demuxer.c new_sh_sub_sid() tried to immediately select the created sub
track for playback if its id matched the "-sid" option value. This was
buggy, as more initialization is needed to properly enable subtitles.
Normally the correct track to play is selected after the demuxer has
been created. It's possible that some DVD use case or such depended on
the removed code to make -sid work with a subtitle track that's not
found at start and only added later (vobsubs probably would start
playing without separate initialization); if so then that needs to be
fixed later in a different way.
Add a framework for subtitle decoder modules that work more like
audio/video decoders do, and change libass rendering of demuxed
subtitles to use the new framework.
The old subtitle code is messy, with details specific to handling
particular subtitle types spread over high-level code. This should
make it easier to clean things up and fix some bugs/limitations.
Change new_demux_packet() and resize_demux_packet() length parameter
type from int to size_t and add a check to abort() if the size is over
1 GB. This should make integer overflow problems leading to memory
corruption in demuxers less likely; and aborting should be no worse
than insane memory consumption. Also make the functions abort() if the
actual allocation fails instead of trying to continue with a
zero-sized buffer.
Add a special option "-leak-report" that enables talloc leak
reporting. It only works if it's given as the first argument.
The code abuses the CONF_TYPE_PRINT option type to make main option
parsing ignore the option. The parser incorrectly consumed the
following commandline argument as a "parameter" for options of this
type when they had the flag to not exit after printing the message.
Fix this. It makes no difference for any previously existing option I
think.
The avsub implementation tries to fall back to MPlayer's other text
subtitle decoding if libavcodec returns text as the 'decoded'
subtitle. The code implementing this is buggy, and as far as I can see
it should not be triggered normally (libavcodec decoding is only
used for xvid, pgs and dvb subtitles, and for those libavcodec should
return bitmaps). Remove the buggy code (don't try to support
non-bitmap results) and simplify things a bit.
The contents of mpcommon.c were quite arbitrary; the most common
reason to place some functions in this file had been "MEncoder happens
to need similar code as MPlayer and we want to share some parts, but
we have no clue whatsoever how to organize things in a sensible way,
so we'll just dump those parts we want to share in mpcommon.c". As a
result of containing an essentially random subset of top-level player
functionality the mpcommon.h header required access to central structs
and was unsuitable for inclusion in lower-level code, but was
nonetheless included there for the mplayer_version symbol.
Move almost all contents from mpcommon.c to mplayer.c. mplayer.c is
already big and should perhaps be split further, but keeping a few
random functions in mpcommon.c would not be an improvement.
Allow negative values of the "deint" parameter and make them select
the same deinterlacing mode as the corresponding positive value, but
without enabling deinterlace on startup. This is useful for
configuration files. Also tweak the overall manpage description of
vo_vdpau a bit to make it sound less as if the VO would be for
hardware acceleration only.
PulseAudio could keep reporting high delay values after a reset of
playing audio. This broke playback after seeking in some cases. Add a
workaround that should make things more robust against such
misbehavior.
Trying to do a framestep while playing an audio-only file would play
the file until the end, then start the next file in paused state. Make
framestep state enter pause again immediately if there is no video.
Also reset framestep state when switching files.
* hr-seek:
input: add default keybindings Shift+[arrow] for small exact seeks
input: support bindings with modifier keys for X input
core: audio: make ogg missing audio timing workaround more complex
core: add support for precise non-keyframe-limited seeks
core: add struct for queued seek info
commands: add generic option -> property wrapper
options: add "choice" option type, use for -pts-association-mode
core: remove looping in update_video(), modify command handling a bit
core: seek: use accurate seek mode with audio-only files
core: avoid using sh_video->pts as "current pts"
libvo: register X11 connection fd in input event system
core: timing: add special handling of long frame intervals
core: move central play loop to a separate function
Conflicts:
DOCS/tech/slave.txt
Add support for binding commands to modifier+key combinations like
"Shift+Left" or "Ctrl+Alt+x", and support reading such combinations
from the output window of X VOs.
The recognized modifier names are Shift, Ctrl, Alt and Meta. Any
combination of those and then a non-modifier key name, separated by
'+', is accepted as a key name in input.conf. For non-special keys
that produce characters shift is ignored as a modifier. For example
"A" is handled as a key without modifiers even if you use shift to
write the capital letter; 'a' vs 'A' already distinguishes the
combinations with a normal keymap, and having separate 'a', 'Shift+A'
and 'A' (written with caps lock for example) would bring more
confusion than benefit.
Currently reading the modifier+key combinations is only supported in
the output window of those VOs that use x11_common.c event
handling. It's not possible to input the key combinations in other VOs
or in a terminal window.
After the addition of exact seeking the code to work around missing
audio timestamps with ogg/ogm needs improvement. Now it's normal to
need adjustment at stream start time 0 (seeking to a position after
start of video but before second keyframe) with any video format, and
for exact seeks with ogg it's now more important not to skip the
sync. Make the check to detect the problem case more precise to avoid
affecting most other formats, and try to decode a second of audio
(hoping to get timestamps for those packets) before giving up.
Add support for seeking to an arbitrary non-keyframe position by
decoding video starting from the previous keyframe. Whether to use
this functionality when seeking is controlled by the new option
-hr-seek and a new third argument to the "seek" command. The default
is to use it for absolute seeks (like chapter seeks) but not for
relative ones. Because there's currently no support for cutting
encoded audio some desync is expected if encoded audio passthrough is
used. Currently precise seeks always go to the first frame with
timestamp equal to or greater than the target position; there's no
support for "matching or earlier" backwards seeks at frame level.
To prepare for the addition of exact seek support, add a struct for
queued seek state and a helper function to update its state. It would
have been cumbersome to update additional state (showing whether the
seek is forced to be exact or non-exact) manually at every point that
handles seeks.
Code in get_metadata() allocated too small a buffer for the text it
wrote (noticed by Clément Bœsch). Make the code cleaner and more
robust by changing it to use talloc_asprintf(). Also make it always
return non-NULL and remove checks on caller side.
Make "-lavdopts threads=0" mean an autodetected number of threads, and
make that the default value of the option. Also increase the upper
limit of the option from 8 to 16. Add new file osdep/numcores.c which
tries to determine the number of cores available on the machine.
numcores.c is based (heavily modified) on public domain numcpus.c by
Philip Willoughby <pgw99@doc.ic.ac.uk>, downloaded from
http://csgsoft.doc.ic.ac.uk/numcpus/
Make seeks backward from a time before the first index entry go to the
first entry instead of failing completely. This change doesn't affect
behavior for most files, because seeks are clamped to 0 from below and
normally files have the first index entry at 0.
Add mp_property_generic_option(), a property function that can be used
for generic option-based properties that do not require any action
beyond manipulating the value of the option variable. Currently it
directly implements GET and SET, plus STEP_UP for "choice" options
only. Use it to add a property for -pts-association-mode (not
particularly useful in normal use, but serves as a test).
Add a "choice" option type. Options of this type take a string as
input and set an int option variable to the value corresponding to the
string. The string->int mapping is option-specific and is given in the
option definition. Strings not found in the mapping are rejected as
invalid option values.
Change the option -pts-association-mode to use this new option type
and accept values "auto, decoder, sort" instead of "0, 1, 2". The
change in accepted values shouldn't cause problems as this option is
not appropriate to use in normal user config files.
Let higher-level code call update_video() again instead of looping
inside it until there's a frame ready to show. Change the conditions
for running user commands somewhat. Overall effect shouldn't be that
big. Now other commands can be executed after a seek before a video
frame is decoded; in this case the seek target time may be used as the
"current position".
Tested to properly decode .nsc files, but not properly tested since
no publically available stream could be found.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32710 b3059339-0415-0410-9bf9-f77b7e298cf2
Remove pointless NULL checks before free.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32715 b3059339-0415-0410-9bf9-f77b7e298cf2
len < 8 is also invalid for 64-bit codec chunk size.
Previous code could cause hang.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32708 b3059339-0415-0410-9bf9-f77b7e298cf2
Avoid using uninitialized data if index read does not return enough data.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32707 b3059339-0415-0410-9bf9-f77b7e298cf2
Improve speex codec pts handling, make audio timestamps work reasonably
even with the native demuxer as long as seeking is not done.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32704 b3059339-0415-0410-9bf9-f77b7e298cf2
Always free before overwriting a pointer with a newly allocated one,
always use calloc instead of realloc when the previous data is not
needed.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32703 b3059339-0415-0410-9bf9-f77b7e298cf2
Add validity check for stream_seek argument to avoid a integer
overflow for huge negative values that would break the internal state
of the stream buffer.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32702 b3059339-0415-0410-9bf9-f77b7e298cf2
Change direct rendering buffer allocation code to treat non-ref frames
like B-frames even if has_b_frames is not set and they are indeed not
B-frames (no reordering). Treating it as an I/P frame would violate
the assumptions of MPlayer's buffering system, which thinks only the
latest previous I/P frame is needed (in addition to one possibly being
decoded). In this case the previous I/P frame will still be needed in
the future, not the non-ref frame being decoded now.
This happens with flv files, as in bug #1079, and this change fixes that
corruption.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32700 b3059339-0415-0410-9bf9-f77b7e298cf2
Replace memcpy with memmove since at least src==dst is possible.
Fixes another issue that is part of bug #1280.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32697 b3059339-0415-0410-9bf9-f77b7e298cf2
