset: 2680
CVS date: 2002/09/18 00:51:33
|
|
CVS patchset: 2679
CVS date: 2002/09/17 11:54:55
|
|
CVS patchset: 2664
CVS date: 2002/09/14 19:04:07
|
|
implemented yet, so please move on, nothing to see here...
CVS patchset: 2660
CVS date: 2002/09/13 18:25:22
|
|
- make most char pointers const in public api
- simpler get_spu_lang / get_audio_lang signature
CVS patchset: 2651
CVS date: 2002/09/11 17:41:07
|
|
CVS patchset: 2637
CVS date: 2002/09/09 19:24:48
|
|
defines (benefit: eases set_property calls - see next commit)
CVS patchset: 2633
CVS date: 2002/09/09 13:55:18
|
|
api (benefit: no remapping is required between dvd's and xine's aspect
codes)
- implement new scaling semantics as intended by Matthias Hopf
CVS patchset: 2627
CVS date: 2002/09/08 16:20:09
|
|
CVS patchset: 2625
CVS date: 2002/09/07 20:09:28
|
|
fix some input plugins that would not copy the mrl on open
CVS patchset: 2623
CVS date: 2002/09/06 18:13:10
|
|
CVS patchset: 2622
CVS date: 2002/09/06 14:36:54
|
|
get_streaminfo instead, added missing parameters/properties noted by daniel
CVS patchset: 2620
CVS date: 2002/09/05 23:20:20
|
|
CVS patchset: 2617
CVS date: 2002/09/05 20:59:14
|
|
CVS patchset: 2614
CVS date: 2002/09/05 16:50:55
|
|
broken now ... matthias/miguel: please fix it :-)
CVS patchset: 2606
CVS date: 2002/09/04 23:31:05
|
|
- new helper module (vo_scale.c) with all common scaling code of video out
drivers. it should greatly simplify drivers development by removing the
boring frame/window calculation, aspect ratio, zooming etc.
obs: not completely tested, may contain bugs and break things. also only
xshm and xv have being simplified so far, others will follow shortly.
CVS patchset: 2455
CVS date: 2002/08/15 03:12:24
|
|
CVS patchset: 2431
CVS date: 2002/08/11 15:06:13
|
|
- removes w32codec syncshutdown hack
- vo_driver api changed to support drivers with persistent overlays. that
will break binary compatibility xine-lib/xine-ui (xine-ui needs to be
recompiled). in order to keep binary compatibility
overlay_begin/overlay_end must be added after redraw_needed.
- video_overlay cleanups
- fix a hard to trigger condition where redraw wouldn't happen (freeing
handles)
CVS patchset: 2427
CVS date: 2002/08/10 21:25:20
|
|
added backwards compatible functions
=> binary compatibility with former libxine0 0.9.11 and 0.9.12
Someone who uses nvtvd should definitely test this!
They should be renamed back when the new API is released.
CVS patchset: 2386
CVS date: 2002/08/03 17:33:06
|
|
CVS patchset: 2308
CVS date: 2002/07/17 21:23:57
|
|
CVS patchset: 2235
CVS date: 2002/07/09 12:45:17
|
|
CVS patchset: 2140
CVS date: 2002/06/23 19:34:36
|
|
CVS patchset: 2138
CVS date: 2002/06/23 16:18:44
|