Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | videomixer: Remove unused link/unlink pad methods | Edward Hervey | 2009-07-10 | 1 | -15/+0 | |
| | ||||||
* | videomixer: I420 mode: Add fast path for 0.0 and 1.0 alpha | Edward Hervey | 2009-07-10 | 1 | -0/+18 | |
| | | | | | If the source alpha is 0.0, we take nothing. If the source alpha is 1.0, we overwrite everything. | |||||
* | videomixer: I420 blending : Fix main algorithm. | Edward Hervey | 2009-07-10 | 1 | -1/+1 | |
| | | | | | | | When blending a source layer with an alpha of 'a' on top of another destination layer we take the sum of: * 'a' percent of the source layer * (100 - 'a') percent of the destination layer (the remainder) | |||||
* | videomixer: Make debugging category global to all the code. | Edward Hervey | 2009-07-10 | 4 | -2/+7 | |
| | ||||||
* | videomixer: improve readability of debugging statements. | Edward Hervey | 2009-07-10 | 1 | -11/+11 | |
| | ||||||
* | rtspsrc: do not leak timeout message | Mark Nauwelaerts | 2009-07-09 | 1 | -3/+3 | |
| | ||||||
* | avi: Don't forward NEWSEGMENT events from upstream | Sebastian Dröge | 2009-07-09 | 1 | -0/+27 | |
| | | | | | | | New ones are generated later and simply forwarding them can result in NEWSEGMENT events of different format going downstream. Fixes bug #587983. | |||||
* | videomixer: Make checker pattern lookup table constant | Sebastian Dröge | 2009-07-08 | 2 | -2/+2 | |
| | ||||||
* | videomixer: Add support for ARGB | Sebastian Dröge | 2009-07-08 | 3 | -143/+156 | |
| | | | | And clean up the caps parsing. | |||||
* | udp: Initialize pointer to NULL | Benjamin Gaignard | 2009-07-08 | 1 | -1/+1 | |
| | | | | | | | Otherwise we're calling free() with some random memory address in error cases. Fixes bug #587982. | |||||
* | qtdemux: sprinkle some more const | Mark Nauwelaerts | 2009-07-08 | 1 | -13/+13 | |
| | ||||||
* | qtdemux: perform some more (careful) data buffering | Mark Nauwelaerts | 2009-07-08 | 1 | -14/+36 | |
| | | | | | | | Once buffering has started (with an mdat atom), continue buffering until moov atom is reached, which handles cases with multiple mdat atoms. Also keep adapter/offset better in sync with upstream and fix some debug statements. Fixes #587426. | |||||
* | avidemux: Replace deprecated GST_DISABLE_DEBUG with correct macro. Fixes #587826 | Philip Jägenstedt | 2009-07-06 | 1 | -7/+7 | |
| | ||||||
* | qtdemux: error out instead of dividing by 0 | Tim-Philipp Müller | 2009-07-01 | 1 | -1/+4 | |
| | | | | Error out if timescale is 0. | |||||
* | Revert "qtdemux: Make sure we don't blacklist streams by wrongly comparing ↵ | Tim-Philipp Müller | 2009-07-01 | 1 | -3/+3 | |
| | | | | | | | | | their" This reverts commit 5503a59a5779b67451d8a271000181790ee76bc7. Reverting this since it causes regressions with a lot of sample files I have, all of which worked fine with the last -good release (#586891). | |||||
* | qtdemux: comment out unused structure | Tim-Philipp Müller | 2009-07-01 | 1 | -3/+3 | |
| | ||||||
* | qtdemux: more size checks, and use g_try_new0() instead of g_new0() | Tim-Philipp Müller | 2009-07-01 | 1 | -3/+21 | |
| | | | | | | | Whenever we alloc something based on a user-supplied size, we should really use g_try_new(), otherwise we can easily be made to abort by passing a ridiculously large number to us for allocing. Fixes problems with some fuzzed files. | |||||
* | qtdemux: guard against bogus atom sizes and short reads | Tim-Philipp Müller | 2009-07-01 | 1 | -8/+49 | |
| | | | | | | | | | | Check the possibly 64-bit atom size more carefully before casting it to an int and passing it to gst_pad_pull_range(), otherwise we might end up pulling 0 bytes, getting an empty buffer as requested and dereferencing not available data whilst thinking we actually asked for and got 0x1000000000000 bytes. Similar fix for push mode operation where neededbytes ends up being 0 bytes, which makes us assert. Fixes crash with broken or fuzzed file (NB #122378). | |||||
* | qtdemux: use 0x prefix when logging numbers in hex | Tim-Philipp Müller | 2009-07-01 | 1 | -2/+2 | |
| | ||||||
* | flacdec: Don't send empty string tags | Edward Hervey | 2009-07-01 | 1 | -1/+2 | |
| | ||||||
* | Don't use sendmsg()-dependent code on Windows | LRN | 2009-06-30 | 1 | -0/+6 | |
| | | | | Fixes #585842 | |||||
* | law: fix caps and negotiation | Wim Taymans | 2009-06-30 | 6 | -129/+170 | |
| | | | | | | Fix the caps to include the depth (instead of width twice) in the caps of audio/x-raw-int. Fix negotiation to not only copy the rate/channels of the first structure. | |||||
* | pulsesink: include "1.0=100%" in volume and change upper limit | Stefan Kost | 2009-06-30 | 1 | -1/+7 | |
| | | | | | | Upper volume limmit was 1000. That appear unneceasrily high. It would also cause sever distortion if accidentialy used. Now its 10 (~ +15db) which is also in sync with volume and playbin2. | |||||
* | pulse: some more trivial cleanups | Wim Taymans | 2009-06-29 | 1 | -20/+19 | |
| | ||||||
* | pulse: trivial cleanups | Wim Taymans | 2009-06-29 | 1 | -13/+2 | |
| | ||||||
* | pulsesink: clear ringbuffer when asked to | Wim Taymans | 2009-06-29 | 1 | -0/+23 | |
| | | | | | | Since we map the ringbuffer to the pulseaudio internal ringbuffer, flush the pulseaudio buffer when we are asked to clear the ringbuffer. This avoids some leftover audio after a seek. | |||||
* | autogen.sh: Actually do the 'echo -n' -> printf change. | Jan Schmidt | 2009-06-26 | 1 | -1/+1 | |
| | ||||||
* | autogen.sh: Use printf instead of 'echo -n'. Check for automake-1.1[01] | Jan Schmidt | 2009-06-26 | 1 | -1/+1 | |
| | | | | | Check for more automake command variants. Use printf instead of 'echo -n' for portability | |||||
* | Automatic update of common submodule | Jan Schmidt | 2009-06-26 | 1 | -0/+0 | |
| | | | | From f810030 to 5845b63 | |||||
* | qtdemux: don't process track_num/track_count tags with a 0 value | Tim-Philipp Müller | 2009-06-26 | 1 | -3/+10 | |
| | | | | | Number/count values of 0 mean they're not set. Don't put those in the taglist. | |||||
* | waveformsink: use 'guint8' instead of 'byte' to fix compilation with MSVC8 | Tim-Philipp Müller | 2009-06-25 | 1 | -1/+1 | |
| | | | | | | We need a cast here for pointer arithmetic to work correctly, but some MSVC versions don't seem to like 'byte', so use guint8 here. Hopefully fixes #585361. | |||||
* | v4l2src: set structs to zero before using them in ioctls | Stefan Kost | 2009-06-25 | 1 | -5/+5 | |
| | | | | This fixes valgrind warnings. | |||||
* | qtdemux: Make sure we don't blacklist streams by wrongly comparing their | Julien Moutte | 2009-06-25 | 1 | -3/+3 | |
| | | | | duration with entire clip duration. | |||||
* | rtpdec: fix some buffer leaks | Krzysztof BÅ‚aszkowski | 2009-06-25 | 1 | -0/+3 | |
| | ||||||
* | flvparse: Add missing break in switch/case. | Edward Hervey | 2009-06-25 | 1 | -0/+1 | |
| | ||||||
* | flvdemux: Remove unused variable, hint branch likeliness, add comments. | Edward Hervey | 2009-06-25 | 1 | -8/+8 | |
| | ||||||
* | avidemux: Removed unused variable | Edward Hervey | 2009-06-25 | 1 | -2/+0 | |
| | ||||||
* | qtdemux: Remove dead assignments and unused variables. | Edward Hervey | 2009-06-25 | 1 | -9/+3 | |
| | | | | Also add branch likeliness macros. | |||||
* | qtdemux: Fix uninitialized variables. Fixes build on macosx | Edward Hervey | 2009-06-25 | 1 | -5/+5 | |
| | ||||||
* | souphttpsrc: free memory in finalize | Stefan Kost | 2009-06-24 | 1 | -12/+6 | |
| | | | | | finalize is called only once. no need to clear pointers there. dispose is for unreffing. | |||||
* | Automatic update of common submodule | Jan Schmidt | 2009-06-24 | 1 | -0/+0 | |
| | | | | From 6ab11d1 to f810030 | |||||
* | avidemux: short-circuit gst_avi_demux_src_convert() when parsing the index | Tim-Philipp Müller | 2009-06-24 | 1 | -37/+51 | |
| | | | | | | | | Don't call gst_avi_demux_src_convert() for each single index entry. Not only do we already have the pointer to the stream context, we also know the formats we want to convert from and to already, so we may just as well use optimised conversion routines that bypass some of the checks and lookups made in gst_avi_demux_src_convert(). | |||||
* | qtdemux: Another round of G_*LIKELY micro-optimisations. | Edward Hervey | 2009-06-24 | 1 | -26/+28 | |
| | ||||||
* | qtdemux: Take last sample duration for dummy segment calculation. | Edward Hervey | 2009-06-24 | 1 | -2/+4 | |
| | | | | | This fixes the cases where files without EDL wouldn't output their last buffer. | |||||
* | avidemux: Sprinkle branch likeliness macros over the code. | Edward Hervey | 2009-06-24 | 1 | -26/+29 | |
| | ||||||
* | raw1394: sprinkle branch likeliness macros accross the code. | Edward Hervey | 2009-06-24 | 2 | -7/+10 | |
| | ||||||
* | qtdemux: Add GST_MEMDUMP statements for unknown atoms. | Edward Hervey | 2009-06-24 | 1 | -2/+13 | |
| | | | | This is to help developers track down and implement unhandled atoms faster. | |||||
* | deinterlace: Remove the interlaced field from the output caps if ↵ | Sebastian Dröge | 2009-06-23 | 1 | -0/+8 | |
| | | | | deinterlacing is enabled | |||||
* | deinterlace: Copy the correct line from correct place in the history | Sebastian Dröge | 2009-06-23 | 1 | -1/+2 | |
| | ||||||
* | rtspsrc: use same protocols after redirect | Wim Taymans | 2009-06-23 | 1 | -0/+11 | |
| | | | | | After a redirect we want to use the same protocols that we were using for the current url. |