Currently entries are added after the current playlist element. This is kinda
confusing, more so given that "loadfile append" appends at the end of the
playlist.
"loadfile filename append-play" will now always append the file to the
playlist, and if nothing is playing yet, start playback. I don't want to
change the semantics of "append" mode, so a new mode is needed.
Probably fixes issue #950.
The MPlayer style syntax ("-mf fps=10:type=png") was removed a while
ago, and now only the flat variants ("--mf-fps=10" etc.) work.
CC: @mpv-player/stable
We told the demuxer that a pipe (if stream cache is enabled) is
seekable. This is because the stream cache is technically seekable, it's
just that seeking may fail at runtime if a non-cached byte range is
requested.
This caused libavformat to issue seeks on initialization (at least when
piping mp4 youtube videos). Initialization failed completely after
spamming tons of error messages.
So, if an unseekable stream is cached, tell the demuxer that the file is
not seekable. This gets reversed later (when printing a message about
caching an unseekable stream), so the user can still try his luck by
issuing a seek command. The important part is that libavformat
initialization will not take code paths that will unnecessarily seek for
whatever reasons.
CC: @mpv-player/stable: regression from 0.3.x
Conflicts:
demux/demux.c
The old FFmpeg API and the new Libav API disagree about mp4 display
rotation direction. Well, whatever, fix it trial-and-error-style.
CC: @mpv-player/stable: add
No reason to wait until the audio has been played. This isn't a problem
with gapless audio disabled, and since gapless is now default, this
behavior might be perceived as regression.
CC: @mpv-player/stable
Logic for this was missing from pull.c. For push.c it was missing if the
driver didn't support it. But even if the driver supported it (such as
with ao_alsa), strange behavior was observed by users. See issue #933.
Always check explicitly whether the AO is in paused mode, and if so,
don't drain.
Possibly fixes#933.
CC: @mpv-player/stable
Happens when playing from a pipe.
Note that seeking forward doesn't work. It would be possible to create a
workaround for that by reading and skipping data until the target
position is reached (and writing the skipped data into the cache file),
but I'm not sure about that.
Fixes#928.
CC: @mpv-player/stable
Key bindings are decided on the "down" event, so if the prefix is not
unique, the first/shortest will be used (e.g. when both "a" and "a-b"
are mapped, "a" will always be chosen).
This also breaks combining multiple mouse buttons. But it seems users
expect it to work, and it's indeed a bit strange that it shouldn't work,
as mouse bindings are emitted on the key "up" event, not "down" (if the
shorter binding didn't emit a command yet, why shouldn't it be
combinable).
Deal with this by clearing the key history when a command is actually
emitted, instead of when a command is decided. This means if both
MOUSE_BTN0 and MOUSE_BTN0-MOUSE_BTN1 are mapped, the sequence of holding
down BTN0 and then BTN1 will redecide the current command. On the other
hand, if BTN0 is released before BTN1 is pressed, the command is
emitted, and the key history is deleted. So the BTN1 press will not
trigger BTN0-BTN1.
For normal keys, nothing should change, because commands are emitted on
the "down" event already, so the key history is always cleared.
Might fix#902.
CC: @mpv-player/stable (if this fix is successful)
Doesn't work quite right, and will pause for the latency duration after
seeking. Some users use --ao=null to disable audio (even though they
should probably use --no-audio), and this use-case is broken by this
issue too.
CC: @mpv-player/stable
When seeking, we violently destroy the filter, because vapoursynth has
no proper API for terminating a video with unknown frame count. This
looks like an error to vapoursynth, and the error is returned via the
frame callbacks. The bug is that we remember this error state across
reinitialization, so on the first filter call after reinitialization, we
thought filtering the current frame failed. This caused a shift by 1
frame on each seek.
CC: @mpv-player/stable
Not all compilers on all platforms have atomics available (even if they
could, technically speaking).
We don't use atomics that much, only the following things rely on it:
1. the audio pull code, and all audio outputs using it
2. updating global msg levels
3. reading log messages through the client API
Just disable 1. and 3. if atomics are not available. For 2., using fake-
atomics isn't too bad; at worst, message levels won't properly update
under certain situations (but most likely, it will work just fine).
This means if atomics are not available, the client API function
mpv_request_log_messages() will do nothing.
CC: @mpv-player/stable
Suggested by tholin on github issue #882.
This is not entirely clean, but the fields we're accessing might be
considered internal to libavformat. On the other hand, existence of the
fields is guaranteed by the ABI, and nothing in the libavformat doxygen
suggestes they're not allowed to be accessed.
CC: @mpv-player/stable
Recently, libavformat added demuxers to open image files like normal
demuxers. This is a good thing, but for now they interfere with the
operation of demux_mf. Add them to the blacklist until there is a proper
solution.
(The list doesn't contain _all_ recognized image formats, just those
that might interfere with demux_mf.)
CC: @mpv-player/stable
Completion now uses "--opt=value" instead of "--opt value". Once the
user presses space and starts a new argument, the option just
completed is out of the picture, whether or not it was given an
argument. This handles options with no arguments or optional arguments
much better; previously, completing such an option would effectively
disable completion for the next argument.
Custom completed options such as "--ao" and friends will no longer
claim to consume an extra argument.
When resizing the cache, the buffer for the DVD timestamps is
initialized with 0. This causes the player to always return playback
position 0 with any file format (not just DVD), and also makes all
relative seeks relative to position 0. Fix this by clearing the
timestamps explicitly.
Closes#899.
CC: @mpv-player/stable
Apparently clearing on every map can cause problems with vdpau when
switching virtual desktops and such. This was observed with at least
XMonad and nvidia-340.17. It's not observed on some other setups without
XMonad.
It's not clear why this happens. Normally, the window background is not
saved, so clearing should have no additional affect. It's a complete
mystery. Possible, the use of legacy X drawing commands (used to clear
the window) interferes with vdpau operation in non-trivial ways.
Work this around by clearing on initial map only. This probably only
hides the underlying issue, but good enough.
Closes#897.
CC: @mpv-player/stable
Some of these might be security relevant.
The RealAudio code was especially bad. I'm not sure if all RealAudio
stuff still plays correctly; I didn't have that many samples for
testing. Some checks might be unnecessary or overcomplicated compared
to the (obfuscated) nature of the code.
CC: @mpv-player/stable
rgain is not an additive value. It's a multiplier/gain.
Previous behaviour produced negative level values in some cases
(when rgain < 1.0) which caused volume to be louder when its value
was lowered.
CC: @mpv-player/stable
Signed-off-by: Mohammad Alsaleh <CE.Mohammad.AlSaleh@gmail.com>
Signed-off-by: wm4 <wm4@nowhere>
These consult the vertical resolution, matching against 576 for
PAL and 480/486 for NTSC. The documentation has also been updated.
Signed-off-by: wm4 <wm4@nowhere>
Notably, we now conform to SMPTE 428-1-2006 when decoding XYZ12 input,
and we can support rendering intents other than colorimetric when
converting between BT.709 and BT.2020, like with :srgb or :icc-profile.
With this change, XYZ input is directly converted to the output
colorspace wherever possible, and to the colorspace specified by the
tags and/or --primaries option, otherwise.
This commit also restructures some of the CMS code in gl_video.c to
hopefully make it clearer which decision is being done where and why.
This also avoids an extra matrix multiplication when using :srgb, making
that path both more efficient and also eliminating more hard-coded
values.
In addition, the previously hard-coded XYZ to RGB matrix will be
dynamically generated.
This add support for reading primary information from lavc, categorized
into BT.601-525, BT.601-625, BT.709 and BT.2020; and passes it on to the
vo. In vo_opengl, we always generate the 3dlut against the wider BT.2020
and transform our source into this colorspace in the shader.
For remarks, pretty much see the manpage additions. Could help with
network streams that require too much seeking (maybe), or might be
extended to help with the use case of watching and downloading a file
at the same time.
In general, it might be a useless feature and could be removed again.
It seems it's generally cleaner to leave this stuff to the BSDs.
Additionally, the NetBSD part wasn't even correct, because it made the
compiler prefer the system include path before local include files.
It is reasonably stable, so all further changes will be versioned.
Also change how the libmpv version number is generated. Fix the patch
version number to 0; I don't think we have a use for this. In
particular, the version doesn't version mpv, just the client API.
Do this simply by clearing the mapped buffer on every draw_image() call
without an actual video frame. (Maybe this is a bit expensive, but at
least not more expensive than regular video display.)
Clear the texture on reconfig(). (We could probably also do this simpler
with a flag, but this is actually less complicated - except that we move
the code to "map" a texture to a separate function.)
With the change to merge osd drawing into video frame drawing, some
bogus logic got in: they skipped drawing the OSD if no video frame is
available. This broke --no-video --force-window mode.
Until now, bitmap subtitles were decoded at "some" point, and then
simply replaced the old subtitle. Although the subtitle is selected
by time (PTS), it could happen that a subtitle was replaced too early.
One consequence is that this might lead to flicker even if the
subtitles are timed to follow each other without a gap (although most
subtitles are explicitly timed to introduce such a gap). With this
commit the past 4 subtitles are kept (instead of 1), so that the
correct one can be picked by time. This should fix the aforementioned
cases, but more importantly will allow demuxing/decoding and video
display to be somewhat asynchronous.
Still missing: somehow making sure the correct range of decoded
subtitles is available, instead of just passing along whatever comes
from the demuxer, and hoping that 4 queued subtitles are enough. But it
should certainly be good enough for now.
This removes a check that resets the subtitles if the PTS is 5 minutes
before the end of the current subtitle; this is probably not needed.
This fixes the build on platform where the atomic calls can't be turned into
lock-free instructions and thus need the external libatomic library (e.g. mips).
Make sure every video filter has valid parameters for input and output.
(This also ensures we don't take possibly invalid decoder output, or
feed invalid decodr/filter output to VOs.)
Also, the updated image size check now (almost) works like the
corresponding check in FFmpeg.
Until now, failure to allocate image data resulted in a crash (i.e.
abort() was called). This was intentional, because it's pretty silly to
degrade playback, and in almost all situations, the OOM will probably
kill you anyway. (And then there's the standard Linux overcommit
behavior, which also will kill you at some point.)
But I changed my opinion, so here we go. This change does not affect
_all_ memory allocations, just image data. Now in most failure cases,
the output will just be skipped. For video filters, this coincidentally
means that failure is treated as EOF (because the playback core assumes
EOF if nothing comes out of the video filter chain). In other
situations, output might be in some way degraded, like skipping frames,
not scaling OSD, and such.
Functions whose return values changed semantics:
mp_image_alloc
mp_image_new_copy
mp_image_new_ref
mp_image_make_writeable
mp_image_setrefp
mp_image_to_av_frame_and_unref
mp_image_from_av_frame
mp_image_new_external_ref
mp_image_new_custom_ref
mp_image_pool_make_writeable
mp_image_pool_get
mp_image_pool_new_copy
mp_vdpau_mixed_frame_create
vf_alloc_out_image
vf_make_out_image_writeable
glGetWindowScreenshot
The error log callback was not thread-safe and not library-safe. And
apparently there were some other details that made it not library-safe,
such as a global lcms plugin registry.
Switch the the thread-safe API provided by lcms2 starting with 2.6.
Remove our approximate thread-safety hacks.
Note that lcms basically provides 2 APIs now, the old functions, and
the thread-safe alternatives whose names end with THR. Some functions
don't change, because they already have a context of some sort. Care
must be taken not to accidentally use old APIs.
When the reader is out of data, it tries to wake up the cache thread to
get more data. In theory, there's a small race condition, which could
cause the cache to miss the wakeup and idle before reaction.
Most certainly didn't cause real issues, because even if this extremely
unlikely race condition happens, the cache won't idle for longer than
1 second (the hardcoded cache idle time).
OSD used to be not thread-safe at all, so a track was used to get it
redrawn. This mostly reverts commit 6a2a8880, because OSD not being
thread-safe was the non-trivial part of it.
Mostly untested, because this code path is used on OSX only, and I don't
have OSX.
The main issue was actually that the OSD callback locked the subtitle
decoder, which does not necessarily work, because the OSD code is
already allowed to lock it. The state was already protected by
unsetting the callback (which involes the OSD lock). So, in summary,
this is probably just a cleanup.
We certainly don't want to maintain and improve the internal converter,
but we still need the internal one for Libav. (In the Libav case,
demux_subreader.c will be used to read the MicroDVD file.)
Let the VOs draw the OSD on their own, instead of making OSD drawing a
separate VO driver call. Further, let it be the VOs responsibility to
request subtitles with the correct PTS. We also basically allow the VO
to request OSD/subtitles at any time.
OSX changes untested.
Subsurfaces are only used by the wayland vo. Thats why it makes sense to move
all osd and subsurface specific parts to the vo_wayland.c
Also destroy the subsurfaces and subcompositor properly.
Not all the hardware supports kCGLPFASupportsAutomaticGraphicsSwitching
(apparently all Mid-2010 and before MacBooks do not work with it), so fallback
to not asking for this attribute in the GL pixel format.
Well, not sure if this really improves anything, but at least it's less
of a WTF to the playback core than always returning the same timestamp
for every frame.
There's really no need to convert this to float and then back. This is
mostly of cosmetic nature, double precision was probably enough to avoid
rounding.
This affects packed RGB formats up to 16 bits per pixel. The old mplayer
names used LSB-to-MSB order, while FFmpeg (and some other libraries) use
MSB-to-LSB.
Nothing should change with this commit, i.e. no bit order or endian bugs
should be added or fixed. In some cases, the name stays the same, even
though the byte order changes, e.g. RGB8->BGR8 and BGR8->RGB8, and this
affects the user-visible names too; this might cause confusion.
These suffixes are annoying when they're redundant, so strip them
automatically. On little endian machines, always strip the "le" suffix,
and on big endian machines vice versa (although I don't think anyone
ever tried to run mpv on a big endian machine).
Since pixel format strings are returned by a certain function and we
can't just change static strings, use a trick to pass a stack buffer
transparently. But this also means the string can't be permanently
stored by the caller, so vf_dlopen.c has to be updated. There seems
to be no other case where this is done, though.
Not sure how this symbol becomes visible in glibc (probably accidental
or mandatory recursive inclusion via the other standard or Linux-
specific headers), but normally this include file is needed to get the
symbol.
The "classic" sub-option stuff is not really needed anymore. The only
remaining use can be emulated in a simpler way. But note that this
breaks the --screenshot option (instead of the "flat" options like
--screenshot-...). This was undocumented and discouraged, so it
shouldn't affect anyone.
Instead of absuing m_option to store the property list, introduce a
separate type for properties. m_option is still used to handle data
types. The property declaration itself now never contains the option
type, and instead it's always queried with M_PROPERTY_GET_TYPE. (This
was already done with some properties, now all properties use it.)
This also fixes that the function signatures did not match the function
type with which these functions were called. They were called as:
int (*)(const m_option_t*, int, void*, void*)
but the actual function signatures were:
int (*)(m_option_t*, int, void*, MPContext *)
Two arguments were mismatched.
This adds one line per property implementation. With additional the
reordering of the parameters, this makes most of the changes in this
commit.
This means use of the min/max fields can be dropped for the flag option
type, which makes some things slightly easier. I'm also not sure if the
client API handled the case of flag not being 0 or 1 correctly, and this
change gets rid of this concern.
Also clarify the semantics.
It seems --idx didn't do anything. Possibly it used to change how the
now removed legacy demuxers like demux_avi used to behave. Or maybe
it was accidental.
--forceidx basically becomes --index=force. It's possible that new
index modes will be added in the future, so I'm keeping it
extensible, instead of e.g. creating --force-index.
Probably "needed" to get the correct alignment, although I'm not aware
of actual breakages or performance issues.
In fact we should probably always just allocate AVPackets, but for now
use the simple fix.
FFmpeg requires a bullshit padding after each input buffer, and they
just increased that padding without warning and without ABI or API bump.
We need this only in one file (although mp_image hardcodes something
similar, for which no FFmpeg API define is available), so drop our own
define.
Does anyone actually use this?
For now, update it, because it's the only case left where an option
points to a global variable (and not a struct offset).
Unfortunately, there's a recursive function call in ao_lavc.c (play
function), leading to a deadlock. The locking is getting a bit messy, so
just make the lock recursive.
This fixes#844.
Squashed from the following mplayer-svn commits. The '#' is removed from
the bug ID to prevent github from doing something stupid. Instead of
adding the mplayer configure check for clock_gettime(), the POSIX
identifiers are used for checking presence of the function.
Use correct type of timestamps when recording from v4l2. Fix 2176
Patch by Jarek Czekalski <jarekczek at poczta onet pl>.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@37222 b3059339-0415-0410-9bf9-f77b7e298cf2
Allow building of v4l2 without clock_gettime().
Add overly verbose message in case monotone timestamps are required by the kernel.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@37223 b3059339-0415-0410-9bf9-f77b7e298cf2
Similar to previous commits.
This also renames --doubleclick-time to --input-doubleclick-time, and
--key-fifo-size to --input-key-fifo-size. We could keep the old names,
but these options are very obscure, and renaming them seems better for
consistency.
While I'm not very fond of "const", it's important for declarations
(it decides whether a symbol is emitted in a read-only or read/write
section). Fix all these cases, so we have writeable global data only
when we really need.
Additionally to removing the global variables, this makes the options
more uniform. --ssf-... becomes --sws-..., and --sws becomes --sws-
scaler. For --sws-scaler, use choices instead of magic integer values.
Usually, each stream driver declares the size and option list of its
private data. This was pretty natural for when most streams still used
global variables to setup their defaults. They did by pointing
priv_defaults to the (mutable) struct containing the option values. But
falls short when storing the option values in MPOpts. So provide a
somewhat inelegant but simple way to let the stream implementation setup
the priv struct at initialization time.
This is done with the get_defaults callback. It should return a copy of
the struct used in MPOpts. (A copy, because if MPOpts is changed, string
fields might be deallocated, and if that field is not described by
stream_info.options, it won't be copied on init.)
Pretty much nothing changes, but using -tv-scan with suboptions doesn't
work anymore (instead of "-tv-scan x" it's "-tv scan-x" now). Flat
options ("-tv-scan-x") stay compatible.
Convert all these commands to properties. (Except tv_last_channel, not
sure what to do with this.) Also, internally, don't access stream
details directly, but dispatch commands with stream ctrls.
Many of the new properties are a bit strange, because they're write-
only. Also remove some OSD output these commands produced, because I
couldn't be bothered to port these.
In general, this makes everything much cleaner, and will also make it
easier to e.g. move the demuxer to its own thread.
Don't bother updating input.conf, but changes.rst documents how old
commands map to the new ones.
Mostly untested, due to lack of hardware.
The Perl script generating the completions actually invokes mpv, and it
runs during the build. This is not sane and breaks at least cross
compilation.
As a workaround, disable the completions by default for now.
Playing a video and then an audio file with cover art kept displaying
the last frame of the video. This was because the hasframe flag was set,
perhaps due to redrawing the last video frame before the cover art image
is decoded.
Basically, this allows gapless playback with similar files (including
the ordered chapter case), while still being robust in general.
The implementation is quite simplistic on purpose, in order to avoid
all the weird corner cases that can occur when creating the filter
chain. The consequence is that it might do not-gapless playback in
more cases when needed, but if that bothers you, you still can use
the normal gapless mode.
Just using "--gapless-audio" or "--gapless-audio=yes" selects the old
mode.
Since redundant wakeups are avoided now, it's easy to miss a wakeup when
creating/setting the pipe/callback after the client API was signalled.
If the client API is signalled, need_wakeup is set to true, and
wakeup_client skips writing to the pipe or calling the client API. That
this can happen is not very obvious to the client API, so trigger a
wakeup right on start in order to remove this special case.
--sub-file is actually a string list, so you can add multipel external
subtitle files. But to be able to set a list, the option value was split
on ",". This made it impossible to add filenames.
One possible solution would be adding escaping. That's probably a good
idea (and some other options already do this), but it's also complicated
both to implement and for the user.
The simpler solution is making --sub-file appending, and make it take
only a single entry.
I'm not quite sure about this yet. It breaks the invariant that if a
value is printed and parsed, you get the same value back. So for now,
just go with the simple solution.
Fixes#840.
Until now, availability of new log messages (through the mechanism
associated with mpv_request_log_messages()) did not wakeup the client
API properly. Commit 3b7402b5 was basically a hack to improve that
somewhat, but it wasn't a solution.
The main problem is that the client API itself is producing messages, so
the message callback would attempt to lock the client API lock,
resulting in a deadlock. Even if the lock was recursive, we'd run into
lock-order issues.
Solve this by using a separate lock for waiting and wakeup. Also, since
it's a natural addition, avoid redundant wakeups. This means the wakeup
callback as well as the wakeup pipe will be triggered only once until
the next mpv_wait_event() call happens.
This might make the wakeup callback be invoked in a reentrant way for
the first time, for example if a mpv_* function prints to a log. Adjust
the docs accordingly. (Note that non-reentrant beheavior was never
guaranteed - basically the wakeup callback is somewhat dangerous and
inconvenient.)
Also remove some traces of unneeded code. ctx->shutdown for one was
never set, and probably a leftover of an abandoned idea.
A bit verbose, but less misleading. In most cases, the API user probably
actually wants mpv_terminate_destroy() instead, so the less-useful
function shouldn't have a simnpler name anyway.
It wasn't necessarily clear how this works. Especially make clear that
the API user shouldn't expect that there's one byte per readable event
in the wakeup pipe. Actually, the reason why this currently won't work
is because property notifications can generate more events than wakeups.
The limit of 4096 is a more fundamental issue, but the event ringbuffer
is currently limited to 1000 entries anyway.
Also add some important comments to mpv_set_wakeup_callback.
mpv_destroy() should perhaps better be called mpv_detach(), because it
destroys only the handle, not necessarily the player. The player is only
terminated if a quit command is sent.
This function quits automatically, and additionally waits until the
player is completely destroyed. It removes the possibility that the
player core is still uninitializing, while all client handles are
already destroyed. (Although in practice, the difference is usually not
important.)
Now a negative timeout mean an infinite timeout. This is similar to the
poll() system call. Apparently this is more intuitive and less confusing
than specifying a "very high" value as timeout if you want to wait
forever.
For callers that never pass negative timeouts, nothing changes.
This was never intended to be installed; waf just picked it up
automagically. There's also a closed ticket on github where someone
complains that the program "simple" is installed, and I didn't realize
at this point that it was actually installed by default when enabling
the client API.
Listening on messages currently uses polling (every time
mpv_wait_event() has no new events, the message buffer is polled and a
message event is possibly created). Improve this situation a bit, and
call the user-supplied wakeup callback.
This will increase the frequency with which the wakeup callback is
called, but the client is already supposed to be able to deal with this
situation. Also, as before, calling mpv_wait_event() from the wakeup
callback is forbidden, so the client can't read new messages from the
callback directly.
The wakeup pipe is written either. Since the wakeup pipe is created
lazily, we can't access the pipe handle without creating a race
condition or a deadlock. (This is actually very silly, since in practice
the race condition won't matter, but for now let's keep it clean.)
It seems we can't really get rid of this. There are no other hints to
remove decorations that work across all reasonable WMs, so we're stuck
with the ugly motif stuff.
But at least we can make the code for it less ugly.
If an API user calls mpv_wait_event() and mpv_observe_property() before
mpv_initialize(), it could happen that a property was accessed before
initialization, which is not ok.
This was always intended this way, and even documented in client.h. Due
to an oversight it was never actually implemented.
The intention is that mpv embedded in applications and "real mpv" don't
conflict. An API user can undo this by setting the "config" option to
"yes", if using the user's mpv config is desired.
(The old "force" choice of that option is renamed to "force-default".)
This allows overriding native ASS script subtitle styles with the style
provided by the --sub-text-* options (like --sub-text-font etc.). This
is disabled by default, and needs to be explicitly enabled with the
--ass-style-override=force option and input property.
This uses in fact exactly the same options (--sub-text-*) and semantics
as the ones used to configure unstyled text subtitles.
It's recommended to combine this with this in the mpv config file:
ass-force-style="ScaledBorderAndShadow=1" # work around dumb libass behavior
Also, adding a key binding to toggle this behavior should be added,
because overriding can easily break:
L cycle ass-style-override
This would cycle override behavior on Shift+L and allows quickly
disabling/enabling style overrides.
Note: ASS should be considered a vector format rather than a subtitle
format. There is no easy or reliable way to determine whether the style
of a given subtitle event can be changed without destroying visuals or
not. This patch relies on a simple heuristic, which often works and
often breaks.
This could for example happen when serving an incomplete file from http,
and the demuxer tries reading data from the end of the file when opening
it (e.g. with avi). Seeking past EOF fails with http, so the file could
never be opened, and the cache would get stuck trying to seek to the
position.
We can't really make the cache report seek failure directly (it would
suck for various reasons), so just make the cache report EOF if seeking
fails.
The previous commit assumed the filter would be 1x1 (then constant
weight is correct) - but our code in fact uses at least a 2x2 filter. A
1x1 filter would generally be useless, except for nearest scaling - so
it didn't exist.
Insteasd of adding such a 1x1 filter, just turn the nearest weight
function into a scare function, which should take care of the issue.
This simply writes the file name as a comment to the top of the watch later
config file.
It can be useful to the user for determining whether a watch later config file
can be manually removed (e.g. in case the corresponding media file has been
deleted) or not.
This accidentally rejected d==0. We can actually deal with sub-normals
fine, we just want to exclude nan and infinity (although infinity is
already accounted for, but anyway).
If a single person complains, I will readd it. But I don't expect that
this will happen.
The main reason for removing this is that it's some of the most unclean
code remaining, it's unmaintained, and I've never ever heard of someone
using it.
Until now, an error was reported only if the command couldn't be parsed.
Attempt to do more fine-grained reporting. This is not necessarily
perfect, but it's an improvement.
This would imply eglGetProcAddress() doesn't work correctly, but using
dlsym() does. For now get rid of it - it won't work in libmpv, and we'll
probably need a better workaround if it's still broken.
This code was in the initial wayland commit.
The functions glXGetProcAddressARB() and glXQueryExtensionsString() were
loaded using dlsym(). This could fail when compiling to libmpv, because
then dlopen(NULL, ...) will look in the main program's list of
libraries, and the libGL linked to libmpv is never considered. (Don't
know if this somehow could be worked around.) The result is that using
vo_opengl with libmpv can fail.
Avoid this by not using dlsym(). glXGetProcAddressARB() was already used
directly in the same file, and that never caused any problems. (Still
add it to the configure test.) glXQueryExtensionsString() is documented
as added in GLX 1.1 - that's ancient.
So the device buffer can be refilled quickly. Fixes dropouts in certain
cases: if all data is moved from the soft buffer to the audio device
buffer, the waiting code thinks it has to enter the mode in which it
waits for new data from the decoder. This doesn't work, because the
get_space() logic tries to keep the total buffer size down. get_space()
will return 0 (or a very low value) because the device buffer is full,
and the decoder can't refill the soft buffer. But this means if the AO
buffer runs out, the device buffer can't be refilled from the soft
buffer. I guess this mess happened because the code is trying to deal
with both AOs with proper event handling, and AOs with arbitrary
behavior.
Unfortunately this increases latency, as the total buffered audio
becomes larger. There are other ways to fix this again, but not today.
Fixes#818.
Apparently this can happen. So actually only return from waiting if ALSA
excplicitly signals that new output is available, or if we are woken up
externally.
This call was used limited the buffer size if installed RAM was below 16
MB. This stopped being useful a decade ago. The check could also
overflow on 32 bit systems. Just get rid of it.
This did not flush remaining audio in the buffer correctly (in case an
AO has an internal block size). So we have to make the audio feed thread
to write the remaining audio, and wait until it's done.
Checking the avoid_ao_wait variable should be enough to be sure that all
data that can be written was written to the AO driver.
This code handles buggy AOs (even if all AOs are bug-free, it's good for
robustness). Move handling of it to the AO feed thread. Now this check
doesn't require magic numbers and does exactly what's it supposed to do.
Until now, we've always calculated a timeout based on a heuristic when
to refill the audio buffers. Allow AOs to do it completely event-based
by providing wait and wakeup callbacks.
This also shuffles around the heuristic used for other AOs, and there is
a minor possibility that behavior slightly changes in real-world cases.
But in general it should be much more robust now.
ao_pulse.c now makes use of event-based waiting. It already did before,
but the code for time-based waiting was also involved. This commit also
removes one awkward artifact of the PulseAudio API out of the generic
code: the callback asking for more data can be reentrant, and thus
requires a separate lock for waiting (or a recursive mutex).
uint_least32_t could be larger than uint32_t, so the return values of
mp_ring_get_wpos/rpos must be adjusted. Actually just use unsigned long
as type instead, because that is less awkward than uint_least32_t.
This was originally added because we thought this would make a good
portable audio API, which would give us good behavior on Windows, Linux,
and OSX. But this hope was disappointed: it's not reliable enough (nice
deadlocks on Linux when seeking, i.e. resetting the audio device),
doesn't have enough features (no channel maps, no digital passthrough),
and in general just is not very good.
There were subtle and minor race conditions in the pull.c code, and AOs
using it (jack, portaudio, sdl, wasapi). Attempt to remove these.
There was at least a race condition in the ao_reset() implementation:
mp_ring_reset() was called concurrently to the audio callback. While the
ringbuffer uses atomics to allow concurrent access, the reset function
wasn't concurrency-safe (and can't easily be made to).
Fix this by stopping the audio callback before doing a reset. After
that, we can do anything without needing synchronization. The callback
is resumed when resuming playback at a later point.
Don't call driver->pause, and make driver->resume and driver->reset
start/stop the audio callback. In the initial state, the audio callback
must be disabled.
JackAudio of course is different. Maybe there is no way to suspend the
audio callback without "disconnecting" it (what jack_deactivate() would
do), so I'm not trying my luck, and implemented a really bad hack doing
active waiting until we get the audio callback into a state where it
won't interfere. Once the callback goes from AO_STATE_WAIT to NONE, we
can be sure that the callback doesn't access the ringbuffer or anything
else anymore. Since both sched_yield() and pthread_yield() apparently
are not always available, use mp_sleep_us(1) to avoid burning CPU during
active waiting.
The ao_jack.c change also removes a race condition: apparently we didn't
initialize _all_ ao fields before starting the audio callback.
In ao_wasapi.c, I'm not sure whether reset really waits for the audio
callback to return. Kovensky says it's not guaranteed, so disable the
reset callback - for now the behavior of ao_wasapi.c is like with
ao_jack.c, and active waiting is used to deal with the audio callback.
We don't need to combine __sync_add_and_fetch with a memory barrier,
since these intrinsics are documented as using a full barrier already.
Use __sync_fetch_and_add instead of __sync_add_and_fetch; this gives
atomic_fetch_add() the correct return value (although we don't use it).
Use __sync_fetch_and_add to emulate atomic_load(). This should enforce
the full barrier semantics better. (This trick is stolen from the
FreeBSD based stdatomic.h emulation.)
In most places where af_fmt2bits is called to get the bits/sample, the
result is immediately converted to bytes/sample. Avoid this by getting
bytes/sample directly by introducing af_fmt2bps.
The i_bps members of the sh_audio and dev_video structs are mostly used
for displaying the average audio and video bitrates. Keeping them in
bits-per-second avoids truncating them to bytes-per-second and changing
them back lateron.
This is incomplete; the video chain will still hold some vaapi objects
after destroying the decoder and thus the vaapi context. This is very
bad. Fixing it would require something like refcounting the vaapi
context, but I don't really want to.
mpv supports two hardware decoding APIs on Linux: vdpau and vaapi. Each
of these has emulation wrappers. The wrappers are usually slower and
have fewer features than their native opposites. In particular the libva
vdpau driver is practically unmaintained.
Check the vendor string and print a warning if emulation is detected.
Checking vendor strings is a very stupid thing to do, but I find the
thought of people using an emulated API for no reason worse.
Also, make --hwdec=auto never use an API that is detected as emulated.
This doesn't work quite right yet, because once one API is loaded,
vo_opengl doesn't unload it, so no hardware decoding will be used if the
first probed API (usually vdpau) is rejected. But good enough.
So, basically this worked only with streams that were not local files,
because stream_dvd.c "intercepts" local files to check whether they
point to DVD images. This means if a stream is not writeable, we have to
try the next stream implementation.
Unbreaks 2-pass encoding.
Sometimes, Matroska files store monotonic PTS for h264 tracks with
b-frames, which means the decoder actually returns non-monotonic PTS.
Handle this with an evil trick: if DTS is missing, set it to the PTS.
Then the existing logic, which deals with falling back to DTS if PTS is
broken. Actually, this trick is not so evil at all, because usually, PTS
has no errors, and DTS is either always set, or always unset. So this
_should_ provoke no regressions (famous last words).
libavformat actually does something similar: it derives DTS from PTS in
ways unknown to me. The result is very broken, but it causes the DTS
fallback to become active, and thus happens to work.
Also, prevent the heuristic from being active if PTS is merely monotonic
instead of strictly-monotonic. Non-unique PTS is broken, but we can't
fallback to DTS anyway in these cases.
The specific mkv file that is fixed with this commit had the following
fields set:
Muxing application: libebml v1.3.0 + libmatroska v1.4.1
Writing application: mkvmerge v6.7.0 ('Back to the Ground') [...]
But I know that this should also fix playback of mencoder produced mkv
files.
When Lua itself prints errors such as:
Error: [string "mp.defaults"]:387: syntax error near 'function'
It's unclear why the location is prefixed with "string ". And for some
reason, it disappears if you prefix the name with '@'. I suppose this is
done for the sake of luaL_loadstring. With the '@' prefix added, the
output is now:
Error: mp.defaults:387: syntax error near 'function'
We pass a pointer to a GLint to sscanf, using the %d format. That format
_always_ takes int, and not GLint (whatever the heck that is). If GLint
is always int, then it doesn't make a difference, but is still better
because it doesn't play russian roulette with pointers.
Don't emit "hard" references to OpenGL functions. Always use the
platform specific function to lookup OpenGL functions, such as
glXGetProcAddress() with GLX (x11).
This actually fixes the build if only Wayland is enabled (e.g. using
--disable-gl-x11 on Linux).
Note that some sources claim that wglGetProcAddress() (win32) does not
return function pointers for OpenGL 1.1 functions (even if they are
valid and necessary in OpenGL 3.0). But if that happens, the fallback
employed in gl_w32.c/w32gpa() should catch this.
Setting this property was added 12 years ago, and the code was always
incorrect. The underlying data type is "long", not "pid_t". It's well
possible that the data types are different, and the pointer to the pid
variable is directly passed to XChangeProperty, possibly invoking
undefined behavior.
It's funny, because in theory using pid_t for PIDs sounds more correct.
Binding multiple commands at once where always considered not
repeatable, because the MP_CMD_COMMAND_LIST wasn't considered
repeatable.
Fixes#807 (probably).
These are actually already included in osdep/io.h, but I think it's
cleaner to repeat them in the file where they are actually needed.
(osdep/io.h needs to have them for other reasons.)
Commit e2e450f9 started making use of luaL_register(), but OF COURSE
this function disappeared in Lua 5.2, and was replaced with a 5.2-only
alternative, slightly different mechanism.
So just NIH our own function. This is actually slightly more correct,
since it forces the user to call "require" to actually make the module
visible for builtin C-only modules other than "mp". Fix autoload.lua
accordingly.
This will load other files in the same directory when a single file is
played. It's an often requested feature, but we definitely don't want it
in the core.
We need this only because Lua's stdlib is so scarce. Lua doesn't intend
to include a complete stdlib - they confine themselves to standard C,
both for portability reasons and to keep the code minimal. But standard
C does not provide much either.
It would be possible to use Lua POSIX wrapper libraries, but that would
be a messy (and unobvious) dependency. It's better to implement the
missing functions ourselves, as long as they're small in number.
A playlist_move command that moves an entry onto itself (both arguments
have the same index) should do nothing, but it did something broken. The
underlying reason is that it checks the prev pointer of the entry which
is temporarily removed for moving.
Also sneak in some cosmetics.
setmode() exists on Windows/msvcrt only, so there's no need for a
config test.
I couldn't reproduce the problem with seekable pipes on wine, so axe
it. (I'm aware that it still could be an issue on real Windows.)
For some reason, we support writeable streams. (Only encoding uses that,
and the use of it looks messy enough that I want to replace it with FILE
or avio today.)
It's a chaos: most streams do not actually check the mode parameter like
they should. Simplify it, and let streams signal availability of write
mode by setting a flag in the stream info struct.
Stop using it in most places, and prefer STREAM_CTRL_GET_SIZE. The
advantage is that always the correct size will be used. There can be no
doubt anymore whether the end_pos value is outdated (as it happens often
with files that are being downloaded).
Some streams still use end_pos. They don't change size, and it's easier
to emulate STREAM_CTRL_GET_SIZE using end_pos, instead of adding a
STREAM_CTRL_GET_SIZE implementation to these streams.
Make sure int64_t is always used for STREAM_CTRL_GET_SIZE (it was
uint64_t before).
Remove the seek flags mess, and replace them with a seekable flag. Every
stream must set it consistently now, and an assertion in stream.c checks
this. Don't distinguish between streams that can only be forward or
backwards seeked, since we have no such stream types.
stream.start_pos was needed for optical media only, and (apparently) not
for very good reasons. Just get rid of it.
For stream_dvd, we don't need to do anything. Byte seeking was already
removed from it earlier.
For stream_cdda and stream_vcd, emulate the start_pos by offsetting the
stream pos as seen by the rest of mpv.
The bits in discnav.c and loadfile.c were for dealing with the code
seeking back to the start in demux.c. Handle this differently by
assuming the demuxer is always initialized with the stream at start
position, and instead seek back if initializing the demuxer fails.
Remove the --sb option, which worked by modifying stream.start_pos. If
someone really wants this option, it could be added back by creating a
"slice" stream (actually ffmpeg already has such a thing).
This is mostly covered by the OSX workaround, if the timeout is very
high. It also means that with systems using 32 bit time_t, the time will
overflow 2036 already, instead of 2037, but we don't consider this a
problem.
_WIN_LAYER is apparently an old GNOME thing (also explains why there is
a function vo_x11_get_gnome_layer() involved in this code). Prefer the
NetWM hints over this. This just moves the NetWM case if-body over the
_WIN_LAYER one.
You can't use identifiers starting with "_" and an uppercase letter in
application programs. They are reserved by the C standard.
Unrelated change: drop unused/misleading vo_wm_NETWM define.
I can only assume the old code was wrong. EWMH does not document
anything with _WIN_LAYER. Instead, you have to toggle the state using a
client message. We also remove these weird non-sense fallbacks, like
using _NET_WM_STATE_BELOW - what the hell?
Small fixes for the OSC, seektooltip now enabled by default.
Option-parser now moved to separate package, can be used from
other scripts, see DOCS/man/en/lua.rst.
OSC config file location moved to lua-settings/osc.conf
Some operating systems apparently can't deal with really long timeouts
in pthread_cond_timedwait(). Passing a time about 300000 in the future
makes the call return immediately. (tv_sec/time_t doesn't overflow in
this situation.) Reduce the wait time to about 100 days, which seems
to work fine.
The list of affected OSes follows: OSX
Probably more correct and better readable. Although the special-casing
of 0x1p63 is weird in terms of readability (the value itself is
INT64_MAX+1, so it's already outside of range, but INT64_MAX is not
exactly representable with double precision).
The quit command has an optional argument that is used as exit code.
Extend that to the quit_watch_later command. Actually, unify the
implementations of the two commands.
Requested in #798.
This usually can't happen, because even if time_us (first input value)
is INT64_MAX, the value added to tv_sec will be about 2^43, and tv_sec
will be <2^31, far below a possible overflow in 64 bits. But should
time_t be 32 bits (32 bit Linux/Windows?), an overflow could happen.
Integrate it with the existing surface allocator in vdpau.c. The changes
are a bit violent, because the vdpau API is so non-orthogonal: compared
to video surfaces, output surfaces use a different ID type, different
format types, and different API functions.
Also, introduce IMGFMT_VDPAU_OUTPUT for VdpOutputSurfaces wrapped in
mp_image, rather than hacking it. This is a bit cleaner.
black_pixel is an (apparently necessary) 1x1 black surface used for
clearing the screen. It was allocated in RGB mode only, but is sometimes
used in YUV mode too.
In at least 2 places, we want to "shift" positional parameters by a
certain amount, or remove all parameters if there are fewer parameters.
bash handled this just fine, but dash vomits upon it. POSIX shell allows
both behaviors, so we have to fix it.
Fix a comparison done with "==" instead of "=".
Cover art is treated like video, but is not really video. In one case,
the audio sync code was accidentally still active. Fixes cover art
playback with --ao=null. (This is due to ao_null's latency emulation.
Although it's not very clear whether that is actually correct...)
vo_vdpau currently has a video queue larger than 1 entry, which causes
the video display code to never queue display the video frame. This is
because we consider cover art an endless stream of frames decoded from
the same source packet, and include special logic to actually only
decode and display 1 frame.
Also, make decode_image() also signal EOF in the cover art case.
Without this change, the compiler uses by default the "talloc.h" file
installed by the package libtalloc within /usr/local/include. Found and
tested on OpenBSD but FreeBSD has the same patch on its ports tree.
This should be more correct. The builtins were made to directly map to
C11, and the way we use them is now relatively close to how gcc
implements atomics in 4.9. In particular, we make use of the load and
store builtins.
I'm not entirely sure why gcc didn't support stdatomic.h in 4.8 already.
Maybe support for the builtins was incomplete or broken - so there's a
lot of room for doubt about the correctness of this.
In my opinion, we shouldn't use atomics at all, but ok.
This switches the mpv code to use C11 stdatomic.h, and for compilers
that don't support stdatomic.h yet, we emulate the subset used by mpv
using the builtins commonly provided by gcc and clang.
This supersedes an earlier similar attempt by Kovensky. That attempt
unfortunately relied on a big copypasted freebsd header (which also
depended on much more highly compiler-specific functionality, defined
reserved symbols, etc.), so it had to be NIH'ed.
Some issues:
- C11 says default initialization of atomics "produces a valid state",
but it's not sure whether the stored value is really 0. But we rely on
this.
- I'm pretty sure our use of the __atomic... builtins is/was incorrect.
We don't use atomic load/store intrinsics, and access stuff directly.
- Our wrapper actually does stricter typechecking than the stdatomic.h
implementation by gcc 4.9. We make the atomic types incompatible with
normal types by wrapping them into structs. (The FreeBSD wrapper does
the same.)
- I couldn't test on MinGW.
Some options change from percentages to number of kilobytes; there are
no cache options using percentages anymore.
Raise the default values. The cache is now 25000 kilobytes, although if
your connection is slow enough, the maximum is probably never reached.
(Although all the memory will still be used as seekback-cache.)
Remove the separate --audio-file-cache option, and use the cache default
settings for it.
This works around an issue in OpenBox: OpenBox apparently sizes the
normal window incorrectly if aspect ratio hints are set, and the window
size is off by 1 pixel. Then, when going fullscreen and leaving
fullscreen again, mpv sets the hints based on OpenBox' broken window
size, and as result, OpenBox sizes the window incorrectly and is off by
1 pixel again - so it's 2 pixels off in total. The error gets more
visible, the more often you toggle fullscreen mode.
Work this around by not setting the window hints if we don't need to.
Actually we only need to do this when the video is resized during
fullscreen, which happens rarely. Under normal circumstances, leaving
fullscreen mode requires that the WM restores the old state.
As such, this commit is not only a workaround, but actually a cleanup.
Note that we do need to set the hints when leaving fullscreen if the
window has resized: even though we set the hints in
vo_x11_highlevel_resize (called by vo_x11_config_vo_window), this
doesn't seem to have an effect (at least on IceWM), so we have to do it
after that.
Side note: ot seems commit 625ad57a strangely triggered the OpenBox
issue according to user reports; I'm not sure why.
Drop: sami, vplayer, rt, pjs, mpsub, aqt, jacosub. None of these seem
to be actually in use, except sami. Sami is very complex, and the
results subreader produces are not very useful.
For all these formats, there are still parsers in FFmpeg. We remove the
subreader implementation, because it might contain security relevant
bugs and such. (This is old, unmaintained C string parsing code, written
in times where absolutely nobody cared about security. The kind of
awesome code.)
We keep the other formats, because they're (mostly) commonly used and
relatively simple, for UTF16 support (still missing in FFmpeg), and for
the sake of Libav.
Before this commit, this was somehow polled (i.e. not the right way).
Also, selects the correct window when doing --wid=0 (which is another
weird special-case).
Enabling DPMS even though you disabled it globally is pretty unfriendly,
so don't do it. Instead, we only disable DPMS if it was enabled, and
only enable it if we disabled it ourselves.
The other way should never happen (disabling DPMS permanently), unless
mpv crashes during playback.
It can easily happen that mp_time_us_to_timespec() gets a time in the
past, and then the time difference will be negative. Regression
introduced in commit f47a4fc3.
Also fix an underflow check in mp_add_timeout().
Currently, it's unused. Still keep the file, because it's not unlikely
we'll need it again, and removing/readding the include statements for
this file is too annoying.
This is probably a good idea, because it would make it easier for
software embedding mpv to configure the mpv parts, without requiring the
host program to provide explicit mechanisms for this (other than calling
mpv_load_config_file()).
The code paths for setting options by string and by direct "raw" value
were too different, which resulted in some weird code. Make the code
paths closer to each other.
Also, use this to remove the weirdness in the mpv_set_option()
implementation.
Use the time as returned by mp_time_us() for mpthread_cond_timedwait(),
instead of calculating the struct timespec value based on a timeout.
This (probably) makes it easier to wait for a specific deadline.
When the player is paused, and video filters are changed, an exact seek
is executed to refresh the display. Increase the exactness of the seek
in this case; this reuses the code used for frame backstepping.
It might help in cases where seeking is very imprecise, such as with
transport streams.
If a property is notified as changed, and then again (before the change
notification is returned to the client), and the second change is a
sporadic change (i.e. nothing actually changed) and the change
notification was associated with with a data type, it could happen that
a change was "overlooked", because it would detect no change on the
second notification.
This is actually a pretty annoying corner case, due to the annoying way
we do things, so just store both the previously returned _and_ the newly
obtained property value. then we always compare with the user value to
check for a change, excluding any possibility of a missed change.
Note that we don't (can't/shouldn't) care if a value changes, and then
changes back; it's fine if that doesn't generate a notification. This is
due to how property notifications are supposed to be coalesced.
Reduces some code-duplication.
Just call DPMSEnable/DPMSDisable, instead of DPMSForceLevel when
reenabling DPMS. "Force" sounds evil, and messing with DPMS is already
pretty evil. I'm not even sure that we should.
XGetWindowProperty is a really bad API, almost as if the NSA designed
it. The wrapper takes care of verifying the return values and handle
corner cases.
The window "gravity" influences how placement interacts with WM added
borders (i.e. from decorations). This is probably what the code removed
in commit c14721c8 was about.
In theory, we'd probably want to set the gravity depending on the
relative placement requested by the user (so that it's possible to line
up the top/left video pixel with the monitor corner, as well as the
bottom/right pixel - but that would be too complicated, and who cares
after all?).
I'm also not sure whether CenterGravity really uses the top/left corner
as reference point (instead of making coordinates relative to the window
center), but empirically it's correct.
Try to get the "new" code path (using NetWM/EWMH) free of hacks done for
the sake of old WMs or the no-WM case.
Implement --fs-screen using _NET_WM_FULLSCREEN_MONITORS.
Keeps the window centered on resize. Seems nicer. (Although it's worse
if 1. the default placement of the WM puts it into a monitor corner,
and 2. you switch to a larger video.)
It was added with 3813c685 in 2004. I'm not really sure why this gravity
stuff would be needed; apparently it has to do with misplacements with
broken WMs and had to be changed on fullscreen. Just get rid of it; it
works perfectly fine without on modern WMs.
The thread discussing this is here:
http://mplayerhq.hu/pipermail/mplayer-dev-eng/2004-July/027674.html
When writing a video to foo.mp3, the user's intention is clearly to drop
the video stream, and similarly, when writing to foo-%d.png, the
intention is clearly to drop the audio stream. Now, explicit
specification of --no-audio or --no-video is no longer necessary in
these cases.
This should get rid of some flickering. Since this actually skips all
the wacky fullscreening code on startup, this might lead to certain
wacky features to stop working. In this case, you'll have to use the
--x11-fstype option, and disable _NETWM_STATE_FULLSCREEN usage.
vo_x11_map_window() was attempting to clear the window on map. However,
it did so immediately after the map request. It probably assumed that
the drawing calls for clearing the window would be queued along with the
map request, and then executed in the right order. However, this
assumption was wrong - the map request first has to go to the window
manager (I guess?), so a lot of things happen before the window is even
mapped.
Fix this by moving the call to the MapNotify message handler, when the
window (apparently) becomes really visible.
I also tried to set CWBackPixel to black instead, but this seemed to
result in flickering on manual resizing.
This blocks everything, until the window is actually reported as mapped.
This fixes the race condition between VO initialization and mapping the
window, which resulted in possibly different window sizes, leading to an
immediate redraw, visible as flashing.
Note that if the map event never comes for some reason, we're out of
luck and will block forever.
It could in theory happen that the filter loop will enter a blocking
wait, even though it could make progress by emptying the list of
already-filtered images. I'm not quite sure if this could actually cause
a real issue - probably not.
VapourSynth won't just filter multiple frames at once on its own. You
have to request multiple frames at once manually. This is what this
commit introduces: a sub-option controls how many frames will be
requested at once. This also changes the semantics of the maxbuffer sub-
option, now renamed to buffered-frames.
The situation has changed a bit since the days of mplayer2, so we can
use more/less diplomatic wording. Merge the two sections listing
changes from MPlayer and mplayer2. Mention the client API and Lua
scripting as alternatives to slave mode.
I'm calling MPlayer code "horrible". This is not meant as an offense,
but after turning around almost every line of MPlayer code, I believe
I have a right to say this. Sorry. I would say that MPlayer has a
surprisingly sane and simple architecture (for what it is), but much
of it drowned under a load of evil hacks or not-cleaned-up-yet code.
This started as a bunch of smaller changes to make the old configure
script maintainable with minimum effort. It ended up as complete
rewrite, because at once point I started to like shell programming (I
hope this sickness is curable), and I wanted to see how small I can
make the configure script. The typical configure test is now 1 or 2
lines big, located in 1 or 2 places, instead of >15 lines and being
spread over 5 or 6 places.
The main "trick" is factoring the tests into a few generic, commonly
needed tests, instead of writing everything manually.
A lot of effort was spent on making the waf based build-system work
properly on all supported platforms, while the old configure script was
neglected. It seems that nobody maintains the non-Linux parts of the
configure script anymore, and all improvements go into the waf scripts.
Thus it makes no sense anymore to maintain the non-Linux parts. They're
just dead weight. Remove them completely.
Also apply some additional simplifications. For example, listing
enabled/disabled VO modules seems like a waste of effort.
These were in the old configure script too.
Two flags are explicitly tested, because I have no idea how widespread
support for them is, and testing them is just easier than trying to look
them up in various gcc/clang manuals. There are people using gcc 4.2
out there, so some caution is warranted.
waf has a proper test, but no test was added to old-configure.
This means that on OSX, old-configure is not supposed to work anymore.
But still allow it to compile cleanly on OSX.
Previous to this commit, read_chunk was not set in stream_smb. The
cache was therefore filled in small 8K chunks. This resulted in poor
performance when compared to, for example, smbnetfs on the same
network.
The value of 128k is chosen both because it is emperically
the "levelling off point" for throughput into mpv's cache, and because
it is the value chosen by smbnetfs when serving smb shares to
mpv.
Note that this change has no effect unless --cache is explicitly
specified as smb:// streams do not activate cache by default. This is
because the default cache size of 320K is so small it actually makes
smb:// perfomance worse. For best results use at least --cache=1024.
This didn't quite work. The main issue was that get_space tries to be
clever to reduce overall buffering, so it will cause the playloop to
decode and queue only as much audio as is needed to refill the AO in
reasonable time. Also, even if ignoring the problem, the logic of the
previous commit was slightly broken. (This required a few retries,
because I couldn't reproduce the issue on my own machine.)
When the audio buffer went low, but could not be refilled yet, it could
happen that the AO playback thread and the decode thread could enter a
wakeup feedback loop, causing up to 100% CPU usage doing nothing. This
happened because the decoder thread would wake up the AO thread when
writing 0 bytes of newly decoded data, and the AO thread in reaction
wakes up the decoder thread after writing 0 bytes to the AO buffer.
Fix this by waking up the decoder thread only if data was actually
played or queued. (This will still cause some redundant wakeups, but
will eventually settle down, reducing CPU usage close to ideal.)
This shouldn't matter, but it's probably better if the code to check is
valid - otherwise an extremely clever compiler might fail to compile it,
and the feature would be misdetected. (Probably.)
Found by cppcheck.
Caused failure to detect .pls files, because they were misdetected as
m3u. The problem is that "forcing playlist files" and "forcing a
specific playlist format" are not really treated separate, and in both
cases p->force is set to true. This made m3u detect all files as m3u
if --playlist was used. So correctly check whether the file format is
actually being probed or not.
These are now equivalent to combining commands with the "cycle pause" or
"set pause" commands, and thus are not needed anymore. They were also
obscure and undocumented.
mpv supports per-file config files, basically filename+".conf". We use
a static buffer for the new filename, and if that buffer is too small,
we print a warning. This is confusing for e.g. long URLs, so just hide
the warning by default.
Why not dynamically allocate the buffer? Who cares.
I don't think this is really a very good idea because it is conceptually
incorrect but other prominent multimedia programs use this approach
(VLC and xbmc), and it seems to make the conversion more robust in certain
cases.
For example it has been reported, that configuring a receiver that can output
7.1 to output 5.1, will make CoreAudio report 8 channel descriptions, and the
last 2 descriptions will be tagged kAudioChannelLabel_Unknown.
Fixes#737
The code was falling back to the full waveext chmap_sel when less than 2
channels were detected. This new code is slightly more correct since it only
fills the chmap_sel with the stereo or mono chmap in the fallback case.
CoreAudio supports 3 kinds of layouts: bitmap based, tag based, and speaker
description based (using either channel labels or positional data).
Previously we tried to convert everything to bitmap based channel layouts,
but it turns out description based ones are the most generic and there are
built-in CoreAudio APIs to perform the conversion in this direction.
Moreover description based layouts support waveext extensions (like SDL and
SDR), and are easier to map to mp_chmaps.
This might shift bits into the sign, which is undefined behavior. Making
the right operand unsigned was supposed to help with this, but it seems
it did nothing, and C99 makes the result type dependent on the left
operand only.
Use the exact floating point value, instead of a broken integer
constant. The expression calculating the constant probably relied on
undefined behavior, because it left-shifts a negative value.
This also changes the type of the constant to double, which is perfectly
fine, and maybe better than an integer constant.
Preparation so that various things related to video can run in different
threads. One part to this is making the video surface pool safe.
Another issue is the preemption mechanism, which continues to give us
endless pain. In theory, it's probably impossible to handle preemption
100% correctly and race-condition free, unless _every_ API user in the
same process uses a central, shared mutex to protect every vdpau API
call. Otherwise, it could happen that one thread recovering from
preemption allocates a vdpau object, and then another thread (which
hasn't recovered yet) happens to free the object for some reason. This
is because objects are referenced by integer IDs, and vdpau will reuse
IDs invalidated by preemption after preemption.
Since this is unreasonable, we're as lazy as possible when it comes to
handling preemption. We don't do any locking around the mp_vdpau_ctx
fields that are normally immutable, and only can change when recovering
from preemption. In practice, this will work, because it doesn't matter
whether not-yet-recovered components use the old or new vdpau function
pointers or device ID. Code calls mp_vdpau_handle_preemption() anyway to
check for the preemption event and possibly to recover, and that
function acquires the lock protecting the preemption state.
Another possible source of potential grandiose fuckup is the fact that
the vdpau library is in fact only a tiny wrapper, and the real driver
lives in a shared object dlopen()ed by the wrapper. The wrapper also
calls dlclose() on the loaded shared object in some situations. One
possible danger is that failing to recreate a vdpau device could trigger
a dlclose() call, and that glibc might unload it. Currently, glibc
implements full unloading of shared objects on the last dlclose() call,
and if that happens, calls to function pointers pointing into the shared
object would obviously crash. Fortunately, it seems the existing vdpau
wrapper won't trigger this case and never unloads the driver once it's
successfully loaded.
To make it short, vdpau preemption opens up endless depths of WTFs.
Another issue is that any participating thread might do the preemption
recovery (whichever comes first). This is easier to implement. The
implication is that we need threadsafe xlib. We just hope and pray that
this will actually work. This also means that once vdpau code is
actually involved in a multithreaded scenario, we have to add
XInitThreads() to the X11 code.
Use the newly provided mp_vdpau_handle_preemption() function, instead of
accessing mp_vdpau_ctx fields directly. Will probably make multithreaded
access to the vdpau context easier.
Mostly unrelated to the actual changes, I've noticed that using hw
decoding with vo_opengl sometimes leads to segfaults inside of nvidia's
libGL when doing the following:
1. use hw decoding + vo_opengl
2. switch to console (will preempt on nvidia systems)
3. switch back to X (mpv will recover, switches to sw decoding)
4. enable hw decoding again
5. exit mpv
Then it segfaults when mpv finally calls exit(). I'll just blame nvidia,
although it seems likely that something in the gl_hwdec_vdpau.c
preemption handling triggers corner cases in nvidia's code.
This was broken for some time, and it didn't recover correctly.
Redo decoder display preemption. Instead of trying to reinitialize the
hw decoder, simply fallback to software decoding. I consider display
preemption a bug in the vdpau API, so being able to _somehow_ recover
playback is good enough.
The approach taking here will probably also make it easier to handle
multithreading.
Move the code that copies the dylib's to the bundle to a new script
(dylib-unhell.py) which is called by osxbundle.py.
dylib-unhell is about 20x faster than the previous implementation. This is
accomplished by removing superflous shell-out operations which are kept track
of using an in memory tree of all the needed dependencies. Moreover the
shell-outs have been further optimized by not requiring a complete shell for
every operation and just using subprocess.call (which is equivalent to Popen).
The many "boxes" in this entry were causing rst2pdf to fail because it
couldn't figure out where to break the page. Make the boxes smaller by
removing semi-redundant examples. Also try and make surrounding text a
little shorter by rewording.
This is done after filters, so things like framerate-doubling
deinterlacing is accounted for.
Unfortunately, framedropping can cause inaccuracies (especially after
precise seeks), and we can't really know when that happens. Even though
we know that the decoder might drop a frame if we request it to do so,
we don't know when the dropped frame will start or stop affecting the
video filter chain. Video filters can have frames buffered, and we
can't tell at which point the dropped frame would have been output.
It's not even possible to mark a discontinuity after seek, because
again we don't know if the filter chain still has the discontinuity
within its buffers.
So we have to live with the fact that the output of this property can
be completely broken after seek, unless --no-hr-seek-framedrop is used.
This allows disabling of decoder framedrop during hr-seek.
It's basically another useless option, but it will help exploring
whether this framedropping really makes seeking faster, or whether
disabling it helps with precise seeking (especially frame backstepping).
Until recently, the VO was an unavoidable part of the seeking code path.
This was because vdpau deinterlacing could double the framerate, and hr-
seek and framestepping etc. all had to "see" the additional frames. But
we've removed the frame doubling from the vdpau VO and moved it into a
video filter (vf_vdpaupp), and there's no reason left why the VO should
participate in seeking.
Instead of queuing frames to the VO during seek and skipping them
afterwards, drop the frames early.
This actually might make seeking with vo_vdpau and software decoding
faster, although I haven't measured it.
Now we avoid calling update_video() twice on reconfig (once to check
whether there are still new frames, and again to actually do the
reconfig). Instead, we check whether there's still something going on
before calling update_video() at all, and depending on that
update_video() will be allowed to reconfig or not.
This will simplify some things later.
It doesn't look like vo_wayland_config() necessarily sets this flag, so
it seems safer to trigger an explicit resize. This accounts for the case
when playing a new file with different size than the one before.
Currently, vo_reconfig() calculates the requested window size and sets
the vo->dwidth/dheight fields _if_ VOCTRL_UPDATE_SCREENINFO is
implemented by the VO or the windowing backend. The window size can be
different from the display size if e.g. the --geometry option is used.
It will also set the vo->dx/dy fields and read vo->xinerama_x/y.
It turned out that this is very backwards and actually requires the
windowing backends to workaround these things. There's also
MPOpts.screenwidth/screenheight, which used to map to actual options,
but is now used only to communicate the screen size to the vo.c code
calculating the window size and position.
Change this by making the window geometry calculations available as
separate functions. This commit doesn't change any VO code yet, and just
emulates the old way using the new functions. VO code will remove its
usage of VOCTRL_UPDATE_SCREENINFO and use the new functions directly.
Because the http playlist URL I had for testing claimed to be m3u by
file extension and mime type, but didn't have the header.
Note that this actually changes behavior only in the case the format is
detected by mime type. Then p->force will be set before calling the
parser, and the header becomes optional.
Changing --softvol-max and then resuming would change the volume level
on resume to something different than the original volume. This is
because the user volume setting is always between 0-100, and 100
corresponds to --softvol-max gain.
Avoid that changing -softvol-max and resuming an older file could lead
to a too loud volume level by refusing to restore if --softvol-max
changed.
Commit 433161 actually broke vo_opengl (and maybe others), because
config_ok is not necessarily set correctly yet _during_ reconfig. So a
vo_get_src_dst_rects() call during reconfig did nothing.
When the VO was not initialized with vo_reconfig(), or if the last
vo_reconfig() failed, changing panscan would cause a crash due to
vo_get_src_dst_rects() dereferencing vo->params (NULL if not
configured).
Just do nothing if that happens, as there is no video that could be
displayed anyway.
Doesn't really seem to be much of use. Get rid of the remaining uses of
it.
Concerning vo_opengl_old, it seems uninitGl() works fine even if called
before initialization.
Make it more suitable for chaining. This means a function formatting a
value to a string using a static buffer can work exactly like
mp_snprintf_append itself.
Also rename it to mp_snprintf_cat, because that's shorter.
Basically, extract the option table from DOCS/man/en/changes.rst, and
search the table if an option wasn't found. If there's an entry about
it, print it. Hopefully this behavior is slightly more userfriendly.
This is strictly bound to option names. It doesn't work for option
values, nor does it attempt to emulate the old option.
The comment says that it wakes up the main thread if 50% has been
played, but in reality the value was 0.74/2 => 37.5%. Correct this. This
probably changes little, because it's a very fuzzy heuristic in the
first place.
Also move down the min_wait calculation to where it's actually used.
mp3 has a hack lowering the probescore for format detection. This is
because detecting mp3s is hard due to their nature, and the fact that
ID3v2 tags are sometimes several megabytes big.
When playing mp3 from network, the mime-type is usually set, and that
matches the format hack entry meant for webradios, overriding the normal
mp3 entry. This can lead to network mp3s not being detected. Lower the
network case to the same probescore as on-disk mp3s. The difference is
that for network mp3s, we don't load the full probe-buffer, and we lower
the amount of audio the demuxer will read to collect data on opening
(0.5 seconds instead of typically 5 seconds).
This was a minor code duplication between vf_vdpaupp.c and vo_vdpau.c.
(In theory, we could always require using vf_vdpaupp with vo_vdpau, but
I think it's better if vo_vdpau can work standalone.)
Also remove MSGL_SMODE and friends.
Note: The indent in options.rst was added to work around a bug in
ReportLab that causes the PDF manual build to fail.
When loading a video, and a script reacts to MPV_EVENT_VIDEO_RECONFIG,
and the script inserts a video filter, the first frame can be skipped.
This happens simply because the first frame is (usually) still queued in
the video filter chain, and changing the filter chain will drop all
queued frames. So this is just a corner case that just happens in a
weird situation.
But it's still annoying when having such a script, and starting
something where the first frame is very visible, and not starting in
paused mode. (All in all, a corner case.) Do this by immediately queuing
1 filtered frame to the VO immediately after reconfig, instead of
leaving it to the video loop doing it as "incremental" work. Simply
fallthrough to the next case. We must not overwrite "r" in this case,
because that contains the current status.
Note that the first frame will not be filtered using the inserted
filter.
This wasn't really fine, and could (perhaps) cause weird corner cases on
reinit or when the player was paused.
Before eb9d20, video_left was also set to true if vo->frame_loaded was
set, and this variable basically indicated whether the previous
update_video() call was successful. This was overlooked when changing
everything. Simply always call update_video(), it should be equivalent.
This can happen when the input stream is somehow blocking on network,
and the user still send input in one way or another, and one of the
commands is a compound command ("cmd a ; cmd b").
Apparently the value of a pointer is "indeterminate" after a free()
call, even if you never dereference the pointer after the free. Since
talloc_free() calls free(), this applies here.
This affects the return value of mp.script_name, the "client name"
(what's returned by mpv_client_name()) and all associated features, as
well as the mpv terminal output module prefix when scripts print
something.
As discussed in #748.
Change how the video decoding loop works. The structure should now be a
bit easier to follow. The interactions on format changes are (probably)
simpler. This also aligns the decoding loop with future planned changes,
such as moving various things to separate threads.
vf_fix_img_params() takes care of overwriting image parameters that are
normally not set correctly by filters. But this makes no sense for input
images. So instead, check that the input is correct.
It still has to be done for the first input image, because that's used
to handle some overrides (see video_reconfig_filters()).
These replace vf_read_output_frame(), although we still emulate that
function. This change is preparation for another commit (and this is
basically just to reduce the diff and signal/noise ratio in that
commit).
Basically, if we feed the filter a new image even after the EOF state
has been reached (e.g. because the input stream "recovered"), we want
the filter to restart, instead of returning an error forever.
Some non-deinterlacing filters (potentially denoising) also use
additional frames for filtering. The vdpau docs suggest providing at
least 1 future and 2 past _fields_, which means we need to provide 1
past frame (the future field is already the other field of the current
field, and both fields are in the same frame).
We can easily achieve this by buffering an additional frame in the non-
deint case.
Remove the special casing of vo_vdpau vs. other VOs. Replace the
complicated interaction between vo.c and vo_vdpau.c with a simple queue
in vo.c. VOs other than vdpau are handled by setting the length of the
queue to 1 (this is essentially what waiting_mpi was).
Note that vo_vdpau.c seems to have buffered only 1 or 2 frames into the
future, while the remaining 3 or 4 frames were past frames. So the new
code buffers 2 frames (vo_vdpau.c requests this queue length by setting
vo->max_video_queue to 2). It should probably be investigated why
vo_vdpau.c kept so many past frames.
The field vo->redrawing is removed. I'm not really sure what that would
be needed for; it seems pointless.
Future directions include making the interface between playloop and VO
simpler, as well as making rendering a frame a single operation, as
opposed to the weird 3-step sequence of rendering, drawing OSD, and
flipping.
Give up on the deint_filters[] array, and probe using explicit code
instead. Add additional checks to test the pixel format to avoid
annoying warnings when a hardware deinterlacer is inserted when the
current video chain is obviously incompatible.
The previous commits changed vo_vdpau so that these options are set by
vf_vdpaupp, and the corresponding vo_vdpau were ignored. But for
compatibility, keep the "old" options working.
The value of this is questionable - maybe the vo_vdpau options should
just be removed. For now, at least demonstrate that it's possible.
The "deint" suboption still doesn't work, because the framerate doubling
logic required for some deint modes was moved to vf_vdpaupp. This
requires more elaborate workarounds.
This is slightly incomplete: the mixer options, such as sharpen and
especially deinterlacing, are ignored. This also breaks automatic
enabling of interlacing with 'D' or --deinterlace. These issues will be
fixed later in the following commits.
Note that we keep all the custom vdpau queue stuff. This will also be
simplified later.
This uses mp_vdpau_mixer_render(). The benefit is that it makes vdpau
deinterlacing just work. One additional minor advantage is that the
video mixer creation code is factored out (although that is a double-
edged sword).
This factors out some code from vo_vdpau.c, especially deinterlacing
handling. The intention is to use this for vo_vdpau.c to make the logic
significantly easier, and to use it for vo_opengl (gl_hwdec_vdpau.c) to
allow selecting deinterlace and postprocessing modes.
As of this commit, the filter actually does nothing, since both vo_vdpau
and vo_opengl treat the generated images as normal vdpau images. This
will change in the following commits.
It now inserts no filters and does nothing until the hot-key is pressed.
This makes it more suitable to be put in ~/.mpv/lua.
When the hot-key is pressed, it now inserts the cropdetect filter and
waits 1 second (or a --lua-opts specified duration) before gathering
the cropdetect metadata and inserting the appropriate crop filter. A
second press of the hotkey removes the crop.
It might have been nice not to do this so that metadata could
accumulate accross seeks, but it seems libavfilter looses its copy
anyway on recreate_graph.
lavfi would segfault due to a NULL dereference if it was asked for its
metadata and none had been allocated (oops). This happens for libav
which has no concept of filter metadata.
Commit 5e4e248 added a mp_image_params field to mp_image, and moved many
parameters to that struct. display_w/h was left redundant with
mp_image_params.d_w/d_h. These fields were supposed to be always in
sync, but it seems some code forgot to do this correctly, such as
vf_fix_img_params() or mp_image_copy_attributes(). This led to the
problem in github issue #756, because display_w/_h could become
incorrect.
It turns out that most code didn't use the old fields anyway. Just
remove them. Note that mp_image_params.d_w/d_h are supposed to be always
valid, so the additional checks for 0 shouldn't be needed. Remove these
checks as well.
Fixes#756.
In theory, returning the screenshot with original pixel aspect would
allow avoiding scaling them with image formats that support non-square
pixels, but in practice this isn't used anyway (nothing seems to
understand e.g. jpeg aspect ratio tags).
We only support them for input. The frame properties of output frames
are ignored (except frame durations).
Properties not set for now: _ChromaLocation, _Field, _FieldBased
Set _DurationNum/_DurationDen on each VS frame, instead of
_AbsoluteTime. The duration is the difference between the timestamp of
the frame and the next frame, and when receiving filtered VS frames, we
convert them back to an absolute PTS by summing them.
We pass the timestamps with microsecond resolution. mpv uses double for
timestamps internally, so we don't know the "real" timebase or FPS. VS
on the other hand uses fractions for frame durations. We can't pass
through the numbers exactly, but microseconds ought to be enough to be
even safe from accumulating rounding errors.
Since this leaks video images, and the player keeps feeding new images
to the fitler even if it fails, this would probably have disastrous
consequences.
Or in other words, add support for properly draining remaining frames
from video filters. vf_yadif is buffering at least one frame, and the
buffered frame was not retrieved on EOF.
For most filters, ignore this for now, and just adjust them to the
changed semantics of filter_ext. But for vf_lavfi (used by vf_yadif),
real support is implemented. libavfilter handles this simply by passing
a NULL frame to av_buffersrc_add_frame(), so we just have to make
mp_to_av() handle NULL arguments.
In load_next_vo_frame(), we first try to output a frame buffered in the
VO, then the filter, and then (if EOF is reached and there's still no
new frame) the VO again, with draining enabled. I guess this was
implemented slightly incorrectly before, because the filter chain still
could have had remaining output frames.
This extracts the scheduling logic to a single function which is nicer to keep
it consistent.
Additionally make sure we don't schedule sync operations from a sync operation
itself since that could cause deadlocks (even if it should not be happening
with the current code).
Previously the window could be made to completly exit the screen with a
combination or moving it close to an edge and halving it's size (via cmd+0).
This commit address the problem in the most simple way possibile by
constraining the window to the closest edge in these edge cases.
This fixes a couple of issues with the Cocoa `--native-fs` mode, primarily:
- A ghost titlebar at the top of the screen in full screen:
This was caused by the window constraining code kicking in during
fullscreen. Simply returning the unconstrained rect from the constraining
method fixes the problem.
- Incorrect behavior when using the titlebar buttons to enter/exit
fullscreen, as opposed to the OSD button.
This was caused by mpv's internal fullscreen state going out of sync with
the NSWindow's one. This was the case because `toggleFullScreen:`
completely bypassed the normal event flow that mpv expects.
Signed-off-by: Ryan Goulden <percontation@gmail.com>
Change style for mpv, simplify and refactor some of the constraining code.
Signed-off-by: Stefano Pigozzi <stefano.pigozzi@gmail.com>
I thought the "_" in "_AbsoluteTime" was part of the documentation
markup.
This still doesn't help us with VS filters that change timing;
apparently you must use frame durations instead.
Make the filter apply the pixel aspect ratio of the input to the output.
This is more useful than forcing 1:1 PAR when playing anamorphic video
such as DVDs.
VapourSynth itself actually allows passing through the aspect ratio, but
it's in a not very useful form for us: it's per video-frame instead of
constant (i.e in VSVideoInfo). As long as we don't have a way to allow a
filter to spontaneously change output parameters, we can't use this.
(And I don't really feel like making this possible.)
Our code currently tries to link -lpthread and adds stuff like -D_REENTRANT
based on the target platform.
GCC actually supports to just pass a -pthread compiler and linker flag that
will automatically enable threading and define the correct symbols for the
platform, so let's try to just use that as our first choice.
clang also supports -pthread but it must be used only as a compiler flag,
so we also take care of that scenario with this commit.
-Wempty-body is not available on all gcc versions but we were using it
unconditionally. Also remove the usage from the clang case. clang still
defines `__GNUC__` so it still gets all the gcc specific flags.
This should fix the build on systems with older gcc versions like OpenBSD which
still comes bundled with gcc 4.2 for license issues.
One problem is that for example stdio functions won't restart syscalls
manually, and instead treat EINTR as an error. So passing SA_RESTART is
the only sane thing to do, unless you have special requirements, which
we don't.
VP9 packets can contain 2 frames in some video packets (from which 1
frame is invisible). Due to a design mismatch between libvpx and the
libavcodec vp9 decoder, libvpx can take the "full" packets, but lavc vp9
can not. The consequence is that we have to split the packets if we want
to feed them to the lavc codec.
This is not entirely correct yet: timestamp handling is missing.
--demuxer=lavf and ffmpeg native utilities have the same problem. We can
fix this only once the ffmpeg VP9 parser is fixed.
This essentially reverts commit cca13efb. The code in the if was
supposed to be run only if the mouse button was down, because in this
case the mouse area is never considered to be left. Since it was run
for every mouse button, mouse_leave wasn't sent.
Fixes#745.
pthread_equal() returns 0 if the threads are not the same, and somehow
I got that wrong. The worst is that I actually explicitly checked the
manpage when I wrote this code.
This used an unnamed union, which is allowed in GNU C and C11, but not
C99. This broke the build with some older compilers.
Replaces pull request #744.
This reverts commit 6e34b0ec1f.
There has always been an error message "proxy already has a listener" and
I couldn't reproduce where it is comming from until now. The display interface
already has a listener and we can't overwrite it. Now remove the code and avoid
this error message.
Conflicts:
video/out/wayland_common.c
The interrupt callback will can be called from another thread if the
cache is enabled, and the stream disconnects. Then stream_reconnect()
will call this function from within the cache thread.
mp_input_check_interrupt() is not thread-safe due to read_events() not
being thread-safe. It will call input callbacks added with
mp_input_add_fd() - these callbacks lead to code not protected by locks,
such as reading X11 events.
Solve this by adding a stupid hack, which checks whether the calling
thread is the main playback thread (i.e. calling the input callbacks
will be safe). We can remove this hack later, but it requires at least
moving the VO to its own thread first.
This used global variables for the asynchronous interrupt callback.
Pick the simple and dumb solution and stuff the callback into
mpv_global. Do this because interrupt checking should also work in the
connect phase, and currently stream creation equates connecting.
Ideally, this would be passed to the stream on creation instead, or
connecting would be separated from creation. But since I don't know yet
which is better, and since moving stream/demuxer into their own thread
is something that will happen later, go with the mpv_global solution.
This is the only function which actually used the time argument of
stream_check_interrupt(). Considering that the whole player freezes
anyway, this is not worth the complication.
Also generally reduce the maximum wait time due to timeout. Introduce
exponential backoff, which makes the first reconnect retries faster, but
still waits up to 500ms in the later retries.
For some reason, some files appear to have broken mp3 packets, or at
least in a form that libavcodec can't deal with. The audio in the sample
file in question could not be decoded using libavcodec.
The problematic file had variable packet sizes, and the libavcodec
decoder kept printing "mp3: Header missing" for each packet it was fed.
Remuxing with mkvmerge fixes the problem. The mp3 data is probably not
VBR, and remuxing resulted in fixed-size mp3 frames. So I don't know why
the sample file was muxed this way - it might just be incorrect.
The sample file had "libmkv 0.6.4" as MuxingApp (although I could not
get mkvinfo to print this element, maybe the file uses an incorrect
element ID), and "HandBrake 0.9.4" as WritingApp.
Note that the libmpg123 decoder does not have any issues with it. It's
probably more robust, because libmpg123 was made to decode whole mp3
files, not just single frames.
Fixes issue #742.
The here documented guarantee might simplify code using this mechanism
a lot, because it becomes unnecessary to invent a separate mechanism to
make the mp_dispatch_queue_process loop exit after processing a dispatch
callback. (Instead, the dispatch callback can set a flag, and the caller
of mp_dispatch_queue_process can check it.)
This allows client API users and Lua scripts to side-step the pretty
horrible video filter string "language" (although it's back and can't be
avoided when using libavfilter).
The wakeup is needed to make mp_dispatch_queue_process() return if
suspension is not needed anymore - which is only the case when the
request count reaches 0.
The assertion added with this commit always has/had to be true.
Note that this mechanism is similarly "unreliable" as for example
pthread_cond_timedwait(). Trying to target the exact wait time will just
make it more complex.
The main use case for this is for threads which either use the dispatch
centrally and want mp_dispatch_queue_process to do a blocking wait for
new work, or threads which have to implement timers. For the former,
anything is fine, as long as they don't have to do active waiting for
new works. For the former, callers are better off recalculating their
deadline after every call.
This is much simpler, leaves fairness isues etc. to the operating
system, and will work better with threading-related debugging tools.
The "trick" to this is that the lock can be acquired and held only while
the queue is in suspend mode. This way we don't need to make sure the
lock is held outside of mp_dispatch_queue_process, which would be quite
messy to get right, because it would have to be in locked state by
default.
mp_dispatch_queue_process() releases the queue->lock mutex while
processing a dispatch callback. But this allowed mp_dispatch_lock() to
grab the "logical" lock represented by queue->locked. Grabbing the
logical lock is not a problem in itself, but it can't be allowed to
happen while the callback is still running.
Fix this by claiming the logical lock while the dispatch callback is
processed. Also make sure that the thread calling mp_dispatch_lock() is
woken up properly.
Fortunately, this didn't matter, because the locking function is unused.
This was part of osdep/threads.c out of laziness. But it doesn't contain
anything OS dependent. Note that the rest of threads.c actually isn't
all that OS dependent either (just some minor ifdeffery to work around
the lack of clock_gettime() on OSX).
Add the event FD after preinit, remove it before destroy. There's no
need to do it on vo_config, and there's no need to remove the event
FD when vo_config fails.
This should fix some issues, such as not being able to set the
"no-video" option with MPV_FORMAT_FLAG.
Note that this changes semantics a bit. Now setting an option strictly
overwrite it, even if the corresponding command line option does not.
For example, if we change --sub to append by default, then setting the
"sub" option via the client API would still never append. (Oddly, this
also applies to --vf-add, which will overwrite the old value when using
the client API.)
I'm doing this because there's no proper separation between the command
line parser and setting an option using the MPV_FORMAT_STRING format.
Maybe the solution to this mess would be adding format aware code (i.e.
m_option_set_node) to every option type, and falling back to strings
only if needed - but this would mean that you couldn't set e.g. an
integer option using MPV_FORMAT_STRING, which doesn't seem to be ideal
either.
In conclusion, the current approach seems to be most robust, but I'm
open to suggestions should someone find that these semantics are a
problem.
glob() is mandated by POSIX. For the only non-POSIX platform we support,
Windows, we have our own replacement. So the ifdeffery is not needed.
Still leave the checks in the configure scripts, because they have to
decide whether to compile the replacement or not. (Although this could
be special cased to mingw-only, the wscript seems to make this hard.)
Unfortunately, if a VO can't display something as intended, we can just
complain to the user, and leave it at it. But it's still better than
silently displaying things differently with different VOs.
For now, this is used for rotation only. Other things that we should
check includes colorspace and colorlevels stuff.
When using rotation with hw decoding, and the VO does not support
rotation, vf_rotate is attempted to be inserted. This will go wrong, and
after that it can't recover because a vf_scale filter was autoinserted.
Just removing all autoinserted filters before reconfig fixes this.
The Windows port uses CommandLineToArgvW, which doesn't expand wildcards
in command line arguments. Use glob to expand them instead, but only for
non-option arguments.
glob-win.c wasn't big, so it was easier to rewrite it. The new version
supports Unicode, handles directories properly, sorts the output and
puts all its allocations in the same talloc context to simplify the
implementation of globfree.
Notably, the old glob had error checking code, but didn't do anything
with the errors since the error reporting code was commented out. The
new glob doesn't copy this behaviour. It just treats errors as if there
were no more matching files, which shouldn't matter for mpv, since it
ignores glob errors too.
To match the other Windows I/O helper functions, the definition is moved
to osdep/io.h.
If the VO can't do rotation, insert a filter to do this. Note that this
doesn't reuse the filter insertion code from command.c (used by "vf"
input command), because that would end up more complicated: we don't
even want to change the user filter option.
This turned out much more complicated than I thought. It's not just a
matter of adjusting the texture coordinates, but you also have to
consider separated scaling and panscan clipping, which make everything
complicated.
This actually still doesn't clip 100% correctly, but the bug is only
visible when rotating (or flipping with --vf=flip), and using something
like --video-pan-x/y at the same time.
For rotation, we assume that the source image will be rotated within the
VO, so the aspect/panscan code needs to calculate its param using
rotated coordinates. VOs which support rotation natively can use this.
This couldn't rotate by 180°. Add this, and also make the parameter in
degrees, instead of magic numbers.
For now, drop the flipping stuff. You can still flip with --vf=flip or
--vf=mirror. Drop the landscape/portrait stuff - I think this is
something almost nobody will use. If it turns out that we need some of
these things, they can be readded later.
Make it use libavfilter. Its vf_transpose implementation looks pretty
simple, except that it uses slice threading and should be much faster.
Often, user configs set options that are not suitable for encoding.
Usually, playback and encoding are pretty different things, so it makes
sense to keep them strictly separate. There are several possible
solutions. The approach taken by this commit is to basically ignore the
default config settings, and switch to an [encoding] config profile
section instead. This also makes it impossible to have --o in a config
file, because --o enables encode mode.
See github issue #727 for discussion.
Fix all include statements of the form:
#include "libav.../..."
These come from MPlayer times, when FFmpeg was somehow part of the
MPlayer build tree, and this form was needed to prefer the local files
over system FFmpeg.
In some cases, the include statement wasn't needed or could be replaced
with mpv defined symbols.
Not needed anymore. I'm not opposed to having asm, but inline asm is too
much of a pain, and it was planned long ago to eventually get rid fo all
inline asm uses.
For the note, the inline asm use that was removed with the previous
commits was almost worthless. It was confined to video filters, and most
video filtering is now done with libavfilter. Some mpv filters (like
vf_pullup) actually redirect to libavfilter if possible.
If asm is added in the future, it should happen in the form of external
files.
No change in speed (or even slightly faster, though I tested with
progressive solid color video only), and normally we use libavformat's
vf_pullup anyway.
I didn't test the speed, but by default, this filter diverts to
libavfilter already. So this would help only if libavfilter is disabled,
or libavfilter doesn't have vf_noise (like on Libav). For these cases,
we still provide the (possibly but not necessarily) slower C
implementation of vf_noise.
This makes it multiple times slower. However, the output format (packed
YUV) isn't handled efficiently by anything to begin with, and I have no
clue we even have this filter. I guess it's one of these filters which
find some use sometimes, but are not of higher importance, which
justifies removing the faster inline asm.
This can be used to easily extent the status line for one's own needs.
I'm not experienced with lua so a few things could probably be done a
better way.
This is for the sake of multi-key commands again. This could break:
SPACE ignore
SPACE-SPACE command
while this worked:
SPACE-SPACE command
SPACE ignore
The reason being that if the shorter command was first in the list,
it would obviously match, and searching was stopped.
This is for the sake of multi-key combinations (see github issue #718).
Now a multi-key sequence isn't matched if any of the previous keys were
actually mapped.
We only need to track key up/down for a single key. There may be some
minor loss of robustness, but this can (probably) happen only if a VO or
user sends strange or complicated sequences of events. In the normal
case, what we do is more than enough. Most simplification comes from the
fact that mpv is not a game console, and users normally execute only one
action at once.
The window close button is usually mapped to the CLOSE_WIN pseudo-key.
Until now, --input-test treated this pseudo-key like any other key (like
the rest of the input handling code), so you couldn't close the window
in this mode. The manpage had silly instructions and warnings how to
deal with this.
Just always quit when CLOSE_WIN is received, and improve the
instructions.
The input code always supported combinations of multiple keys (even in
MPlayer, although there the code was active really only for mouse
buttons). This was arcance and also made the code more complicated. I
only know of a single person who ever made use of this feature.
Remove this feature, and repurpose some of the support code (e.g.
parsing, display of key combinations, etc.) to handle such multi-
combinations as sequences, instead of keys to be pressed at the same
time. This is much simpler and implements the feature requested in
github issue #718.
This commit will probably cause a bunch of regressions, since the input
handling code has some weird corner cases. I couldn't find any problems
when testing, though.
This avoids too many realloc() calls if the caller is appending to an
audo buffer. This case is actually quite noticeable when using something
that buffers a large amount of audio.
For some reason, the buffered_audio variable was used to "cache" the
ao_get_delay() result. But I can't really see any reason why this should
be done, and it just seems to complicate everything.
One reason might be that the value should be checked only if the AO
buffers have been recently filled (as otherwise the delay could go low
and trigger an accidental EOF condition), but this didn't work anyway,
since buffered_audio is set from ao_get_delay() anyway at a later point
if it was unset. And in both cases, the value is used _after_ filling
the audio buffers anyway.
Simplify it. Also, move the audio EOF condition to a separate function.
(Note that ao_eof_reached() probably could/should whether the last
ao_play() call had AOPLAY_FINAL_CHUNK set to avoid accidental EOF on
underflows, but for now let's keep the code equivalent.)
This replaces translate_key_input with a solution that gives mpv more
control over how keyboard input is converted to unicode. As a result:
- Key up/down events are generated the same way for all keys.
- Dead keys generate their base character instead of being combined with
the following character.
- Many Ctrl and Ctrl+Alt key combinations that were previously broken
are fixed, since it's possible to discover the base keys.
- AltGr doesn't produce special characters when mp_input_use_alt_gr is
false.
This also fixes some logic to do with detecting AltGr and adds proper
UTF-16 decoding.
This was reported with PulseAudio 2.1. Apparently it still has problems
with reporting the correct delay. Since ao_pulse.c still has our custom
get_delay implementation, there's a possibility that this is our fault,
but this seems unlikely, because it's full of workarounds for issues
like this. It's also possible that this problem doesn't exist on
PulseAudio 5.0 anymore (I didn't explicitly retest it).
The check is general and works for all push based AOs. For pull based
AOs, this can't happen as pull.c implements all the logic correctly.
This should probably be an AO function, but since the playloop still has
some strange stuff (using the buffered_audio variable instead of calling
ao_get_delay() directly), just leave it and make it more explicit.
This collects statistics and other things. The option dumps raw data
into a file. A script to visualize this data is included too.
Litter some of the player code with calls that generate these
statistics.
In general, this will be helpful to debug timing dependent issues, such
as A/V sync problems. Normally, one could argue that this is the task of
a real profiler, but then we'd have a hard time to include extra
information like audio/video PTS differences. We could also just
hardcode all statistics collection and processing in the player code,
but then we'd end up with something like mplayer's status line, which
was cluttered and required a centralized approach (i.e. getting the data
to the status line; so it was all in mplayer.c). Some players can
visualize such statistics on OSD, but that sounds even more complicated.
So the approach added with this commit sounds sensible.
The stats-conv.py script is rather primitive at the moment and its
output is semi-ugly. It uses matplotlib, so it could probably be
extended to do a lot, so it's not a dead-end.
This was accidentally completely destroyed with commit 24f1878e. I
didn't notice it when testing, because forward seeking still worked
mostly.
The issue was that dvd_seek_to_time() actually called stream_seek(),
which was supposed to call the byte-level seek function dvd_seek(). So
we have to restore this function, and replace all generic stream calls
with stream_dvd.c internal ones. This also affects stream->pos (now a
random number as far as stream_dvd.c is concerned) and stream_skip().
The recent change of waking up the playback thread using a wakeup pipe
doesn't work on windows, because windows is horrible. So use a condition
variable instead to wake up the thread. To make things worse, jackaudio
is also horrible and "disallows" the use of mutexes, so all we can do is
implementing a half-solution that is not race condition free.
It would probably better to give up on this lock-free crap in the pull-
API audio path.
Mostly untested.
The audio subsystem now wakes up the playback thread explicitly, and we
don't need this anymore.
It still could cause dropouts and such if there are bugs in the recently
introduced audio changes, so this is a thing to watch out for.
Same change as in e2184fcb, but this time for pull based AOs. This is
slightly controversial, because it will make a fast syscall from e.g.
ao_jack. And according to JackAudio developers, syscalls are evil and
will destroy realtime operation. But I don't think this is an issue at
all.
Still avoid locking a mutex. I'm not sure what jackaudio does in the
worst case - but if they set the jackaudio thread (and only this thread)
to realtime, we might run into deadlock situations due to priority
inversion and such. I'm not quite sure whether this can happen, but I'll
readily follow the cargo cult if it makes hack happy.
I'm not quite sure why ao_pulse needs this. It was broken when a thread
to fill audio buffers was added to AO - the pulseaudio callback was
waking up the playback thread, not the audio thread. But nobody noticed,
so it can't be very important. In any case, this change makes it wake up
the audio thread instead (which in turn wakes up the playback thread if
needed).
And also add a function ao_need_data(), which AO drivers can call if
their audio buffer runs low.
This change intends to make it easier for the playback thread: instead
of making the playback thread calculate a timeout at which the audio
buffer should be refilled, make the push.c audio thread wakeup the core
instead.
ao_need_data() is going to be used by ao_pulse, and we need to
workaround a stupid situation with pulseaudio causing a deadlock because
its callback still holds the internal pulseaudio lock.
For AOs that don't call ao_need_data(), the deadline is calculated by
the buffer fill status and latency, as before.
qscale export has been completely removed from Libav 10, and FFmpeg has
an alternative API, so this code does nothing and only causes
deprecation warnings on Libav.
And slightly adjust the semantics of MPV_EVENT_PAUSE/MPV_EVENT_UNPAUSE.
The real pause state can now be queried with the "core-idle" property,
the user pause state with the "pause" property, whether the player is
paused due to cache with "paused-for-cache", and the keep open event can
be guessed with the "eof-reached" property.
This property is set to "yes" if playback was paused due to --keep-open.
The change notification might not always be perfect; maybe that should
be improved.
Currently this is (probably) equivalent to "paused-for-cache", but the
latter is a bit special, while this new property is a bit more general.
One case where they might actually be different is dvdnav menus, but I
haven't checked.
Also add property change notifications for these two properties.
We were relying on vsscript_freeScript() to take care of proper
termination. But it doesn't do that: it doesn't wait for the filters to
finish and exit at all. Instead, it just destroys all objects, which
causes the worker threads to crash sometimes.
Also, we're supposed to wait for the frame callback to finish before
freeing the associated node.
Handle this by explicitly waiting as far as we can. Probably fixes
crashes on seeking, although VapourSynth itself might also need some
work to make this case completely stable.
It's a bit strange to allow this, so get rid of it.
This probably breaks a bunch of user config files.
The client API still allows setting them with MPV_FORMAT_FLAG with a
value of 1 (i.e. true), but I guess this is tolerable.
The most user visible change is that "420p" is now displayed as
"yuv420p". This is what FFmpeg uses (almost), and is also less confusing
since "420p" is often confused with "420 pixels vertical resolution".
In general, we return the FFmpeg pixel format name. We still use our own
old mechanism to keep a list of exceptions to provide compatibility for
a while.
Also, never return NULL for image format names. If the format is unset
(0/IMGFMT_NONE), return "none". If the format has no name (probably
never happens, FFmpeg seems to guarantee that a name is set), return
"unknown".
This re-allows the previous behaviour of being able to reencode with
metadata removed, which is useful when encoding "inconsistently" tagged
data for a device/player that shows file names when tags are not
present.
Before this commit, the filter attempted to keep the vsscript state
(p->se) even when the script was reloaded. Change it to destroy the
script state too on reloading. Now no workaround for LoadPlugin is
necessary, and this also fixes a weird theoretical race condition when
destroying and recreating the mpv source filter.
I hate tabs.
This replaces all tabs in all source files with spaces. The only
exception is old-makefile. The replacement was made by running the
GNU coreutils "expand" command on every file. Since the replacement was
automatic, it's possible that some formatting was destroyed (but perhaps
only if it was assuming that the end of a tab does not correspond to
aligning the end to multiples of 8 spaces).
This is a read-only property that uses VFCTRL_GET_METADATA
to retrieve mp_tags metadata from a filter specified by label
Signed-off-by: wm4 <wm4@nowhere>
These playlist parsers are all what's left from the old mplayer playlist
parsing code. All of it is old code that does little error checking; the
type of C string parsing code that gives you nightmare.
Some playlist parsers have been rewritten and are located in
demux_playlist.c. The removed formats were not reimplemented. ASX and
SMIL use XML, and since we don't want to depend on a full blown XML
parser, this is not so easy. Possibly these formats could be supported
by writing a very primitive XML-like lexer, which would lead to success
with most real world files, but I haven't attempted that. As for NSC, I
couldn't find any URL that worked with MPlayer, and in general this
formats seems to be more than dead.
Move playlist_parse_file() to playlist.c. It's pretty small now, and
basically just opens a stream and a demuxer. No use keeping
playlist_parser.c just for this.
mpv was resizing to the same size before it went to fullscreen, we don't need to schedule a resize because the compositor will send a configure event with the new dimensions and thats when we should do it.
Basically a cheap hack to fix that the --msgmodule prefix will cause an
unwanted linebreak by making the line too long.
Suggested by Hamuko in github issue #710.
Fixes#710.
This is needed if you want to reimplement the status line in lua
I could only test drop-frame-count because I didn't find an easy way to
trigger paused-for-cache and total-avsync-change
Signed-off-by: wm4 <wm4@nowhere>
Mainly meant to apply simple VapourSynth filters to video at runtime.
This has various restrictions, which are listed in the manpage.
Additionally, this actually copies video frames when converting frame
references from mpv to VapourSynth, and a second time when going from
VapourSynth to mpv. This is inefficient and could probably be easily
improved. But for now, this is simpler, and in fact I'm not sure if
we even can references VapourSynth frames after the core has been
destroyed.
Commit 2c2c1203 sorted the output of --list-options, but the same code
ias also used for listing sub-options, such as --vo=scale:help. For sub-
options, the order actually matters.
Until now, --list-options printed options in random order. There
literally wasn't any logic in its order, they just appeared as they were
declared. So just sort them.
Note that we can't sort them in advance, because for certain things
internal to m_config, the order actually matters.
Also we're using strcasecmp(), which is bad (locale dependent), but this
is output intended for human consumption, so it's not a problem.
This used to display the property type, but it was not always correct or
even available. The way the property mechanism works, we can know this
only at runtime.
Otherwise, the client API user could not know why playback was stopped.
Regarding the fact that 0 is used both for normal EOF and EOF on error:
this is because mplayer traditionally did not distinguish these, and in
general it's hard to tell the real reason. (There are various weird
corner cases which make it hard.)
Although this is something really basic, Lua's standard library doesn't
provide anything like this. Probably because there are too many ways to
do it right or wrong.
This code tries to be really careful when dealing with mixed
arrays/maps, e.g. when a table has integer keys starting from 1, making
it look like an array, but then also has other keys.
The stats were retrieved and written on every encode call, instead of
every encode call that actually returned a packet. ffmpeg.c also does it
this way, so it must be "more correct". Fixes 2-pass encoding.
Our own tables have size for only 8 chars, so these sequences must be
rejected. It seems strings of length 8 are still ok, because the code
uses memcmp and not strcmp, so still allow these.
Based on mplayer-svn commit r37129.
I have some doubts that short reads are even allowed/
possible for /dev/js*, does someone know for sure?
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@37132 b3059339-0415-0410-9bf9-f77b7e298cf2
We needed this because the OSD rendering path used GBRP for RGB
rendering, and not all swscale versions supported this conversion. But
recently we've dropped support for very old ffmpeg/libav versions, so
this isn't needed anymore.
This was broken at some unknown point (even before the recent cache
changes). There are several problems:
- stream_dvd returning a random stream position, confusing the cache
layer (cached data and stream data lost their 1:1 corrospondence by
position)
- this also confused the mechanism added with commit a9671524, which
basically triggered random seeking (although this was not the only
problem)
- demux_lavf requesting seeks in the stream layer, which resulted in
seeks in the cache or the real stream
Fix this by completely removing byte-based seeking from stream_dvd. This
already works fine for stream_dvdnav and stream_bluray. Now all these
streams do time-based seeks, and pretend to be infinite streams of data,
and the rest of the player simply doesn't care about the stream byte
positions.
resize_cache() checks the size itself and clamps the size to the valid
range if necessary, so we don't need these checks. In fact, the checks
are different. Also, output the cache size after clamping, instead of
before.
Use NtQueryVolumeInformationFile instead of GetDriveType for detecting
remote filesystems on Windows. This has the advantage of working
directly on the file handle instead of needing a path and it works
unmodified in Cygwin where the previous code wouldn't understand Cygwin
paths or symlinks.
There is some risk in using NtQueryVolumeInformationFile, since it's an
internal function and its behaviour could change at any time or it could
be removed in a future version of Windows, however it's documented[1] in
the WDK and it's used successfully by Cygwin, so it should be fine. If
it's removed, the code should fail gracefully by treating all files as
local.
[1]: http://msdn.microsoft.com/en-us/library/windows/hardware/ff567070.aspx
Signed-off-by: wm4 <wm4@nowhere>
Merge the cache_read function into cache_fill_buffer, since there's
not much reason to keep them separate. Also, simply call read_buffer()
to see if there's any readable data, instead of checking for the
condition manually.
The only tricky part is keeping the cache contents, which is made simple
by allocating the new cache while still keeping the old cache around,
and then copying the old data.
To explain the "Don't use this when playing DVD or Bluray." comment: the
cache also associates timestamps to blocks of bytes, but throws away the
timestamps on seek. Thus you will experience strange behavior after
resizing the cache until the old cached region is exhausted.
The only difference is that the MP_DBG message is not printed anymore if
the current user read position is outside of the current cache range.
(In order to handle seek_limit==0 gracefully in the normal case of
linear reading, change the comparison from ">=" to ">".)
Until now, this could never happen, because new data was simply always
appended to the end of the cache. But for making stream cache resizing
easier, doing it this way seems advantageous. It also makes it harder to
make the internal state inconsistent. (Before this change it could
happen that cache and stream position went out of sync if the read
position was adjusted "inappropriately".)
Until now, cache_read() (which calls read_buffer()) could return short
reads. This was a simplification allowed by the stream interface. But
for cache resizing, it will be more practical to make read_buffer() do
a full read.
Seems like a good idea. One possible bad effect would be slowing down
uncached controls, but they're already slow. The good thing is that
many controls make intrusive changes to the stream (at least controls
which do write accesses), so the cached parameters should be updated.
Some of these property implementations already send notifications on
their own, but most don't. This takes care of them.
Of course this still doesn't handle all propertry changes - this is
impossible without special-casing each property that can change on its
own.
This might be a good idea in order to prevent queuing a frame too far in
the future (causing apparent freezing of the video display), or dropping
an infinite number of frames (also apparent as freezing).
I think at this point this is most of what we can do if the vdpau time
source is unreliable (like with Mesa). There are still inherent race
conditions which can't be fixed.
The strange thing about this code was the shift parameter of the
prev_vs2 function. The parameter is used to handle timestamps before the
last vsync, since the % operator handles negative values incorrectly.
Most callers set shift to 0, and _usually_ pass a timestamp after the
last vsync. One caller sets it to 16, and can pass a timestamp before
the last timestamp.
The mystery is why prev_vs2 doesn't just compensate for the % operator
semantics in the most simple way: if the result of the operator is
negative, add the divisor to it. Instead, it adds a huge value to it
(how huge is influenced by shift). If shift is 0, the result of the
function will not be aligned to vsyncs.
I have no idea why it was written in this way. Were there concerns about
certain numeric overflows that could happen in the calculations? But I
can't think of any (the difference between ts and vc->recent_vsync_time
is usually not that huge). Or is there something more clever about it,
which is important for the timing code? I can't think of anything
either.
So scrap it and simplify it.
vo_vdpau used a somewhat complicated and fragile mechanism to convert
the vdpau time to internal mpv time. This was fragile as in it couldn't
deal well with Mesa's (apparently) random timestamps, which can change
the base offset in multiple situations. It can happen when moving the
mpv window to a different screen, and somehow it also happens when
pausing the player.
It seems this mechanism to synchronize the vdpau time is not actually
needed. There are only 2 places where sync_vdptime() is used (i.e.
returning the current vdpau time interpolated by system time).
The first call is for determining the PTS used to queue a frame. This
also uses convert_to_vdptime(). It's easily replaced by querying the
time directly, and adding the wait time to it (rel_pts_ns in the patch).
The second call is pretty odd: it updates the vdpau time a second time
in the same function. From what I can see, this can matter only if
update_presentation_queue_status() is very slow. I'm not sure what to
make out of this, because the call merely queries the presentation
queue. Just assume it isn't slow, and that we don't have to update the
time.
Another potential issue with this is that we call VdpPresentationQueueGetTime()
every frame now, instead of every 5 seconds and interpolating the other
calls via system time. More over, this is per video frame (which can be
portantially dropped, and not per actually displayed frame. Assume this
doesn't matter.
This simplifies the code, and should make it more robust on Mesa. But
note that what Mesa does is obviously insane - this is one situation
where you really need a stable time source. There are still plenty of
race condition windows where things can go wrong, although this commit
should drastically reduce the possibility of this.
In my tests, everything worked well. But I have no access to a Mesa
system with vdpau, so it needs testing by others.
See github issues #520, #694, #695.
This turned out ridiculously complex. I think it will have to be
simplified some day. Main reason for the complexity are:
- filtering properties by forcing clients to observe individual
properties explicitly
(to avoid spamming clients with changes they don't want)
- optional retrieval of property value with the notification
(the basic idea was that this is more user friendly)
- allowing to the client to specify a format in which the value
should be retrieved
(because if a property changes its type, the client API couldn't
convert it properly, and compatibility would break)
I don't know yet which of these are important, and everything could
change. In particular, the interface and semantics should be adjusted
to reduce the implementation complexity.
While I consider the API complete, there could (and probably will) be
bugs left. Also while the implementation is complete, it's inefficient.
The complexity of the property matching is O(a*b*c) with a clients,
b observed properties, and c properties changing at once. I threw away
an earlier implementation using bitmasks, because it was too unwieldy.
Remove the use of mp_ring and use a simple array and a bunch of
variables instead. This is way less awkwad.
The change in reserve_reply fixes incorrect tracking of free events.
Now they can be paused and resumed.
Since pausing and disabling the timer is essentially the same underlying
operation, we also just provide one method for it.
mp.cancel_timer probably still works, but I'm considering this
deprecated, and it's removed from the manpage. (We didn't have a release
with this function yet, so no formal deprecation.)
This commit adds support for automatic selection of color profiles based on
the display where mpv is initialized, and automatically changes the color
profile when display is changed or the profile itself is changed from
System Preferences.
@UliZappe was responsible with the testing and implementation of a lot of this
commit, including the original implementation of `cocoa_get_icc_profile_path`
(See #594).
Fixes#594
The --ass-styles option is implemented by calling ass_read_styles().
This function can take a codepage (so libass will use iconv to convert
it). This was implemented before our --subcp option was changed, and
this code was not updated. Now libass fails opening iconv, because
--subcp is not always (and not by default) a valid iconv codepage.
Just always pass NULL, which means the file passed to --ass-styles must
be in UTF-8. The --ass-styles option is a fringe option anyway (and will
destroy your subtitles), so having codepage support for it isn't
important at all.
Change the type of the property from a string list (alternating
key/value entries) to a map. Using the client API, this will return
MPV_FORMAT_NODE_MAP, while Lua mp.get_property_native returns a
dictionary-like table.
The function tag_property() in command.c passed a key action with empty
path to m_property_read_list. This is normally not valid, because key
actions are supposed to access sub-paths. But it's kind of inconvenient
to check for this case in tag_property(). So make it valid by providing
a m_property_unkey() function, which turns a key access to "" into a
top-level action.
We've just checked whether a sub-path started with "name/", but that
changes behavior whether the property name has a trailing '/' or not.
Using a helper function to split of path components avoids this problem.
Also fix a format string mistake in a log call using it.
I wonder if this code shouldn't use FormatMessage, but it looks kind
of involved [1], so: no, thanks.
[1] http://support.microsoft.com/kb/256348/en-us
There are two kind of encryption for Blu-ray disc, AACS and BD+,
and both of them can be checked through BLURAY_DISC_INFO object.
This commit makes the bluray and bdnav streams refuse playback
if AACS/BD+ is detected and decryption is failed.
This played the file at a wrong sample rate if the rate was out of
certain bounds.
A comment says this was for the sake of libaf/af_resample.c. This
resampler has been long removed. Our current resampler
(libav/swresample) checks supported sample rates on reconfiguration, and
will error out if a sample rate is not supported. And I think that is
the correct behavior.
The angles should be set and queried only if a valid title is
selected. Also, in navigation mode, there are some limitations
which make it impossible to query current title/angle.
This commit introduces new stream protocols: bdnav(and others).
bdnav stream shares lots of codes with original bluray stream, so
it's not separated in different source file.
Major difference from bluray is that bdnav does not support longest
title because there is no way to query that information.
bdnav://menu and bdnav://first correspond to top menu title and
first play title respectively, though they often point same title.
Also, binary position based seeking has been removed, because it
didn't have no point.
Reduce most dependencies on struct mp_csp_details, which was a bad first
attempt at dealing with colorspace stuff. Instead, consistently use
mp_image_params.
Code which retrieves colorspace matrices from csputils.c still uses this
type, though.
This is pretty obscure, so it didn't matter much. It still breaks
switching output levels at runtime, because the video output is not
reinitialized with the new params.
This was accidentally broken in commit b72ba3f7. I somehow made the
wild assumption that replaygain adjusted the volume relative to 0%
instead of 100%.
The detach suboption was similarly broken.
This was actually supposed to be removed with pull reuqest #671, but
I accidentally re-added it with a rebasing mistake.
This probably also coincidentally fixes compilation with older
libbluray (issue #672).
Use bd_get_playlist_info() instead of bd_get_title_info(). The
previous implementation couldn't query current playlist and this
made it impossible to call bd_get_playlist_info() which is more
desirable than bd_get_title_info() because, for Blu-rays, playlist
is the unit of playback not title. This commit fixes that.
The cost of calling bd_get_title_info() is quite expensive and
requires lots of CPU usage. Using BD_EVENT_PLAYLIST and
BD_EVENT_TITLE, it's possible to cache BLURAY_TITLE_INFO object for
current title and BD_EVENT_ANGLE handler caches current angle. In
my test case, with this commit, CPU usage can be saved about 15-20%.
demux_mf.c explicitly checks for the stream type to check whether images
are opened via pattern (mf://..., i.e. stream_mf.c) or directly. Of
course the stream type is not set to STREAMTYPE_MF if the stream is
wrapped through the cache, so it tried to open the pattern directly as
file, which failed.
Fix this by disabling caching for mf://. The cache doesn't make sense
here anyway, because each file is opened and closed every frame (perhaps
to avoid memory bloat).
It's possible that MPContext has a chapter list, but the demuxer
doesn't. In this case, accesing the chapter-metadata property would
lead to invalid accesses.
(This fixes the out of bound access, but in theory, the returned data
can still be incorrect, since MPContext chapters don't need to map
directly to demuxer chapters.)
This cd_info_t struct was practically unused. The only thing it did was
storing the track name of the form "Track %d" in a very roundabout way.
Remove it. (It made more sense when there was still CDDB support.)
Don't use an integer division to get the time, since that would round on
second boundaries. Also round up the time by sector size. Seeking rounds
down due to alignment constraints, but if we round up the time, we can
make it land on the exact destination sector.
This fixes that the track change code printed the previous track when
seeking by chapter.
This was usually handled at the end of the switch statement, so if
something returns from the function before that, the event has to be
freed explicitly.
Previous implementation updated video resolution when highlight
event was given. However, this may not work if highlight event
is given before video size is queried.
This commit adds checking routine into rendering function, too.
dvdnav.c did not handle event in regular sequence. Usually this
does not make any trouble except around MP_NAV_EVENT_RESET_ALL.
Those events should be handled in regular sequence. If they're
mixed, it can make wrong result.
For instance, MP_NAV_EVENT_HIGHLIGHT right after
MP_NAV_EVENT_RESET_ALL should not be ignored but it might be
because MP_NAV_EVENT_RESET_ALL makes the demuxer reloaded and osd
hidden.
And consistently use MP_NOPTS_VALUE as error value for the users of this
function. This is better than using -1, especially because negative
values can be valid timestamps.
Instead of comparing the current chapter every time, set the playback
end timestamp to the chapter end. Likewise, don't execute an extra seek
for the start chapter.
Maybe we could also use the timeline facility to restrict playback to
the given chapter range, but this would be strange when using
--chapter=N to start playback at a given chapter. Then you couldn't seek
back, which is possibly not what the user wants.
Instead, always use the mpctx->chapters array. Before this commit, this
array was used only for ordered chapters and such, but now it's always
populated if there are chapters.
Stream-level chapters (like DVD etc.) did potentially not have
timestamps for each chapter, so STREAM_CTRL_SEEK_TO_CHAPTER and
STREAM_CTRL_GET_CURRENT_CHAPTER were needed to navigate chapters. We've
switched everything to use timestamps and that seems to work, so we can
simplify the code and remove this old mechanism.
Report the time for each chapter (tracks are treated as chapters). This
allows us to get rid of the "old" chapter mechanism, and also behaves
better with the frontend.
This makes assumptions about the audio formats, but that format is
hardcoded anyway in the rawaudio demuxer defaults (and always was).
Libav 10 was released, so we can enable testing the stable Libav version
again.
FFmpeg 2.2 was also released, but since we still support 2.1.4, we stick
with the older version. This is better for testing.
There were some bad interactions with the OSC.
For one, dragging the OSC bar, and then moving the mouse outside of the
OSC (while mouse button still held) would suddenly initiate window
dragging. This was because win_drag_button1_down was not reset when
sending a normal mouse event, which means the window dragging code can
become active even after we've basically decided that the preceding
click didn't initiate window dragging.
Second, dragging the window and clicking on the OSC bar after that did
nothing. This was because no mouse button up event was sent to the core,
even though a mouse down event was sent. So make sure the key state is
erased with MP_INPUT_RELEASE_ALL.
We don't check whether the WM supports _NET_WM_MOVERESIZE_MOVE, but
if it doesn't, nothing bad happens. There might be a race condition
when pressing a button, and then moving the mouse and releasing the
button at the same time; then the WM might get the message to initiate
moving the window after the mouse button has been released, in which
case the result will probably be annoying. This could possibly be fixed
by sending _NET_WM_MOVERESIZE_CANCEL on button release, but on the
other hand, we probably won't receive a button release event in this
situation, so ignore this problem.
The dragging is initiated only when moving the mouse pointer after a
click in order to reduce annoying behavior when the user is e.g.
doubleclicking.
Closes#608.
The title for stream_bluray DID start from 1 and I misunderstood
that it started from 0 because mpv accepted bd://0 as a proper
argument. In fact, 0 title was an alias for the longest title but
it was not handled as a special value. This commit fixes these
behavious. 'disc-title' property for Blu-ray now starts from 0 and
the default title can be specified by 'longest' title just like
stream_dvdnav: bd://longest. Of course, 'longest' can be omitted.
This commit makes 'disc-title' property writable using
STREAM_CTRL_SET_CURRENT_TITLE. This commit also contains
implementation of STREAM_CTRL_SET_CURRENT_TITLE for stream_bluray.
Currently, 'disc-title' is writable only for stream_dvdnav and
stream_bluray and stream_dvd is not supported.
This commit makes 'disc-title' properties for DVDs start from 0.
There was an inconsistency around 'disc-title' property between
DVDs (from 1) and Blu-rays (from 0). This fixes#648.
Signed-off-by: wm4 <wm4@nowhere>
This check incorrectly passed on Cygwin. While Cygwin has the
statfs.f_type field, it contains the volume's FileSystemAttributes,
which aren't useful for detecting network mounts.
"enable-osc" will make the OSC appear at any time (although it'll
quickly disappear again if the mouse is not inside the OSC). "disable-
osc" will make it disappear permanently.
Also, if the OSC is visible, force remap the DEL key to make the OSC
disappear.
Change script_message to broadcast the message to all clients. Add a new
script_message_to command, which does what the old script_message
command did.
This is intended as simplification, although it might lead to chaos too.
VAAPI has some ambiguous image formats, like VA_FOURCC_I420,
VA_FOURCC_IYUV, VA_FOURCC_YV12 (the latter exactly the same as the first
two, just with swapped planes). There is potentially a problem when one
specific VAAPI format was picked, and converting it to a mpv format and
back to a VAAPI FourCC would result in a numerically different format
(even if it's actually the same). Then it could e.g. happen that
functions like va_surface_upload() reallocate the underlying VAImage,
which would be inefficient. Change the code so that this can't happen.
(Probably not a problem in practice with the current VAAPI usage.)
It's not really needed to be public. Other code can just use mp_image.
The only disadvantage is that the other code needs to call an accessor
to get the VASurfaceID.
Although I at first thought it would be better to have a separate
implementation for hwaccels because the difference to software images
are too large, it turns out you can actually save some code with it.
Note that the old implementation had a small memory management bug. This
got painted over in commit 269c1e1, but is hereby solved properly.
Also note that I couldn't test vf_vavpp.c (due to lack of hardware), and
I hope I didn't accidentally break it.
The plan is to get rid of the custom VAAPI and possibly VDPAU surface
allocators.
Add custom surface allocation, because hwaccel surfaces are allocated
completely differently from software surfaces.
Add optional LRU allocation, which is (probably) helpful for hwaccel,
but (probably) less optimal for software surfaces.
mp_image_pool_get_no_alloc() is specifically for VAAPI, which can't
allocate new decoder surfaces after decoder init.
They were used by ancient libavcodec versions. This also removes the
need to distinguish vdpau image formats at all (since there is only
one), and some code can be simplified.
Image formats used to be FourCCs, so unsigned int was better. But now
it's annoying and the only difference is that unsigned int is more to
type than int.
Instead of doing it on every seek (libavcodec calls get_format on every
seek), reinitialize the decoder only if the video resolution changes.
Note that this may be relatively naive, since we e.g. (or: in
particular) don't check for profile changes. But it's not worse than the
state before the get_format change, and at least it paints over the
current vaapi breakage (issue #646).
Set refcounted_frames, because in some versions of libavcodec mixing the
new AVFrame API and non-refcounted decoding could cause memory
corruption. Likewise, it's probably still required to unref a frame
before calling the decoder.
This follows the (only slowly progressing) plan to replace all internal
video filters with libavfilter.
All what's left in vf_gradfun.c is the weird wrapper around vf_lavfi.c.
I have no tolerance for this garbage anymore. There are tons of issues
with it (see e.g. previous commit), and there is no reason to use it
either. Use Libav git, or Libav 10 when it's released.
This also drops support for earlier FFmpeg release, which have exactly
the same issues as Libav 9. FFmpeg 2.1.4 is still supported, because
it's the latest release, and is reasonably recent. (Although this will
probably also be dropped as soon as FFmpeg 2.2 is released.)
Assumed version table (lots of nonsensical numbers):
FFmpeg 2.1.4 FFmpeg (n2.2-rc2) Libav (v10_beta2)
lavu 52.48.101 52.66.100 53.3.0
lavc 55.39.101 55.52.102 55.34.1
lavf 55.19.104 55.33.100 55.12.0
lsws 2.5.101 2.5.101 2.1.2
lavi 3.90.100 4.2.100 4.2.0
lswr 0.17.104 0.18.100 -
lavr 1.1.0 1.2.0 1.1.0
libpostproc and libavdevice are not interesting.
Following this commit, code needed just to support old Libav versions
will start to be removed.
This "sometimes" crashed when seeking. The fault apparently lies in
libavcodec: the decoder returns an unreferenced frame! This is
completely insane, but somehow I'm apparently still expected to
work this around. As a reaction, I will drop Libav 9 support in the
next commit. (While this commit will go into release/0.3.)
Instead of parsing the ASS file in demux_libass.c and trying to pass the
ASS_Track to the subtitle renderer, just read all file data in
demux_libass.c, and let the subtitle renderer pass the file contents to
ass_process_codec_private(). (This happens to parse full files too.)
Makes the code simpler, though it also relies harder on the (messy)
probe logic in demux_libass.c.
They're strictly DVD-only, so it's better to mark them as such. This
also documentes the "title" (now renamed to "dvd-title") property.
This also avoids collision with the --title option. (Technically, there
was no problem. But it might be confusing for users, since we have a
policy of naming properties and options the same if they refer to the
same underlying functionality.)
This commit adds new property 'title' which indicates current
playing title of disc. This property is useful when using a stream
whose title can be changed during playback, e.g., dvdnav.
This commit provides impelmentation of STREAM_CTRL_GET_NUM_TITLES
for dvdnav stream. Other streams for DVD or Blu-ray are already
provide STREAM_CTRL_GET_NUM_TITLES.
Maybe this should be default. On the other hand, this filter does
something even if the volume is neutral: it clips samples against the
allowed range, should the decoder or a previous filter output garbage.
Currently, both replaygain adjustment and user volume control (if
softvol is enabled) share the same variable. Sharing the variable would
cause especially if --volume is used; then the replaygain volume would
always be overwritten.
Now both gain values are simple added right before doing filtering.
Xlib.h (included from x11_common.h) defines a macro 'Status' as 'int'.
This messed up a bunch of definitions in windows.h and broke the build
in Cygwin. Including windows.h first seems to solve the problem.
This commit also removes the definition of an unused flag.
This adds the options replaygain-track and replaygain-album. If either is set,
the replaygain track or album gain will be automatically read from the track
metadata and the volume adjusted accordingly.
This only supports reading REPLAYGAIN_(TRACK|ALBUM)_GAIN tags. Other formats
like LAME's info header would probably require support from libav.
The main incompatibility was that Libav didn't have av_opt_set_int_list.
But since that function is excessively ugly and idiotic (look how it
handles types), I'm not missing it much. Use an aformat filter instead
to handle the functionality that was indirectly provided by it. This is
similar to how vf_lavfi works.
The other incompatibility was channel handling. Libav consistently uses
channel layouts only, why ffmpeg still requires messing with channel
counts to some degree. Get rid of most channel count uses (and hope
channel layouts are "exact" enough). Only in one case FFmpeg fails with
a runtime check if we feed it AVFrames with channel count unset.
Another issue were AVFrame accessor functions. FFmpeg introduced these
for ABI compatibility with Libav. I refuse to use them, and it's not my
problem if FFmpeg doesn't manage to provide a stable ABI for fields
provided both by FFmpeg and Libav.
And make some additions.
I'm not really sure why we list dependencies that are not ours, but that
of external libraries. Still might be useful, though.
I've tried not to be too detailed (because it's not a reference, just
some guidelines), but it still got relatively long. Also contains
conventions for sending patches.
The window doesn't recieve a WM_LBUTTONUP message after it's dragged,
probably because it's swallowed by the modal loop. To stop the button
from sticking, release it manually when the drag is complete.
Mouse buttons can get stuck down if the button is pressed inside the
video window and released outside. Avoid this by capturing mouse input
when a button is pressed.
Linux also has fstatfs(), and the test relied on certain system include
files being available on BSD, but not on Linux. It would break if Linux
added the missing includes for some reason.
Make it a bit stricter, and check for the struct statfs field the code
needs.
This reads MPV_CLIENT_API_VERSION from the source header, and turns it
into a 3 part version number.
E.g. if MPV_CLIENT_API_VERSION were 0x12abcdef, this would result in
"18.171.773615" (8 bits, 8 bits, 16 bits).
We'll see if this is actually useful, or if it's too clever.
Rename it to --enable-libmpv-shared. The option name didn't really
tell much. When we add the possibility to create a static library,
it would also be bad if that were named --enable-static (because it
would sound like it does what --static-build does).
The volume controls in mpv now affect the session's volume (the
application's volume in the mixer). Since we do not request a
non-persistent session, the volume and mute status persist across mpv
invocations and system reboots.
In exclusive mode, WASAPI doesn't have access to a mixer so the endpoint
(sound card)'s master volume is modified instead. Since by definition
mpv is the only thing outputting audio in exclusive mode, this causes no
conflict, and ao_wasapi restores the last user-set volume when it's
uninitialized.
Due to the COM Single-Threaded Apartment model, the thread owning the
objects will still do all the actual method calls (in the form of
message dispatches), but at least this will be COM's problem rather than
having to set up several handles and adding extra code to the event
thread.
Since the event thread still needs to own the WASAPI handles to avoid
waiting on another thread to dispatch the messages, the init and uninit
code still has to run in the thread.
This also removes a broken drain implementation and removes unused
headers from each of the files split from the original ao_wasapi.c.
ao_wasapi.c was almost entirely init code mixed with option code and
occasionally actual audio handling code. Split most things to
ao_wasapi_utils.c and keep the audio handling code in ao_wasapi.c.
Gets rid of the internal ring buffer and get_buffer. Corrects an
implementation error in thread_reset.
There is still a possible race condition on reset, and a few refactors
left to do. If feasible, the thread that handles everything
WASAPI-related will be made to only handle feed events.
This reverts commit 75dd3ec210.
This broke seeking with ordered chapters in some situations. While
the reverted commit was perfectly fine for playback of normal files,
it overlooked that in the ordered chapters case switching segments
actually reinitialized the audio chain completely, including the
decoder. And decoders still read packets on initialization. We can
restore the original commit as soon as decoders stop doing this.
Apparently the "right" place to initialize the hardware decoder is in
the libavcodec get_format callback.
This doesn't change vda.c and vdpau_old.c, because I don't have OSX, and
vdpau_old.c is probably going to be removed soon (if Libav ever manages
to release Libav 10). So for now the init_decoder callback added with
this commit is optional.
This also means vdpau.c and vaapi.c don't have to manage and check the
image parameters anymore.
This change is probably needed for when libavcodec VDA supports gets a
new iteration of its API.
This updates the logic for the new, somewhat unified behavior of SRGB
and 3DLUT since 34bf9be (not that it was particularly correct even that
change) and checks for the presence of corresponding extensions only in
the cases in which they're needed.
This commit:
- Changes some of the #define and variable names for clarification and
adds comments where appropriate.
- Unifies :srgb and :icc-profile, making them fit into the same step of
the decoding process and removing the weird interactions between both
of them.
- Makes :icc-profile take precedence over :srgb (to significantly reduce
the number of confusing and useless special cases)
- Moves BT709 decompanding (approximate or actual) to the shader in all
cases, making it happen before upscaling (instead of the old 0.45
gamma function). This is the simpler and more proper way to do it.
- Enables the approx gamma function to work with :srgb as well due to
this (since they now share the gamma expansion code).
- Renames :icc-approx-gamma to :approx-gamma since it is no longer tied
to the ICC options or LittleCMS.
- Uses gamma 2.4 as input space for the actual 3DLUT, this is now a
pretty arbitrary factor but I picked 2.4 mainly because a higher pure
power value here seems to produce visually better results with wide
gamut profiles, rather then the previous 1.95 or BT.709.
- Adds the input gamma space to the 3dlut cache header in case we change
it more in the future, or even make it user customizable (though I
don't see why the latter would really be necessary).
- Fixes the OSD's gamma when using :srgb, which was previously still
using the old (0.45) approximation in all cases.
- Updates documentation on :srgb, it was still mentioning the old
behavior from circa a year ago.
This commit should serve to both open up and make the CMS/shader code much
more accessible and less confusing/error-prone and simultaneously also
improve the performance of 3DLUTs with wide gamut color spaces.
I would liked to have made it more modular but almost all of these
changes are interdependent, save for the documentation updates.
Note: Right now, the "3DLUT takes precedence over SRGB" logic is just
coded into gl_lcms.c's compile_shaders function. Ideally, this should be
done earlier, when parsing the options (by overriding the actual
opts.srgb flag) and output a warning to the user.
Note: I'm not sure how well this works together with real-world
subtitles that may need to be color corrected as well. I'm not sure
whether :approx-gamma needs to apply to subtitles as well. I'll need to
test this on proper files later.
Note: As of now, linear light scaling is still intrinsically tied to
either :srgb or :icc-profile. It would be thinkable to have this as an
extra option, :linear-scaling or similar, that could be used with or
without the two color management options.
Assume obtained.samples contains the number of samples the SDL audio
callback will request at once. Then make sure ao.c will set the buffer
size at least to 3 times that value (or more).
Might help with bad SDL audio backends like ESD, which supposedly uses a
500ms buffer.
This obviously doesn't work. It wasn't much of a problem in the past
because most passthrough formats use 2 channels, which is also the
default for downmix.
In general, we don't need to have a large hw audio buffer size anymore,
because we can quickly fill it from the soft buffer.
Note that this probably doesn't change much anyway. On my system (dmix
enabled), the buffer size is only 170ms, and ALSA won't give more. Even
when using a hardware device the buffer size seems to be limited to
341ms.
This AO pretended to support volume operations when in spdif passthrough
mode, but actually did nothing. This is wrong: at least the GET
operations must write their argument. Signal that volume is unsupported
instead.
This was probably a hack to prevent insertion of volume filters or so,
but it didn't work anyway, while recovering after failed volume filter
insertion does work, so this is not needed at all.
Since the addition of the AO feed thread, 200ms of latency (MIN_BUFFER)
was added to all push-based AOs. This is not so nice, because even AOs
with relatively small buffering (e.g. ao_alsa on my system with ~170ms
of buffer size), the additional latency becomes noticable when e.g.
toggling mute with softvol.
Fix this by trying to keep not only 200ms minimum buffer, but also 200ms
maximum buffer. In other words, never buffer beyond 200ms in total. Do
this by estimating the AO's buffer fill status using get_space and the
initially known AO buffer size (the get_space return value on
initialization, before any audio was played). We limit the maximum
amount of data written to the soft buffer so that soft buffer size and
audio buffer size equal to 200ms (MIN_BUFFER).
To avoid weird problems with weird AOs, we buffer beyond MIN_BUFFER if
the AO's get_space requests more data than that, and as long as the soft
buffer is large enough.
Note that this is just a hack to improve the latency. When the audio
chain gains the ability to refilter data, this won't be needed anymore,
and instead we can introduce some sort of buffer replacement function in
order to update data in the soft buffer.
It is possible to have ao->reset() called between ao->pause() and
ao->resume() when seeking during the pause. If the underlying PCM
supports pausing, resuming an already reset PCM will produce an error.
Avoid that by explicitly checking PCM state before calling
snd_pcm_pause().
Signed-off-by: wm4 <wm4@nowhere>
The uint64_t math would cause overflow at long enough system uptimes
(...such as 3 days), and any precision error given by the double math will
be under one milisecond.
One strange issue is that we apparently can't stop the audio API on
audio reset (ao_driver.reset). We could use SDL_PauseAudio, but that
doesn't specify whether remaining audio is dropped. We also could use
SDL_LockAudio, but holding that over a long time will probably be bad,
and it probably doesn't drop audio. This means we simply play silence
after a reset, instead of stopping the callback completely. (The
existing code ran into an underrun in this situation.)
The delay estimation works about the same. We simply assume that the
callback is locked to audio timing (like ao_jack), and that 1 callback
corresponds to 1 period. It seems this (removed) code fragment assumes
there 1 one period size delay:
// delay subcomponent: remaining audio from the next played buffer, as
// provided by the callback
buffer_interval += callback_interval;
so we explicitly do that too.
Apparently, this is always _really_ monotonic, despite what the Linux
manpages say. So this should be much better than gettimeofday(). (At
times there were kernel bugs which broke the monotonic property.)
From the perspective of the player, time can still be discontinuous
(you could just stop the process with ^Z), but at least it's guaranteed
to be monotonic without further hacks required.
Also note that clock_gettime() returns the time in nanoseconds. We want
microseconds only, because that's the unit we chose internally. Another
problem is that nanoseconds can wrap pretty quickly (less than 300 years
in 63 bits), so it's just better to use microseconds. The devision won't
make the code that much slower (compilers can avoid a real division).
Note: this expects that the system provides clock_gettime() as well as
CLOCK_MONOTONIC. Both are optional according to POSIX. The only system
I know which doesn't have these, OSX, has seperate timer code anyway,
but I still don't know whether more obscure (yet supported) platforms
have a problem with this, so I'm playing safely. But this still expects
that CLOCK_MONOTONIC always works at runtime if it's defined.
This is probably "safer". Without it, we will play 1 sample, because the
logic was written in a way to decode 1 sample if audio is paused. 1
sample usually will initialize the audio PTS, but not play any real
audio. Also see previous commit.
In ancient times, this actually used 1 byte (instead of 1 sample), so
clearly no sample was written, unless the audio was 8-bit mono.
Remove the ao_buffer_playable_samples field. This contained the number
of samples that fill_audio_out_buffers() wanted to write to the AO (i.e.
this data was supposed to be played at some point), but ao_play()
rejected it due to partial fill.
This could happen with many AOs, notably those which align all written
data to an internal period size (often called "outburst" in the AO
code), and the accepted number of samples is rounded down to period
boundaries. The left-over samples at the end were still kept in
mpctx->ao_buffer, and had to be played later.
The reason ao_buffer_playable_samples had to exist was to make sure that
at EOF, the correct number of left-over samples was played (and not
possibly other data in the buffer that had to be sliced off due to
endpts in fill_audio_out_buffers()). (You'd think you could just slice
the entire buffer, but I suspect this wasn't done because the end time
could actually change due to A/V sync changes. Maybe that was the reason
it's so complicated.)
Some commits ago, ao.c gained internal buffering, and ao_play() will
never return partial writes - as long as you don't try to write more
samples than ao_get_space() reports. This is always the case. The only
exception is filling the audio buffers while paused. In this case, we
decode and play only 1 sample in order to initialize decoding (e.g. on
seeking). Actually playing this 1 sample is in fact a bug, but even of
the AO doesn't have period size alignment, you won't notice it. In
summary, this means we can safely remove the code.
Until now, this was always conflated with uninit. This was ugly, and
also many AOs emulated this manually (or just ignored it). Make draining
an explicit operation, so AOs which support it can provide it, and for
all others generic code will emulate it.
For ao_wasapi, we keep it simple and basically disable the internal
draining implementation (maybe it should be restored later).
Tested on Linux only.
Same deal as with the previous commit. We don't lose any functionality,
except for waiting "properly" on audio end, instead of waiting using the
delay estimate.
This removes the ringbuffer management from the code, and uses the
generic code added with the previous commit. The result should be
pretty much the same.
The "estimate" sub-option goes away. This estimation is now always
active. The new code for delay estimation is slightly different, and
follows the claim of the jack framework that callbacks are timed
exactly.
This has 2 goals:
- Ensure that AOs have always enough data, even if the device buffers
are very small.
- Reduce complexity in some AOs, which do their own buffering.
One disadvantage is that performance is slightly reduced due to more
copying.
Implementation-wise, we don't change ao.c much, and instead "redirect"
the driver's callback to an API wrapper in push.c.
Additionally, we add code for dealing with AOs that have a pull API.
These AOs usually do their own buffering (jack, coreaudio, portaudio),
and adding a thread is basically a waste. The code in pull.c manages
a ringbuffer, and allows callback-based AOs to read data directly.
Since the AO will run in a thread, and there's lots of shared state with
encoding, we have to add locking.
One case this doesn't handle correctly are the encode_lavc_available()
calls in ao_lavc.c and vo_lavc.c. They don't do much (and usually only
to protect against doing --ao=lavc with normal playback), and changing
it would be a bit messy. So just leave them.
We want to move the AO to its own thread. There's no technical reason
for making the ao struct opaque to do this. But it helps us sleep at
night, because we can control access to shared state better.
This field will be moved out of the ao struct. The encoding code was
basically using an invalid way of accessing this field.
Since the AO will be moved into its own thread too and will do its own
buffering, the AO and the playback core might not even agree which
sample a PTS timestamp belongs to. Add some extrapolation code to handle
this case.
Use QueryPerformanceCounter to improve the accuracy of
IAudioClock::GetPosition.
While this is mainly for "realtime correctness" (usually the delay is a
single sample or less), there are cases where IAudioClock::GetPosition
takes a long time to return from its call (though the documentation doesn't
define what a "long time" is), so correcting its value might be important in
case the documented possible delay happens.
The lack of device latency made get_delay report latencies shorter than
they should; on systems with fast enough drivers, the delay is not
perceptible, but high enough invisible delays would cause desyncs.
I'm not yet completely sure whether this is 100% accurate, there are
some issues involved when repeatedly pausing+unpausing (the delay might
jump around by several dozen miliseconds), but seeking seems to be
working correctly now.
The player didn't quit when the end of a file was reached. The reason
for this is that jack reported a constant audio delay even when all
audio was done playing. Whether that was recognized as EOF by the player
depended whether the exact value was higher or lower than the player's
threshhold for what it considers no more audio.
get_delay() should return amount of time it takes until the last sample
written to the audio buffer reaches the speaker. Therefore, we have to
track the estimated time when the last sample is done, and subtract it
from the calculated latency. Basically, the latency is the only amount
of time left in the delay, and it should go towards 0 as audio reaches
ths speakers.
I'm not sure if this is correct, but at least it solves the problem. One
suspicious thing is that we use system time to estimate the end of the
audio time. Maybe using jack_frame_time() would be more correct. But
apart from this, there doesn't seem to be a better way to handle this.
For example, consider the case when audio initialization fails. Then the
audio track is deselected. Before this commit, this would have been
equivalent to the user disabling audio. This is bad when multiple files
are played at once (the next file would have audio disabled, even if it
works), or if playback resume is used (if e.g. audio output failed to
initialize, then audio would be disabled when resuming, even if the
system's audio driver was fixed).
Setting string options to strings over the m_option fallback (i.e.
M_PROPERTY_SET_STRING is called if the option type is CONF_TYPE_STRING)
failed. This was because m_option_parse() returns 0. 0 still means
success, but the property code tried to be clever, and considered 0 not
a success in order to disallow setting flags to an emtpy string (which
in turn is allowed, because the command line allows flag options without
parameters).
Fix this by removing the overly clever code.
This could happen when e.g. using the "set" command on options/title (a
string option), and also was a problem for the client API.
Closes#610.
The OSC used significant CPU time while the player was paused. It turned
out that the "tick" event sent during pause is the problem. The OSC
accesses the player core when receiving a tick event, which in turn will
cause the core to send another tick event, leading to infinite feedback.
Fix this by sending an idle tick only every 500ms. This is not very
proper, but the idea behind the tick event isn't very clean to begin
with (and the OSC should use timers instead).
The playloop usually waits in select(), using a timeout needed for
refilling audio and video buffers. This means the client API needs
a separate mechanism to interrupt the select() call. This mechanism
exists, but I forgot to use it. This commit fixes it.
If it works, this will make the client API react faster, epsecially
in audio-only mode. If video is enabled, the reaction time is capped
to 50ms (or somewhat faster if the framerate is >20 fps), because
the playloop stops reacting to anything in order to render and time
the next video frame. (This will be fixed later by moving the VO
to its own thread.)
--ass-style-override=force now attempts to override the 'Default' style.
May or may not work. In some situations it will work, but also mess up
seemingly unrelated things like signs typeset with ASS.
luaL_loadstring(), which was used until now, uses the start of the Lua
code itself as chunk name. Since the chunk name shows up even with
runtime errors triggered by e.g. Lua code loaded from user scripts, this
looks a but ugly. Switch to luaL_loadbuffer(), which is almost the same
as luaL_loadstring(), but allows setting a chunk name.
Will be helpful to track down strange wait times and such issues, as
well when you have develop something timing related. (Then you may print
timestamps in your debug output, and the --msgtime timestamps will help
giving context.)
Sending an asynchronous request and then calling mpv_destroy() would
crash the player when trying to send the reply to the removed client.
Fix this by waiting until all remaining replies have been sent.
The step argument for "add volume <step>" was ignored until now. Fix it.
There is one problem: by defualt, "add volume" should use the value set
with --volstep. This value is 3 by default. Since the default volue for
the step argument is always 1 (and we don't really want to make the
generic code more complicated by introducing custom step sizes), we
simply multiply the step argument with --volstep to keep it compatible.
The --volstep option should probably be just removed in the future.
The value range is 0-100, so fractional values don't make much sense.
But the underlying data type is probably float to avoid getting "stuck"
when doing small volume increments. So step this around and pretend it's
an integer just on display.
The previous version of the gamma suboption was pretty useless. It could
be used to disable delayed gamma enabling, which is a mechanism to avoid
having to adjust gamma in the shader by default.
Repurpose the suboption and allow setting an exact gamma value with it.
You can already override gamma with the --gamma option as well as the
gamma input property, but these use a weird curve to create the
impression of a linear perceived brightness change when changing the
value. This suboption now allows setting an exact gamma value.
This used to be absolute colorimetric, but relative colorimetric is a
saner default due to the arguments presented in issue #595.
A short summary: In general it doesn't affect much because our eyes
adapt to the white point either way, but if running in windowed mode it
would make the whites seem inconsistent/tinted. For fullscreen
projection it's also undesirable since it reduces the dynamic range
without much benefit (again, since our eyes adapt either way) and it
also breaks calibration against ambient lighting.
This shouldn't change much, since most profile types that aren't 3DLUTs
aren't capable of either of those transforms, and most displays are
calibrated against D65 (same as BT.709 source) either way.
Lua doesn't distinguish between arrays and maps on the language level;
there are just tables. Use metatables to mark these tables with their
actual types. In particular, it allows distinguishing empty arrays from
empty tables.
When passing a very large timeout to mpthread_cond_timed_wait(), the
calculations could overflow, setting tv_sec to a negative value, and
making the pthread_cond_timed_wait() call return immediately. This
accidentally made Lua support poll and burn CPU for no reason.
The existing overflow check was ineffective on 32 bit systems. tv_sec is
usually a long, so adding INT_MAX to it will usually not overflow on 64
bit systems, but on 32 bit systems it's guaranteed to overflow. Simply
fix by clamping against a relatively high value. This will work until 1
week before the UNIX time wraps around in 32 bits.
Always map MPV_FORMAT_STRING to setting property value directly through
M_PROPERTY_SET_STRING, instead of trying to go through
M_PROPERTY_SET_NODE.
This treats a direct MPV_FORMAT_STRING query differently from a
MPV_FORMAT_STRING wrapped in a mpv_node. This was already the case in
mpv_get_property(). The reason for all this is that mpv_node is supposed
to be the exact type, while a direct MPV_FORMAT_STRING goes through all
possible conversions.
Not sure if these semantics are good.
This is a bit weird: m_option_string types (i.e. char*) can be NULL. But
they're supposed to be treated just like empty strings. So don't make
the m_option_type.print function return NULL for these values. Returning
NULL would mean failure.
This didn't matter much before, but was quite visible through the client
API.
1. Cannot set option after initialized: it seems that this bug has
existed since libmpv was introduced first. Maybe just a typo.
2. Crash when setting property with native format: mpv_set_property
just causes a crash when using a native format. I found an invalid
casting and fixed it.
3. Wrong error value for mpv_get_property: when an error occurred,
mpv_get_property always returns wrong format error because every
error for property except M_PROPERTY_NOT_IMPLEMENTED is just ignored.
Signed-off-by: wm4 <wm4@nowhere>
Closes pull request #593. Does not incldue the first fix, which was not
correct. The underlying bug will be fixed by a later commit.
Commit message extracted from pull request and slightly edited.
E.g. binding MOUSE_BTN0 always used the user defined binding. While it
is ok that the user can override mouse_move and mouse_leave (for
whatever reasons), we want to strictly override the bindings when input
is sent to the OSC itself.
Regression since 03624a1.
This lowers the number of data stored in the resume config a bit
further, because some properties can't be read at program start and when
e.g. the VO wasn't created yet.
Some fields still need to be read from a property (actually only
"volume-restore-data", a hack to save the full volume information). So
abuse the "options/" property, and make use of the fact that changing
things at runtime also changes the options.
This is approximate: we read each option value on program start
(before starting playback of a file), and when writing the resume
config, compare each value to the current state. This also means
when a value is changed and then changed back, it's not stored. In
particular, option values set in config files and on the command
line are considered the default.
This should help reducing the numbers of options overridden by the
resume config. If too much is overridden, it becomes an inconvenience,
because changes in config files will apparently have no effect when
resuming a file.
Also see github issue #574.
This might be helpful if we ever want cascading config files. Also, we
will probably need it if we change the default input.conf bindings, and
want to provide compatibility input.conf files.
This created an essentially empty config file. This is not really
needed and probably causes more trouble than it solves (such as
littering the home directory with crap), so get rid of it.
This was broken by commit bb6b543812. Note that the original pull
request was fine, but it was broken by my own stupidity when I was
"improving" it.
The problem is that the new loadfile argument was not considered
optional anymore after my changes. The original pull request did handle
this by setting .defval to a dummy value, but I removed that part.
Fix it again by introducing a flag that designates that the parameter is
optional. (I didn't want to add it to m_option.h, because technically,
all options are optional, and it's not possible to have non-optional
options.)
Not sure about this... might redo.
At least this provides a case of a broadcasted event, which requires
per-event data allocation.
See github issue #576.
May or may not be useful in some ways.
We require a context parameter for this just to be sure, even if the
internal implementation currently doesn't.
That's one less mpv internal function for the Lua wrapper.
Also mention that NULL isn't valid. Although I'm not sure whether the
implementation strictly follows this (it should, but there are some
wacky corner cases).
There are some complications because the client API distinguishes
between integers and floats, while Lua has only "numbers" (which are
usually floats). But I think this should work now.
This is only half-implemented: actually the option will first be
converted from mpv_node to its native type, then it's converted to a
string, and then back to its native type. This is because the option
API was made for strings and not anything else.
Other than being grossly inelegant, the only downside is probably with
string lists and key/value lists, which don't escape strings containing
syntax elements correctly.
This automatically allows accessing properties like chapter-list and
track-list to be read as mpv_node. This affects all properties which use
m_property_read_sub() and m_property_read_list().
This actually makes use of the client.h declarations and the mpv_node
mechanisms added some commits ago.
For now, using MPV_FORMAT_STRING will usually fallback to explicit
string conversion, but not in the other cases. E.g. reading a numeric
property as string will work, but not reading a string property as
number. Other than that, only MPV_FORMAT_INT64->MPV_FORMAT_DOUBLE does
an automatic conversion.
I'm not sure whether these semantics and API are good, so comments and
suggestions are welcome.
Allows retrieving properties by their native values (or something close
to it), rather than having to go through string conversion. The caller
could actually just copy the value itself and then use the m_option
functions to convert it to mpv_node, but maybe it's more flexible this
way.
m_option is basically the mechanism to handle C data types in a dynamic
way. Add functions to convert values to and from mpv_node. For example,
string lists are turned into mpv_node using MPV_FORMAT_NODE_ARRAY, and
so on.
This adds declarations for new formats. The implementation will be added
in the following commits. (It still compiles and runs with this commit,
because it adds constants only.)
The obvious new types are MPV_FORMAT_FLAG, MPV_FORMAT_INT64,
MPV_FORMAT_DOUBLE. MPV_FORMAT_FLAG is a boolean, but to avoid nasty ABI
issues or with languages that don't have a bool data type (C89), it uses
int. Thus the format is not named MPV_FORMAT_BOOL, to avoid confusion.
The MPV_FORMAT_NONE type (mpv_node) is a generic structured type, like a
variant or, say, JSON. It can store strings/bools/numbers, as well as
arrays and key/value pairs (with string keys only).
The MPV_FORMAT_NODE_ARRAY and MPV_FORMAT_NODE_MAP types are used
internally by mpv_node only and can't be used with most of the other API
(like mpv_set_property()) directly.
With mpv_set_property(h, "property", MPV_FORMAT_STRING, ptr), ptr now
has to be of type char** instead of char*. This makes it more consistent
with mpv_get_property() and also non-pointer formats, which will be
introduced in the following commits. mpv_set_property() of course does
not change its interface (only its implementation is adjusted to keep
its interface).
This also affects mpv_set_option(), but again not
mpv_set_option_string().
This is allowed in C99 and C++11, but apparently not in C89 and C++98.
Make it conform to the older standards, since we want the client API
header to be highly portable.
This uses the value of 1.95 as an approximation for the exact gamma
curve, which replicates the behavior of popular video software including
anything in the Apple ecosystem, as per issue #534.
Until now, strings were the only allowed dynamically allocated argument
type in input commands. Extend it so that it works for any type. (The
string expansion in command.c is of course still string specific.)
If this function could return the input value (i.e. the == case was
correct), then macros like MP_GROW_ARRAY would have been incorrect. The
implementation was correct though, so there's no bug.
Some code accessed m_option.name to get the property name. (Maybe only
show_property_osd() had a significant use of it.) Remove that, and
remove setting names and dummy names as well.
The old code usually assumed that the name was set, and
show_property_osd() used it to get the proper name of deprecated
aliases.
The "vf" property was listed as "vf*". Not sure why that was done, but
it works without anyway.
Doesn't require other code to care about this, which will allow us to
simplify the property code.
Only "wildcard" options like "vf" and string lists used this, and
m_option_list_findb() (which is excused).
M_OPT_PARSE_ESCAPES was pretty stupid, and broke the (useful) assumption
that string variables contain exactly the same value as set by the
option. Simplify it, and move escape handling to the place where it's
used.
Escape handling itself is not terribly useful, but still allows useful
things like multiline custom OSD with "\n".
MP_CMD_COMMAND_LIST commands (used to implement key bindings with
multiple commands) were not checked for abort commands. Implement it.
Remove the remarks about multi-commands being special from the manpage.
Seek coalescing is handled differently now, and the issue with abort
commands is fixed with this commit.
Current code stolen from waf's extras, only supported 'pe' and 'elf'. OS X
uses the 'Mach-O' binary format (which waf calls 'mac-o'... go figure).
Add support for generating the global symbols file with nm and using it from
clang.
The old way still works, and is fine to use. Still discourage it,
because it might conflict with other ways to access this property, such
as the one added in the next commit.
This is a regression introduced from moving Lua scripts (including the
OSC) to their own threads. Now OSC and dvdnav can add their bindings at
the same time without coordination, which seems to result in the OSC
winning most time, and thus overriding the dvdnav menu bindings.
Fix this by adding a flag that makes dvdnav menu bindings take priority
over all other bindings.
When timeline was used, and the --start option was not used, the initial
seek (needed to switch to the first timeline segment) seeked to -1 due
to an oversight.
This is simply not important enough to warrant so much space, and it's
perhaps also very confusing.
Although I'm not fully sure, since this is about the only way that
allows a user to interact with a script, besides key bindings and static
options.
Detected 'protocols' are AFP, nfs, smb and webdav. This can be extended on
request.
This is currently only implemented for BSD systems (using fstatfs). This
addresses issue #558 on the above platforms.
There was already an undocumented mechanism provided by
mp.set_key_bindings and other functions, but this was relatively
verbose, and also weird. It was mainly to make the OSC happy (including
being efficient and supporting weird corner cases), while the new
functions try to be a bit simpler.
This also provides a way to let users rebind script-provided commands.
(This mechanism is less efficient, because it's O(n^2) for n added key
bindings, but it shouldn't matter.)
- Adds description of and uses $JOBS envvar in MXE instructions
- Adds MXE_TARGETS to command line instead of echoing it to settings.mk
- Prettify and sentence usage
This adds a mechanism for easier export of sub-properties. The following
commits will make use of it to export fine grained information about
certain things. The sub-property mechanism reduces the amount of code
needed to export a data value to 1 line.
* 'master' of git://github.com/mpv-player/mpv:
win32: restore support for exe directory as config directory
crosscompile-mingw: improve instructions for MXE
sd_lavc: handle subtitles with no subtitle resolution set
options: make --no-config block all auto-loaded configuration files
lua: auto-load scripts from ~/.mpv/lua/
lua: make register_event() not overwrite previous event handler
It seems that it was causing issues with certain perl setups (such as
the one on issue #549). It also turns out that it was not behaving correctly
(not all constants were being promoted to big nums as they should), so we
use explicit objects to derive the constants.
There were also precedence issues. I wonder if this even worked right to
begin with.
The 'double' path (8-byte floats) is untested, as I couldn't easily find
a file with such a field.
Closes#549.
Set subtitle resolution to video resolution when avctx->width and
avctx->height are zero.
This can happen with broken vobsubs that have no size set in their
.idx file (or Matroska extradata). At least with the test file provided
in issue #551, using the video resolution as fallback instead of what
guess_resolution() does is better.
Note that these files clearly are broken. It seems this particular
file was created by trying to use ffmpeg to transcode DVB subtitles
to vobsub, and ffmpeg "forgot" to set the subtitle resolution in the
destination file. On the other hand, ffmpeg DVB and PGS decoders set
the resolution on the first subtitle packet (or somewhere close), so
it's not really clear what to do here.
Closes#551.
Signed-off-by: wm4 <wm4@nowhere>
Patch by xylosper, rewritten commit message by wm4.
Until now, the --no-config was explicitly checked in multiple places to
suppress loading of config files.
Add such a check to the config path code itself, and refuse to resolve
_any_ configuration file locations if the option is set.
osc.lua needs a small fixup, because it didn't handle the situation when
no path was returned. There may some of such cases in the C code too,
but I didn't find any on a quick look.
Instead, chain them.
Note that there's no logic to prevent the other event handlers to be run
from an event handler (like it's popular in GUI toolkits), because I
think that's not very useful for this purpose.
Cygwin's libc (newlib) doesn't obey a lot of unix feature test macros,
including _GNU_SOURCE; as a result, a lot of functions and defines get
masked out -- important defines such as M_PI and strcasecmp. Work around
it by undefining __STRICT_ANSI__ on cygwin systems.
This will still cause compilation issues on any non-cygwin system that
uses newlib, but hopefully nobody does that, or if they do, they will
find this commit message and know to add -U__STRICT_ANSI__ to their
CFLAGS. Hopefully.
The mplayer decoder (spudec.c) actually handled this. There was explicit
code for binary palettes (16 32 bit values), and the subtitle resolution
was handled by video resolution coincidentally matching the subtitle
resolution.
Whoever puts vobsub into mp4 should be punished.
Fixes the sample gundam_sample.mp4, closes github issue #547.
In particular, this affects drag & drop of subtitles, which uses sub_add
internally. This will make the subtitles show up immediately, instead of
requiring manual selection of the added subtitle.
Might be not so ideal when adding multiple subtitles at once, because
that leads to multiple sub_add commands, and will end up with the last
subtitle instead of the first selected. But this is a minor detail.
The minimum required version was bumped in the old configure script, but
for the waf build system is was somehow forgotten or overlooked.
Probably happened while the waf build system was developed in a separate
branch.
Closes#546.
This is the same issue as addressed by 257d9f1, except this time for
the :srgb option as well. (257d9f1 only addressed :icc-profile)
The conditions of the srgb_compand mix() call are also flipped to
prevent an off-by-one error.
Use a list instead of a table. This makes it easier to provide extended
information about a property, and doesn't require you to fiddle with rhe
RST ASCII-art tables.
Also, extend some property descriptions.
The initialization code was split and refactored for the libmpv changes.
One change, moving a part of cocoa initialization, accidentally broke
--force-window on OSX, which creates a VO in a certain initialization
stage. We still don't know how cocoa should behave with libmpv, so fix
this with a hack to beat it back into working. Untested.
I was unhappy with the old way of handling buffers, especially resizing. But my
original plan to use wl_shm_pool_resize wasn't as good as I initially thought.
I might get back to it.
With the new buffer pools it now possible to select triple buffering. Also the
buffer pools are also needed for the upcoming subsurfaces for osd and subtitles.
I hope this change was worth it.
Use of these is "discouraged", but they're there to select these special
cases with the "aspect" property. They really should use some sort of
choice option type, but since it would be some work to make these work
with float values, the simple and dumb alternative was picked.
Return the error Lua-style, instead of raising it as Lua error. This is
better, because raising errors is reserved for more "fatal" conditions.
Pretending they're exceptions and trying to do exception-style error
handling will just lead to pain in this language.
Print a warning if a library has mismatched compile time and link time
versions.
Refuse to work if the compile time and link time versions are a mix of
ffmpeg and libav. We print an error message and call exit(). Since we'd
randomly crash anyway, I think this is ok.
This doesn't catch the case if you e.g. use a ffmpeg libavcodec and a
libav libavformat, which would of course just crash as quickly, but I
think this checks enough already.
This library will export the client API functions.
Note that this doesn't allow compiling the command line player to link
against this library yet. The reason is that there's lots of weird stuff
required to setup the execution environment (mostly Windows and OSX
specifics), as well as things which are out of scope of the client API
and every application has to do on its own. However, since the mpv
command line player basically reuses functions from the mpv core to
implement these things, it's not very easy to separate the command
line player form the mpv core.
This is partial only, and it still accesses some MPContext internals.
Specifically, chapter and track lists are still read directly, and OSD
access is special-cased too.
The OSC seems to work fine, except using the fast-forward/backward
buttons. These buttons behave differently, because the OSC code had
certain assumptions how often its update code is called.
The Lua interface changes slightly.
Note that this has the odd property that Lua script and video start
at the same time, asynchronously. If this becomes an issue, explicit
synchronization could be added.
Add a client API, which is intended to be a stable API to get some rough
control over the player. Basically, it reflects what can be done with
input.conf commands or the old slavemode. It will replace the old
slavemode (and enable the implementation of a new slave protocol).
This avoids trouble if another mpv instance is initialized in the same
process.
Since timeBeginPeriod/timeEndPeriod are hereby not easily matched
anymore, use an atexit() handler to call timeEndPeriod, so that we
can be sure these calls are matched, even if we allow multiple
initializations later when introducing the client API.
It's quite possible to overflow the calculation by setting the timeout
to high values. Limit it to INT_MAX, which should be safe. The issue is
mainly the secs variable.
timespec.tv_sec will normally be 64 bit on sane systems, and we assume
it can't overflow by adding INT_MAX to it.
This skipped all audio packets before the first video key frame was
found. I'm not really sure why this would be needed; most likely it
isn't. So get rid of it. Even if audio packets are returned to the
player too soon, the player will sync the audio start to the video
start by decoding and discarding audio data.
Note that although the removed code was just added in the previous
commit, it merely kept the old keeping semantics which demux_mkv
always followed. This commit removes these special semantics.
v_skip_to_keyframe is set to true while non-keyframe video packets are
skipped. Until now, audio packets were also skipped when doing this. I
can't see any good reason why this would be done, but for now I want to
keep the old logic when audio+video seeks are done.
However, for audio-only mode, do proper seeking, which also fixes
behavior when trying to seek past the end of the file: playback is
terminated properly, instead of starting playback on the start of the
last cluster.
Note that a_no_timecode_check is used only for audio+video seek. I'm
not sure what this is needed for, but it might influence A/V sync after
seeking.
This sometimes happened when changing playback speed (= reinitializing
audio) after seeking of playback start. The assertion in audio.c:441 was
triggered, because buffer_playable_samples wasn't reset correctly when
the audio buffer was cleared or shortened. The assertion is correct and
should hold up any time.
I could not see any difference whatsoever, but for usage with a 3DLUT
there's zero performance difference so we might as well follow the spec to
the letter.
On Windows, no ANSI control sequences are available, so we can't easily
clear lines, move the cursor, etc. It's yet to be decided how this
should be handled (emulate ANSI escapes in osdep/terminal-win.c, or
provide abstracted terminal API functions to unify the Linux and Windows
code).
For now, this fixes the regression that was introduced earlier by the
status line rewrite. It doesn't fix all aspects of status line and
terminal OSD handling, as can be clearly seen by the unconditional use
of terminal_erase_to_end_of_line further down the changed code.
Fixes github issue #499 (sort of).
Trying to set a non-existent flag (like +keepside on Libav) causes
libavutil print an incomprehensible warning (something about eval;
probably the overengineered libavutil option parser tripping over the
'+' normally used for flags, and trying to interpret it as formula).
There's apparently no easy way to check for the existence of a flag,
so add some more ifdeffery to shut it up.
The code removed from handle_input_and_seek_coalesce() did two things:
1. If there's a queued seek, stop accepting non-seek commands, and delay
them to the next playloop iteration.
2. If a seek is executing (i.e. the seek was unqueued, and now it's
trying to decode and display the first video frame), stop accepting
seek commands (and in fact all commands that were queued after the
first seek command). This logic is disabled if seeking started longer
than 300ms ago. (To avoid starvation.)
I'm not sure why 1. would be needed. It's still possible that a command
immediately executed after a seek command sees a "seeking in progress"
state, because it affects queued seeks only, and not seeks in progress.
Drop this code, since it can easily lead to input starvation, and I'm
not aware of any disadvantages.
The logic in 2. is good to make seeking behave much better, as it
guarantees that the video display is updated frequently. Keep the core
idea, but implement it differently. Now this logic is applied to seeks
only. Commands after the seek can execute freely, and like with 1., I
don't see a reason why they couldn't. However, in some cases, seeks are
supposed to be executed instantly, so queue_seek() needs an additional
parameter to signal the need for immediate update.
One nice thing is that commands like sub_seek automatically profit from
the seek delay logic. On the other hand, hitting chapter seek multiple
times still does not update the video on chapter boundaries (as it
should be).
Note that the main goal of this commit is actually simplification of the
input processing logic and to allow all commands to be executed
immediately.
Legacy GL context creation (glXCreateContext) explicitly requires a X
visual, while the modern one (glXCreateContextAttribsARB) does not for
some reason. So fail only on the legacy code path if we don't find a
visual. Note that vo_x11_config_vo_window() will select a default visual
if a NULL visual is passed to it.
This fixes issue #504. For some reason, glXChooseFBConfig() will return
a fbconfig with no associated visual. (I'm not sure if this allowed.
They don't always have a visual, but since GLX_X_RENDERABLE is set
and GLX_DRAWABLE_TYPE is (implicitly) set to GLX_WINDOW_BIT, why would
there be no visual?)
Even worse, a test program seems to show that a 16 bit fbconfig is
selected (instead of 24/32 bit), which doesn't sound nice at all. Since
there _are_ better fbconfigs available, glXChooseFBConfig() should
normally sort them by quality, and return the better ones first. It's
worth noting that this function should also prefer GLX_TRUE_COLOR
over anything else, although this comes last in the sort order.
Whatever is going on, requesting GLX_X_VISUAL_TYPE with GLX_TRUE_COLOR
seems to fix it.
There is some logic to discard packets from streams that are not
selected. Run the metadata update code before this, just to make 100%
sure that no metadata updates can be lost when streams are deselected.
(I'm not sure why this logic would be needed, since both libavformat and
the generic demuxer code do this already. But a quick test shows that
av_read_frame() can return a packet from a stream even if the stream has
AVStream.discard set to AVDISCARD_ALL. This happened after stream
switching. Maybe libavformat doesn't discard already queued packets.)
Instead of printing lines like:
Demuxer info GENRE changed to Alternative Rock
Just output all tags once they change. The assumption is that individual
tags rarely change, while all tags change in the common case.
This changes tag updates to use polling. This could be fixed later,
although the ICY stuff makes it a bit painful, so maybe it will remain
this way.
Also remove DEMUXER_CTRL_UPDATE_INFO. This was intended to check for tag
updates, but now we use a different approach.
Use an arbitrary constant instead, which is as good as PATH_MAX.
This helps us to avoid having to think about pull request #523.
Also fix a case where a potentially signed char was passed to isspace().
This was done incorrectly in the previous commit: the fallback size used
the window size as requested with the first config call, which is the
size of the hidden window in the vo_opengl case. (That damn hidden
window again...)
This code essentially does nothing. As far as I could find out, this
actually used to do something. Then it was removed with commit efe7c39f,
leaving some leftover code that didn't do anything useful. This happened
12 years ago!
Also remove a commented debug printf.
vo_opengl creates a hidden X11 window to probe the OpenGL context. It
must do that before creating a visible window, because VO creation and
VO config are separate phases.
There's a race condition involving the hidden window: when starting with
--fs, and then leaving fullscreen, the unfullscreened window is
sometimes set to the aspect ratio of the hidden window. I'm not sure why
the window size itself uses the correct size (but corrupted by the wrong
aspect), but that's perhaps because the window manager is free to ignore
the size hint while honoring the aspect, or something equally messed up.
It turns out this happens because x11_common.c thinks the size of the
hidden window is the size of the unfullscreened window. This in turn
happens because vo_x11_update_geometry() reads the size of the hidden
window when called in vo_x11_fullscreen() (called from
vo_x11_config_vo_window()) when mapping the fullscreen window. At that
point, the window could be mapped, but not necessarily. If it's not
mapped, it will get the size of the unfullscreened window... I think.
One could fix this by actively waiting until the window is mapped. Try
to pick a less hacky approach instead, and never read the window size
until MapNotify is received.
vo_x11_create_window() needs a hack, because we'd possibly set the VO's
size to 0, resulting e.g. in vdpau to fail initialization. (It'll print
error messages until a proper resize is received.)
This fixes a weird bug with aspect ratio handling. It has to do with
float handling: with -std=gnu99, gcc implicitly enables broken non-
standard semantics giving float variables excess precision. This can for
example make this fail in theory: "float a = 0.1; assert(a == a);"
While standard C allows excess precision _within_ expressions, it
requires truncation when storing float values in variables of types
"float" or "double". The "gnu99" mode breaks this. It can be unbroken by
using "c99", or by specifying -fexcess-precision=standard. The former
seems less likely to break compilers other than modern gcc. Note that
-ffloat-store would also fix this, but also makes float expressions less
efficient and less precise for no reason.
The code that mistakenly fails because of this is dec_video.c line 393.
It caused the container aspect to be ignored in some or all situations,
depending how the compiler optimizes. For example, on gcc-4.6 with -Os,
the aspect is always ignored.
In future, we should probably just get rid of storing aspects as floats.
Note that this still happens in the stream level, so we can't have
nice highlevel behavior restricting seeking. Instead, if a seek leads
to the demuxer requesting data outside of the cached range, the seek
will simply fail. This might confuse the demuxer, and the resulting
behavior is not necessarily useful.
Note that this also doesn't try to skip data on a forward seek. This
would just freeze the stream with slow unseekable streams.
One nice thing is that stream.h has a separate function for merely
skipping data (separate from seeking forward), which is pretty useful
in this case: we want skipping of data to work, even if we reject
seeking forward by skipping data as too expensive. This probably is
or will be useful for demux_mkv.c.
Usually, you have to call pthread_cond_timedwait() in a loop (because it
can wake up sporadically). If this function is used by another higher
level function, which uses a relative timeout, we actually have to
reduce the timeout on each iteration - or, simpler, compute the
"deadline" at the beginning of the function, and always pass the same
absolute time to the waiting function.
Might be unsafe if the system time is changed. On the other hand, this
is a fundamental race condition with these APIs.
This used to work; I'm not sure when or why it regressed. When setting
AVProbeData.filename to NULL, libavformat will crash in rtp_probe() by
unconditionally accessing the string.
We used to set the filename to NULL to prevent probing by file extension
when we don't deem it as necessary. Using an empty string also works for
this purpose.
Try to make it more intuitive by not requiring hex values. The new way
uses float values in the range 0.0-1.0, separated by '/' (':' was
suggested, but that wouldn't allow color options in sub-options).
Example: --osd-color=1.0/0.0/0.0/0.75
Using the range 0.0-1.0 has the advantage that it could be easily
extended to colors beyond 8 bit.
Details see manpage.
Suggestions for alternative syntax or value ranges are welcome, but be
quick with it.
This avoids stray newlines when:
1. Some (non-status line) text was output
2. Then an empty status line is output
According to the logic, 2. should print an empty line to show the blank
status line. Don't do that, and instead output nothing in this case.
This caused problems with mpv_identify.sh, and also looked ugly when
using --quiet.
Larger sizes can introduce overflows, depending on the image format. In
the worst case, something larger than 16000x16000 with 8 bytes per pixel
will overflow 31 bits.
Maybe there should be a proper failure path instead of a hard crash, but
not yet. I imagine anything that sets a higher image size than a known
working size should be forced to call a function to check the size (much
like in ffmpeg/libavutil).
Not everything in the OSD path handles 0x0 sized sub-bitmaps well. At
least the code implementing --sub-gray had severe problems with it.
Fix this by skipping such bitmaps.
RGB565 is one of the fastest and most supported formats on low end consumer
devices, but ffmpeg spams warning when using it. Make it opt-in instead of
opt-out.
The problem seems to have solved itself. I guess the previous changes to
resizing and commit ba101ab made this possible. Consider me happy for removing
that crap.
Windows applications that use LoadLibrary are vulnerable to DLL
preloading attacks if a malicious DLL with the same name as a system DLL
is placed in the current directory. mpv had some code to avoid this in
ao_wasapi.c. This commit just moves it to main.c, since there's no
reason it can't be used process-wide.
This change can affect how plugins are loaded in AviSynth, but it
shouldn't be a problem since MPC-HC also does this and it's a very
popular AviSynth client.
Windows users expect this when a program crashes. Without it, the
program just disappears. Also change the SetErrorMode call to use macros
instead of a hardcoded constant.
If, for some reason, the subtitle renderer attempts to render a
subtitle before SD_CTRL_SET_VIDEO_PARAMS was called, it passed a
value calculated from invalid values. This can happen with --vf=sub
and --start. The crash happens if 1. there was a subtitle packet that
falls into the timestamp of the rendered video frame, 2. the playloop
hasn't informed the subtitle decoder about the video resolution yet
(normally unneeded, because that is used for weird corner cases only,
so this code is a bit fuzzy), and 3. something actually requests a
frame to be drawn from the subtitle renderer, like with vf_sub.
The actual crash was due to passing NaN as pixel aspect to libass,
which then created glyphs with ridiculous sizes, involving a few
integer overflows and unchecked mallocs.
The sd_lavc.c and sd_spu.c cases probably don't crash, but I'm not
sure, and it's better fix them anyway.
Not bothering with sd_spu.c, this crap is for compatibility and will
be removed soon.
Note that this would have been no problem, had the code checked whether
SD_CTRL_SET_VIDEO_PARAMS was actually called. This commit adds such a
check (although it basically checks after using the parameters).
Regression since 49caa0a7 and 633fde4a.
This generally affects mp3 files that don't have any (or many) mp3
frames in the first 2 MB. 2 MB is the maximum probe size, and
libavformat returns a low probescore even if we give it the full 2 MB.
Trying to probe a larger buffer (or even the full file) doesn't work for
mysterious reasons.
The workaround consists in accepting a very weak probescore if the
format is detected as mp3 and we probed already 2 MB.
Restructure it a bit, so we can use the format hack list even if no mime
type applies. Shouldn't change anything functionally yet. Preparation
for the next commit.
Still untested, because now it crashes inside of libSDL for unknown
reasons. (This also happens with mpv git from yesterday - probably an
installation problem, or SDL doing weird things it shouldn't be doing.)
The main difference between the old and new callbacks is that the old
callbacks required passing the window size, which is and always was very
inconvenient and confusing, since the window size is already in
vo->dwidth and vo->dheight.
Balance controls as used by mixer.c was broken, because af_pan.c stopped
accepting its arguments. We have to allow 0 channels explicitly. Also,
fix null pointer access if the matrix parameter is not used.
Regression from commit 82983970.
If there's more than one edition, print the list of editions, including
the edition name, whether the edition is selected, whether the edition
is default, and the command line option to select the edition. (Similar
to stream list.)
Move reading the tags to a separate function process_tags(), which is
called when all other state is parsed. Otherwise, that tags will be lost
if chapters are read after the tags.
Pretty worthless. This is called from the seek code, which will
reinitialize these anyway. Even if seeking somehow decides to fail, the
new values are still valid.
One could say a failed seek (if that happens) should jump back to the
original position, and thus it would be better to make sure the state
is restored. But then demux_mkv_seek needs to do this correctly,
including not setting up skipping to the target timestamp. But not
bothering with this.
Extremely obscure corner case with concatenated segments, in which EOF
wasn't recognized correctly, and it tried to demux clusters from the
next segment.
See [MKV]_Editions,_Linked_Segments,_&_Tracksets.mkv from the CCCP test
file collection.
This basically used to be part of the user interface, before mpv moved
printing the track list to the frontend, and this code was raised to
verbose output level.
For some reason, if an error happened when reading headers, it merely
stopped reading the headers, and then continued normally. (It looks like
the case to exit hard (-2) was mainly used for skipping unwanted ordered
chapter segments.)
I can't comprehend this. Always exit on error when reading headers.
(Maybe some more error tolerance would be good, but I have no test case,
and there's some danger of entering endless loops.)
This makes everything more robust, and also somewhat simpler (even if
the diffstat isn't very impressive).
Instead of recursively following SeekHeads while reading headers, just
read the headers until the first cluster, and then possibly use
SeekHeads to read the remaining missing headers.
Rename vo_get_src_dst_rects() to mp_get_src_dst_rects() and make it
independent from the VO (it takes a comical amount of parameters now to
pass all required state). Add a convenience wrapper with the name
vo_get_src_dst_rects() to vo.c. Replace all aspdat and vo usages with
immediate parameters.
Functionally, nothing should change, except that the window size is
clamped to a minimum of size 1 much earlier, and some log messages
change the prefix (don't bother with vo.vo_log stuff).
The plan is to make all the code in aspect.c independent from vo.c,
which should make the code easier to understand, will allow removal of
vo->aspdat, and reduces the amount of code that accesses weird mutable
struct vo fields.
vo->aspdat is basically an outdated version of vo->params, plus some
weirdness. Get rid of it, which will allow further cleanups and which
will make multithreading easier (less state to care about).
Also, simplify some VO code by using mp_image_set_attributes() instead
of caring about display size, colorspace, etc. manually. Add the
function osd_res_from_image_params(), which is often needed in the case
OSD renders into an image.
Some mpv builds identify with e.g. "mpv b'0.3.3' ". The version looks
like str() was called on a Python byte string. I couldn't reproduce it
on my machine (I tried with both Python 2 and 3), so I'm not exactly
sure what's going on here, but I'm hoping this commit does fix it.
This was broken yesterday: the playlist demuxer will always fall back to
plaintext playlist files, which will cause the ASX playlist parser and
some others never to be called.
This is probably ok. Probing could hit this case very often, since it'll
mean running this function on potentially binary data, but on the other
hand, probing usually uses a memory stream (to limit the amount of data
read), and memory streams have s->log silenced (details see
open_memory_stream()).
This simplifies the implementation and should make it more robust. For
example, we return an error if a line is longer than the provided buffer
(instead of splitting the line).
The code is much shorter, because now finding the new line and reading
characters is done in one go.
As of this commit, stream_read_line() can't actually error (except in
the case the passed in buffer is 0, which never happens here). This
commit is preparation for the following commit, which checks harder
whether the read data is actually text. Before this commit, an error
was treated as end-of-file, but the data read so far was considered
valid.
Signed-off-by: wm4 <wm4@nowhere>
This merges pull request #496. The problem was that at least the
initialization of the distance[] array accessed af_fmtstr_table[]
entries that were out of bounds. Small cosmetic changes applied to
the original pull request.
Doesn't affect the generated code, but avoids confusion
in both humans and newer Coverity versions.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@36623 b3059339-0415-0410-9bf9-f77b7e298cf2
The struct we need to copy is actually a cdrom_msf0, not cdrom_msf.
Even though the kernel for no good reason reads it in as a
cdrom_msf struct, but only uses the part shared with cdrom_msf0 -
this is probably a kernel bug.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@36622 b3059339-0415-0410-9bf9-f77b7e298cf2
Do two things:
1. add locking to struct osd_state
2. make struct osd_state opaque
While 1. is somewhat simple, 2. is quite horrible. Lots of code accesses
lots of osd_state (and osd_object) members. To make sure everything is
accessed synchronously, I prefer making osd_state opaque, even if it
means adding pretty dumb accessors.
All of this is meant to allow running VO in their own threads.
Eventually, VOs will request OSD on their own, which means osd_state
will be accessed from foreign threads.
The plan is to make the whole OSD thread-safe, and we start with this.
We just put locks on all entry points (fortunately, dec_sub.c and all
sd_*.c decoders are very closed off, and only the entry points in
dec_sub.h let you access it). I think this is pretty ugly, but at least
it's very simple.
There's a special case with sub_get_bitmaps(): this function returns
pointers to decoder data (specifically, libass images). There's no way
to synchronize this internally, so expose sub_lock/sub_unlock functions.
To make things simpler, and especially because the lock is sort-of
exposed to the outside world, make the locks recursive. Although the
only case where this is actually needed (although trivial) is
sub_set_extradata().
One corner case are ASS subtitles: for some reason, we keep a single
ASS_Renderer instance for subtitles around (probably to avoid rescanning
fonts with ordered chapters), and this ASS_Renderer instance is not
synchronized. Also, demux_libass.c loads ASS_Track objects, which are
directly passed to sd_ass.c. These things are not synchronized (and
would be hard to synchronize), and basically we're out of luck. But I
think for now, accesses happen reasonably serialized, so there is no
actual problem yet, even if we start to access OSD from other threads.
These were needed before the last commit, but now they don't do anything
anymore. (They were used to decide whether to replace or stack the
previous OSD message when a new one was displayed.)
If certain OSD messages were displayed at the same time, the hidden
messages were put on the stack, and displayed again once the higher
priority messages disappeared. The idea was probably that lower priority
messages could not hide higher priority ones, and also that the lower
messages did not get lost.
But in practice, this gives confusing results with OSD messages randomly
reappearing for a brief time. Remove it.
Showing subtitles on terminal used the OSD message stack (which uses a
stack to "pile up" messages that were displayed at the same time). This
had a bunch of weird and annoying consequences. This accessed a certain
osd_state field, which is a minor annoyance since I want to make that
struct opaque. Implement this differently.
The values set by this new option can be queried by Lua scripts using
the mp.getopt() function. The function takes a string parameter, and
returns the value of the first key that matches. If no key matches, nil
is returned.
When the Lua code was written, the core didn't have names for log levels
yet (just numbers). The only user visible change is that "verbose"
becomes "v", since this level had different names.
Adds the following Lua function to enable message events:
mp.enable_messages(size, level)
size is the maximum number of messages the ringbuffer consists of. level
is the minimum log level for a message to be added to the ringbuffer,
and uses the same values as the mp.log() function. (Actually not yet,
but this will be fixed in the following commit.)
The messages will be delivered via the mp_event() in the user script,
using "message" as event name. The event argument is a table with the
following fields:
level: log level of the message (string as in mp.log())
prefix: string identifying the module of origin
text: contents of the message
As of currently, the message text will contain newline characters. A
message can consist of several lines. It is also possible that a
message doesn't end with a newline, and a caller can use multiple
messages to "build" a line. Most messages will contain exactly 1 line
ending with a single newline character, though.
If the message buffer overflows (messages are not read quickly enough),
new messages are lost until the queued up messages are read. At the
point of the overflow, a special overflow message is inserted. It will
have prefix set to "overflow", and the message text is set to "".
Care should be taken not to print any messages from the message event
handler. This would lead to an infinite loop (the event handler would be
called again after returning, because a new message is available). This
includes mp.log() and all mp.msg.* functions. Keep in mind that the Lua
print() function is mapped to mp.msg.info().
Until now, mp_msg output always went to the terminal. There was no way
to grab the stream of output messages. But this will be needed by
various future changes: Lua scripts, slave mode, client library...
This commit allows registering a ring buffer. A callback would be more
straight-forward, but since msg.c sits at the bottom of the lock
hierarchy (it's used by virtually everything), this would probably be a
nightmare. A ring buffer will be simpler and more predictable in the
long run.
We allocate new memory for each ringbuffer entry, which is probably a
bit expensive. We could try to be clever and somehow pack the data
directly into the buffer, but I felt like this wouldn't be worth the
complexity. You'd have to copy the data a bunch of times anyway. I'm
hoping that we can get away with using the ringbuffer mechanism for
low frequency important messages only (and not e.g. for high volume
debug messages), so the cost doesn't matter that much.
A ringbuffer has a simple, single log level. I considered allowing
--msglevel style per-prefix configuration for each ringbuffer, but
that would have been pretty complicated to implement, and wouldn't
have been that useful either.
This makes
mp_msg(x, y, "a\nb\n")
behave the same as
mp_msg(x, y, "a\n")
mp_msg(x, y, "b\n")
which is probably what one would expect. Before this commit, the "b"
line didn't have a prefix when using ths single mp_msg call.
Before that, it just returned -1.
The print case is inconsistent with that, but I'll leave it for now,
because it's consistent with status line / show_progress behavior.
This is a bit of a hack, but in order to prevent TranslateMessage from
seeing WM_KEYDOWN messages that we already know how to decode, move the
decoding logic to the event loop. This should fix#476, since it stops
the generation of extraneous WM_CHAR messages that were triggering more
than one action on keydown.
Starting a network stream could stall by executing uncacheable stream
control requests (STREAM_CTRL_GET_LANG and STREAM_CTRL_GET_DVD_INFO).
Being uncacheable means the player has to wait until the cache is done
reading the current block of data. These requests can't be cached
because they're too complicated, so the only way to avoid them is
special casing the DVD and Bluray streams (which are the only things
which need these requests), and not doing them in other cases.
(This is kind of inelegant, but so is the rest of the DVD/BD code.)
Application icon was added to the Dock only when run inside of a bundle. That
was handled automatically by OS X using the Info.plist definition.
To add the Application icon when run as a CLI program, I used the samme
approach in the X11 code and loaded the icon as a static binary blob inside
of mpv's binary. This is the simplest approach as it avoid headackes when
relocating the binary and such.
Many ebml_read_* functions have a length int pointer parameter, which
returns the number of bytes skipped. Nothing actually needed this
(anymore), and code using it was rather hard to understand, so get rid
of them.
Matroska makes it pretty hard to resync correctly on broken files:
random data returns "valid" EBML IDs with a high probability, and when
trying to skip them it's likely that you skip a random amount of data
(instead of considering the element length invalid).
Improve upon this by skipping known level 1 elements only. Consider
everything else invalid and call the resync code. This might result in
annoying behavior when Matroska adds new level 1 elements, although it
won't be particularly harmful. Matroska doesn't really allow us to do
better (even mkvtoolnix explicitly checks for known level 1 elements).
Since we now don't always want to combine EBML element skipping and
resyncing, remove ebml_read_skip_or_resync_cluster(), and make
ebml_read_skip() more tolerant against skipping broken elements.
Also, don't resync when reading sub-elements, and instead do resyncing
when reading them results in an error.
Until now, corrupted files were detected if the size of an element (that
should be skipped) was larger than the remaining file. This still could
skip larger regions of the file itself if the broken size happened to be
within the file.
Change it so that it's never allowed to skip outside the parent's
element.
The terminal OSD code includes the handling of the terminal status line,
showing player OSD messages on the terminal, and showing subtitles on
terminal (the latter two only if there is no video window, or if
terminal OSD is forced).
This didn't handle some corner cases correctly. For example, showing an
OSD message on the terminal always cleared the previous line, even if
the line was an important message (or even just the command prompt, if
most other messages were silenced).
Attempt to handle this correctly by keeping track of how many lines the
terminal OSD currently consists of. Since there could be race conditions
with other messages being printed, implement this in msg.c. Now msg.c
expects that MSGL_STATUS messages rewrite the status line, so the caller
is forced to use a single mp_msg() call to set the status line.
Instead of littering print_status() all over the place, update the
status only once per playloop iteration in update_osd_msg(). In audio-
only mode, the status line might now be a little bit off, but it's
perhaps ok.
Print the status line only if it has changed, or if another message was
printed. This might help with extremely slow terminals, although in
audio+video mode, it'll still be updated very often (A-V sync display
changes on every frame).
Instead of hardcoding the terminal sequences, use
terminfo/termcap to get the sequences. Remove the --term-osd-esc option,
which allowed to override the hardcoded escapes - it's useless now.
The fallback for terminals with no escape sequences for moving the
cursor and clearing a line is removed. This somewhat breaks status line
display on these terminals, including the MS Windows console: instead of
querying the terminal size and clearing the line manually by padding the
output with spaces, the line is simply not cleared. I don't expect this
to be a problem on UNIX, and on MS Windows we could emulate escape
sequences. Note that terminal OSD (other than the status line) was
broken anyway on these terminals.
In osd.c, the function get_term_width() is not used anymore, so remove
it. To remind us that the MS Windows console apparently adds a line
break when writint the last column, adjust screen_width in terminal-
win.c accordingly.
Seeking usually show the status on OSD. In terminal OSD mode, no status
is shown, because there is already a separate status line.
Unfortunately, the mechanism for showing the status was still active,
which forced showing no message while the code for showing seek status
was active.
Insane .ass subtitle scripts can cause severe slowdown (depending on the
speed of the machine, or the insanity of the script), so mention how to
test without subtitles. This is mainly to make the user aware that
subtitle rendering can be a problem. For longwinded explanation, there
isn't enough space.
Doesn't make any sense anymore. X11 (which was mentioned in the manpage)
autodetects it, and everything else ignored the option values.
Since for incomprehensible reasons the backends and vo.c still need to
exchange information about the screensize using the option fields,
they're not removed yet.
For some reason, this made all VO backends both set the screen
resolution in opts->screenwidth/height, and call
aspect_save_screenres(). Remove the latter. Move the code to calculate
the PAR-corrected window size from aspect.c to vo.c, and make it so that
the monitor PAR is recalculated when it makes sense.
When using --monitoraspect, but either the screen width or height or
both are unknown, a fallback is applied. This is a completely useless
obscure corner case that's going to help nobody, so get it out of the
way.
For a long time the cocoa backend set the xinerama_x/y and used dx/dy from the
VO instance. This somewhat worked with some workarounds but wasn't really
what was supposed to be happening. Moreover 27e4360, which touched this
workaround introduced a regression.
New code doesn't set the xinerama_x/y values so that dx/dy are offsets in the
current screen (not a virtual screen composed of all the screens). The screen
reference detected during VOCTRL_UPDATE_SCREENINFO is also passed down to the
window initialization code.
Fixes#472
Like with the previous commit, this is probably not needed, but it's
unclear whether that really is the case. Most likely, it used to be
needed by some demuxer, and now the only demuxer left that could
_possibly_ trigger this is demux_mkv.c.
Note that mjpeg is the only decoder that reads the extra_huff option,
and nothing in libavformat actually sets the option. So maybe it's
fundamentally not needed anymore.
This case can't happen with the normal realvideo codepath in
demux_mkv.c, because the code would errors out if the extradata is too
small, and everything would be broken anyway in the case the vd_lavc.c
condition is actually triggered.
It still might happen with VfW-muxed realvideo in Matroska, though.
Basically, I'm hoping this doesn't matter anyway, and that the vd_lavc.c
code was for other old demuxers, like demux_avi or demux_rm. Following
the commit history, it's not really clear for what demuxer this code
was added.
MicroDVD files _can_ contain real timestamps instead of frame timestamps
if they declare a FPS. But this seems to be rare, so ignore that if the
FPS happens to match with the libavformat microdvd parser's default FPS.
This might actually break files that declare 23.976 FPS, but the video
file is not 23.976 FPS, but the chance that this happens is probably
very low, and the commit fixes the more common breakage with 25 FPS
video.
Subtitle formats with frame based timing require using the video FPS to
compute proper subtitle timestamps. But it looks like the calculation to
do that was inversed.
This basically reverts the default as set by commit 812798c5. This seems
to be a matter of taste, but personally I think keeping the pause
setting is better.
Use the video chain for this instead. This is for facilitating coming
changes, which will clean up the vo->aspdat stuff, and this code would
be in the way.
On X11, if no wayland compositor is running, wl_list_init() will never
be called. This will cause destroy_display() to segfault when trying to
iterate over the list.
This fixes two things:
1. Dropping files on the VO window will auto-load subtitles (since most
drag & drop code prefixes the filenames with 'file://', and the
subtitle auto-load code considers 'file://' non-local)
2. Fix behavior of the %x screenshot filename template (similar problem)
One could force all that code to special-case 'file://' URLs, but just
replacing the filename on playback start is simpler.
The user_data is passed on add_listener and can later be changed with
set_user_data. But because we don't want to change it later and because it is
the same object remove the set_user_data call.
This might be a copy&paste leftover from the initial draft for the wayland
backend.
This finally gets rid of the LaTeX dependency.
We should actually be using docultils directly here, but I didn't
do this because of all the potential Python 2/3 breakage.
1000ms is a bit insane. It makes behavior on playback speed changes
worse (because the player has to catch up the dropped audio due to
audio-chain reset), and perhaps makes seeking slower.
Note that the problem of playback speed changes misbehaving will be
fixed in the future, but even then we don't want to have a buffer that
large.
I added enough logic to never set ontop or fullscreen twitce.
This commit keeps also the size of the video if multiple videos are played.
If the ratio differs the width will be kept at the same size and only the
height changes.
libwayland-client contains the following code [1]:
runtime_dir = getenv("XDG_RUNTIME_DIR");
if (!runtime_dir) {
fprintf(stderr,
"error: XDG_RUNTIME_DIR not set in the environment.\n");
This means this message will unconditionally and unavoidably be printed
if XDG_RUNTIME_DIR is not set. Since mpv is a terminal program, and we
want to avoid unnecessary output, work it around by not attempting to
use wayland if this environment variable is not set.
[1] http://cgit.freedesktop.org/wayland/wayland/tree/src/wayland-client.c#n636
(cd0dccd01e16fa404e03974d30ded3aebdb1c4bc)
This commonly happens when initializing vo_opengl on a X11-only system.
Unfortunately, most wl_*_destroy() functions appear not to accept NULL
pointers, making partial deinitialization a pain: you have to add your
own NULL checks everywhere to avoid crashes.
xkb.context is uninitialized separately, because you can initialize it
just fine, even if the rest of input initialization fails.
When both libavresample and libswresample were detected, the script
enabled both at the same time. This is not supported; although nothing
bad happened apparently. Make the dependencies both mutually exclusive.
This was inconsistent: the actual statusline used [statusline] as
message prefix, while other parts of the terminal OSD used [cplayer]
(and MSGL_STATUS). This commit makes it consistent.
Note that we can't use mp_msg, because it's not async-signal safe (we
might be running other threads while forking, so only functions
specified to be async-signal safe can be called, and this doesn't
include stdio; mp_msg acquires a mutex too).
Also, always print a \n before running the program to flush the status
line. The effect is that a program running successfully as well as the
error message will effectively start on a new line.
Quvi subtitles are considered external subtitles (simply because they're
separate from the audio/video stream), but for the sake of subtitle
auto-selection, they should not be considered external.
Change this so that quvi subtitles are treated like muxed subtitles
(with default flag never set). This means subtitles won't be selected by
default, unless explicitly requested with --sid or --slang.
Because of this commit there were problems displaying the frmase in their right
order.
This reverts commit 96e75d234a.
Conflicts:
video/out/gl_wayland.c
video/out/wayland_common.h
The changes in the vo_wayland_ontop function have no effect on the workaround.
Somehow the problem just disappeared. I guess it is because of the new control
function in gl_wayland.c where the resize happens immediatly after the event
dispatch/flush.
I don't like this function at all, but it's basically a trick to get the
input's mp_log instance in a case the mp_input_parse_cmd_strv() is
almost certainly not going to output anything. But still make it
somewhat more consistent with mp_input_parse_cmd_strv() - why force the
caller to always use MP_ON_OSD_AUTO?
Both X11 and Wayland support the same format for drag & drop operations
(text/uri-list), and the code for that was copied from x11_common.c to
wayland_common.c. Factor it out.
demux_subreader.c contains the old MPlayer subtitle parser, and I have
absolutely no confidence in this (very crappy) code. There might be
one or two security risks associated with running that code on
arbitrary input.
This solves the issue where we would not receive any frame events. The
difference to my earlier tests is that now it looks like eglSwapBuffers uses
it's own event queue or something similiar along the lines. Becaues the
performance is the same as without any redraw callback.
At the moment there are visual glitches when we resize the window. This happens
because in wayland there a special function for resizing EGL windows.
To prevent the glitches move the egl_context to the wayland state in
wayland_common.h and add a new control function to gl_wayland.c to wrap the
vo_wayland_control function to check for resize events.
With the new control wrapper the glitches are gone and the resizing is fluid.
The reason a segmentation happend here was because we couldn't get the
requested minor version. The major version is enough for differentiating
between OpenGL 3 and OpenGL 2. If it fails there is still a fallback to any
version available.
Also add a warning if we use the fallback.
Note that we don't try to be clever about detecting the files as
subtitles: we just check the file extension. We could go all the way and
check the files by opening them with a demuxer, but that would probably
do more bad than good.
Drag and drop is pretty complicated (just note how the number of Atoms
in use almost doubles), so I'm not sure whether this works everywhere.
This has been written by looking at the specification [1] or what looks
like the specification, and some external example code [2]. (The latter
one has no code license, but we didn't copy any code.)
We completely ignore the "requirement" of the spec. that the filename
"must" include username and hostname, e.g. "file://user@host/path/file".
In theory, this is required because X is network transparent, but at
this point the so called network transparency is a complete joke, and
Konqueror for one didn't include hostnames in "file://" URIs.
Tested with konqueror as drop source.
[1] http://www.newplanetsoftware.com/xdnd/
[2] http://www.edwardrosten.com/code/dist/x_clipboard-1.1/paste.cc
This is necessary to start mpv without forcing a console window,
but also breaks console usability. A workaround is to call mpv
from a wrapper process that uses the console subsystem and helps
redirecting the standard streams and WriteConsole output to where
they belong.
Streams like CDDA have special requirements in what quantities data can
be read: you can only read a sector at once, not more and not less. The
stream_peek() function didn't respect that and set less (used internal
buffer size of 2048 bytes, instead of CD sector size of 2352 bytes), so
no data was read and EOF was accidentally set, making playback with
cdda:// fail.
This is a regression since commit 9a723f, but that commit merely exposed
the issue (the redundant seek would clear the EOF flag).
Add check in old-configure as well. Reformat the check to use a maximum of 80
columns in the wscript.
Signed-off-by: Stefano Pigozzi <stefano.pigozzi@gmail.com>
This code is shared between input.conf parser and option parser. Until
now, the performance didn't really matter. But I want to use this code
for JSON parsing too, and since JSON will have to be parsed a lot, it
should probably try to avoid realloc'ing too much.
This commit moves parsing of C-style escaped strings into a common
function, and allows using it in a way realloc can be completely
avoided, if the already allocated buffer is large enough.
Apparently this can be really useful when being paranoid and trying to
avoid too much malloc/realloc, since it can be used to appending into a
buffer (with transparent realloc only if the buffer is too small).
@@ -201,7 +201,7 @@ void af_window_kaiser(int n, FLOAT_TYPE* w, FLOAT_TYPE b)
{
FLOAT_TYPEtmp;
FLOAT_TYPEk1=1.0/besselizero(b);
intk2=1-(n&1);
intk2=1-(n&1);
intend=(n+1)>>1;
inti;
Loaded 100 of 466 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Alessandro Ghedini