mkvtoolnix/mkvmerge.cpp
2003-06-20 21:49:24 +00:00

1602 lines
45 KiB
C++

/*
mkvmerge -- utility for splicing together matroska files
from component media subtypes
mkvmerge.cpp
Written by Moritz Bunkus <moritz@bunkus.org>
Distributed under the GPL
see the file COPYING for details
or visit http://www.gnu.org/copyleft/gpl.html
*/
/*!
\file
\version $Id$
\brief command line parameter parsing, looping, output handling
\author Moritz Bunkus <moritz@bunkus.org>
*/
#include "os.h"
#include <errno.h>
#include <ctype.h>
#if defined(SYS_UNIX) || defined(COMP_CYGWIN)
#include <signal.h>
#endif
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <time.h>
#include <unistd.h>
#include <iostream>
#include <string>
#include <vector>
#if __GNUC__ == 2
#include <typeinfo>
#endif
#include "EbmlHead.h"
#include "EbmlSubHead.h"
#include "EbmlVersion.h"
#include "EbmlVoid.h"
#include "FileKax.h"
#include "KaxBlock.h"
#include "KaxCluster.h"
#include "KaxClusterData.h"
#include "KaxCues.h"
#include "KaxInfo.h"
#include "KaxInfoData.h"
#include "KaxSeekHead.h"
#include "KaxSegment.h"
#include "KaxTags.h"
#include "KaxTracks.h"
#include "KaxTrackEntryData.h"
#include "KaxTrackAudio.h"
#include "KaxTrackVideo.h"
#include "KaxVersion.h"
#include "mkvmerge.h"
#include "mm_io.h"
#include "cluster_helper.h"
#include "common.h"
#include "iso639.h"
#include "r_aac.h"
#include "r_ac3.h"
#include "r_dts.h"
#include "r_avi.h"
#include "r_mp3.h"
#include "r_wav.h"
#ifdef HAVE_OGGVORBIS
#include "r_ogm.h"
#endif
#include "r_srt.h"
#include "r_ssa.h"
#include "r_matroska.h"
#include "r_mp4.h"
using namespace LIBMATROSKA_NAMESPACE;
using namespace std;
typedef struct {
char *ext;
int type;
char *desc;
} file_type_t;
typedef struct filelist_tag {
char *name;
mm_io_c *fp;
int type, status;
packet_t *pack;
generic_reader_c *reader;
track_info_t *ti;
} filelist_t;
typedef struct {
int status;
packet_t *pack;
generic_packetizer_c *packetizer;
} packetizer_t;
vector<packetizer_t *> packetizers;
vector<filelist_t *> files;
// Variables set by the command line parser.
char *outfile = NULL;
int64_t file_sizes = 0;
int max_blocks_per_cluster = 65535;
int max_ms_per_cluster = 1000;
bool write_cues = true, cue_writing_requested = false;
bool video_track_present = false;
bool write_meta_seek = true;
int64_t meta_seek_size = 0;
bool no_lacing = false, no_linking = false;
int64_t split_after = -1;
bool split_by_time = false;
int split_max_num_files = 65535;
float video_fps = -1.0;
int default_tracks[3];
cluster_helper_c *cluster_helper = NULL;
KaxSegment *kax_segment;
KaxInfo *kax_infos;
KaxTracks *kax_tracks;
KaxTrackEntry *kax_last_entry;
KaxCues *kax_cues;
EbmlVoid *kax_seekhead_void = NULL;
KaxDuration *kax_duration;
KaxSeekHead *kax_seekhead = NULL;
// Actual pass. 0 for normal operation. 1 for first pass of splitting
// (only find split points, output goes to /dev/null). 2 for the second
// pass of the splitting run.
int pass = 0;
int file_num = 1;
// Specs say that track numbers should start at 1.
int track_number = 1;
mm_io_c *out;
bitvalue_c seguid_prev(128), seguid_current(128), seguid_next(128);
bitvalue_c *seguid_link_previous = NULL, *seguid_link_next = NULL;
file_type_t file_types[] =
{{"---", TYPEUNKNOWN, "<unknown>"},
{"demultiplexers:", -1, ""},
{"mkv", TYPEMATROSKA, "general Matroska files"},
#ifdef HAVE_OGGVORBIS
{"ogg", TYPEOGM, "general OGG media stream, audio/video embedded in OGG"},
#endif // HAVE_OGGVORBIS
{"avi", TYPEAVI, "AVI (Audio/Video Interleaved)"},
{"wav", TYPEWAV, "WAVE (uncompressed PCM)"},
{"srt", TYPESRT, "SRT text subtitles"},
{"ssa", TYPESSA, "SSA/ASS text subtitles"},
// {" ", TYPEMICRODVD, "MicroDVD text subtitles"},
// {"idx", TYPEVOBSUB, "VobSub subtitles"},
{"mp3", TYPEMP3, "MPEG1 layer III audio (CBR and VBR/ABR)"},
{"ac3", TYPEAC3, "A/52 (aka AC3)"},
{"dts", TYPEDTS, "DTS (Digital Theater System)"},
{"aac", TYPEAAC, "AAC (Advanced Audio Coding)"},
{"output modules:", -1, ""},
#ifdef HAVE_OGGVORBIS
{" ", -1, "Vorbis audio"},
#endif // HAVE_OGGVORBIS
{" ", -1, "Video (not MPEG1/2)"},
{" ", -1, "uncompressed PCM audio"},
{" ", -1, "simple text subtitles"},
// {" ", -1, "VobSub subtitles"},
{" ", -1, "MP3 audio"},
{" ", -1, "AC3 audio"},
{" ", -1, "DTS audio"},
{" ", -1, "AAC audio"},
{NULL, -1, NULL}};
static void usage(void) {
fprintf(stdout,
"mkvmerge -o out [global options] [options] <file1> [@optionsfile ...]"
"\n\n Global options:\n"
" -v, --verbose verbose status\n"
" -q, --quiet suppress status output\n"
" -o, --output out Write to the file 'out'.\n"
" --cluster-length <n[ms]> Put at most n data blocks into each cluster.\n"
" If the number is postfixed with 'ms' then\n"
" put at most n milliseconds of data into each\n"
" cluster.\n"
" --no-cues Do not write the cue data (the index).\n"
" --no-meta-seek Do not write any meta seek information.\n"
" --meta-seek-size <d> Reserve d bytes for the meta seek entries.\n"
" --no-lacing Do not use lacing.\n"
" --split <d[K,M,G]|HH:MM:SS|ns>\n"
" Create a new file after d bytes (KB, MB, GB)\n"
" or after a specific time.\n"
" --split-max-files <n> Create at most n files.\n"
" --dont-link Don't link splitted files.\n"
" --link-to-previous <UID> Link the first file to the given UID.\n"
" --link-to-next <UID> Link the last file to the given UID.\n"
"\n Options for each input file:\n"
" -a, --atracks <n,m,...> Copy audio tracks n,m etc. Default: copy all\n"
" audio tracks.\n"
" -d, --vtracks <n,m,...> Copy video tracks n,m etc. Default: copy all\n"
" video tracks.\n"
" -s, --stracks <n,m,...> Copy subtitle tracks n,m etc. Default: copy\n"
" all subtitle tracks.\n"
" -A, --noaudio Don't copy any audio track from this file.\n"
" -D, --novideo Don't copy any video track from this file.\n"
" -S, --nosubs Don't copy any text track from this file.\n"
" -y, --sync <TID:d[,o[/p]]>\n"
" Synchronize, delay the audio track with the\n"
" id TID by d ms. \n"
" d > 0: Pad with silent samples.\n"
" d < 0: Remove samples from the beginning.\n"
" o/p: Adjust the timecodes by o/p to fix\n"
" linear drifts. p defaults to 1000 if\n"
" omitted. Both o and p can be floating point\n"
" numbers.\n"
" --default-track <TID> Sets the 'default' flag for this track.\n"
" --cues <TID:none|iframes|all>\n"
" Create cue (index) entries for this track:\n"
" None at all, only for I frames, for all.\n"
" --language <TID:lang> Sets the language for the track (ISO639-2\n"
" code, see --list-languages).\n"
"\n Options that only apply to video tracks:\n"
" -f, --fourcc <FOURCC> Forces the FourCC to the specified value.\n"
" Works only for video tracks.\n"
" --aspect-ratio <f|a/b> Sets the aspect ratio.\n"
"\n Options that only apply to text subtitle tracks:\n"
" --sub-charset <charset> Sets the charset the text subtitles are\n"
" written in for the conversion to UTF-8.\n"
"\n\n Other options:\n"
" -i, --identify <file> Print information about the source file.\n"
" -l, --list-types Lists supported input file types.\n"
" --list-languages Lists all ISO639 languages and their\n"
" ISO639-2 codes.\n"
" -h, --help Show this help.\n"
" -V, --version Show version information.\n"
" @optionsfile Reads additional command line options from\n"
" the specified file (see man page).\n"
"\n\nPlease read the man page/the HTML documentation to mkvmerge. It\n"
"explains several details in great length which are not obvious from\n"
"this listing.\n"
);
}
#if defined(SYS_UNIX) || defined(COMP_CYGWIN)
static void sighandler(int signum) {
#ifdef DEBUG
if (signum == SIGUSR1)
debug_c::dump_info();
#endif // DEBUG
}
#endif
static int get_type(char *filename) {
mm_io_c *mm_io;
off_t size;
int type;
try {
mm_io = new mm_io_c(filename, MODE_READ);
mm_io->setFilePointer(0, seek_end);
size = mm_io->getFilePointer();
mm_io->setFilePointer(0, seek_current);
} catch (exception &ex) {
fprintf(stderr, "Error: could not open source file or seek properly in it "
"(%s).\n", filename);
exit(1);
}
if (avi_reader_c::probe_file(mm_io, size))
type = TYPEAVI;
else if (mkv_reader_c::probe_file(mm_io, size))
type = TYPEMATROSKA;
else if (wav_reader_c::probe_file(mm_io, size))
type = TYPEWAV;
else if (mp4_reader_c::probe_file(mm_io, size)) {
fprintf(stderr, "Error: MP4/Quicktime files are not supported (%s).\n",
filename);
exit(1);
}
#ifdef HAVE_OGGVORBIS
else if (ogm_reader_c::probe_file(mm_io, size))
type = TYPEOGM;
#endif // HAVE_OGGVORBIS
else if (srt_reader_c::probe_file(mm_io, size))
type = TYPESRT;
else if (mp3_reader_c::probe_file(mm_io, size))
type = TYPEMP3;
else if (ac3_reader_c::probe_file(mm_io, size))
type = TYPEAC3;
else if (dts_reader_c::probe_file(mm_io, size))
type = TYPEDTS;
else if (aac_reader_c::probe_file(mm_io, size))
type = TYPEAAC;
else if (ssa_reader_c::probe_file(mm_io, size))
type = TYPESSA;
else
type = TYPEUNKNOWN;
delete mm_io;
file_sizes += size;
return type;
}
static int display_counter = 1;
static void display_progress(int force) {
filelist_t *winner;
int i;
if (((display_counter % 500) == 0) || force) {
display_counter = 0;
winner = files[0];
for (i = 1; i < files.size(); i++)
if (files[i]->reader->display_priority() >
winner->reader->display_priority())
winner = files[i];
winner->reader->display_progress();
}
display_counter++;
}
void add_packetizer(generic_packetizer_c *packetizer) {
packetizer_t *pack = (packetizer_t *)safemalloc(sizeof(packetizer_t));
pack->packetizer = packetizer;
pack->status = EMOREDATA;
pack->pack = NULL;
packetizers.push_back(pack);
}
static void parse_tracks(char *s, vector<int64_t> *tracks) {
char *comma;
int64_t tid;
tracks->clear();
comma = strchr(s, ',');
while ((comma != NULL) && (*s != 0)) {
*comma = 0;
if (!parse_int(s, tid)) {
fprintf(stderr, "Error: Invalid track ID '%s'.\n", s);
exit(1);
}
tracks->push_back(tid);
s = &comma[1];
comma = strchr(s, ',');
}
if (*s != 0) {
if (!parse_int(s, tid)) {
fprintf(stderr, "Error: Invalid track ID '%s'.\n", s);
exit(1);
}
tracks->push_back(tid);
}
}
static void parse_sync(char *s, audio_sync_t &async) {
char *linear, *div, *colon;
double d1, d2;
// Extract the track number.
if ((colon = strchr(s, ':')) == NULL) {
fprintf(stderr, "Error: Invalid sync option. No track ID specified (%s)."
"\n", s);
exit(1);
}
*colon = 0;
if (!parse_int(s, async.id)) {
fprintf(stderr, "Error: Invalid track ID specified.\n");
exit(1);
}
s = &colon[1];
if (*s == 0) {
fprintf(stderr, "Error: Invalid sync option specified.\n");
exit(1);
}
// Now parse the actual sync values.
if ((linear = strchr(s, ',')) != NULL) {
*linear = 0;
linear++;
div = strchr(linear, '/');
if (div == NULL)
async.linear = strtod(linear, NULL) / 1000.0;
else {
*div = 0;
div++;
d1 = strtod(linear, NULL);
d2 = strtod(div, NULL);
if (d2 == 0.0) {
fprintf(stderr, "Error: linear sync: division by zero?\n");
exit(1);
}
async.linear = d1 / d2;
}
if (async.linear <= 0.0) {
fprintf(stderr, "Error: linear sync value may not be <= 0.\n");
exit(1);
}
} else
async.linear = 1.0;
async.displacement = atoi(s);
}
static float parse_aspect_ratio(char *s) {
char *div;
float w, h;
div = strchr(s, '/');
if (div == NULL)
return strtod(s, NULL);
*div = 0;
div++;
if (*s == 0) {
fprintf(stderr, "Error: aspect ratio: missing dividend.\n");
exit(1);
}
if (*div == 0) {
fprintf(stderr, "Error: aspect ratio: missing divisor.\n");
exit(1);
}
w = strtod(s, NULL);
h = strtod(div, NULL);
if (h == 0.0) {
fprintf(stderr, "Error: aspect ratio: divisor is 0.\n");
exit(1);
}
return w/h;
}
static void parse_split(const char *arg) {
int64_t modifier;
char *s, *p;
if ((arg == NULL) || (arg[0] == 0) || (arg[1] == 0)) {
fprintf(stderr, "Error: Invalid format for --split.\n");
exit(1);
}
s = safestrdup(arg);
// HH:MM:SS
if ((strlen(s) == 8) && (s[2] == ':') && (s[5] == ':') &&
isdigit(s[0]) && isdigit(s[1]) && isdigit(s[3]) &&
isdigit(s[4]) && isdigit(s[6]) && isdigit(s[7])) {
int secs, mins, hours;
s[2] = 0;
s[5] = 0;
parse_int(s, hours);
parse_int(&s[3], mins);
parse_int(&s[6], secs);
split_after = secs + mins * 60 + hours * 3600;
if ((hours < 0) || (mins < 0) || (mins > 59) ||
(secs < 0) || (secs > 59) || (split_after < 10)) {
fprintf(stderr, "Error: Invalid time for --split.\n");
exit(1);
}
split_after *= 1000;
split_by_time = true;
safefree(s);
return;
}
// Number of seconds: e.g. 1000s or 4254S
if ((s[strlen(s) - 1] == 's') || (s[strlen(s) - 1] == 'S')) {
s[strlen(s) - 1] = 0;
if (!parse_int(s, split_after)) {
fprintf(stderr, "Error: Invalid time for --split.\n");
exit(1);
}
split_after *= 1000;
split_by_time = true;
safefree(s);
return;
}
// Size in bytes/KB/MB/GB
p = &s[strlen(s) - 1];
if ((*p == 'k') || (*p == 'K'))
modifier = 1024;
else if ((*p == 'm') || (*p == 'M'))
modifier = 1024 * 1024;
else if ((*p == 'g') || (*p == 'G'))
modifier = 1024 * 1024 * 1024;
else
modifier = 1;
if (modifier != 1)
*p = 0;
if (!parse_int(s, split_after)) {
fprintf(stderr, "Error: invalid split size.\n");
exit(1);
}
split_after *= modifier;
if (split_after <= (1024 * 1024)) {
fprintf(stderr, "Error: invalid split size.\n");
exit(1);
}
safefree(s);
split_by_time = false;
}
static void parse_cues(char *s, cue_creation_t &cues) {
char *colon;
// Extract the track number.
if ((colon = strchr(s, ':')) == NULL) {
fprintf(stderr, "Error: Invalid cues option. No track ID specified (%s)."
"\n", s);
exit(1);
}
*colon = 0;
if (!parse_int(s, cues.id)) {
fprintf(stderr, "Error: Invalid track ID specified.\n");
exit(1);
}
s = &colon[1];
if (*s == 0) {
fprintf(stderr, "Error: Invalid cues option specified.\n");
exit(1);
}
if (!strcmp(s, "all"))
cues.cues = CUES_ALL;
else if (!strcmp(s, "iframes"))
cues.cues = CUES_IFRAMES;
else if (!strcmp(s, "none"))
cues.cues = CUES_NONE;
else {
fprintf(stderr, "Error: '%s' is an unsupported argument for --cues.\n", s);
exit(1);
}
}
static void parse_language(char *s, language_t &lang) {
char *colon;
// Extract the track number.
if ((colon = strchr(s, ':')) == NULL) {
fprintf(stderr, "Error: Invalid language option. No track ID specified "
"(%s).\n", s);
exit(1);
}
*colon = 0;
if (!parse_int(s, lang.id)) {
fprintf(stderr, "Error: Invalid track ID specified.\n");
exit(1);
}
s = &colon[1];
if (*s == 0) {
fprintf(stderr, "Error: Invalid language specified.\n");
exit(1);
}
if (!is_valid_iso639_2_code(s)) {
fprintf(stderr, "Error: '%s' is not a valid ISO639-2 code. See "
"'mkvmerge --list-languages'.\n", s);
exit(1);
}
lang.language = s;
}
static void render_headers(mm_io_c *out, bool last_file, bool first_file) {
EbmlHead head;
int i;
try {
EDocType &doc_type = GetChild<EDocType>(head);
*static_cast<EbmlString *>(&doc_type) = "matroska";
EDocTypeVersion &doc_type_ver = GetChild<EDocTypeVersion>(head);
*(static_cast<EbmlUInteger *>(&doc_type_ver)) = 1;
EDocTypeReadVersion &doc_type_read_ver =
GetChild<EDocTypeReadVersion>(head);
*(static_cast<EbmlUInteger *>(&doc_type_read_ver)) = 1;
head.Render(*out);
kax_infos = &GetChild<KaxInfo>(*kax_segment);
KaxTimecodeScale &time_scale = GetChild<KaxTimecodeScale>(*kax_infos);
*(static_cast<EbmlUInteger *>(&time_scale)) = TIMECODE_SCALE;
kax_duration = &GetChild<KaxDuration>(*kax_infos);
*(static_cast<EbmlFloat *>(kax_duration)) = 0.0;
string version = string("libebml v") + EbmlCodeVersion +
string(" + libmatroska v") + KaxCodeVersion;
*((EbmlUnicodeString *)&GetChild<KaxMuxingApp>(*kax_infos)) =
cstr_to_UTFstring(version.c_str());
*((EbmlUnicodeString *)&GetChild<KaxWritingApp>(*kax_infos)) =
cstr_to_UTFstring(VERSIONINFO);
GetChild<KaxDateUTC>(*kax_infos).SetEpochDate(time(NULL));
// Generate the segment UIDs.
if (first_file) {
seguid_current.generate_random();
if (!last_file)
seguid_next.generate_random();
} else {
seguid_prev = seguid_current;
seguid_current = seguid_next;
if (!last_file)
seguid_next.generate_random();
}
// Set the segment UIDs.
KaxSegmentUID &kax_seguid = GetChild<KaxSegmentUID>(*kax_infos);
kax_seguid.CopyBuffer(seguid_current.data(), 128 / 8);
if (!no_linking) {
if (!first_file) {
KaxPrevUID &kax_prevuid = GetChild<KaxPrevUID>(*kax_infos);
kax_prevuid.CopyBuffer(seguid_prev.data(), 128 / 8);
} else if (seguid_link_previous != NULL) {
KaxPrevUID &kax_prevuid = GetChild<KaxPrevUID>(*kax_infos);
kax_prevuid.CopyBuffer(seguid_link_previous->data(), 128 / 8);
}
if (!last_file) {
KaxNextUID &kax_nextuid = GetChild<KaxNextUID>(*kax_infos);
kax_nextuid.CopyBuffer(seguid_next.data(), 128 / 8);
} else if (seguid_link_next != NULL) {
KaxNextUID &kax_nextuid = GetChild<KaxNextUID>(*kax_infos);
kax_nextuid.CopyBuffer(seguid_link_next->data(), 128 / 8);
}
}
kax_segment->WriteHead(*out, 5);
// Reserve some space for the meta seek stuff.
if (write_cues && write_meta_seek) {
kax_seekhead = new KaxSeekHead();
kax_seekhead_void = new EbmlVoid();
if (meta_seek_size == 0)
if (video_track_present)
meta_seek_size =
(int64_t)((float)file_sizes * 1.5 / 10240.0);
else
meta_seek_size =
(int64_t)((float)file_sizes * 3 / 4096.0);
kax_seekhead_void->SetSize(meta_seek_size);
kax_seekhead_void->Render(*out);
}
kax_infos->Render(*out);
kax_tracks = &GetChild<KaxTracks>(*kax_segment);
kax_last_entry = NULL;
for (i = 0; i < files.size(); i++)
files[i]->reader->set_headers();
kax_tracks->Render(*out);
} catch (exception &ex) {
fprintf(stderr, "Error: Could not render the track headers.\n");
exit(1);
}
}
static void create_readers() {
filelist_t *file;
int i;
for (i = 0; i < files.size(); i++) {
file = files[i];
try {
switch (file->type) {
case TYPEMATROSKA:
file->reader = new mkv_reader_c(file->ti);
break;
#ifdef HAVE_OGGVORBIS
case TYPEOGM:
file->reader = new ogm_reader_c(file->ti);
break;
#endif // HAVE_OGGVORBIS
case TYPEAVI:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs)
fprintf(stderr, "Warning: -s/-S are ignored for AVI files.\n");
file->reader = new avi_reader_c(file->ti);
break;
case TYPEWAV:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"WAVE files.\n");
file->reader = new wav_reader_c(file->ti);
break;
case TYPESRT:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"SRT files.\n");
file->reader = new srt_reader_c(file->ti);
break;
case TYPEMP3:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"MP3 files.\n");
file->reader = new mp3_reader_c(file->ti);
break;
case TYPEAC3:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"AC3 files.\n");
file->reader = new ac3_reader_c(file->ti);
break;
case TYPEDTS:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"DTS files.\n");
file->reader = new dts_reader_c(file->ti);
break;
case TYPEAAC:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"AAC files.\n");
file->reader = new aac_reader_c(file->ti);
break;
case TYPESSA:
if ((file->ti->stracks->size() != 0) || file->ti->no_subs ||
(file->ti->atracks->size() != 0) || file->ti->no_audio ||
(file->ti->vtracks->size() != 0) || file->ti->no_video)
fprintf(stderr, "Warning: -a/-A/-d/-D/-s/-S are ignored for "
"SSA/ASS files.\n");
file->reader = new ssa_reader_c(file->ti);
break;
default:
fprintf(stderr, "Error: EVIL internal bug! (unknown file type)\n");
exit(1);
break;
}
} catch (error_c error) {
fprintf(stderr, "Error: Demultiplexer failed to initialize:\n%s\n",
error.get_error());
exit(1);
}
}
}
static void identify(const char *filename) {
track_info_t ti;
filelist_t *file;
memset(&ti, 0, sizeof(track_info_t));
ti.audio_syncs = new vector<audio_sync_t>;
ti.cue_creations = new vector<cue_creation_t>;
ti.default_track_flags = new vector<int64_t>;
ti.languages = new vector<language_t>;
ti.aspect_ratio = 1.0;
ti.atracks = new vector<int64_t>;
ti.vtracks = new vector<int64_t>;
ti.stracks = new vector<int64_t>;
file = (filelist_t *)safemalloc(sizeof(filelist_t));
file->name = safestrdup(filename);
file->type = get_type(file->name);
ti.fname = file->name;
if (file->type == TYPEUNKNOWN) {
fprintf(stderr, "Error: File %s has unknown type. Please have a look "
"at the supported file types ('mkvmerge --list-types') and "
"contact me at moritz@bunkus.org if your file type is "
"supported but not recognized properly.\n", file->name);
exit(1);
}
file->fp = NULL;
file->status = EMOREDATA;
file->pack = NULL;
file->ti = duplicate_track_info(&ti);
files.push_back(file);
verbose = 0;
create_readers();
file->reader->identify();
}
static void parse_args(int argc, char **argv) {
track_info_t ti;
int i, j;
filelist_t *file;
char *s;
audio_sync_t async;
cue_creation_t cues;
int64_t id;
language_t lang;
memset(&ti, 0, sizeof(track_info_t));
ti.audio_syncs = new vector<audio_sync_t>;
ti.cue_creations = new vector<cue_creation_t>;
ti.default_track_flags = new vector<int64_t>;
ti.languages = new vector<language_t>;
ti.aspect_ratio = 1.0;
ti.atracks = new vector<int64_t>;
ti.vtracks = new vector<int64_t>;
ti.stracks = new vector<int64_t>;
// Check if only information about the file is wanted. In this mode only
// two parameters are allowed: the --identify switch and the file.
if ((argc == 2) &&
(!strcmp(argv[0], "-i") || !strcmp(argv[0], "--identify"))) {
identify(argv[1]);
exit(0);
}
// First parse options that either just print some infos and then exit
// or that are needed right at the beginning.
for (i = 0; i < argc; i++)
if (!strcmp(argv[i], "-V") || !strcmp(argv[i], "--version")) {
fprintf(stdout, VERSIONINFO "\n");
exit(0);
} else if (!strcmp(argv[i], "-h") || !strcmp(argv[i], "-?") ||
!strcmp(argv[i], "--help")) {
usage();
exit(0);
} else if (!strcmp(argv[i], "-o") || !strcmp(argv[i], "--output")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -o lacks a file name.\n");
exit(1);
} else if (outfile != NULL) {
fprintf(stderr, "Error: only one output file allowed.\n");
exit(1);
}
outfile = safestrdup(argv[i + 1]);
i++;
} else if (!strcmp(argv[i], "-l") || !strcmp(argv[i], "--list-types")) {
fprintf(stdout, "Known file types:\n ext description\n"
" --- --------------------------\n");
for (j = 1; file_types[j].ext; j++)
fprintf(stdout, " %s %s\n", file_types[j].ext, file_types[j].desc);
exit(0);
} else if (!strcmp(argv[i], "--list-languages")) {
list_iso639_languages();
exit(0);
} else if (!strcmp(argv[i], "-i") || !strcmp(argv[i], "--identify")) {
fprintf(stderr, "Error: --identify can only be used with a file name. "
"No other options are allowed.\n");
exit(1);
}
if (outfile == NULL) {
fprintf(stderr, "Error: no output file given.\n\n");
usage();
exit(1);
}
for (i = 0; i < argc; i++) {
// Ignore the options we took care of in the first step.
if (!strcmp(argv[i], "-o") || !strcmp(argv[i], "--output")) {
i++;
continue;
}
// Global options
if (!strcmp(argv[i], "-q"))
verbose = 0;
else if (!strcmp(argv[i], "-v") || !strcmp(argv[i], "--verbose"))
verbose = 2;
else if (!strcmp(argv[i], "--split")) {
if (((i + 1) >= argc) || (argv[i + 1][0] == 0)) {
fprintf(stderr, "Error: --split lacks the size.\n");
exit(1);
}
parse_split(argv[i + 1]);
i++;
} else if (!strcmp(argv[i], "--split-max-files")) {
if (((i + 1) >= argc) || (argv[i + 1][0] == 0)) {
fprintf(stderr, "Error: --split-max-files lacks the number of files."
"\n");
exit(1);
}
if (!parse_int(argv[i + 1], split_max_num_files) ||
(split_max_num_files < 2)) {
fprintf(stderr, "Error: Wrong argument to --split-max-files.\n");
exit(1);
}
i++;
} else if (!strcmp(argv[i], "--dont-link")) {
no_linking = true;
} else if (!strcmp(argv[i], "--link-to-previous")) {
if (((i + 1) >= argc) || (argv[i + 1][0] == 0)) {
fprintf(stderr, "Error: --link-to-previous lacks the previous UID."
"\n");
exit(1);
}
if (seguid_link_previous != NULL) {
fprintf(stderr, "Error: Previous UID was already given.\n");
exit(1);
}
try {
seguid_link_previous = new bitvalue_c(argv[i + 1], 128);
} catch (exception &ex) {
fprintf(stderr, "Error: Unknown format for the previous UID.\n");
exit(1);
}
i++;
} else if (!strcmp(argv[i], "--link-to-next")) {
if (((i + 1) >= argc) || (argv[i + 1][0] == 0)) {
fprintf(stderr, "Error: --link-to-next lacks the previous UID."
"\n");
exit(1);
}
if (seguid_link_next != NULL) {
fprintf(stderr, "Error: Next UID was already given.\n");
exit(1);
}
try {
seguid_link_next = new bitvalue_c(argv[i + 1], 128);
} catch (exception &ex) {
fprintf(stderr, "Error: Unknown format for the next UID.\n");
exit(1);
}
i++;
} else if (!strcmp(argv[i], "--cluster-length")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --cluster-length lacks the length.\n");
exit(1);
}
s = strstr("ms", argv[i + 1]);
if (s != NULL) {
*s = 0;
if (!parse_int(argv[i + 1], max_ms_per_cluster) ||
(max_ms_per_cluster < 0) ||
(max_ms_per_cluster > 65535)) {
fprintf(stderr, "Error: Cluster length out of range (0..65535).\n");
exit(1);
}
max_blocks_per_cluster = 65535;
} else {
if (!parse_int(argv[i + 1], max_blocks_per_cluster) ||
(max_blocks_per_cluster < 0) ||
(max_blocks_per_cluster > 65535)) {
fprintf(stderr, "Error: Cluster length out of range (0..65535).\n");
exit(1);
}
max_ms_per_cluster = 65535;
}
i++;
} else if (!strcmp(argv[i], "--meta-seek-size")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --meta-seek-size lacks the size argument.\n");
exit(1);
}
if (!parse_int(argv[i + 1], meta_seek_size) || (meta_seek_size < 1)) {
fprintf(stderr, "Error: Invalid size given for --meta-seek-size.\n");
exit(1);
}
i++;
} else if (!strcmp(argv[i], "--no-cues"))
write_cues = false;
else if (!strcmp(argv[i], "--no-meta-seek"))
write_meta_seek = false;
else if (!strcmp(argv[i], "--no-lacing"))
no_lacing = true;
// Options that apply to the next input file only.
else if (!strcmp(argv[i], "-A") || !strcmp(argv[i], "--noaudio"))
ti.no_audio = true;
else if (!strcmp(argv[i], "-D") || !strcmp(argv[i], "--novideo"))
ti.no_video = true;
else if (!strcmp(argv[i], "-S") || !strcmp(argv[i], "--nosubs"))
ti.no_subs = true;
else if (!strcmp(argv[i], "-a") || !strcmp(argv[i], "--atracks")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -a lacks the stream number(s).\n");
exit(1);
}
parse_tracks(argv[i + 1], ti.atracks);
i++;
} else if (!strcmp(argv[i], "-d") || !strcmp(argv[i], "--vtracks")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -d lacks the stream number(s).\n");
exit(1);
}
parse_tracks(argv[i + 1], ti.vtracks);
i++;
} else if (!strcmp(argv[i], "-s") || !strcmp(argv[i], "--stracks")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -s lacks the stream number(s).\n");
exit(1);
}
parse_tracks(argv[i + 1], ti.stracks);
i++;
} else if (!strcmp(argv[i], "-f") || !strcmp(argv[i], "--fourcc")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -f lacks the FourCC.\n");
exit(1);
}
if (strlen(argv[i + 1]) != 4) {
fprintf(stderr, "Error: The FourCC must be exactly four chars "
"long.\n");
exit(1);
}
memcpy(ti.fourcc, argv[i + 1], 4);
ti.fourcc[4] = 0;
i++;
} else if (!strcmp(argv[i], "--aspect-ratio")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --aspect-ratio lacks the aspect ratio.\n");
exit(1);
}
ti.aspect_ratio = parse_aspect_ratio(argv[i + 1]);
i++;
} else if (!strcmp(argv[i], "-y") || !strcmp(argv[i], "--sync")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: -y lacks the audio delay.\n");
exit(1);
}
parse_sync(argv[i + 1], async);
ti.audio_syncs->push_back(async);
i++;
} else if (!strcmp(argv[i], "--cues")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --cues lacks its argument.\n");
exit(1);
}
parse_cues(argv[i + 1], cues);
ti.cue_creations->push_back(cues);
i++;
} else if (!strcmp(argv[i], "--default-track")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --default-track lacks the track ID.\n");
exit(1);
}
if (!parse_int(argv[i + 1], id)) {
fprintf(stderr, "Error: Invalid track ID specified.\n");
exit(1);
}
ti.default_track_flags->push_back(id);
i++;
} else if (!strcmp(argv[i], "--language")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --language lacks its argument.\n");
exit(1);
}
parse_language(argv[i + 1], lang);
ti.languages->push_back(lang);
i++;
} else if (!strcmp(argv[i], "--sub-charset")) {
if ((i + 1) >= argc) {
fprintf(stderr, "Error: --sub-charset lacks its argument.\n");
exit(1);
}
ti.sub_charset = argv[i + 1];
i++;
}
// The argument is an input file.
else {
if ((ti.atracks->size() != 0) && ti.no_audio) {
fprintf(stderr, "Error: -A and -a used on the same source file.\n");
exit(1);
}
if ((ti.vtracks->size() != 0) && ti.no_video) {
fprintf(stderr, "Error: -D and -d used on the same source file.\n");
exit(1);
}
if ((ti.stracks->size() != 0) && ti.no_subs) {
fprintf(stderr, "Error: -S and -s used on the same source file.\n");
exit(1);
}
file = (filelist_t *)safemalloc(sizeof(filelist_t));
file->name = argv[i];
file->type = get_type(file->name);
ti.fname = argv[i];
if (file->type == TYPEUNKNOWN) {
fprintf(stderr, "Error: File %s has unknown type. Please have a look "
"at the supported file types ('mkvmerge --list-types') and "
"contact me at moritz@bunkus.org if your file type is "
"supported but not recognized properly.\n", file->name);
exit(1);
}
file->fp = NULL;
if (file->type != TYPECHAPTERS) {
file->status = EMOREDATA;
file->pack = NULL;
file->ti = duplicate_track_info(&ti);
files.push_back(file);
} else
safefree(file);
delete ti.atracks;
delete ti.vtracks;
delete ti.stracks;
delete ti.audio_syncs;
delete ti.cue_creations;
delete ti.default_track_flags;
delete ti.languages;
memset(&ti, 0, sizeof(track_info_t));
ti.audio_syncs = new vector<audio_sync_t>;
ti.cue_creations = new vector<cue_creation_t>;
ti.default_track_flags = new vector<int64_t>;
ti.languages = new vector<language_t>;
ti.aspect_ratio = 1.0;
ti.atracks = new vector<int64_t>;
ti.vtracks = new vector<int64_t>;
ti.stracks = new vector<int64_t>;
}
}
if (files.size() == 0) {
usage();
exit(1);
}
if (no_linking) {
if ((seguid_link_previous != NULL) || (seguid_link_next != NULL)) {
fprintf(stderr, "Error: '--dont-link' cannot be used together with "
"'--link-to-previous' or '--link-to-next'.\n");
exit(1);
}
if (split_after <= 0)
fprintf(stderr, "Warning: '--dont-link' is only useful in combination "
"with '--split'.\n");
}
}
static char **add_string(int &num, char **values, char *new_string) {
values = (char **)saferealloc(values, (num + 1) * sizeof(char *));
values[num] = safestrdup(new_string);
num++;
return values;
}
static char *strip(char *s) {
char *p;
if (s[0] == 0)
return s;
p = &s[strlen(s) - 1];
while ((p != s) && ((*p == '\n') || (*p == '\r') || isspace(*p))) {
*p = 0;
p--;
}
p = s;
while ((*p != 0) && isspace(*p))
p++;
memmove(s, p, strlen(p) + 1);
return s;
}
static char **read_args_from_file(int &num_args, char **args, char *filename) {
mm_io_c *mm_io;
char buffer[8192], *space;
try {
mm_io = new mm_io_c(filename, MODE_READ);
} catch (exception &ex) {
fprintf(stderr, "Error: Could not open file '%s' for reading command line "
"arguments from.", filename);
exit(1);
}
while (!mm_io->eof() && (mm_io->gets(buffer, 8191) != NULL)) {
buffer[8191] = 0;
strip(buffer);
if ((buffer[0] == '#') || (buffer[0] == 0))
continue;
space = strchr(buffer, ' ');
if ((space != NULL) && (buffer[0] == '-')) {
*space = 0;
space++;
} else
space = NULL;
args = add_string(num_args, args, buffer);
if (space != NULL)
args = add_string(num_args, args, space);
}
delete mm_io;
return args;
}
static void handle_args(int argc, char **argv) {
int i, num_args;
char **args;
args = NULL;
num_args = 0;
for (i = 1; i < argc; i++)
if (argv[i][0] == '@')
args = read_args_from_file(num_args, args, &argv[i][1]);
else
args = add_string(num_args, args, argv[i]);
parse_args(num_args, args);
for (i = 0; i < num_args; i++)
safefree(args[i]);
if (args != NULL)
safefree(args);
}
static void setup() {
#if ! defined(COMP_MSC)
if (setlocale(LC_CTYPE, "") == NULL) {
fprintf(stderr, "Error: Could not set the locale properly. Check the "
"LANG, LC_ALL and LC_CTYPE environment variables.\n");
exit(1);
}
#endif
#if defined(SYS_UNIX) || defined(COMP_CYGWIN)
signal(SIGUSR1, sighandler);
nice(2);
#endif
srand(time(NULL));
cc_local_utf8 = utf8_init(NULL);
cluster_helper = new cluster_helper_c();
}
static void init_globals() {
track_number = 1;
cluster_helper = NULL;
kax_seekhead = NULL;
kax_seekhead_void = NULL;
video_fps = -1.0;
default_tracks[0] = 0;
default_tracks[1] = 0;
default_tracks[2] = 0;
}
static void destroy_readers() {
int i;
filelist_t *file;
for (i = 0; i < files.size(); i++) {
file = files[i];
if (file->reader != NULL) {
delete file->reader;
file->reader = NULL;
}
}
while (packetizers.size()) {
safefree(packetizers[packetizers.size() - 1]);
packetizers.pop_back();
}
}
static void cleanup() {
delete cluster_helper;
filelist_t *file;
destroy_readers();
while (files.size()) {
file = files[files.size() - 1];
free_track_info(file->ti);
safefree(file);
files.pop_back();
}
safefree(outfile);
utf8_done();
}
// Transform the output filename and insert the current file number.
// Rules and search order:
// 1) %d
// 2) %[0-9]+d
// 3) . ("-%03d" will be inserted before the .)
// 4) "-%03d" will be appended
string create_output_name() {
int p, p2, i;
string s(outfile);
bool ok;
char buffer[20];
// First possibility: %d
p = s.find("%d");
if (p >= 0) {
sprintf(buffer, "%d", file_num);
s.replace(p, 2, buffer);
return s;
}
// Now search for something like %02d
ok = false;
p = s.find("%");
if (p >= 0) {
p2 = s.find("d", p + 1);
if (p2 >= 0)
for (i = p + 1; i < p2; i++)
if (!isdigit(s[i]))
break;
ok = true;
}
if (ok) { // We've found e.g. %02d
string format(&s.c_str()[p]), len = format;
format.erase(p2 - p + 1);
len.erase(0, 1);
len.erase(p2 - p - 1);
char buffer[strtol(len.c_str(), NULL, 10) + 1];
sprintf(buffer, format.c_str(), file_num);
s.replace(p, format.size(), buffer);
return s;
}
sprintf(buffer, "-%03d", file_num);
// See if we can find a '.'.
p = s.rfind(".");
if (p >= 0)
s.insert(p, buffer);
else
s.append(buffer);
return s;
}
void create_next_output_file(bool last_file, bool first_file) {
string this_outfile;
kax_segment = new KaxSegment();
kax_cues = new KaxCues();
kax_cues->SetGlobalTimecodeScale(TIMECODE_SCALE);
if (pass == 1) {
// Open the a dummy file.
try {
out = new mm_null_io_c();
} catch (exception &ex) {
die("mkvmerge.cpp/create_next_output_file(): Could not create a dummy "
"output class.");
}
cluster_helper->set_output(out);
render_headers(out, last_file, first_file);
return;
}
if (pass == 2)
this_outfile = create_output_name();
else
this_outfile = outfile;
// Open the output file.
try {
out = new mm_io_c(this_outfile.c_str(), MODE_CREATE);
} catch (exception &ex) {
fprintf(stderr, "Error: Couldn't open output file %s (%s).\n",
this_outfile.c_str(), strerror(errno));
exit(1);
}
if (verbose)
fprintf(stdout, "Opened '%s\' for writing.\n", this_outfile.c_str());
cluster_helper->set_output(out);
render_headers(out, last_file, first_file);
file_num++;
}
void finish_file() {
int64_t old_pos;
int i;
// Render the cues.
if (write_cues && cue_writing_requested) {
if (verbose == 1)
fprintf(stdout, "Writing cue entries (the index)...");
kax_cues->Render(*out);
if (verbose == 1)
fprintf(stdout, "\n");
}
// Write meta seek information if it is not disabled.
if (write_meta_seek) {
if (write_cues && cue_writing_requested)
kax_seekhead->IndexThis(*kax_cues, *kax_segment);
kax_seekhead->UpdateSize();
if (kax_seekhead_void->ReplaceWith(*kax_seekhead, *out, true) == 0) {
fprintf(stdout, "Warning: Could not update the meta seek information "
"as the space reserved for them was too small. Re-run "
"mkvmerge with the additional parameters '--meta-seek-size "
"%lld'.\n", kax_seekhead->ElementSize() + 100);
if (write_cues && cue_writing_requested) {
delete kax_seekhead;
kax_seekhead = new KaxSeekHead();
kax_seekhead->IndexThis(*kax_cues, *kax_segment);
kax_seekhead->UpdateSize();
kax_seekhead_void->ReplaceWith(*kax_seekhead, *out, true);
}
}
}
// Now re-render the kax_duration and fill in the biggest timecode
// as the file's duration.
old_pos = out->getFilePointer();
out->setFilePointer(kax_duration->GetElementPosition());
*(static_cast<EbmlFloat *>(kax_duration)) =
(cluster_helper->get_max_timecode() -
cluster_helper->get_first_timecode()) * 1000000.0 / TIMECODE_SCALE;
kax_duration->Render(*out);
out->setFilePointer(old_pos);
// Set the correct size for the segment.
if (kax_segment->ForceSize(out->getFilePointer() -
kax_segment->GetElementPosition() -
kax_segment->HeadSize()))
kax_segment->OverwriteHead(*out);
delete out;
delete kax_segment;
delete kax_cues;
if (write_meta_seek) {
delete kax_seekhead_void;
delete kax_seekhead;
}
for (i = 0; i < packetizers.size(); i++)
packetizers[i]->packetizer->reset();;
}
void main_loop() {
packet_t *pack;
int i;
packetizer_t *ptzr, *winner;
// Let's go!
while (1) {
// Step 1: Make sure a packet is available for each output
// as long we havne't already processed the last one.
for (i = 0; i < packetizers.size(); i++) {
ptzr = packetizers[i];
while ((ptzr->pack == NULL) && (ptzr->status == EMOREDATA) &&
(ptzr->packetizer->packet_available() < 2))
ptzr->status = ptzr->packetizer->read();
if (ptzr->pack == NULL)
ptzr->pack = ptzr->packetizer->get_packet();
if ((ptzr->pack != NULL) && !ptzr->packetizer->packet_available())
ptzr->pack->duration_mandatory = 1;
}
// Step 2: Pick the packet with the lowest timecode and
// stuff it into the Matroska file.
winner = NULL;
for (i = 0; i < packetizers.size(); i++) {
ptzr = packetizers[i];
if (ptzr->pack != NULL) {
if ((winner == NULL) || (winner->pack == NULL))
winner = ptzr;
else if (ptzr->pack &&
(ptzr->pack->timecode < winner->pack->timecode))
winner = ptzr;
}
}
if ((winner != NULL) && (winner->pack != NULL))
pack = winner->pack;
else // exit if there are no more packets
break;
// Step 3: Add the winning packet to a cluster. Full clusters will be
// rendered automatically.
cluster_helper->add_packet(pack);
winner->pack = NULL;
// display some progress information
if (verbose >= 1)
display_progress(0);
}
// Render all remaining packets (if there are any).
if ((cluster_helper != NULL) && (cluster_helper->get_packet_count() > 0))
cluster_helper->render();
if (verbose == 1) {
display_progress(1);
fprintf(stdout, "\n");
}
}
int main(int argc, char **argv) {
init_globals();
setup();
handle_args(argc, argv);
if (split_after > 0) {
fprintf(stdout, "Pass 1: finding split points. This may take a while."
"\n\n");
create_readers();
pass = 1;
create_next_output_file(true, true);
main_loop();
finish_file();
fprintf(stdout, "\nPass 2: merging the files. This will take even longer."
"\n\n");
delete cluster_helper;
destroy_readers();
init_globals();
cluster_helper = new cluster_helper_c();
cluster_helper->find_next_splitpoint();
create_readers();
pass = 2;
create_next_output_file(cluster_helper->get_next_splitpoint() >=
cluster_helper_c::splitpoints.size(), true);
main_loop();
finish_file();
} else {
create_readers();
pass = 0;
create_next_output_file(true, true);
main_loop();
finish_file();
}
cleanup();
return 0;
}