EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
/*
|
2013-07-12 20:12:02 +00:00
|
|
|
* Original author: Uoti Urpala
|
|
|
|
*
|
2015-04-13 07:36:54 +00:00
|
|
|
* This file is part of mpv.
|
|
|
|
*
|
Relicense some non-MPlayer source files to LGPL 2.1 or later
This covers source files which were added in mplayer2 and mpv times
only, and where all code is covered by LGPL relicensing agreements.
There are probably more files to which this applies, but I'm being
conservative here.
A file named ao_sdl.c exists in MPlayer too, but the mpv one is a
complete rewrite, and was added some time after the original ao_sdl.c
was removed. The same applies to vo_sdl.c, for which the SDL2 API is
radically different in addition (MPlayer supports SDL 1.2 only).
common.c contains only code written by me. But common.h is a strange
case: although it originally was named mp_common.h and exists in MPlayer
too, by now it contains only definitions written by uau and me. The
exceptions are the CONTROL_ defines - thus not changing the license of
common.h yet.
codec_tags.c contained once large tables generated from MPlayer's
codecs.conf, but all of these tables were removed.
From demux_playlist.c I'm removing a code fragment from someone who was
not asked; this probably could be done later (see commit 15dccc37).
misc.c is a bit complicated to reason about (it was split off mplayer.c
and thus contains random functions out of this file), but actually all
functions have been added post-MPlayer. Except get_relative_time(),
which was written by uau, but looks similar to 3 different versions of
something similar in each of the Unix/win32/OSX timer source files. I'm
not sure what that means in regards to copyright, so I've just moved it
into another still-GPL source file for now.
screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but
they're all gone.
2016-01-19 17:36:06 +00:00
|
|
|
* mpv is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU Lesser General Public
|
|
|
|
* License as published by the Free Software Foundation; either
|
|
|
|
* version 2.1 of the License, or (at your option) any later version.
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
*
|
2015-04-13 07:36:54 +00:00
|
|
|
* mpv is distributed in the hope that it will be useful,
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
Relicense some non-MPlayer source files to LGPL 2.1 or later
This covers source files which were added in mplayer2 and mpv times
only, and where all code is covered by LGPL relicensing agreements.
There are probably more files to which this applies, but I'm being
conservative here.
A file named ao_sdl.c exists in MPlayer too, but the mpv one is a
complete rewrite, and was added some time after the original ao_sdl.c
was removed. The same applies to vo_sdl.c, for which the SDL2 API is
radically different in addition (MPlayer supports SDL 1.2 only).
common.c contains only code written by me. But common.h is a strange
case: although it originally was named mp_common.h and exists in MPlayer
too, by now it contains only definitions written by uau and me. The
exceptions are the CONTROL_ defines - thus not changing the license of
common.h yet.
codec_tags.c contained once large tables generated from MPlayer's
codecs.conf, but all of these tables were removed.
From demux_playlist.c I'm removing a code fragment from someone who was
not asked; this probably could be done later (see commit 15dccc37).
misc.c is a bit complicated to reason about (it was split off mplayer.c
and thus contains random functions out of this file), but actually all
functions have been added post-MPlayer. Except get_relative_time(),
which was written by uau, but looks similar to 3 different versions of
something similar in each of the Unix/win32/OSX timer source files. I'm
not sure what that means in regards to copyright, so I've just moved it
into another still-GPL source file for now.
screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but
they're all gone.
2016-01-19 17:36:06 +00:00
|
|
|
* GNU Lesser General Public License for more details.
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
*
|
Relicense some non-MPlayer source files to LGPL 2.1 or later
This covers source files which were added in mplayer2 and mpv times
only, and where all code is covered by LGPL relicensing agreements.
There are probably more files to which this applies, but I'm being
conservative here.
A file named ao_sdl.c exists in MPlayer too, but the mpv one is a
complete rewrite, and was added some time after the original ao_sdl.c
was removed. The same applies to vo_sdl.c, for which the SDL2 API is
radically different in addition (MPlayer supports SDL 1.2 only).
common.c contains only code written by me. But common.h is a strange
case: although it originally was named mp_common.h and exists in MPlayer
too, by now it contains only definitions written by uau and me. The
exceptions are the CONTROL_ defines - thus not changing the license of
common.h yet.
codec_tags.c contained once large tables generated from MPlayer's
codecs.conf, but all of these tables were removed.
From demux_playlist.c I'm removing a code fragment from someone who was
not asked; this probably could be done later (see commit 15dccc37).
misc.c is a bit complicated to reason about (it was split off mplayer.c
and thus contains random functions out of this file), but actually all
functions have been added post-MPlayer. Except get_relative_time(),
which was written by uau, but looks similar to 3 different versions of
something similar in each of the Unix/win32/OSX timer source files. I'm
not sure what that means in regards to copyright, so I've just moved it
into another still-GPL source file for now.
screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but
they're all gone.
2016-01-19 17:36:06 +00:00
|
|
|
* You should have received a copy of the GNU Lesser General Public
|
|
|
|
* License along with mpv. If not, see <http://www.gnu.org/licenses/>.
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <stdbool.h>
|
|
|
|
#include <string.h>
|
2015-02-17 22:48:39 +00:00
|
|
|
#include <inttypes.h>
|
2020-02-15 17:09:28 +00:00
|
|
|
#include <limits.h>
|
2015-02-17 22:48:39 +00:00
|
|
|
#include <math.h>
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
|
2016-01-11 18:03:40 +00:00
|
|
|
#include "mpv_talloc.h"
|
2015-02-17 22:48:39 +00:00
|
|
|
|
2015-05-24 21:26:54 +00:00
|
|
|
#include "demux.h"
|
|
|
|
#include "timeline.h"
|
2015-02-17 22:48:39 +00:00
|
|
|
#include "common/msg.h"
|
|
|
|
#include "options/path.h"
|
|
|
|
#include "misc/bstr.h"
|
|
|
|
#include "common/common.h"
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
#include "stream/stream.h"
|
|
|
|
|
2013-11-21 14:59:00 +00:00
|
|
|
#define HEADER "# mpv EDL v0\n"
|
2013-11-19 21:23:41 +00:00
|
|
|
|
2015-02-17 22:48:39 +00:00
|
|
|
struct tl_part {
|
|
|
|
char *filename; // what is stream_open()ed
|
|
|
|
double offset; // offset into the source file
|
|
|
|
bool offset_set;
|
|
|
|
bool chapter_ts;
|
2019-10-06 21:31:16 +00:00
|
|
|
bool is_layout;
|
2015-02-17 22:48:39 +00:00
|
|
|
double length; // length of the part (-1 if rest of the file)
|
2018-07-31 17:46:19 +00:00
|
|
|
char *title;
|
2015-02-17 22:48:39 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
struct tl_parts {
|
2019-01-05 07:43:58 +00:00
|
|
|
bool disable_chapters;
|
2020-02-15 17:09:28 +00:00
|
|
|
bool dash, no_clip, delay_open;
|
2017-01-30 18:38:43 +00:00
|
|
|
char *init_fragment_url;
|
2020-02-20 23:19:17 +00:00
|
|
|
struct sh_stream **sh_meta;
|
|
|
|
int num_sh_meta;
|
2015-02-17 22:48:39 +00:00
|
|
|
struct tl_part *parts;
|
|
|
|
int num_parts;
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
struct tl_parts *next;
|
2015-02-17 22:48:39 +00:00
|
|
|
};
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
struct tl_root {
|
|
|
|
struct tl_parts **pars;
|
|
|
|
int num_pars;
|
|
|
|
};
|
|
|
|
|
2015-02-17 22:49:38 +00:00
|
|
|
struct priv {
|
|
|
|
bstr data;
|
|
|
|
};
|
|
|
|
|
2020-02-15 14:07:52 +00:00
|
|
|
// Static allocation out of laziness.
|
|
|
|
#define NUM_MAX_PARAMS 20
|
|
|
|
|
|
|
|
struct parse_ctx {
|
|
|
|
struct mp_log *log;
|
|
|
|
bool error;
|
|
|
|
bstr param_vals[NUM_MAX_PARAMS];
|
|
|
|
bstr param_names[NUM_MAX_PARAMS];
|
|
|
|
int num_params;
|
|
|
|
};
|
|
|
|
|
|
|
|
// This returns a value with bstr.start==NULL if nothing found. If the parameter
|
|
|
|
// was specified, bstr.str!=NULL, even if the string is empty (bstr.len==0).
|
|
|
|
// The parameter is removed from the list if found.
|
|
|
|
static bstr get_param(struct parse_ctx *ctx, const char *name)
|
2015-02-17 22:48:39 +00:00
|
|
|
{
|
2020-02-15 14:07:52 +00:00
|
|
|
bstr bname = bstr0(name);
|
|
|
|
for (int n = 0; n < ctx->num_params; n++) {
|
|
|
|
if (bstr_equals(ctx->param_names[n], bname)) {
|
|
|
|
bstr res = ctx->param_vals[n];
|
|
|
|
int count = ctx->num_params;
|
|
|
|
MP_TARRAY_REMOVE_AT(ctx->param_names, count, n);
|
|
|
|
count = ctx->num_params;
|
|
|
|
MP_TARRAY_REMOVE_AT(ctx->param_vals, count, n);
|
|
|
|
ctx->num_params -= 1;
|
|
|
|
if (!res.start)
|
|
|
|
res = bstr0(""); // keep guarantees
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return (bstr){0};
|
|
|
|
}
|
|
|
|
|
|
|
|
// Same as get_param(), but return C string. Return NULL if missing.
|
|
|
|
static char *get_param0(struct parse_ctx *ctx, void *ta_ctx, const char *name)
|
|
|
|
{
|
|
|
|
return bstrdup0(ta_ctx, get_param(ctx, name));
|
|
|
|
}
|
|
|
|
|
2020-02-15 17:09:28 +00:00
|
|
|
// Optional int parameter. Returns the parsed integer, or def if the parameter
|
|
|
|
// is missing or on error (sets ctx.error on error).
|
|
|
|
static int get_param_int(struct parse_ctx *ctx, const char *name, int def)
|
|
|
|
{
|
|
|
|
bstr val = get_param(ctx, name);
|
|
|
|
if (val.start) {
|
|
|
|
bstr rest;
|
|
|
|
long long ival = bstrtoll(val, &rest, 0);
|
|
|
|
if (!val.len || rest.len || ival < INT_MIN || ival > INT_MAX) {
|
|
|
|
MP_ERR(ctx, "Invalid integer: '%.*s'\n", BSTR_P(val));
|
|
|
|
ctx->error = true;
|
|
|
|
return def;
|
|
|
|
}
|
|
|
|
return ival;
|
|
|
|
}
|
|
|
|
return def;
|
|
|
|
}
|
|
|
|
|
2020-02-15 14:07:52 +00:00
|
|
|
// Optional time parameter. Currently a number.
|
|
|
|
// Returns true: parameter was present and valid, *t is set
|
|
|
|
// Returns false: parameter was not present (or broken => ctx.error set)
|
|
|
|
static bool get_param_time(struct parse_ctx *ctx, const char *name, double *t)
|
|
|
|
{
|
|
|
|
bstr val = get_param(ctx, name);
|
|
|
|
if (val.start) {
|
|
|
|
bstr rest;
|
|
|
|
double time = bstrtod(val, &rest);
|
|
|
|
if (!val.len || rest.len || !isfinite(time)) {
|
|
|
|
MP_ERR(ctx, "Invalid time string: '%.*s'\n", BSTR_P(val));
|
|
|
|
ctx->error = true;
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
*t = time;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
static struct tl_parts *add_part(struct tl_root *root)
|
|
|
|
{
|
|
|
|
struct tl_parts *tl = talloc_zero(root, struct tl_parts);
|
|
|
|
MP_TARRAY_APPEND(root, root->pars, root->num_pars, tl);
|
|
|
|
return tl;
|
|
|
|
}
|
|
|
|
|
2020-02-20 23:19:17 +00:00
|
|
|
static struct sh_stream *get_meta(struct tl_parts *tl, int index)
|
|
|
|
{
|
|
|
|
for (int n = 0; n < tl->num_sh_meta; n++) {
|
|
|
|
if (tl->sh_meta[n]->index == index)
|
|
|
|
return tl->sh_meta[n];
|
|
|
|
}
|
|
|
|
struct sh_stream *sh = demux_alloc_sh_stream(STREAM_TYPE_COUNT);
|
|
|
|
talloc_steal(tl, sh);
|
|
|
|
MP_TARRAY_APPEND(tl, tl->sh_meta, tl->num_sh_meta, sh);
|
|
|
|
return sh;
|
|
|
|
}
|
|
|
|
|
2015-02-17 22:48:39 +00:00
|
|
|
/* Returns a list of parts, or NULL on parse error.
|
|
|
|
* Syntax (without file header or URI prefix):
|
|
|
|
* url ::= <entry> ( (';' | '\n') <entry> )*
|
|
|
|
* entry ::= <param> ( <param> ',' )*
|
|
|
|
* param ::= [<string> '='] (<string> | '%' <number> '%' <bytes>)
|
|
|
|
*/
|
2020-02-15 14:07:52 +00:00
|
|
|
static struct tl_root *parse_edl(bstr str, struct mp_log *log)
|
2015-02-17 22:48:39 +00:00
|
|
|
{
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
struct tl_root *root = talloc_zero(NULL, struct tl_root);
|
|
|
|
struct tl_parts *tl = add_part(root);
|
2015-02-17 22:48:39 +00:00
|
|
|
while (str.len) {
|
2019-03-02 23:14:57 +00:00
|
|
|
if (bstr_eatstart0(&str, "#")) {
|
2015-02-17 22:48:39 +00:00
|
|
|
bstr_split_tok(str, "\n", &(bstr){0}, &str);
|
2019-03-02 23:14:57 +00:00
|
|
|
continue;
|
|
|
|
}
|
2015-02-17 22:48:39 +00:00
|
|
|
if (bstr_eatstart0(&str, "\n") || bstr_eatstart0(&str, ";"))
|
|
|
|
continue;
|
2017-01-30 18:38:43 +00:00
|
|
|
bool is_header = bstr_eatstart0(&str, "!");
|
2020-02-15 14:07:52 +00:00
|
|
|
struct parse_ctx ctx = { .log = log };
|
2015-02-17 22:48:39 +00:00
|
|
|
int nparam = 0;
|
|
|
|
while (1) {
|
|
|
|
bstr name, val;
|
|
|
|
// Check if it's of the form "name=..."
|
|
|
|
int next = bstrcspn(str, "=%,;\n");
|
|
|
|
if (next > 0 && next < str.len && str.start[next] == '=') {
|
|
|
|
name = bstr_splice(str, 0, next);
|
|
|
|
str = bstr_cut(str, next + 1);
|
2019-01-05 07:40:23 +00:00
|
|
|
} else if (is_header) {
|
|
|
|
const char *names[] = {"type"}; // implied name
|
|
|
|
name = bstr0(nparam < 1 ? names[nparam] : "-");
|
2015-02-17 22:48:39 +00:00
|
|
|
} else {
|
|
|
|
const char *names[] = {"file", "start", "length"}; // implied name
|
|
|
|
name = bstr0(nparam < 3 ? names[nparam] : "-");
|
|
|
|
}
|
|
|
|
if (bstr_eatstart0(&str, "%")) {
|
|
|
|
int len = bstrtoll(str, &str, 0);
|
|
|
|
if (!bstr_startswith0(str, "%") || (len > str.len - 1))
|
|
|
|
goto error;
|
|
|
|
val = bstr_splice(str, 1, len + 1);
|
|
|
|
str = bstr_cut(str, len + 1);
|
|
|
|
} else {
|
|
|
|
next = bstrcspn(str, ",;\n");
|
|
|
|
val = bstr_splice(str, 0, next);
|
|
|
|
str = bstr_cut(str, next);
|
|
|
|
}
|
2020-02-15 14:07:52 +00:00
|
|
|
if (ctx.num_params >= NUM_MAX_PARAMS) {
|
|
|
|
mp_err(log, "Too many parameters, ignoring '%.*s'.\n",
|
|
|
|
BSTR_P(name));
|
2019-01-05 07:40:23 +00:00
|
|
|
} else {
|
2020-02-15 14:07:52 +00:00
|
|
|
ctx.param_names[ctx.num_params] = name;
|
|
|
|
ctx.param_vals[ctx.num_params] = val;
|
|
|
|
ctx.num_params += 1;
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
|
|
|
nparam++;
|
|
|
|
if (!bstr_eatstart0(&str, ","))
|
|
|
|
break;
|
|
|
|
}
|
2017-01-30 18:38:43 +00:00
|
|
|
if (is_header) {
|
2020-02-15 14:07:52 +00:00
|
|
|
bstr f_type = get_param(&ctx, "type");
|
2019-01-05 07:40:23 +00:00
|
|
|
if (bstr_equals0(f_type, "mp4_dash")) {
|
2017-01-30 18:38:43 +00:00
|
|
|
tl->dash = true;
|
2020-02-15 14:07:52 +00:00
|
|
|
tl->init_fragment_url = get_param0(&ctx, tl, "init");
|
2019-01-11 11:07:55 +00:00
|
|
|
} else if (bstr_equals0(f_type, "no_clip")) {
|
|
|
|
tl->no_clip = true;
|
2019-01-05 07:40:23 +00:00
|
|
|
} else if (bstr_equals0(f_type, "new_stream")) {
|
2020-02-15 14:22:05 +00:00
|
|
|
// (Special case: ignore "redundant" headers at the start for
|
|
|
|
// general symmetry.)
|
|
|
|
if (root->num_pars > 1 || tl->num_parts)
|
|
|
|
tl = add_part(root);
|
2019-01-05 07:43:58 +00:00
|
|
|
} else if (bstr_equals0(f_type, "no_chapters")) {
|
|
|
|
tl->disable_chapters = true;
|
2020-02-15 17:09:28 +00:00
|
|
|
} else if (bstr_equals0(f_type, "track_meta")) {
|
2020-02-20 23:19:17 +00:00
|
|
|
int index = get_param_int(&ctx, "index", -1);
|
|
|
|
struct sh_stream *sh = index < 0 && tl->num_sh_meta
|
|
|
|
? tl->sh_meta[tl->num_sh_meta - 1]
|
|
|
|
: get_meta(tl, index);
|
2020-02-15 17:09:28 +00:00
|
|
|
sh->lang = get_param0(&ctx, sh, "lang");
|
|
|
|
sh->title = get_param0(&ctx, sh, "title");
|
|
|
|
sh->hls_bitrate = get_param_int(&ctx, "byterate", 0) * 8;
|
2020-02-21 13:16:26 +00:00
|
|
|
bstr flags = get_param(&ctx, "flags");
|
|
|
|
bstr flag;
|
|
|
|
while (bstr_split_tok(flags, "+", &flag, &flags) || flag.len) {
|
|
|
|
if (bstr_equals0(flag, "default")) {
|
|
|
|
sh->default_track = true;
|
|
|
|
} else if (bstr_equals0(flag, "forced")) {
|
|
|
|
sh->forced_track = true;
|
|
|
|
} else {
|
|
|
|
mp_warn(log, "Unknown flag: '%.*s'\n", BSTR_P(flag));
|
|
|
|
}
|
|
|
|
}
|
2020-02-15 17:09:28 +00:00
|
|
|
} else if (bstr_equals0(f_type, "delay_open")) {
|
2020-02-20 23:19:17 +00:00
|
|
|
struct sh_stream *sh = get_meta(tl, tl->num_sh_meta);
|
2020-02-15 17:09:28 +00:00
|
|
|
bstr mt = get_param(&ctx, "media_type");
|
|
|
|
if (bstr_equals0(mt, "video")) {
|
|
|
|
sh->type = sh->codec->type = STREAM_VIDEO;
|
|
|
|
} else if (bstr_equals0(mt, "audio")) {
|
|
|
|
sh->type = sh->codec->type = STREAM_AUDIO;
|
|
|
|
} else if (bstr_equals0(mt, "sub")) {
|
|
|
|
sh->type = sh->codec->type = STREAM_SUB;
|
|
|
|
} else {
|
|
|
|
mp_err(log, "Invalid or missing !delay_open media type.\n");
|
|
|
|
goto error;
|
|
|
|
}
|
|
|
|
sh->codec->codec = get_param0(&ctx, sh, "codec");
|
|
|
|
if (!sh->codec->codec)
|
|
|
|
sh->codec->codec = "null";
|
|
|
|
sh->codec->disp_w = get_param_int(&ctx, "w", 0);
|
|
|
|
sh->codec->disp_h = get_param_int(&ctx, "h", 0);
|
2020-02-21 13:47:50 +00:00
|
|
|
sh->codec->fps = get_param_int(&ctx, "fps", 0);
|
|
|
|
sh->codec->samplerate = get_param_int(&ctx, "samplerate", 0);
|
2020-02-15 17:09:28 +00:00
|
|
|
tl->delay_open = true;
|
2019-01-05 07:41:09 +00:00
|
|
|
} else {
|
2020-02-15 14:07:52 +00:00
|
|
|
mp_err(log, "Unknown header: '%.*s'\n", BSTR_P(f_type));
|
2019-01-05 07:41:09 +00:00
|
|
|
goto error;
|
2017-01-30 18:38:43 +00:00
|
|
|
}
|
2019-01-05 07:40:23 +00:00
|
|
|
} else {
|
2020-02-15 14:07:52 +00:00
|
|
|
struct tl_part p = { .length = -1 };
|
|
|
|
p.filename = get_param0(&ctx, tl, "file");
|
|
|
|
p.offset_set = get_param_time(&ctx, "start", &p.offset);
|
|
|
|
get_param_time(&ctx, "length", &p.length);
|
|
|
|
bstr ts = get_param(&ctx, "timestamps");
|
|
|
|
if (bstr_equals0(ts, "chapters")) {
|
|
|
|
p.chapter_ts = true;
|
|
|
|
} else if (ts.start && !bstr_equals0(ts, "seconds")) {
|
|
|
|
mp_warn(log, "Unknown timestamp type: '%.*s'\n", BSTR_P(ts));
|
|
|
|
}
|
|
|
|
p.title = get_param0(&ctx, tl, "title");
|
|
|
|
bstr layout = get_param(&ctx, "layout");
|
|
|
|
if (layout.start) {
|
|
|
|
if (bstr_equals0(layout, "this")) {
|
|
|
|
p.is_layout = true;
|
|
|
|
} else {
|
|
|
|
mp_warn(log, "Unknown layout param: '%.*s'\n", BSTR_P(layout));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!p.filename) {
|
|
|
|
mp_err(log, "Missing filename in segment.'\n");
|
2019-01-05 07:40:23 +00:00
|
|
|
goto error;
|
2020-02-15 14:07:52 +00:00
|
|
|
}
|
2019-01-05 07:40:23 +00:00
|
|
|
MP_TARRAY_APPEND(tl, tl->parts, tl->num_parts, p);
|
2017-01-30 18:38:43 +00:00
|
|
|
}
|
2020-02-15 14:07:52 +00:00
|
|
|
if (ctx.error)
|
|
|
|
goto error;
|
2020-02-21 11:05:29 +00:00
|
|
|
for (int n = 0; n < ctx.num_params; n++) {
|
|
|
|
mp_warn(log, "Unknown or duplicate parameter: '%.*s'\n",
|
|
|
|
BSTR_P(ctx.param_names[n]));
|
|
|
|
}
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
2020-02-15 14:07:52 +00:00
|
|
|
assert(root->num_pars);
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
for (int n = 0; n < root->num_pars; n++) {
|
2020-02-15 14:07:52 +00:00
|
|
|
if (root->pars[n]->num_parts < 1) {
|
|
|
|
mp_err(log, "EDL specifies no segments.'\n");
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
goto error;
|
2020-02-15 14:07:52 +00:00
|
|
|
}
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
}
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
return root;
|
2015-02-17 22:48:39 +00:00
|
|
|
error:
|
2020-02-15 14:07:52 +00:00
|
|
|
mp_err(log, "EDL parsing failed.\n");
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
talloc_free(root);
|
2015-02-17 22:48:39 +00:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
static struct demuxer *open_source(struct timeline *root,
|
|
|
|
struct timeline_par *tl, char *filename)
|
2015-02-17 22:48:39 +00:00
|
|
|
{
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
for (int n = 0; n < tl->num_parts; n++) {
|
|
|
|
struct demuxer *d = tl->parts[n].source;
|
2019-12-17 00:57:42 +00:00
|
|
|
if (d && d->filename && strcmp(d->filename, filename) == 0)
|
2015-02-17 22:48:39 +00:00
|
|
|
return d;
|
|
|
|
}
|
2017-01-30 18:38:43 +00:00
|
|
|
struct demuxer_params params = {
|
|
|
|
.init_fragment = tl->init_fragment,
|
stream, demux: redo origin policy thing
mpv has a very weak and very annoying policy that determines whether a
playlist should be used or not. For example, if you play a remote
playlist, you usually don't want it to be able to read local filesystem
entries. (Although for a media player the impact is small I guess.)
It's weak and annoying as in that it does not prevent certain cases
which could be interpreted as bad in some cases, such as allowing
playlists on the local filesystem to reference remote URLs. It probably
barely makes sense, but we just want to exclude some other "definitely
not a good idea" things, all while playlists generally just work, so
whatever.
The policy is:
- from the command line anything is played
- local playlists can reference anything except "unsafe" streams
("unsafe" means special stream inputs like libavfilter graphs)
- remote playlists can reference only remote URLs
- things like "memory://" and archives are "transparent" to this
This commit does... something. It replaces the weird stream flags with a
slightly clearer "origin" value, which is now consequently passed down
and used everywhere. It fixes some deviations from the described policy.
I wanted to force archives to reference only content within them, but
this would probably have been more complicated (or required different
abstractions), and I'm too lazy to figure it out, so archives are now
"transparent" (playlists within archives behave the same outside).
There may be a lot of bugs in this.
This is unfortunately a very noisy commit because:
- every stream open call now needs to pass the origin
- so does every demuxer open call (=> params param. gets mandatory)
- most stream were changed to provide the "origin" value
- the origin value needed to be passed along in a lot of places
- I was too lazy to split the commit
Fixes: #7274
2019-12-20 08:41:42 +00:00
|
|
|
.stream_flags = root->stream_origin,
|
2017-01-30 18:38:43 +00:00
|
|
|
};
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
struct demuxer *d = demux_open_url(filename, ¶ms, root->cancel,
|
|
|
|
root->global);
|
2015-02-20 20:56:55 +00:00
|
|
|
if (d) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_TARRAY_APPEND(root, root->sources, root->num_sources, d);
|
2015-02-20 20:56:55 +00:00
|
|
|
} else {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_ERR(root, "EDL: Could not open source file '%s'.\n", filename);
|
2015-02-20 20:56:55 +00:00
|
|
|
}
|
2015-02-17 22:48:39 +00:00
|
|
|
return d;
|
|
|
|
}
|
|
|
|
|
|
|
|
static double demuxer_chapter_time(struct demuxer *demuxer, int n)
|
|
|
|
{
|
|
|
|
if (n < 0 || n >= demuxer->num_chapters)
|
|
|
|
return -1;
|
|
|
|
return demuxer->chapters[n].pts;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Append all chapters from src to the chapters array.
|
|
|
|
// Ignore chapters outside of the given time range.
|
|
|
|
static void copy_chapters(struct demux_chapter **chapters, int *num_chapters,
|
|
|
|
struct demuxer *src, double start, double len,
|
|
|
|
double dest_offset)
|
|
|
|
{
|
|
|
|
for (int n = 0; n < src->num_chapters; n++) {
|
|
|
|
double time = demuxer_chapter_time(src, n);
|
|
|
|
if (time >= start && time <= start + len) {
|
|
|
|
struct demux_chapter ch = {
|
|
|
|
.pts = dest_offset + time - start,
|
2015-08-12 09:10:45 +00:00
|
|
|
.metadata = mp_tags_dup(*chapters, src->chapters[n].metadata),
|
2015-02-17 22:48:39 +00:00
|
|
|
};
|
|
|
|
MP_TARRAY_APPEND(NULL, *chapters, *num_chapters, ch);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void resolve_timestamps(struct tl_part *part, struct demuxer *demuxer)
|
|
|
|
{
|
|
|
|
if (part->chapter_ts) {
|
|
|
|
double start = demuxer_chapter_time(demuxer, part->offset);
|
|
|
|
double length = part->length;
|
|
|
|
double end = length;
|
|
|
|
if (end >= 0)
|
|
|
|
end = demuxer_chapter_time(demuxer, part->offset + part->length);
|
|
|
|
if (end >= 0 && start >= 0)
|
|
|
|
length = end - start;
|
|
|
|
part->offset = start;
|
|
|
|
part->length = length;
|
|
|
|
}
|
|
|
|
if (!part->offset_set)
|
|
|
|
part->offset = demuxer->start_time;
|
|
|
|
}
|
|
|
|
|
2019-01-11 13:10:41 +00:00
|
|
|
static struct timeline_par *build_timeline(struct timeline *root,
|
|
|
|
struct tl_parts *parts)
|
2015-02-17 22:48:39 +00:00
|
|
|
{
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
struct timeline_par *tl = talloc_zero(root, struct timeline_par);
|
|
|
|
MP_TARRAY_APPEND(root, root->pars, root->num_pars, tl);
|
|
|
|
|
2017-01-30 18:38:43 +00:00
|
|
|
tl->track_layout = NULL;
|
|
|
|
tl->dash = parts->dash;
|
2019-01-11 11:07:55 +00:00
|
|
|
tl->no_clip = parts->no_clip;
|
2020-02-15 17:09:28 +00:00
|
|
|
tl->delay_open = parts->delay_open;
|
|
|
|
|
|
|
|
// There is no copy function for sh_stream, so just steal it.
|
2020-02-20 23:19:17 +00:00
|
|
|
for (int n = 0; n < parts->num_sh_meta; n++) {
|
|
|
|
MP_TARRAY_APPEND(tl, tl->sh_meta, tl->num_sh_meta,
|
|
|
|
talloc_steal(tl, parts->sh_meta[n]));
|
|
|
|
parts->sh_meta[n] = NULL;
|
|
|
|
}
|
|
|
|
parts->num_sh_meta = 0;
|
2017-01-30 18:38:43 +00:00
|
|
|
|
|
|
|
if (parts->init_fragment_url && parts->init_fragment_url[0]) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_VERBOSE(root, "Opening init fragment...\n");
|
stream, demux: redo origin policy thing
mpv has a very weak and very annoying policy that determines whether a
playlist should be used or not. For example, if you play a remote
playlist, you usually don't want it to be able to read local filesystem
entries. (Although for a media player the impact is small I guess.)
It's weak and annoying as in that it does not prevent certain cases
which could be interpreted as bad in some cases, such as allowing
playlists on the local filesystem to reference remote URLs. It probably
barely makes sense, but we just want to exclude some other "definitely
not a good idea" things, all while playlists generally just work, so
whatever.
The policy is:
- from the command line anything is played
- local playlists can reference anything except "unsafe" streams
("unsafe" means special stream inputs like libavfilter graphs)
- remote playlists can reference only remote URLs
- things like "memory://" and archives are "transparent" to this
This commit does... something. It replaces the weird stream flags with a
slightly clearer "origin" value, which is now consequently passed down
and used everywhere. It fixes some deviations from the described policy.
I wanted to force archives to reference only content within them, but
this would probably have been more complicated (or required different
abstractions), and I'm too lazy to figure it out, so archives are now
"transparent" (playlists within archives behave the same outside).
There may be a lot of bugs in this.
This is unfortunately a very noisy commit because:
- every stream open call now needs to pass the origin
- so does every demuxer open call (=> params param. gets mandatory)
- most stream were changed to provide the "origin" value
- the origin value needed to be passed along in a lot of places
- I was too lazy to split the commit
Fixes: #7274
2019-12-20 08:41:42 +00:00
|
|
|
stream_t *s = stream_create(parts->init_fragment_url,
|
|
|
|
STREAM_READ | root->stream_origin,
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
root->cancel, root->global);
|
2019-10-08 21:55:05 +00:00
|
|
|
if (s) {
|
|
|
|
root->is_network |= s->is_network;
|
|
|
|
root->is_streaming |= s->streaming;
|
2017-01-30 18:38:43 +00:00
|
|
|
tl->init_fragment = stream_read_complete(s, tl, 1000000);
|
2019-10-08 21:55:05 +00:00
|
|
|
}
|
2017-01-30 18:38:43 +00:00
|
|
|
free_stream(s);
|
|
|
|
if (!tl->init_fragment.len) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_ERR(root, "Could not read init fragment.\n");
|
2017-01-30 18:38:43 +00:00
|
|
|
goto error;
|
|
|
|
}
|
2018-09-07 20:40:12 +00:00
|
|
|
struct demuxer_params params = {
|
|
|
|
.init_fragment = tl->init_fragment,
|
stream, demux: redo origin policy thing
mpv has a very weak and very annoying policy that determines whether a
playlist should be used or not. For example, if you play a remote
playlist, you usually don't want it to be able to read local filesystem
entries. (Although for a media player the impact is small I guess.)
It's weak and annoying as in that it does not prevent certain cases
which could be interpreted as bad in some cases, such as allowing
playlists on the local filesystem to reference remote URLs. It probably
barely makes sense, but we just want to exclude some other "definitely
not a good idea" things, all while playlists generally just work, so
whatever.
The policy is:
- from the command line anything is played
- local playlists can reference anything except "unsafe" streams
("unsafe" means special stream inputs like libavfilter graphs)
- remote playlists can reference only remote URLs
- things like "memory://" and archives are "transparent" to this
This commit does... something. It replaces the weird stream flags with a
slightly clearer "origin" value, which is now consequently passed down
and used everywhere. It fixes some deviations from the described policy.
I wanted to force archives to reference only content within them, but
this would probably have been more complicated (or required different
abstractions), and I'm too lazy to figure it out, so archives are now
"transparent" (playlists within archives behave the same outside).
There may be a lot of bugs in this.
This is unfortunately a very noisy commit because:
- every stream open call now needs to pass the origin
- so does every demuxer open call (=> params param. gets mandatory)
- most stream were changed to provide the "origin" value
- the origin value needed to be passed along in a lot of places
- I was too lazy to split the commit
Fixes: #7274
2019-12-20 08:41:42 +00:00
|
|
|
.stream_flags = root->stream_origin,
|
2018-09-07 20:40:12 +00:00
|
|
|
};
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
tl->track_layout = demux_open_url("memory://", ¶ms, root->cancel,
|
|
|
|
root->global);
|
2017-01-30 18:38:43 +00:00
|
|
|
if (!tl->track_layout) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_ERR(root, "Could not demux init fragment.\n");
|
2017-01-30 18:38:43 +00:00
|
|
|
goto error;
|
|
|
|
}
|
2019-01-20 14:45:55 +00:00
|
|
|
MP_TARRAY_APPEND(root, root->sources, root->num_sources, tl->track_layout);
|
2017-01-30 18:38:43 +00:00
|
|
|
}
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
tl->parts = talloc_array_ptrtype(tl, tl->parts, parts->num_parts);
|
2015-02-17 22:48:39 +00:00
|
|
|
double starttime = 0;
|
|
|
|
for (int n = 0; n < parts->num_parts; n++) {
|
|
|
|
struct tl_part *part = &parts->parts[n];
|
2017-01-30 18:38:43 +00:00
|
|
|
struct demuxer *source = NULL;
|
2015-02-17 22:48:39 +00:00
|
|
|
|
2017-01-30 18:38:43 +00:00
|
|
|
if (tl->dash) {
|
|
|
|
part->offset = starttime;
|
|
|
|
if (part->length <= 0)
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_WARN(root, "Segment %d has unknown duration.\n", n);
|
2017-01-30 18:38:43 +00:00
|
|
|
if (part->offset_set)
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_WARN(root, "Offsets are ignored.\n");
|
2017-02-05 14:51:36 +00:00
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
if (!tl->track_layout)
|
|
|
|
tl->track_layout = open_source(root, tl, part->filename);
|
2020-02-15 17:09:28 +00:00
|
|
|
} else if (tl->delay_open) {
|
|
|
|
if (n == 0 && !part->offset_set) {
|
|
|
|
part->offset = starttime;
|
|
|
|
part->offset_set = true;
|
|
|
|
}
|
|
|
|
if (part->chapter_ts || (part->length < 0 && !tl->no_clip)) {
|
|
|
|
MP_ERR(root, "Invalid specification for delay_open stream.\n");
|
|
|
|
goto error;
|
|
|
|
}
|
2017-01-30 18:38:43 +00:00
|
|
|
} else {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_VERBOSE(root, "Opening segment %d...\n", n);
|
2015-02-17 22:48:39 +00:00
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
source = open_source(root, tl, part->filename);
|
2017-01-30 18:38:43 +00:00
|
|
|
if (!source)
|
|
|
|
goto error;
|
2015-02-17 22:48:39 +00:00
|
|
|
|
2017-01-30 18:38:43 +00:00
|
|
|
resolve_timestamps(part, source);
|
|
|
|
|
2017-06-20 11:57:58 +00:00
|
|
|
double end_time = source->duration;
|
2017-01-30 18:38:43 +00:00
|
|
|
if (end_time >= 0)
|
|
|
|
end_time += source->start_time;
|
|
|
|
|
|
|
|
// Unknown length => use rest of the file. If duration is unknown, make
|
|
|
|
// something up.
|
|
|
|
if (part->length < 0) {
|
|
|
|
if (end_time < 0) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_WARN(root, "EDL: source file '%s' has unknown duration.\n",
|
2017-01-30 18:38:43 +00:00
|
|
|
part->filename);
|
|
|
|
end_time = 1;
|
|
|
|
}
|
|
|
|
part->length = end_time - part->offset;
|
|
|
|
} else if (end_time >= 0) {
|
|
|
|
double end_part = part->offset + part->length;
|
|
|
|
if (end_part > end_time) {
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_WARN(root, "EDL: entry %d uses %f "
|
2017-01-30 18:38:43 +00:00
|
|
|
"seconds, but file has only %f seconds.\n",
|
|
|
|
n, end_part, end_time);
|
|
|
|
}
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
|
|
|
|
2019-01-05 07:43:58 +00:00
|
|
|
if (!parts->disable_chapters) {
|
|
|
|
// Add a chapter between each file.
|
|
|
|
struct demux_chapter ch = {
|
|
|
|
.pts = starttime,
|
|
|
|
.metadata = talloc_zero(tl, struct mp_tags),
|
|
|
|
};
|
|
|
|
mp_tags_set_str(ch.metadata, "title",
|
|
|
|
part->title ? part->title : part->filename);
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
MP_TARRAY_APPEND(root, root->chapters, root->num_chapters, ch);
|
2015-02-17 22:48:39 +00:00
|
|
|
|
2019-01-05 07:43:58 +00:00
|
|
|
// Also copy the source file's chapters for the relevant parts
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
copy_chapters(&root->chapters, &root->num_chapters, source,
|
2019-01-05 07:43:58 +00:00
|
|
|
part->offset, part->length, starttime);
|
|
|
|
}
|
2017-01-30 18:38:43 +00:00
|
|
|
}
|
2015-02-17 22:48:39 +00:00
|
|
|
|
2015-02-17 22:48:47 +00:00
|
|
|
tl->parts[n] = (struct timeline_part) {
|
2015-02-17 22:48:39 +00:00
|
|
|
.start = starttime,
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
.end = starttime + part->length,
|
2015-02-17 22:48:39 +00:00
|
|
|
.source_start = part->offset,
|
|
|
|
.source = source,
|
2017-01-30 18:38:43 +00:00
|
|
|
.url = talloc_strdup(tl, part->filename),
|
2015-02-17 22:48:39 +00:00
|
|
|
};
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
starttime = tl->parts[n].end;
|
2017-02-02 17:38:16 +00:00
|
|
|
|
2019-10-06 21:31:16 +00:00
|
|
|
if (source && !tl->track_layout && part->is_layout)
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
tl->track_layout = source;
|
2019-12-17 00:57:42 +00:00
|
|
|
|
|
|
|
tl->num_parts++;
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
|
2020-02-15 14:24:58 +00:00
|
|
|
if (tl->no_clip && tl->num_parts > 1)
|
|
|
|
MP_WARN(root, "Multiple parts with no_clip. Undefined behavior ahead.\n");
|
|
|
|
|
2019-10-06 21:31:16 +00:00
|
|
|
if (!tl->track_layout) {
|
|
|
|
// Use a heuristic to select the "broadest" part as layout.
|
|
|
|
for (int n = 0; n < parts->num_parts; n++) {
|
|
|
|
struct demuxer *s = tl->parts[n].source;
|
|
|
|
if (!s)
|
|
|
|
continue;
|
|
|
|
if (!tl->track_layout ||
|
|
|
|
demux_get_num_stream(s) > demux_get_num_stream(tl->track_layout))
|
|
|
|
tl->track_layout = s;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-02-15 17:09:28 +00:00
|
|
|
if (!tl->track_layout && !tl->delay_open)
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
goto error;
|
|
|
|
if (!root->meta)
|
|
|
|
root->meta = tl->track_layout;
|
|
|
|
|
2019-12-17 00:57:42 +00:00
|
|
|
assert(tl->num_parts == parts->num_parts);
|
2019-01-11 13:10:41 +00:00
|
|
|
return tl;
|
2015-02-17 22:48:39 +00:00
|
|
|
|
|
|
|
error:
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
root->num_pars = 0;
|
2019-01-11 13:10:41 +00:00
|
|
|
return NULL;
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
|
|
|
|
2020-01-02 22:31:02 +00:00
|
|
|
static void fix_filenames(struct tl_parts *parts, char *source_path)
|
|
|
|
{
|
|
|
|
if (bstr_equals0(mp_split_proto(bstr0(source_path), NULL), "edl"))
|
|
|
|
return;
|
|
|
|
struct bstr dirname = mp_dirname(source_path);
|
|
|
|
for (int n = 0; n < parts->num_parts; n++) {
|
|
|
|
struct tl_part *part = &parts->parts[n];
|
2020-02-15 14:11:05 +00:00
|
|
|
if (!mp_is_url(bstr0(part->filename))) {
|
|
|
|
part->filename =
|
|
|
|
mp_path_join_bstr(parts, dirname, bstr0(part->filename));
|
|
|
|
}
|
2020-01-02 22:31:02 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-02-17 22:48:47 +00:00
|
|
|
static void build_mpv_edl_timeline(struct timeline *tl)
|
2015-02-17 22:48:39 +00:00
|
|
|
{
|
2015-02-17 22:49:38 +00:00
|
|
|
struct priv *p = tl->demuxer->priv;
|
|
|
|
|
2020-02-15 14:07:52 +00:00
|
|
|
struct tl_root *root = parse_edl(p->data, tl->log);
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
if (!root) {
|
2015-02-17 22:48:47 +00:00
|
|
|
MP_ERR(tl, "Error in EDL.\n");
|
2015-02-17 22:48:39 +00:00
|
|
|
return;
|
|
|
|
}
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
|
2019-01-11 13:10:41 +00:00
|
|
|
bool all_dash = true;
|
|
|
|
bool all_no_clip = true;
|
|
|
|
bool all_single = true;
|
|
|
|
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
for (int n = 0; n < root->num_pars; n++) {
|
|
|
|
struct tl_parts *parts = root->pars[n];
|
2020-01-02 22:31:02 +00:00
|
|
|
fix_filenames(parts, tl->demuxer->filename);
|
2019-01-11 13:10:41 +00:00
|
|
|
struct timeline_par *par = build_timeline(tl, parts);
|
|
|
|
if (!par)
|
demux_edl, cue, mkv: clean up timeline stuff slightly
Remove the singly linked list hack, replace it with a slightly more
proper data structure. This probably gets rid of a few minor bugs along
the way, caused by the awkward nonsensical sharing/duplication of some
fields.
Another change (because I'm touching everything related to timeline
anyway) is that I'm removing the special semantics for parts[num_parts].
This is now strictly out of bounds, and instead of using the start time
of the next/beyond-last part, there is an end time field now.
Unfortunately, this also requires touching the code for cue and mkv
ordered chapters. From some superficial testing, they still seem to
mostly work.
One observable change is that the "no_chapters" header is per-stream
now, which is arguably more correct, and getting the old behavior would
require adding code to handle it as special-case, so just adjust
ytdl_hook.lua to the new behavior.
2019-01-11 12:52:29 +00:00
|
|
|
break;
|
2019-01-11 13:10:41 +00:00
|
|
|
all_dash &= par->dash;
|
|
|
|
all_no_clip &= par->no_clip;
|
|
|
|
all_single &= par->num_parts == 1;
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
}
|
2019-01-11 13:10:41 +00:00
|
|
|
|
|
|
|
if (all_dash) {
|
|
|
|
tl->format = "dash";
|
|
|
|
} else if (all_no_clip && all_single) {
|
|
|
|
tl->format = "multi";
|
|
|
|
} else {
|
|
|
|
tl->format = "edl";
|
|
|
|
}
|
|
|
|
|
demux, demux_edl: add extension for tracks sourced from separate streams
This commit adds an extension to mpv EDL, which basically allows you to
do the same as --audio-file, --external-file, etc. in a single EDL file.
This is a relatively quick & dirty implementation. The dirty part lies
in the fact that several shortcuts are taken. For example, struct
timeline now forms a singly linked list, which is really weird, but also
means the other timeline using demuxers (cue, mkv) don't need to be
touched. Also, memory management becomes even worse (weird object
ownership rules that are just fragile WTFs). There are some other
dubious small changes, mostly related to the weird representation of
separate streams.
demux_timeline.c contains the actual implementation of the separate
stream handling. For the most part, most things that used to be on the
top level are now in struct virtual_source, of which one for each
separate stream exists. This is basically like running multiple
demux_edl.c in parallel. Some changes could strictly speaking be split
into a separate commit, such as the stream_map type change.
Mostly untested. Seems to work for the intended purpose. Potential for
regressions for other timeline uses (like ordered chapters) is probably
low. One thing which could definitely break and which I didn't test is
the pseudo-DASH fragmented EDL code, of which ytdl can trigger various
forms in obscure situations. (Uh why don't we have a test suite.)
Background:
The intention is to use this for the ytdl wrapper. A certain streaming
site from a particularly brain damaged and plain evil Silicon Valley
company usually provides streams as separate audio and video streams.
The ytdl wrapper simply does use audio-add (i.e. adding it as external
track, like with --audio-file), which works mostly fine. Unfortunately,
mpv manages caching completely separately for external files. This has
the following potential problems:
1. Seek ranges are rendered incorrectly. They always use the "main"
stream, in this case the video stream. E.g. clicking into a cached range
on the OSC could trigger a low level seek if the audio stream is
actually not cached at the target position.
2. The stream cache bloats unnecessarily. Each stream may allocate the
full configured maximum cache size, which is not what the user intends
to do. Cached ranges are not pruned the same way, which creates disjoint
cache ranges, which only use memory and won't help with fast seeking or
playback.
3. mpv will try to aggressively read from both streams. This is done
from different threads, with no regard which stream is more important.
So it might happen that one stream starves the other one, especially if
they have different bitrates.
4. Every stream will use a separate thread, which is an unnecessary
waste of system resources.
In theory, the following solutions are available (this commit works
towards D):
A. Centrally manage reading and caching of all streams. A single thread
would do all I/O, and decide from which stream it should read next. As
long as the total TCP/socket buffering is not too high, this should be
effective to avoid starvation issues. This can also manage the cached
ranges better. It would also get rid of the quite useless additional
demuxer threads. This solution is conceptually simple, but requires
refactoring the entire demuxer middle layer.
B. Attempt to coordinate the demuxer threads. This would maintain a
shared cache and readahead state to solve the mentioned problems
explicitly. While this sounds simple and like an incremental change,
it's probably hard to implement, creates more messy special cases,
solution A. seems just a better and simpler variant of this. (On the
other hand, A. requires refactoring more code.)
C. Render an intersection of the seek ranges across all streams. This
fixes only problem 1.
D. Merge all streams in a dedicated wrapper demuxer. The general demuxer
layer remains unchanged, and reading from separate streams is handled as
special case. This effectively achieves the same as A. In particular,
caching is simply handled by the usual demuxer cache layer, which sees
the wrapper demuxer as a single stream of interleaved packets. One
implementation variant of this is to reuse the EDL infrastructure, which
this commit does.
All in all, solution A would be preferable, because it's cleaner and
works for all external streams in general.
Some previous commit tried to prepare for implementing solution A. This
could still happen. But it could take years until this is finally
seriously started and finished. In any case, this commit doesn't block
or complicate such attempts, which is also why it's the way to go.
It's worth mentioning that original mplayer handles external files by
creating a wrapper demuxer. This is like a less ideal mixture of A. and
D. (The similarity with A. is that extending the mplayer approach to be
fully dynamic and without certain disadvantages caused by the wrapper
would end up with A. anyway. The similarity with D. is that due to the
wrapper, no higher level code needs to be changed.)
2019-01-04 12:09:02 +00:00
|
|
|
talloc_free(root);
|
2015-02-17 22:48:39 +00:00
|
|
|
}
|
|
|
|
|
2013-07-12 19:58:11 +00:00
|
|
|
static int try_open_file(struct demuxer *demuxer, enum demux_check check)
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
{
|
2016-12-04 22:15:31 +00:00
|
|
|
if (!demuxer->access_references)
|
|
|
|
return -1;
|
|
|
|
|
2015-02-17 22:49:38 +00:00
|
|
|
struct priv *p = talloc_zero(demuxer, struct priv);
|
|
|
|
demuxer->priv = p;
|
2015-02-18 20:10:43 +00:00
|
|
|
demuxer->fully_read = true;
|
2015-02-17 22:49:38 +00:00
|
|
|
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
struct stream *s = demuxer->stream;
|
2017-02-02 17:24:27 +00:00
|
|
|
if (s->info && strcmp(s->info->name, "edl") == 0) {
|
2015-02-17 22:49:38 +00:00
|
|
|
p->data = bstr0(s->path);
|
2013-11-19 21:26:35 +00:00
|
|
|
return 0;
|
|
|
|
}
|
2013-07-12 19:58:11 +00:00
|
|
|
if (check >= DEMUX_CHECK_UNSAFE) {
|
stream: turn into a ring buffer, make size configurable
In some corner cases (see #6802), it can be beneficial to use a larger
stream buffer size. Use this as argument to rewrite everything for no
reason.
Turn stream.c itself into a ring buffer, with configurable size. The
latter would have been easily achievable with minimal changes, and the
ring buffer is the hard part. There is no reason to have a ring buffer
at all, except possibly if ffmpeg don't fix their awful mp4 demuxer, and
some subtle issues with demux_mkv.c wanting to seek back by small
offsets (the latter was handled with small stream_peek() calls, which
are unneeded now).
In addition, this turns small forward seeks into reads (where data is
simply skipped). Before this commit, only stream_skip() did this (which
also mean that stream_skip() simply calls stream_seek() now).
Replace all stream_peek() calls with something else (usually
stream_read_peek()). The function was a problem, because it returned a
pointer to the internal buffer, which is now a ring buffer with
wrapping. The new function just copies the data into a buffer, and in
some cases requires callers to dynamically allocate memory. (The most
common case, demux_lavf.c, required a separate buffer allocation anyway
due to FFmpeg "idiosyncrasies".) This is the bulk of the demuxer_*
changes.
I'm not happy with this. There still isn't a good reason why there
should be a ring buffer, that is complex, and most of the time just
wastes half of the available memory. Maybe another rewrite soon.
It also contains bugs; you're an alpha tester now.
2019-11-06 20:36:02 +00:00
|
|
|
char header[sizeof(HEADER) - 1];
|
|
|
|
int len = stream_read_peek(s, header, sizeof(header));
|
|
|
|
if (len != strlen(HEADER) || memcmp(header, HEADER, len) != 0)
|
2013-07-12 19:58:11 +00:00
|
|
|
return -1;
|
|
|
|
}
|
2015-02-17 22:49:38 +00:00
|
|
|
p->data = stream_read_complete(s, demuxer, 1000000);
|
|
|
|
if (p->data.start == NULL)
|
2013-07-11 18:08:12 +00:00
|
|
|
return -1;
|
2015-02-17 22:49:38 +00:00
|
|
|
bstr_eatstart0(&p->data, HEADER);
|
demux: change hack for closing subtitle files early
Subtitles (and a few other file types, like playlists) are not streamed,
but fully read on opening. This means keeping the file handle or network
socket open is a waste of resources and could cause other weird
behavior. This is why there's a hack to close them after opening.
Change this hack to make the demuxer itself do this, which is less
weird. (Until recently, demuxer->stream ownership was more complex,
which is why it was done this way.)
There is some evil shit due to a huge ownership/lifetime mess of various
objects. Especially EDL (the currently only nested demuxer case)
requires being careful about mp_cancel and passing down stream pointers.
As one defensive programming measure, stop accessing the "stream"
variable in open_given_type(), even where it would still work. This
includes removing a redundant line of code, and removing the peak call,
which should not be needed anymore, as the remaining demuxers do this
mostly correctly.
2018-09-07 21:02:36 +00:00
|
|
|
demux_close_stream(demuxer);
|
2013-07-11 18:08:12 +00:00
|
|
|
return 0;
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
const struct demuxer_desc demuxer_desc_edl = {
|
|
|
|
.name = "edl",
|
2013-11-19 21:23:41 +00:00
|
|
|
.desc = "Edit decision list",
|
2013-07-12 19:58:11 +00:00
|
|
|
.open = try_open_file,
|
2015-02-17 22:48:47 +00:00
|
|
|
.load_timeline = build_mpv_edl_timeline,
|
EDL: add support for new EDL file format
The timeline code previously added to support Matroska ordered
chapters allows constructing a playback timeline from segments picked
from multiple source files. Add support for a new EDL format to make
this machinery available for use with file formats other than Matroska
and in a manner easier to use than creating files with ordered
chapters.
Unlike the old -edl option which specifies an additional file with
edits to apply to the video file given as the main argument, the new
EDL format is used by giving only the EDL file as the file to play;
that file then contains the filename(s) to use as source files where
actual video segments come from. Filename paths in the EDL file are
ignored. Currently the source files are only searched for in the
directory of the EDL file; support for a search path option will
likely be added in the future.
Format of the EDL files
The first line in the file must be "mplayer EDL file, version 2".
The rest of the lines belong to one of these classes:
1) lines specifying source files
2) empty lines
3) lines specifying timeline segments.
Lines beginning with '<' specify source files. These lines first
contain an identifier used to refer to the source file later, then the
filename separated by whitespace. The identifier must start with a
letter. Filenames that start or end with whitespace or contain
newlines are not supported.
On other lines '#' characters delimit comments. Lines that contain
only whitespace after comments have been removed are ignored.
Timeline segments must appear in the file in chronological order. Each
segment has the following information associated with it:
- duration
- output start time
- output end time (= output start time + duration)
- source id (specifies the file the content of the segment comes from)
- source start time (timestamp in the source file)
- source end time (= source start time + duration)
The output timestamps must form a continuous timeline from 0 to the
end of the last segment, such that each new segment starts from the
time the previous one ends at. Source files and times may change
arbitrarily between segments.
The general format for lines specifying timeline segments is
[output time info] source_id [source time info]
source_id must be an identifier defined on a '<' line. Both the time
info parts consists of zero or more of the following elements:
1) timestamp
2) -timestamp
3) +duration
4) *
5) -*
, where "timestamp" and "duration" are decimal numbers (computations
are done with nanosecond precision). Whitespace around "+" and "-" is
optional. 1) and 2) specify start and end time of the segment on
output or source side. 3) specifies duration; the semantics are the
same whether this appears on output or source side. 4) and 5) are
ignored on the output side (they're always implicitly assumed). On the
source side 4) specifies that the segment starts where the previous
segment _using this source_ ended; if there was no previous segment
time 0 is used. 5) specifies that the segment ends where the next
segment using this source starts.
Redundant information may be omitted. It will be filled in using the
following rules:
- output start for first segment is 0
- two of [output start, output end, duration] imply third
- two of [source start, source end, duration] imply third
- output start = output end of previous segment
- output end = output start of next segment
- if "*", source start = source end of earlier segment
- if "-*", source end = source start of a later segment
As a special rule, a last zero-duration segment without a source
specification may appear. This will produce no corresponding segment
in the resulting timeline, but can be used as syntax to specify the
end time of the timeline (with effect equal to adding -time on the
previous line).
Examples:
----- begin -----
mplayer EDL file, version 2
< id1 filename
0 id1 123
100 id1 456
200 id1 789
300
----- end -----
All segments come from the source file "filename". First segment
(output time 0-100) comes from time 123-223, second 456-556, third
789-889.
----- begin -----
mplayer EDL file, version 2
< f filename
f 60-120
f 600-660
f 30- 90
----- end -----
Play first seconds 60-120 from the file, then 600-660, then 30-90.
----- begin -----
mplayer EDL file, version 2
< id1 filename1
< id2 filename2
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
+10 id1 *
+10 id2 *
----- end -----
This plays time 0-10 from filename1, then 0-10 from filename1, then
10-20 from filename1, then 10-20 from filename2, then 20-30 from
filename1, then 20-30 from filename2.
----- begin -----
mplayer EDL file, version 2
< t1 filename1
< t2 filename2
t1 * +2 # segment 1
+2 t2 100 # segment 2
t1 * # segment 3
t2 *-* # segment 4
t1 3 -* # segment 5
+0.111111 t2 102.5 # segment 6
7.37 t1 5 +1 # segment 7
----- end -----
This rather pathological example illustrates the rules for filling in
implied data. All the values can be determined by recursively applying
the rules given above, and the full end result is this:
+2 0-2 t1 0-2 # segment 1
+2 2-4 t2 100-102 # segment 2
+0.758889 4-4.758889 t1 2-2.758889 # segment 3
+0.5 4.4758889-5.258889 t2 102-102.5 # segment 4
+2 5.258889-7.258889 t1 3-5 # segment 5
+0.111111 7.258889-7.37 t2 102.5-102.611111 # segment 6
+1 7.37-8.37 t1 5-6 # segment 7
2011-02-14 11:05:35 +00:00
|
|
|
};
|