Change "cache not filling" warning to include a hint to increase the
cache size.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32815 b3059339-0415-0410-9bf9-f77b7e298cf2
cosmetics: Move header_check helper function.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32803 b3059339-0415-0410-9bf9-f77b7e298cf2
Add return_check() helper function and use it to simplify some checks.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32805 b3059339-0415-0410-9bf9-f77b7e298cf2
Add return_statement_check() helper function and use it to simplify some checks.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32806 b3059339-0415-0410-9bf9-f77b7e298cf2
Add MakeMKV to list of webservers always supporting ranges even when
they claim otherwise.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32785 b3059339-0415-0410-9bf9-f77b7e298cf2
Fix hack to check for MakeMKV server being inverted.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32807 b3059339-0415-0410-9bf9-f77b7e298cf2
Fix printf format string warning
fixes this warning:
asfinfo.c: In function 'print_video_header':
asfinfo.c:158: warning: format '%.4s' expects type 'char *', but argument 3 has type 'long int *'
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32774 b3059339-0415-0410-9bf9-f77b7e298cf2
add missing fclose
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32775 b3059339-0415-0410-9bf9-f77b7e298cf2
add missing fclose()
found by cppcheck:
[avisubdump.c:199]: (error) Resource leak: f
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32776 b3059339-0415-0410-9bf9-f77b7e298cf2
This version is the last OS version available for certain classes of
PPC Macs, but it lacks AudioDeviceIOProcID, so it has to be emulated.
patch by Aaro Koskinen, aaro.koskinen iki fi
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32770 b3059339-0415-0410-9bf9-f77b7e298cf2
Add some notes about BE/LE colorspaces and fix a small typo.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32769 b3059339-0415-0410-9bf9-f77b7e298cf2
Use limits.h to get the maximum length instead of hardcoding it.
Original patch by Sang-Uok Kum.
Signed-off-by: Tobias Diedrich <ranma@google.com>
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32766 b3059339-0415-0410-9bf9-f77b7e298cf2
Define _XOPEN_SOURCE to 600 instead of 500 in mkstemp check, this is
consistent to the use in libavcodec, consistent with other
_XOPEN_SOURCE defines in tests and it fixes detection on Solaris that
refuses to compile system headers when combining std=c99 with
_XOPEN_SOURCE 500.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32759 b3059339-0415-0410-9bf9-f77b7e298cf2
Add define_statement_check function and use it to simplify some checks.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32758 b3059339-0415-0410-9bf9-f77b7e298cf2
Simplify a bunch of configure checks with the statement_check function.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32760 b3059339-0415-0410-9bf9-f77b7e298cf2
Use getfps helper function everywhere, simplifies code and avoids a
possible division by 0.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32755 b3059339-0415-0410-9bf9-f77b7e298cf2
Also move the profiles to the bottom of the example configuration file
as the original remarks in the file suggested.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32754 b3059339-0415-0410-9bf9-f77b7e298cf2
Simplify vector declarations and fast inttypes check with statement_check().
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32744 b3059339-0415-0410-9bf9-f77b7e298cf2
Drop unnecessary Xlib.h #include from Xss screensaver extensions check.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32745 b3059339-0415-0410-9bf9-f77b7e298cf2
configure: Simplify *BSD ioctl_meteor.h check.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32746 b3059339-0415-0410-9bf9-f77b7e298cf2
Add sys/time.h #include to videodev.h/videodev2.h checks.
Older kernels had broken videodev* headers that lacked the required #include.
patch by Michael Lampe, mlampe0 googlemail com
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32747 b3059339-0415-0410-9bf9-f77b7e298cf2
Revert previous wrong simplification of AltiVec vector declarations check.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32749 b3059339-0415-0410-9bf9-f77b7e298cf2
cosmetics: Drop some unnecessary end-of-line backslashes after &&/||.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32752 b3059339-0415-0410-9bf9-f77b7e298cf2
cosmetics: Get rid of some backslashes at the end of lines.
This is done by either moving the logical &&/|| operators to the
preceding line or by merging the two lines together.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32753 b3059339-0415-0410-9bf9-f77b7e298cf2
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32741 b3059339-0415-0410-9bf9-f77b7e298cf2
Remove AAC/FAAD2 installation instructions.
There is nothing special about building and installing FAAD2, so there is
no longer a need to keep maintaining instructions for it.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32742 b3059339-0415-0410-9bf9-f77b7e298cf2
We do not need them and they cause compilation errors with cygwin's
mingw-w64 cross-compilers (32 bit target) since its _mingw.h has a
"#define __ptr32", which makes these typedefs an invalid syntax.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32728 b3059339-0415-0410-9bf9-f77b7e298cf2
configure: Compilation fixes for current Cygwin
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32724 b3059339-0415-0410-9bf9-f77b7e298cf2
Do not #define _WIN32 on the command line for Cygwin.
Newer Cygwin versions no longer do this and hopefully we should be able
to survive without this hack as well. This change necessitates adapting
two #ifdefs in the MPlayer codebase. It is committed untested as I do
not have access to a Cygwin system.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32763 b3059339-0415-0410-9bf9-f77b7e298cf2
llvmpipe, while only slow instead of insanely slow, is a software
renderer. Recognize it as such and disable -vo gl_nosw on it.
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@32719 b3059339-0415-0410-9bf9-f77b7e298cf2
With extreme playback speed changes it was possible to trigger an
overflow in code calculating frame timing. This could break the VDPAU
frame scheduling mechanism and lead to the shown picture not changing
until reset by events such as seeking. Add an extra check to prevent
the overflow.
Add a definition to show the playback speed property on screen, and
change the separate commands (speed_mult etc) to use the property
mechanism. Now 'set_property_osd speed' can show the value on OSD, and
the separate commands respect the range limits of the option.
* sub:
sub/OSD: move some related files to sub/
subtitles: options: enable -ass by default
subtitles: change default libass rendering style
demux_mkv, chapters: change millisecond arithmetic to ns
cleanup: rename ass_* functions to mp_ass_*
subs: use correct font aspect ratio for libass + converted subs
cleanup: some random minor code simplification and cleanup
vf_vo: fix EOSD change detection bug
sd_ass: remove subreader use, support plaintext markup
subtitles: style support for common SubRip tags and MicroDVD
core: ordered chapters: fix bad subtitle parameter
subs/demux: don't try to enable sub track when creating it
subtitles/demux: store duration instead of endpts in demux packets
subtitles: add framework for subtitle decoders
options: add special -leak-report option
subtitles: remove code trying to handle text subs with libavcodec
cleanup: move MP_NOPTS_VALUE definition to mpcommon.h
subtitles: move global ass_track to struct osd_state
core: move most mpcommon.c contents to mplayer.c
core: move global "subdata" and "vo_sub_last" to mpctx
subtitles: remove sub_last_pts hack
options: move -noconfig to option struct, simplify
There are still some problems with -ass. For example some other
subtitle options won't work the same way or at all with it enabled,
and inserting the video filter for VOs that lack native rendering
support won't work with different colorspaces. However I think that
the benefit from styling support outweights those disadvantages, and
also that the "discoverability" of features is better this way; it
will be easier for people who encounter problems to find -noass than
for people who see no styling to find out that they could add it with
-ass.
Enable -ass by default. Also fix other outdated information in the
manpage entries for -ass and -fontconfig.
Change the default style used for rendering plaintext subtitles with
libass. The following attributes change:
* FontSize: Increase the size multiplier used with the default
-subfont-autoscale 3 from 1.4 to 1.7, thus making the font 21%
larger. Actually implementing the -subfont-autoscale modes other
than 1 would give a multiplier of 5/3=1.67 for 4:3 aspect ratio
video, sqrt(337)/9=2.04 for 16:9.
* PrimaryColour: Change color from yellow to white.
* Outline: Before this was 2 regardless of user font size choice. Make
it FontSize/16 instead. With default font size this is about half
the previous width.
* Kerning: Enable kerning for the track. There won't be styling
designed for VSFilter's lack of kerning, and hopefully people won't
use broken fonts with bad kerning tables for the default font
either.
demux_mkv kept various integer timestamps in millisecond units.
Matroska timestamp arithmetic is however specified in nanoseconds
(even though files typically use 1 ms precision), and using ms units
instead of that only made things more complex. Based on the demux_mkv
example the general demuxer-level chapter structure also used ms
units. Change the demux_mkv arithmetic and demuxer chapter structures
to use nanoseconds instead. This also fixes a seeking problem in
demux_mkv with files using a TimecodeScale other than the usual
1000000 (confusion between ms and TimecodeScale*ns units).
The various ass_* functions were created when libass was part of the
MPlayer tree and the distinction between MPlayer-specific and other
functions was less clear. Now that libass is a clearly separate
library, using the same ass_* namespace for player functions is ugly.
Rename the functions to use mp_ass_ prefix instead.
Rendering of ASS subtitles tries to be bug compatible with VSFilter
and stretches fonts when the video is anamorphic (some scripts try to
compensate for this VSFilter behavior, so trying to render them
"correctly" would give the wrong result). However this behavior is not
appropriate for subtitles we converted to ASS format ourselves for
libass rendering, as they certainly don't have VSFilter bug
workarounds. Change the code to use different behavior for "native"
ASS tracks and converted ones. It's questionable whether the
VSFilter-compatible behavior is appropriate for external .ass files
either, as there could be anamorphic and non-anamorphic versions of
the same video and the bug-compatible behavior can only be correct for
one alternative at most. However it's probably better to keep it as a
default at least, so that extracting a muxed subtitle track and using
that does not give behavior different from the original muxed one.
The aspect ratio setting is per ASS_Renderer, and changing it resets
libass caches. For that reason this commit adds separate renderer
instances to use for the "correct" and "VSFilter bug compatible"
cases.
Remove some definitions from generated config.h/config.mak that were
only used when building internal FFmpeg. Remove Solaris mlib test.
Also fix definition of --disable-ffmpeg in configure help output.
Also remove an old mention of "Subversion" from comments and fix an
error in non-git-repo version string generation (which hasn't been
used for anything).
This really should have been added to the manpage at the start of the
previous decade. There's still bad advice about -playlist in the XML
docs, but I'll remove that later when purging more obsolete/wrong
stuff from there.
demux_ty relied on demuxer->filepos being initially set to 0, but
demuxer.c has been changed to initialize it to -1. This caused a
"Invalid seek to negative position!" error message when running the
demux_ty file format check (so it occurred for any file which had not
been recognized as another type before that). Fix by making demux_ty
initialize filepos to 0.
When trying to determine the format of an input stream, demux_lavf
retries the probe with a larger buffer size up to some limit if the
match score is low, but when reaching the size limit it accepted the
best match (if any) regardless of its score. Change it to require a
score of at least AVPROBE_SCORE_MAX/4 to accept a match at all.
Uoti Urpala
diego
reimar
cboesch
siretart
cehoyos
compn
ranma
jrash
Clément Bœsch