sort by:
Revision Author Date Message Commit Date
f1b9413 gstdecode: support alloc'ing vlc pictures with padding Allocate the output vlc pictures with dimensions padded, as requested by the decoder (for alignments). This further increases the chances of direct rendering. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 11 March 2016, 14:57 UTC
6c813cb video_output: decklink: fix build with format macros 11 March 2016, 09:00 UTC
797eb9b demux: adaptive: fix builds with format macros 11 March 2016, 09:00 UTC
d82758a sftp: public key authentication 11 March 2016, 08:32 UTC
9d955be Contribs: fix GNUTls store lookup for tizen 10 March 2016, 16:28 UTC
c8d93f1 GnuTLS: update to 3.3.22 This is a move from 3.2 to 3.3. EGD is reactivated 10 March 2016, 15:35 UTC
29c5c69 Revert "lib: Add libvlc_dialog.h to vlc.h" This reverts commit 1e0f167c78121a4c790bd92e762e8e306cf5075a. 10 March 2016, 14:37 UTC
b42a03f demux: ts: set filter only after pcr repicking 10 March 2016, 10:52 UTC
994b93c demux: ts: simplify CAM setting And comment 10 March 2016, 10:52 UTC
c021c52 libvlc: re-order dialog callback arguments The opaque p_data is now the first argument for the sake of consistency. 10 March 2016, 10:45 UTC
b37866a dialog: re-order callback arguments The opaque p_data is now the first argument for the sake of consistency. 10 March 2016, 10:45 UTC
86ca39d demux: ts: create missing dvb subs (refs #5267) 10 March 2016, 10:07 UTC
1e0f167 lib: Add libvlc_dialog.h to vlc.h 10 March 2016, 09:36 UTC
42d180b directdraw: more debug info for overlay issues Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 10 March 2016, 09:23 UTC
6e44b5d contrib/fontconfig: require at least 2.10.92 Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 10 March 2016, 09:23 UTC
569f928 demux: dash: add missing config.h 10 March 2016, 09:21 UTC
3925299 demux: dash: add missing config.h 10 March 2016, 09:17 UTC
518c88e avahi: remove domain from rtsp service type 10 March 2016, 08:49 UTC
42568e3 avahi: remove useless avahi_threaded_poll_stop() from error path Error path is never triggered after avahi_threaded_poll_start(). Even if it was, thread should probably be stopped before freeing avahi client ... 10 March 2016, 08:26 UTC
3cbc233 avahi: remove AvahiServiceBrowser from sys Only last created service browser was stored. avahi_client_free() should free also all service browsers. 10 March 2016, 08:16 UTC
d6b6f0f demux: smooth: fix include order 09 March 2016, 20:50 UTC
e87ac8f demux: hls: fix include order 09 March 2016, 20:50 UTC
e9c488a demux: dash: fix include order 09 March 2016, 20:50 UTC
f1a57d0 demux: adaptive: fix include order 09 March 2016, 20:50 UTC
f47aace libvlc: add missing include in umbrella header 09 March 2016, 18:03 UTC
8cf54bf contrib/live555: update to current version 09 March 2016, 17:56 UTC
8d119e0 mkv: fix build failure when compiled with llvm LLVM treats a typedef with forward-declaration inside a local struct as a forward-declaration for a nested type of that struct, effectively erroring on the implementation of the dispatchers. This patch fixes that issue by forcing the name to referr to a type in the sourrounding scope (by introducing a proper forward-declaration). Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 17:26 UTC
c7ccd77 Contribs: update libass to 0.13.2 09 March 2016, 17:14 UTC
02d5157 mkv: fixed C++03 restriction regarding IndexFinder It somehow slipped my mind that C++03 disallows local types as template-arguments, which makes one of the earlier commits invalid when compiled as C++03. This patch correctly moves the type in question to the global namespace (inside an anonymous namespace to not pollute the global linkage scope) so that everything is 100% legal C++03. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 12:32 UTC
b9341d8 Fixed vlm show to display next launch for schedules with endless repeats Currently vlm show does not show the next launch of endlessly repeated schedules after the first launch Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 12:00 UTC
e41757a mkv: minor clean-up + c-style casts => c++ casts Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
86e13b3 mkv: made MkvTree match legacy code in terms of indentation Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
166bbc1 mkv: use EbmlTypeDispatcher in matroska_segment_c::ParseChapterAtom Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
dba3e1d mkv: removed unnecessary variable and cleaned up expression Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
2e185a2 mkv: added debug-helper to matroska_segment_c::ParseInfo Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
f85f390 mkv: removed unnecessary dynamic allocation Replaced the dynamic allocation with an equivalent construction on the stack (we do not need dynamic life time since it is unconditionally destroyed at the end of the block). Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
fc393f4 mkv: replaced reallocation of EbmlParser with EbmlParser::reconstruct Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:58 UTC
6487b49 mkv: Added reconstruct function to EbmlParser The two overloads of EbmlParser::reconstruct has been added due to the fact that there are _a lot_ of places in the code following the below: delete ep; ep = new EbmlParser (a, b, c, b); The above will, unless the compiler feels cocky and optimizes it to what is included in this patch. First free the memory used, and then allocate new memory for a new EbmlParser. Instead of doing what is effectively a reallocation, this patch introduces EbmlParser::reconstruct which will reconstruct the object in-place (without actually having to reallocate the underlying storage). Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
1b41c8e mkv: tweaked diagnostic when seeking Outputting the percentage when we do not have any (-1.f) is of no use, this patch properly makes use diagnose what we actually have (either a fixed location, or the percentage). Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
4f65458 mkv: removed legacy code from matroska_segment.cpp The macro and the disabled code blocks are not required, as such they have been removed. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
d6f4169 mkv: replaced loop with (already present) helper BlockFindTrackIndex does what the explicit loop is doing; we should use it. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
cc84e98 mkv: use binary- instead of linear-search in Seek Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
0a7d6eb mkv: introduced helper functions related to indexes Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:57 UTC
6a735f6 mkv: Use StringDispatcher in matroska_segment_c::TrackInit Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:52 UTC
4a80d93 mkv: Added StringDispatcher Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:52 UTC
597b289 i420_yuy2: fix the line remaining offset for other YUY2 variants Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 11:45 UTC
befae51 yuy2_i420: only handle visible pixels Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 10:59 UTC
bcf4484 yuy2_i420: only adjust the line ending when the line was actually used -- must have been tested with samples where the shift was 0 this fixes http://streams.videolan.org/samples/V-codecs/V422.AVI playing on a I420 surface Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 10:59 UTC
351cd5a directdraw: in overlay mode we need to use DDSCAPS_FLIP and DDSCAPS_FRONTBUFFER We use DDSCAPS_BACKBUFFER to do overlay. According to the documentation we need these 2 flags on the front surface (the only one we have). https://msdn.microsoft.com/en-us/library/windows/desktop/gg426110%28v=vs.85%29.aspx#DDSCAPS_BACKBUFFER https://msdn.microsoft.com/en-us/library/windows/desktop/gg426110%28v=vs.85%29.aspx#DDSCAPS_FRONTBUFFER Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 09 March 2016, 10:55 UTC
88521ef url: test previous commit 09 March 2016, 05:04 UTC
99d97ef url: validate host name and path This avoids passing embedding unexpected garbage into network requests, leading to weird errors, or in the worst cases, injection (e.g. white spaces in HTTP/1.x requests). 09 March 2016, 05:03 UTC
736e7e7 i420_yuy2: only handle visible pixels Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 16:51 UTC
e29200c i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 16:51 UTC
ffcc062 mkv: use EbmlTypeDispatcher in matroska_segment_c::BlockGet 08 March 2016, 16:51 UTC
f5cdfd6 mkv: Use EbmlTypeDispatcher in `matroska_segment_c::ParseInfo` 08 March 2016, 16:51 UTC
2d8949d mkv: Cleaned up code to issue debug diagnostics - helper function to print diagnostics within MetaDataHandlers introduced - fixed so that `matroska_segment_c::ParseTracks` is using `MkvTree` for diagnostics 08 March 2016, 16:51 UTC
0503738 mkv: Use EbmTypeDispatcher in `matroska_segment_c::{ParseTrackEntry, ParseTracks}` Introduced `EbmlTypeDispatcher` to the functions in question, making them more readable (as well as faster). This commit is a very good example of how the new header `Ebml_dispatcher.hpp` allows for better and faster code. Also removed some legacy code that has been disabled due to, by matroska, deprecated types. 08 March 2016, 16:51 UTC
421416d mkv: Added EbmlTypeDispatcher The EbmlTypeDispatcher is a type used to "dispatch" an object of type EbmlElement* to an appropriate callback depending on the dynamic type of the referred to object. It can effectivelly replace the massive if-else branches found throughout the module by not only making the code easier to understand and maintain, but also by making it _a lot_ faster (benchmarks show a speed increase between 450 and 700% in terms of lookup). 08 March 2016, 16:51 UTC
155fbc6 mediacodec: fix crash when reporting a hw error 08 March 2016, 14:07 UTC
bfb5a9c i420_yuy2: fix using buffer areas not visible fixes #16686 Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 13:58 UTC
ca3e55c i420_yuy2: remove useless lines Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 13:58 UTC
b26c633 directdraw: list the preferred formats not supported in debug builds Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 10:35 UTC
c51d1a8 directdraw: better debugging of the pixel format used for surfaces Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 10:35 UTC
21d8776 bluray: fix copy-paste error Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 08 March 2016, 08:25 UTC
e3d09f9 Only add items in the root playlist to the recent elements 07 March 2016, 16:15 UTC
6abe049 video_output: better logging of filters needed Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 14:15 UTC
7cdabac Qt: fix potential crash in UpdateVout() 07 March 2016, 13:36 UTC
f4055f8 contrib: update libdsm 07 March 2016, 13:34 UTC
ca78be7 dsm: update with last API changes libdsm functions return now a negative number in case of error. Add NT_STATUS error check to detect permission errors in order to ask for credentials only in that case. 07 March 2016, 13:34 UTC
106a325 Add shortcut to set focus on playlist search field Signed-off-by: Guillaume Gomez <guillaume1.gomez@gmail.com> Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:12 UTC
fcedf52 ogg codecs: use checked realloc() instead of xrealloc() Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:07 UTC
37a9c77 oggspots: Add OggSpots codec module Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:05 UTC
98448d3 mkv: revert 8317c24270d45b4938acaf58104e18f5039eade0 This reverts commit 8317c24270d45b4938acaf58104e18f5039eade0. it introduced issues with CPU intensive files dropping a lot of frames Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:04 UTC
ce5d53a sftp: Factorize common cleanup code in Open() and Close() Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:02 UTC
2a7f738 sftp: fix browsing home directory ( ex. vlc sftp://127.0.0.1 ) Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:02 UTC
61ffa3f avahi: remove unused default port Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:02 UTC
32e98b0 avahi: add RTSP Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 07 March 2016, 13:02 UTC
b6a3dd5 demux: ts: add debuging for arib descriptors From untranslated TR-B14 06 March 2016, 22:36 UTC
24f3724 demux: ts: rename transport type field and add ignored one, fully blacklist dsm cc 06 March 2016, 21:33 UTC
a9aea03 demux: ts: show psip current program progress 06 March 2016, 21:04 UTC
7283809 demux: ts: remove pmt access control test Should already have been evaluated on pat 06 March 2016, 21:03 UTC
c7497d0 demux: ts: remove indirect eit callbacks 06 March 2016, 20:22 UTC
c91def4 epg: fix current replacement and add tests 06 March 2016, 20:13 UTC
f28e899 demux: ts: use fail proof realloc 06 March 2016, 20:13 UTC
7ce220a demux: ts: fix current event/running event TS 101 211, undefined status event should be handled as running. Should fix channels without real event status like France 5 06 March 2016, 20:13 UTC
8521672 demux: ts: simplify eit offset (arib) and add network time 06 March 2016, 20:13 UTC
b051a4b posix fs: define vlc_cloexec when mkostemp does not exist On Android, for 64 bits, O_CLOEXEC is defined, accept4 exists, but not mkostemp. Therefore, in vlc_mkstemp we go to the fallback that calls vlc_cloexec, but we don't define it, because we have accept4 Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:40 UTC
2e5cbf3 ogg: Add OggSpots video codec support Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:39 UTC
c7ff5cc mkv: fixed seek request diagnostic The diagnostic's intention was to print where we are going to seek in procent, though the actual output was previously in the range [0.0, 1.0]. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:21 UTC
d3f83dc mkv: Reimplemented MkvTree and moved it to `Util.hpp` `MkvTree` can be used all across the module, meaning that it better belongs in `Util.hpp` than with static linkage in `matroska_segment_parse.cpp`. - A helper function `MkvTree_va` has been introduced so that one can easily call the function from functions that accepts a variadic number of arguments (using va_list). - The implementation has been optimized so that we do not need to allocate storage for the format string on every invocation if it would fit inside a fixed length buffer. - An error check has been added to signal if we run out of memory. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:21 UTC
5fc6cd6 mkv: removed unnecessary memory-allocation in `matroska_segment_parse.cpp` Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:21 UTC
7d32c88 mkv: Added error diagnostic if i_track does not match seekpoint Added an if-check to make sure that we do not crash on the lines that follow if we, for some reason, end up in a state where there is no matching seekpoint for the specified `i_track`. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:21 UTC
fce2945 mkv: replaced (manual) linked-list with std::vector in `matroska_segment_c::Seek` Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:20 UTC
2eca52d mkv: removed manual memory management from Tags + SimpleTags In order to provide better runtime safety manual management of dynamic memory has been removed from `class Tags` and `class SimpleTags`. This includes introducing usage of `std::vector` as well as `std::string`, while also removing data-members that are no longer required. - SimpleTag::b_default has been removed since it was written to (at once place), but never read; the write has been replaced by a comment. - The ParseSimpleTag function has been changed to return `bool` instead of a pointer to signal whether parsing was successful. The result will be written to `pout_simple` (instead of being indirectly returned through a pointer). Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:20 UTC
e527210 mkv: removed unnecessary dynamic memory allocations As written in the commit title, a few uses of dynamic memory management has been removed, effectivelly fixing a memory leak as well as making the code cleaner/faster. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:20 UTC
19c66b8 mkv: `MKV_CHECKED_PTR_DECL` introduced to `matroska_segment.cpp` See previous commit for information regarding the macro. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:20 UTC
d8b179e mkv: Introduced MKV_CHECKED_PTR_DECL `MKV_IS_ID` is a macro to check whether the object hiding behind a pointer has a certain dynamic type and it is being used all over the module. There are however a lot of cases that is followed by a static_cast to the appropriate type; `MKV_CHECKED_PTR_DECL` has been introduced to shorten code that has such usage. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:20 UTC
553febf mkv: replaced manual memory-management with std::vector in matroska_segment_c Manually managing memory comes at a cost of both maintainability (in terms of safety) and performance, as such I have replaced `p_indexes` with a `std::vector` with equivalent functionality. Three helper member-functions have been introduced in order to clean up the usage of the functionality, as well as removal of two now obsolete member-variables. A `typedef` has also been introduced to aid future development. The changes in `mkv.cpp` are due to the fact that it needs access to the indexes present in `matroska_segment_c`; this should be refactored away in the future. Also fixed a bug where you would access index out of bounds if there are no known indexes. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:15 UTC
881299b mkv: removed indirection when initializing std::string There really is no need to allocate memory for a temporary buffer, copy the data to this buffer, and then use the buffer to initialize a std::string, before freeing the buffer. One of the constructors of std::string can handle the above for us. Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:14 UTC
d55cd99 mkv: stream_io_callback.cpp c-style casts => c++ casts Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org> 05 March 2016, 21:14 UTC
back to top