summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2004-07-03hmmm, don't ask me why it worked on some machines...Thibaut Mattern
CVS patchset: 6767 CVS date: 2004/07/03 19:22:16
2004-07-03Fix accumulation buffer management.Thibaut Mattern
CVS patchset: 6766 CVS date: 2004/07/03 16:47:55
2004-07-02Fix compilation with gcc 2.95.Thibaut Mattern
CVS patchset: 6763 CVS date: 2004/07/02 17:44:28
2004-07-01Use avcodec_align_dimensions for querying required frame dimension alignmentsJames Stembridge
CVS patchset: 6762 CVS date: 2004/07/01 20:56:15
2004-07-01Fix av sync for RV10 and RV20 streamsJames Stembridge
CVS patchset: 6761 CVS date: 2004/07/01 20:16:24
2004-06-29Mac OS X XineVideoWindow: move setKeepsAspectRatio: and keepsAspectRatioAndre Pang
methods to NSWindow class, since they're generic to all windows CVS patchset: 6760 CVS date: 2004/06/29 05:25:28
2004-06-28Fix a stupid bug causing a segfault with huffyuv_sample.avi test stream.Thibaut Mattern
Let's see tomorrow how many segfaults are gone ;-) CVS patchset: 6759 CVS date: 2004/06/28 22:44:57
2004-06-27automake madness...Michael Roitzsch
CVS patchset: 6758 CVS date: 2004/06/27 17:22:13
2004-06-27only issue the warning on unavailable audio devices under the conditionMichael Roitzsch
(in_buf && in_buf->num_frames), otherwise, the dummy packet send by ao_exit() to wake up the audio out loop can trigger this message (because it has an undefined buf->format.rate, which, when 0 causes driver->open() to fail and the message gets emitted) we also send the message to the streams attached to this audio out only, not to all streams known to the engine CVS patchset: 6757 CVS date: 2004/06/27 13:33:57
2004-06-27include the new goom version 2k4-dev15Michael Roitzsch
everybody: test, test, test CVS patchset: 6756 CVS date: 2004/06/27 11:58:47
2004-06-27calling XInitThreads() is definitely the frontend's job, because it hasMichael Roitzsch
to call Xlib functions before xine to properly fill x11_visual_t CVS patchset: 6755 CVS date: 2004/06/27 11:16:20
2004-06-26Mac OS X video_window.m: Change #include to #importAndre Pang
CVS patchset: 6753 CVS date: 2004/06/26 15:10:10
2004-06-26Mac OS X video_window.h: #import <Cocoa/Cocoa.h> so the header file canAndre Pang
compile :) CVS patchset: 6752 CVS date: 2004/06/26 15:09:35
2004-06-26patch from Stefan Kristiansson (slightly modified) to move Xv initializationMichael Roitzsch
from class to plugin context CVS patchset: 6750 CVS date: 2004/06/26 14:47:34
2004-06-26mark the old deinterlace setting as deprecatedMichael Roitzsch
CVS patchset: 6749 CVS date: 2004/06/26 13:52:20
2004-06-26do not assume YV12 and YUY2 are the only image formatsMichael Roitzsch
CVS patchset: 6748 CVS date: 2004/06/26 13:51:13
2004-06-26remove statement with no effect (state->cmd_ptr is already set to next_seq)Michael Roitzsch
CVS patchset: 6747 CVS date: 2004/06/26 13:49:54
2004-06-25- removed a printf() which causes the cause of the printf() to occur moreDaniel Mack
often which then again causes the printf() again... - destroy the mutex in ..._close(); CVS patchset: 6746 CVS date: 2004/06/25 16:19:08
2004-06-25- trying to fix the kernel panics by wrapping some functions with a mutex.Daniel Mack
maybe that helps. - some indentation fixes CVS patchset: 6745 CVS date: 2004/06/25 16:14:20
2004-06-25exit pause loop, when video loop is terminatedMichael Roitzsch
(fixed latest deadlock report by Stefan Holst) CVS patchset: 6744 CVS date: 2004/06/25 16:09:31
2004-06-25beautify last commitMichael Roitzsch
CVS patchset: 6743 CVS date: 2004/06/25 16:04:59
2004-06-25have CVS compile againMichael Roitzsch
CVS patchset: 6742 CVS date: 2004/06/25 15:52:21
2004-06-25some cleanups, still hunting the evil kernel-killerDaniel Mack
CVS patchset: 6741 CVS date: 2004/06/25 11:14:27
2004-06-25Move IA32 specific code out of the generic code into the ISA-specific ↵Robin KAY
arch_accel(). On Solaris/SPARC, use sysinfo(2) to detect supported extensions rather than by trapping SIGILL. CVS patchset: 6740 CVS date: 2004/06/25 07:32:22
2004-06-24this->buffers must be signed to hold a negative value on vram_alloc() ↵Robin KAY
failure. Must not set procs to NULL directly as this could occur in the window between the video out loop testing and calling them. Cleanup the code for handling insufficient video memory. CVS patchset: 6739 CVS date: 2004/06/24 13:06:57
2004-06-23* fix the realloc-bug introduced by me and found by FrantisekMichael Roitzsch
* since plugin catalog assumes a limit on plugin count, we should actually enforce this limit on plugin loading * reduce DECODER_MAX from 256 to 128, because PLUGIN_MAX is already 256 and having only decoder plugins is unlikely CVS patchset: 6738 CVS date: 2004/06/23 19:45:52
2004-06-23src/video_out/macosx/video_window.m should export XineViewDidResizeNotificationAndre Pang
as a global symbol, since end-applications which use xine-lib need to access this data. (e.g. grep for 'Notification' in /System/Library/Frameworks/WebKit.framework/Headers/ to see how this works). CVS patchset: 6737 CVS date: 2004/06/23 02:25:42
2004-06-23bugfix: changed call to seek from old to new behaviourAndreas Heinchen
CVS patchset: 6736 CVS date: 2004/06/23 00:05:38
2004-06-22moved a check to *maybe* get rid of a kernel panic caused by drawRectDaniel Mack
CVS patchset: 6735 CVS date: 2004/06/22 22:03:11
2004-06-22higher output gainDaniel Mack
CVS patchset: 6734 CVS date: 2004/06/22 22:01:25
2004-06-22Compiler warning fixesAndre Pang
CVS patchset: 6733 CVS date: 2004/06/22 16:08:07
2004-06-22Mac OS X XineOpenGLView:Andre Pang
Fix dealloc behaviour Set fields to nil after releasing, to avoid stray messages CVS patchset: 6732 CVS date: 2004/06/22 16:06:38
2004-06-22File input module now groks file://localhost/... and file://127.0.0.1/... MRLsAndre Pang
CVS patchset: 6731 CVS date: 2004/06/22 14:31:40
2004-06-22Mac OS X XineOpenGLView:Andre Pang
Added dealloc method Removed unused opengl_context field Added NSCoding functions, in preparation for Interface Builder palettization Mac OS X XineVideoWindow: Added dealloc method CVS patchset: 6729 CVS date: 2004/06/22 08:25:24
2004-06-21remember the last opened device for ejecting so when using more than oneMichael Roitzsch
DVD drive, the correct one is ejected; should close bug #976347; also done some cleanup, removed currently unused structure members CVS patchset: 6728 CVS date: 2004/06/21 18:48:17
2004-06-21* add one file missing in release (not critical, since unneeded)Michael Roitzsch
* remove two unneeded files CVS patchset: 6727 CVS date: 2004/06/21 16:37:00
2004-06-21removing unused SPU_DVD_SUBTYPE_SUBP_CONTROL and the stream_filter stuffMichael Roitzsch
in SPU decoders which is obsoleted by that removal CVS patchset: 6726 CVS date: 2004/06/21 16:19:40
2004-06-21CoreAudio output: Should call CloseComponent() on the component instances, notAndre Pang
the component itself CVS patchset: 6725 CVS date: 2004/06/21 16:13:10
2004-06-21CoreAudio output driver: build fix for libtool 1.5Andre Pang
CVS patchset: 6724 CVS date: 2004/06/21 15:09:53
2004-06-21CoreAudio output driver: renamed 'CloseAComponent' function to 'CloseComponent'Andre Pang
CVS patchset: 6723 CVS date: 2004/06/21 15:09:24
2004-06-21Mac OS X XineVideoWindow:Andre Pang
Added #ifdef guards around video_window.h Removed setContentSize: method from XineVideoWindow interface, since this method is already present in NSWindow Changed setKeepsAspectRatio: & keepAspectRatio's int arguments to BOOL Added initWithContentSize: method Added init method CVS patchset: 6722 CVS date: 2004/06/21 15:07:00
2004-06-21Mac OS X XineVideoWindow:Andre Pang
Replaced fitToScreen method with windowWillUseStandardFrame:defaultFrame:, since NSWindow's native zoom: method uses that to determine the zoomed side Made new windowWillUseStandardFrame:defaultFrame: method center the screen CVS patchset: 6719 CVS date: 2004/06/21 03:24:51
2004-06-20* merge latest CVS of libdvdnav fixes dvd:/<title>.<part> MRLsMichael Roitzsch
* small cleanup in input_dvd, removing unused structure member 'mode' * remove CVS log from input_dvd since this is uncommon in xine-lib and enlarges the file too much; people can use "cvs log" if they want the info CVS patchset: 6716 CVS date: 2004/06/20 16:59:13
2004-06-20this is supposedly a booleanMichael Roitzsch
CVS patchset: 6715 CVS date: 2004/06/20 15:04:19
2004-06-20housekeepingMichael Roitzsch
CVS patchset: 6714 CVS date: 2004/06/20 10:10:01
2004-06-19modified patch from Reinhard Nissl to report the length of the out fifosMichael Roitzsch
CVS patchset: 6713 CVS date: 2004/06/19 20:07:15
2004-06-19rename part of config entry keys from "plugin" to "subsection", because forMichael Roitzsch
the users it is better if we structure by intuitive categories, not strictly by plugins CVS patchset: 6712 CVS date: 2004/06/19 19:54:59
2004-06-19this comment is outdated since decoders provide the discontinuity() functionMichael Roitzsch
CVS patchset: 6711 CVS date: 2004/06/19 19:53:29
2004-06-19force slave metronoms to reinit when a new master is attached; otherwise aMichael Roitzsch
visualization plugin's metronom gets confused when you wire different streams to the plugin's input CVS patchset: 6710 CVS date: 2004/06/19 19:51:25
2004-06-19readd the decoder's name to the description of the config entry, but doMichael Roitzsch
it right this time: use a different pointer for every decoder by putting the descriptions into a dynamically growing chunk of heap memory the name of the decoder in the description is necessary, because frontends should not have to display the config entry key and this is currently the only string to contain the decoder's name CVS patchset: 6709 CVS date: 2004/06/19 19:48:42