February 2013 - DVDnav-discuss - lists.mplayerhq.hu

[PATCH] do not close file twice
by Reimar Döffinger 05 Oct '13

05 Oct '13

There is another close further down that will always be reached. Current code will result in a double close. Index: vm/vm.c =================================================================== --- vm/vm.c (revision 1243) +++ vm/vm.c (working copy) @@ -176,7 +176,6 @@ off = lseek( fd, 32 * (off_t) DVD_VIDEO_LB_LEN, SEEK_SET ); if( off == ( 32 * (off_t) DVD_VIDEO_LB_LEN ) ) { off = read( fd, data, DVD_VIDEO_LB_LEN ); - close(fd); if (off == ( (off_t) DVD_VIDEO_LB_LEN )) { fprintf(MSG_OUT, "libdvdnav: DVD Title: "); for(i=25; i < 73; i++ ) {

5 13

[PATCH] fix dvdnav_convert_time to work for full range
by Reimar Döffinger 05 Oct '13

05 Oct '13

Hello, currently the calculation will use int and thus overflow for anything larger or equal to 20 hours. Not sure if this is the best/nicest way, personally I prefer to do this kind of thing with the result = hour >> 4; result *= 10; result += hour; result *= 6; result += minute >> 4; result *= 10; .... scheme, but this is the minimal change required. Index: dvdnav.c =================================================================== --- dvdnav.c (revision 1243) +++ dvdnav.c (working copy) @@ -203,7 +203,7 @@ int64_t result; int64_t frames; - result = (time->hour >> 4 ) * 10 * 60 * 60 * 90000; + result = (time->hour >> 4 ) * 10 * 60 * 60 * 90000ull; result += (time->hour & 0x0f) * 60 * 60 * 90000; result += (time->minute >> 4 ) * 10 * 60 * 90000; result += (time->minute & 0x0f) * 60 * 90000;

3 2

[patch] Fix SIGSEGV in dvdnav_describe_title_chapters
by Sylvain HENRY 05 Oct '13

05 Oct '13

Hi, Please consider checking and applying the attached patch. It fixes a segfault occuring with some DVDs. It seems to be the same problem as the one mentionned here: https://bugs.launchpad.net/ubuntu/+source/libdvdnav/+bug/934471 In my case, "cellnr" and "endcellnr" variables were set to 0 when the crash occured. Cheers Sylvain PS: I'm not registered on the mailing-list, please answer directly to me if necessary

3 8

Re: [DVDnav-discuss] Error in libdvdnav when reading iso images
by Fabian Keil 05 Oct '13

05 Oct '13

Oncaphillis wrote in http://lists.mplayerhq.hu/pipermail/dvdnav-discuss/2012-May/001722.html: > I used to create DVD iso images via k3b and played them via VLC. > With isos of relatively newly released DVD this stops to work and > VLC fails with the message > > libdvdnav: demux error! 00 00 00 (should be 0x000001) > > Playing directly from /dev/dvd still works > > The distro is Fedora16 > + k3b == 2.0.2 > + libdvdcss == 1.2.11 ( I also tested 1.2.12) > + vlc = 1.1.13 > + libdvdnav 4.1.4. I also tried the current SVN build > > I get the same error when playing with mplayer and > I also tried brasero for iso creation. > > Since direct play works it seems like brasero/k3b have > problems to create a proper iso and the problems sits > somewhere in the libs. The above message can be generated when the ripping program has generously padded regions with unreadable sectors with binary zeros and one of the padded blocks is actually referenced in the ifos. In this case it's not the result of a bug in the libs (assuming you meant libdvdnav or libdvdread) but the result of a limitation in the program that is using them in combination with companies that interpret the relevant standards somewhat liberal. I'm not familiar with k3b or brasero and thus don't know if this is the cause of the issue, but at least it would explain why the real disc works and the copy doesn't. I've never seen the issue with discs that haven't obviously been mastered incorrectly, but I suppose it can also happen if the disc is scratched. In my experience, copies of intentionally incorrectly mastered discs that cause this issue when padding with zeroes, can usually be played correctly with VLC when padding with other blocks from the region that have been read correctly instead. It's not perfect, but can be sufficient to solve the issue. I'm still experimenting with this, though. With some of those copies I can trigger at least four different crashes in libdvdnav 4.2.0's dvdnav_describe_title_chapters(), that aren't triggered when ripping with zero padding which lets VLC rejects the disc right away. Recent VLC versions use this function, but one can work around it by undefining HAVE_DVDNAV_DESCRIBE_TITLE_CHAPTERS. VLC 1.1.3 doesn't seem to use the function yet and thus correctly plays the copies out of the box (I only confirmed the latter). One of these issues has already been fixed after the libdvdnav 4.2.0 release (https://bugs.launchpad.net/ubuntu/+source/libdvdnav/+bug/934471) but as far as I can tell the others still exist, or at least don't seem to be fixed in https://github.com/microe/libdvdnav and I haven't seen them mentioned when crawling the archives either. I attached the patches that seem to prevent these crashes for me. I'd appreciate comments on whether or not the skip conditions are reasonable, especially the one in patch 3. Note that I ran into the issues with copies where I now know that the padding has been done somewhat sloppy. I haven't tested yet if these crashes can be reached by playing the original discs. If not, I suspect that with a fancier padding algorithms the copies might play correctly as well. Fabian

4 4

[PATCH] Sanitize PTT start offsets
by John Stebbins 05 Oct '13

05 Oct '13

If the start offset points past the last_byte (outside the region read) stop reading PTTs and adjust nr_of_srpts. Fixes Transformers 3 -- John GnuPG fingerprint: D0EC B3DB C372 D1F1 0B01 83F0 49F1 D7B2 60D4 D0F7

3 5

[PATCH] "dvdnav_jump_to_sector" as an alternative to "dvdnav_time_search" (REV 5: combined args to structs)
by gnosygnu 29 Jul '13

29 Jul '13

This revision includes the following changes: 1) combined multiple arguments into structures (See dvdnav_pos_data_t, dvdnav_cell_data_t, dvdnav_jump_args_t) 2) changed multi-line statements to remove leading commas 3) removed tabs/trailing whitespace 4) replaced "discontinuity tmap indexes" approach with simpler "use entry sector of current cell" to handle DVDs with initial dummy cells. Thanks to all who have read through the patch and/or offered comments. As far as I know, there are no other known pending changes/requests. I am hoping that this patch is at a near-final state. Please let me know what else should be done to get this patch accepted. Thanks.

3 19

Still timeout workaround in vm.c
by Richard 22 Feb '13

22 Feb '13

Hi, In vm_position_get there is a workaround for discs that indicate a still by "a cell playback time higher than the time the frames in this cell actually take to play (like 1 frame with 1 minute playback time)", and that these discs are considered to be broken. I don't think they *are* broken. It's unusual but I don't think there's any reason why a VOBU needs to be more than one sector (i.e. the NAV packet) long. Take audio DVDs, for example (and here I don't mean "proper" audio DVDs, with data in the AUDIO_TS directory, I mean DVDs that are basically audio-only with data in the VIDEO_TS directory and playable on a normal DVD player). These DVDs often show a single frame over several minutes worth of music. The first VOBU of a cell on such a disc will typically contain a single video frame along with audio frames. Subsequent VOBUs will only contain audio packets (ignoring the NAV packets). As I understand it, the initial video frame should be shown either until the next video frame is found or until the end of the cell. Through a combination of the 'pci_gi_t.vobu_se_e_ptm', 'dsi_gi_t.vobu_1stref_ea' and 'vobu_sri_t.next_video' fields in the NAV packets, it can be determined when the last frame will come, and when the next should be expected. So, having determined that, if you remove the audio stream, you have the exact scenario that the workaround was implemented for, except that it doesn't appear to be a broken or invalid scenario to me. Unusual, certainly, but not invalid. It would also allow timeouts of greater than 254 seconds (although that could probably also be achieved via other means, but would still be a valid alternative). In this scenario, it would be up to the player to ensure the correct timing is maintained by syncing itself to the start and end PTMs in each NAV packet rather than video or audio data. The problem with the workaround (apart from the duration issue) is that a player following the above behaviour will likely get into trouble as dvdnav_get_next_cache_block will return DVDNAV_STILL_FRAME at the end of the cell, and cause it to show the frame for too long (at best twice as long, at worst, indefinitely). Does that explanation make sense? Comments? Richard.

1 1

Problems with libdvdread 4.2
by Guy Dawson 18 Feb '13

18 Feb '13

Hi In common with many other people, I appear to be having problems with libdvdread. Whilst I am in no tearing hurry to resolve this, and I am currently happy to await a resolution, I thought it might be helpful to share my experience. I have been using VLC under Fedora for some time now, and have had no major problems until I upgraded to F17. With the current setup, I cannot play DVD using VLC. A similar problem occurs if I try lsdvd, which give this result ; <<<< [root@blitz ~]# lsdvd /dev/sr1 libdvdread: Using libdvdcss version 1.2.11 for DVD access *** libdvdread: CHECK_VALUE failed in /builddir/build/BUILD/libdvdread-4.2.0/src/ifo_read.c:1251 *** *** for vts_ptt_srpt->title[i].ptt[j].pgn != 0 *** *** libdvdread: CHECK_VALUE failed in /builddir/build/BUILD/libdvdread-4.2.0/src/ifo_read.c:1980 *** *** for pgci_ut->nr_of_lus < 100 *** >>>> The point I want to make here is that this happens with Disney's Toy Story, which is a disc that I have defineatly played in the past using VLC. This implies that whatever the problem is, it is a regression rather that a new bug. Note that this happend on both X86_64 & i686 systems. I this information helps... UBD

1 0

[PATCH] Fix playback of DVDs with menu broken by revision 1245
by Erik Auerswald 08 Feb '13

08 Feb '13

Hi, svn revision 1245 broke playback of DVDs _with_ menu by incorrectly translating an assertion into an early return: ---[cut here]--- $ svn log -r1244:1245 ------------------------------------------------------------------------ r1245 | rathann | 2012-12-09 22:13:14 +0100 (Sun, 09 Dec 2012) | 4 lines Fix failed assertion when there's no menu. Patch by Erik Hovland. ------------------------------------------------------------------------ $ svn diff -r1244:1245 Index: vm/vm.c =================================================================== --- vm/vm.c (revision 1244) +++ vm/vm.c (revision 1245) @@ -1762,7 +1762,8 @@ pgcit_t *pgcit; pgcit = get_PGCIT(vm); - assert(pgcit != NULL); /* ?? Make this return -1 instead */ + if (pgcit != NULL) + return 0; if(pgcN < 1 || pgcN > pgcit->nr_of_pgci_srp) { #ifdef TRACE ---[cut here]--- To return early if the assertion fails would be ---[cut here]--- Index: vm/vm.c =================================================================== --- vm/vm.c (revision 1250) +++ vm/vm.c (working copy) @@ -1762,7 +1762,7 @@ pgcit_t *pgcit; pgcit = get_PGCIT(vm); - if (pgcit != NULL) + if (pgcit == NULL) return 0; if(pgcN < 1 || pgcN > pgcit->nr_of_pgci_srp) { ---[cut here]--- This simple change restores playback of DVDs containing a menu. The current code skips the actual functionality of vm.c::set_PGCN() if the call to vm.c::get_PGCIT() was successful! There were quite a few bug reports about MPlayer (svn version) dvdnav:// playback being broken, see this MPlayer-users thread: http://lists.mplayerhq.hu/pipermail/mplayer-users/2013-January/085825.html My patch (inline above, attached as well) fixes this regression (tested with current MPlayer from svn) and fits the commit log of revision 1245. Please consider applying. Thanks, Erik

3 5

libdvdnav fork
by Jean-Baptiste Kempf 06 Feb '13

06 Feb '13

Hello, As I am a bit annoyed by the lack of response and maintenance on this project, and also because I am not happy when the programs are the ones pointed at when there is a crash inside this library, I am, for _now_, forking the project on the videolan infrastructure. So far, the git has most of Erik's patches, Fabian's ones and a revert of the obvious mistake on the assert. People are welcome to share patches and join the project. People who have access to libbluray should already have write access to this new repo. People who have write access to libdvdcss, FFmpeg, VLC or libaacs are very quick to add. git://git.videolan.org/libdvdnav.git Sorry to have to do this, Best regards, -- Jean-Baptiste Kempf http://www.jbkempf.com/ - +33 672 704 734 Sent from my Electronic Device

7 10