January 2018 - MPlayer-dev-eng - lists.mplayerhq.hu

[PATCH] stream_ffmpeg: add -lavfstreamopts option.
by Reimar Döffinger 01 Nov '20

01 Nov '20

Allows specifying options to ffmpeg-based streams. --- DOCS/man/en/mplayer.1 | 4 ++++ cfg-common.h | 1 + stream/stream.h | 1 + stream/stream_ffmpeg.c | 19 ++++++++++++++++++- 4 files changed, 24 insertions(+), 1 deletion(-) diff --git a/DOCS/man/en/mplayer.1 b/DOCS/man/en/mplayer.1 index 2a143cb..d8e9b3b 100644 --- a/DOCS/man/en/mplayer.1 +++ b/DOCS/man/en/mplayer.1 @@ -1806,6 +1806,10 @@ Skip the proxy for IPv6 addresses. It will still be used for IPv4 connections. . .TP +.B \-lavfstreamopts <options> +Specify extra options for libavformat based streams. +. +.TP .B \-loadidx <index file> The file from which to read the video index data saved by \-saveidx. This index will be used for seeking, overriding any index data diff --git a/cfg-common.h b/cfg-common.h index 3a422cf..ca4e9c9 100644 --- a/cfg-common.h +++ b/cfg-common.h @@ -545,6 +545,7 @@ const m_option_t common_opts[] = { #ifdef CONFIG_FFMPEG {"lavdopts", lavc_decode_opts_conf, CONF_TYPE_SUBCONFIG, 0, 0, 0, NULL}, {"lavfdopts", lavfdopts_conf, CONF_TYPE_SUBCONFIG, CONF_GLOBAL, 0, 0, NULL}, + {"lavfstreamopts", &lavfstreamopts, CONF_TYPE_STRING, CONF_GLOBAL, 0, 0, NULL}, #endif #ifdef CONFIG_XVID4 {"xvidopts", xvid_dec_opts, CONF_TYPE_SUBCONFIG, 0, 0, 0, NULL}, diff --git a/stream/stream.h b/stream/stream.h index 615f579..388c353 100644 --- a/stream/stream.h +++ b/stream/stream.h @@ -392,6 +392,7 @@ extern char *dvd_device; extern const m_option_t dvbin_opts_conf[]; extern char *rtsp_destination; +extern char *lavfstreamopts; typedef struct { int id; // 0 - 31 mpeg; 128 - 159 ac3; 160 - 191 pcm diff --git a/stream/stream_ffmpeg.c b/stream/stream_ffmpeg.c index 57d45e4..1883af3 100644 --- a/stream/stream_ffmpeg.c +++ b/stream/stream_ffmpeg.c @@ -27,6 +27,8 @@ #include "av_helpers.h" #include "libmpdemux/demuxer.h" +char *lavfstreamopts; + static int fill_buffer(stream_t *s, char *buffer, int max_len) { int r = avio_read(s->priv, buffer, max_len); @@ -90,6 +92,7 @@ static int open_f(stream_t *stream, int mode, void *opts, int *file_format) { int flags = 0; const char *filename; + AVDictionary *avopts = NULL; AVIOContext *ctx = NULL; int res = STREAM_ERROR; int64_t size; @@ -125,9 +128,23 @@ static int open_f(stream_t *stream, int mode, void *opts, int *file_format) dummy = !strncmp(filename, "rtsp:", 5) || !strncmp(filename, "dummy:", 6); mp_msg(MSGT_OPEN, MSGL_V, "[ffmpeg] Opening %s\n", filename); - if (!dummy && avio_open(&ctx, filename, flags) < 0) + if (lavfstreamopts && av_dict_parse_string(&avopts, lavfstreamopts, "=", ",", 0) < 0) { + mp_msg(MSGT_HEADER,MSGL_ERR, "Your options /%s/ look like gibberish to me pal\n", lavfstreamopts); + goto out; + } + + if (!dummy && avio_open2(&ctx, filename, flags, NULL, &avopts) < 0) goto out; + if (!dummy && av_dict_count(avopts)) { + AVDictionaryEntry *e = NULL; + while ((e = av_dict_get(avopts, "", e, AV_DICT_IGNORE_SUFFIX))) { + mp_msg(MSGT_HEADER,MSGL_ERR,"Unknown option %s\n", e->key); + } + goto out; + } + av_dict_free(&avopts); + stream->priv = ctx; size = dummy ? 0 : avio_size(ctx); if (size >= 0) -- 2.7.0

1 1

[PATCH] aalib subtitle fix
by za3k＠za3k.com 15 Apr '19

15 Apr '19

For some time now, aalib rendering has had broken subtitles. -vo aa is designed to use human-readable text at the size of one character per character. In actuality, it's somehow switched to trying to render a standard font using dithering since its introduction. This patch fixes it. Explanation: vo_font doesn't seem to get used anywhere, because OSD rendering is actually done in a completely separate way from other video output, but I kept it in just in case. As a side note, someone may want to investigate unicode support for subtitles in -vo aa. *** /var/src/MPlayer-1.3.0/libvo/vo_aa.c 2016-02-12 12:23:39.000000000 -0800 --- vo_aa.c 2017-05-07 02:07:11.900548663 -0700 *************** static int osdx, osdy; *** 92,97 **** --- 92,98 ---- static int osd_text_length = 0; int aaconfigmode=1; font_desc_t* vo_font_save = NULL; + font_desc_t* sub_font_save = NULL; static struct SwsContext *sws=NULL; /* configuration */ *************** config(uint32_t width, uint32_t height, *** 249,256 **** /* now init our own 'font' */ if(!vo_font_save) vo_font_save = vo_font; ! if(vo_font == vo_font_save) { ! vo_font=malloc(sizeof(font_desc_t));//if(!desc) return NULL; memset(vo_font,0,sizeof(font_desc_t)); vo_font->pic_a[0]=malloc(sizeof(raw_file)); memset(vo_font->pic_a[0],0,sizeof(raw_file)); --- 250,258 ---- /* now init our own 'font' */ if(!vo_font_save) vo_font_save = vo_font; ! if(!sub_font_save) sub_font_save = vo_font; ! if(vo_font == vo_font_save || sub_font == sub_font_save) { ! vo_font=sub_font=malloc(sizeof(font_desc_t));//if(!desc) return NUL L; memset(vo_font,0,sizeof(font_desc_t)); vo_font->pic_a[0]=malloc(sizeof(raw_file)); memset(vo_font->pic_a[0],0,sizeof(raw_file));

3 4

[PATCH] Read scratched DVDs, warn for corrupt titles and report number of cells in title
by Micah Richert 05 May '18

05 May '18

This patch does 3 things: Continues to next cell when DVD sector read fails instead of the current behavior of failing/signaling EOF. This is useful for playing scratched DVDs. Reports the total number of cells in a DVD title. This is useful for finding the longest title when title duration information is inaccurate. Warns when packs are out of order in the current title. This is useful for finding titles that are corrupt/invalid. If desired, I can separate this patch into two, one for the sector read failure handling and the second to report cell count and pack order. Thank you, Micah Richert

2 2

mplayer 1.4.0 release
by Compn 15 Apr '18

15 Apr '18

i see development has dropped a bit... are we going to get a 1.4.0 release this year? -compn

3 3

[PATCH] MNG detection fails with libjpeg-turbo
by The Wanderer 16 Jan '18

16 Jan '18

With current SVN (and for as long as I can remember checking), the configure check for the availability of libmng fails. Today I finally got around to looking into why. The test performs a compile check against libmng.h, which includes jpeglib.h. With libjpeg-turbo (the source of jpeglib.h in current Debian), this test fails because FILE is undefined. This is typedef is normally expected to come from stdio.h. https://github.com/libjpeg-turbo/libjpeg-turbo/issues/17 indicates that this is intentional, for backwards compatibility with libjpeg, and that "it is actually a design requirement of the libjpeg API that you include stdio.h before including jpeglib.h.". This would seem to indicate that this has been broken for a very long time, and quite possibly no one noticed (which would imply that no one bothered trying to compile with MNG support). The last MNG-related commits seem to have been in 2012; support was originally added in 2011. The compilation check is performed by the return_statement_check function, which only permits passing one header-file argument. There is an existing statement_check_broken function, specifically for use with headers which don't include some of their own dependencies; however, it hardcodes a 'return 0', whereas return_statement_check permits passing in an expression to be used in the return statement. Rather than try to add an ' 0' to all callers of statement_check_broken, I've duplicated it into return_statement_check_broken, which adds the return-statement argument from return_statement_check. With that, detection of libmng succeeds. With an include of stdio.h added to vo_mng.c, compilation also succeeds. The attached patch makes both changes. -- The Wanderer The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. -- George Bernard Shaw

2 11

compilation issue with recent ffmpeg
by sylvain.bertrand＠gmail.com 13 Jan '18

13 Jan '18

Hi, While compiling today mplayer snapshot with a today git ffmpeg, I get: -------------------------------------------------------------------------------- libavformat/os_support.c:49:12: error: static declaration of 'inet_aton' follows non-static declaration static int inet_aton(const char *str, struct in_addr *add) ^~~~~~~~~ In file included from libavformat/network.h:69:0, from libavformat/os_support.c:44: /include/arpa/inet.h:73:12: note: previous declaration of 'inet_aton' was here extern int inet_aton (const char *__cp, struct in_addr *__inp) __THROW; ^~~~~~~~~ libavformat/os_support.c:49:12: warning: 'inet_aton' defined but not used [-Wunused-function] static int inet_aton(const char *str, struct in_addr *add) ^~~~~~~~~ make[1]: *** [ffbuild/common.mak:60: libavformat/os_support.o] Error 1 -------------------------------------------------------------------------------- I got this error compiling a standalone ffmpeg, but it was fixed very recently by ffmpeg devs. regards, -- Sylvain

4 6

[PATCH][RFC] configure: Invoke FFmpeg's configure
by Alexander Strasser 09 Jan '18

09 Jan '18

This is not a finished patch. I want to gather opinions on going this route to pave the way for further development. One downside of this patch is that configure will be slower. On my tests running configure took twice as long. Though I think, if we take into account the frustration it saves for users building MPlayer and the time it saves for developers, it should be OK. Here is a probably incomplete list of TODOs: * pass on build tools related options (like selected compiler, crosscompiling etc.) * better merging of the additionally needed libs into extra_ldflags * use libs autodetected by MPlayer's configure to enable features in FFmpeg * remove some stuff that is no longer needed to be done in MPlayer's configure (as long as it isn't too time consuming to do) * show summarized results of FFmpeg's configure in MPlayer's configure output * test more thoroughly * test on more platforms HOWEVER after completion of those points, this will still only be a short to mid term solution. It should give us time to discuss and find a long term direction and eventually implement that step by step. That new discussion should be started in a dedicated mail thread. I can do it after we have decided about this approach. Here follows the promised data on how often we change configure: * Starting from 2017-01-01 to 2017-01-07 there were 109 commits in total * 78/109 commits (~71.6%) were *not* touching configure * 31/109 commits (~28.4%) were touching configure * 26/31 commits (~83.9%) related to FFmpeg's build system somehow * 5/31 commits (~16,1%) not related to FFmpeg's build system somehow Some of that 26 commits related to FFmpeg's build system are not a direct response to a change in FFmpeg but probably wouldn't have been needed if we would run FFmpeg's configure as is proposed in this patch. I didn't closely look at all commit diffs, and there might be some errors in how I processed or catagorized the data. I will post the data in subsequent reply, so others can have a look and/or investigate on their own. I am eagerly waiting for your comments. Let's move forward tackling the recurring problems with the build system. Alexander --- configure | 85 +++++++++------------------------------------------------------ 1 file changed, 11 insertions(+), 74 deletions(-) diff --git a/configure b/configure index f62865248..4832c1a4b 100755 --- a/configure +++ b/configure @@ -8235,6 +8235,17 @@ else fi echores "$_dvdnav" +if test "$ffmpeg_a" != "no" ; then + echocheck "Configuring internal FFmpeg" + res_internal_ffmpeg_configured="no" + if (cd ffmpeg && ./configure --enable-gpl >/dev/null 2>&1) then + res_internal_ffmpeg_configured="yes" + fi + ffmpeg_extra_libs="$(awk -F= '/^EXTRALIBS/ { if ($2 != "") s[$2] } END { for (extralibs in s) { all_libs = all_libs " " extralibs } ; print all_libs }' ffmpeg/ffbuild/config.mak)" + extra_ldflags="$extra_ldflags $ffmpeg_extra_libs" + echores "$res_internal_ffmpeg_configured" +fi + extra_ldflags="$extra_ldflags $libm" # DO NOT ADD ANY TESTS THAT USE LINKER FLAGS HERE (like cc_check). @@ -9498,80 +9509,6 @@ cmp -s "$TMPH" config.h || mv -f "$TMPH" config.h ############################################################################ -# Create FFmpeg config files only for internal ffmpeg -if test "$ffmpeg_a" = yes; then - -# Create avconfig.h for FFmpeg. -cat > "$TMPH" << EOF -/* Generated by mpconfigure */ -#ifndef AVUTIL_AVCONFIG_H -#define AVUTIL_AVCONFIG_H -$def_av_bigendian -$def_av_fast_unaligned -#define AV_HAVE_INCOMPATIBLE_LIBAV_ABI 0 -#endif /* AVUTIL_AVCONFIG_H */ -EOF - -# Do not overwrite an unchanged avconfig.h to avoid superfluous rebuilds. -cmp -s "$TMPH" ffmpeg/libavutil/avconfig.h || mv -f "$TMPH" ffmpeg/libavutil/avconfig.h - - -if x86; then -# Create a skeleton config.asm with just the ARCH_ and *CODER definitions for FFmpeg. -> "$TMPS" -echo "%define CONFIG_GPL 1" >> "$TMPS" -echo "%define HAVE_ALIGNED_STACK 1" >> "$TMPS" -echo "$(ff_config_enable "$arch_all" "$arch" "%" "ARCH")" >> "$TMPS" -echo "$(ff_config_enable "$subarch_all" "$subarch" "%" "ARCH")" >> "$TMPS" -echo "$(ff_config_enable "$cpuexts_all" "$cpuexts_external" "%" "HAVE" "_EXTERNAL")" >> "$TMPS" -echo "$(ff_config_enable "$cpuexts_all" "$cpuexts" "%" "HAVE" "_INLINE")" >> "$TMPS" -echo "$(ff_config_enable "$yasm_features_all" "$yasm_features" "%" "HAVE")" >> "$TMPS" -echo "$(ff_config_enable "$libavdecoders_all" "$libavdecoders" "%" "CONFIG")" >> "$TMPS" -echo "$(ff_config_enable "$libavencoders_all" "$libavencoders" "%" "CONFIG")" >> "$TMPS" - -cmp -s "$TMPS" ffmpeg/config.asm || mv -f "$TMPS" ffmpeg/config.asm -fi - - -# Create a config.mak for FFmpeg that includes MPlayer's config.mak. -cat > ffmpeg/ffbuild/config.mak << EOF -ifndef FFMPEG_CONFIG_MAK -FFMPEG_CONFIG_MAK = 1 -include ../config.mak -endif # FFMPEG_CONFIG_MAK -EOF - -# TODO: temporary support for older ffmpeg -cp ffmpeg/ffbuild/config.mak ffmpeg/config.mak - -cat > ffmpeg/config.h << EOF -#include "../config.h" -EOF -touch ffmpeg/.config - -# generate the lists of enabled components for ffmpeg -print_enabled_components(){ - file=$1 - struct_name=$2 - name=$3 - shift 3 - list=$(echo $* | tolower) - echo "static const $struct_name *$name[] = {" > $TMPH - for c in $list; do - printf " &ff_%s,\n" $c >> $TMPH - done - echo " NULL };" >> $TMPH - cmp -s $TMPH ffmpeg/$file && return - cp $TMPH ffmpeg/$file -} - -print_enabled_components libavcodec/bsf_list.c AVBitStreamFilter bitstream_filters $libavbsfs -print_enabled_components libavformat/protocol_list.c URLProtocol url_protocols $libavprotocols - -fi - -############################################################################# - cat << EOF Config files successfully generated by ./configure $configuration ! --

3 5

Adapt config.h's generation with some ffmpeg's commits
by Etienne Buira 08 Jan '18

08 Jan '18

Adapt config.h's generation with some ffmpeg's commits ffmpeg's related commits: e0d56f097f42bcdbe6c3b2f57df62a4da63f2094 - CONFIG_LINUX_PERF 5de85c1029881970dc69e529dc48e277f3b5336c - HAVE_LOCAL_ALIGNED c708b5403346255ea5adc776645616cc7c61f078 - HAVE_MACH_MACH_TIME_H & HAVE_MACH_ABSOLUTE_TIME 8f86e6623811f7713d5e72c13797e20fffb3df62 - HAVE_SIMD_ALIGN_64 b7a3d1f249c9da9f21d10ea9137e304961da02ba - HAVE_AVX512 1f4c62e84e586c52282a58ff878022a79feac4d1 - HAVE_DIRENT_H 465becbc4f1fd6dc7973fb69a2d46e8c8580e719 - HAVE_ACCESS d65b9114f35c1afe2a7061f0a1ec957d33ba02b5 - HAVE_LSTAT 6e073dd127f09f203bd21d40a1eb423afd183e7a - HAVE_STRUCT_STAT_ST_MTIM_TV_NSEC 534a2231f69638c3fdd998853ba4158ed79322a8 - HAVE_FCNTL Regards

4 8

[PATCH] Remove needless memsets after calloc
by Michael McConville 05 Jan '18

05 Jan '18

Hi guys. Unless, I’m missing something obvious, this is pretty self-explanatory. Calloc zeroes memory, so there’s no need to re-zero it, and modern compilers often miss this optimization. Thanks for your time, Michael McConville University of Utah Index: libmpdemux/muxer_avi.c =================================================================== --- libmpdemux/muxer_avi.c (revision 38012) +++ libmpdemux/muxer_avi.c (working copy) @@ -120,7 +120,6 @@ si->idx=calloc(si->idxsize, sizeof(struct avi_odmlidx_entry)); si->riffofssize=16; si->riffofs=calloc((si->riffofssize+1), sizeof(off_t)); - memset(si->riffofs, 0, sizeof(off_t)*si->riffofssize); switch(type){ case MUXER_TYPE_VIDEO: @@ -619,7 +618,6 @@ si->superidxsize = si->superidxpos; si->superidx = calloc(si->superidxsize, sizeof(*si->superidx)); - memset(si->superidx, 0, sizeof(*si->superidx) * si->superidxsize); idxpos = 0; for (j=0; j<si->superidxpos; j++) {

2 1