DVB Tables
Locked
DVB Tables
 
|
I thought I'd take another go at getting an EPG from a DVB-T/DVB-T2
device. I am sure there are many instances of doing this sensibly and even in C++, but Google seems not to have indexed them. So not a lot of code to "rip off" :-) I have EN 300 468 1.15.1 which is dry and boring, but contains the definite statement of what should be. Reading https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gst- plugins-bad-libs/html/gst-plugins-bad-libs-DVB-variants-of-MPEG-TS- sections.html seems to indicate that the GStreamer EIT section for example is not structured as the EIT table is specified, and all the names are a bit different. This apparent lack of consistency between the standard and the GStreamer stuff is making this much harder work that I thought it should be. Unless I am just misunderstanding. -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
signature.asc (849 bytes)
Re: DVB Tables
|
|
Re: DVB Tables
|
|
On Thu, 2017-05-25 at 18:32 +0000, Nuno Mota wrote:
> I would suggest looking into dvblast. It will give you EIT tables in > XML > format under a C source code. > I've done a sample to extract some of the EIT tags to a file a couple > of > years ago, that's the best way I found at the time. > > Been using this ever since. > https://mailman.videolan.org/pipermail/dvblast-devel/2013-October/001 > 232.html > of an EIT, but the code is VideoLAN rather than GStreamer so can't really give an insight into how the GStreamer sections need to be handled to get the data specified in the standard. GNOME DVBDaemon was a code I was looking at but it is Vala and doesn't deal with the streams directly in code it just creates a shell pipeline as a subprocess and handles the results from that. -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
In reply to this post by Russel Winder
Hi,
On Thu, 2017-05-25 at 13:38 +0100, Russel Winder wrote: > I thought I'd take another go at getting an EPG from a DVB-T/DVB-T2 > device. I am sure there are many instances of doing this sensibly and > even in C++, but Google seems not to have indexed them. So not a lot > of > code to "rip off" :-) > > I have EN 300 468 1.15.1 which is dry and boring, but contains the > definite statement of what should be. > > Reading https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gs > t- > plugins-bad-libs/html/gst-plugins-bad-libs-DVB-variants-of-MPEG-TS- > sections.html seems to indicate that the GStreamer EIT section for > example is not structured as the EIT table is specified, and all the > names are a bit different. This apparent lack of consistency between > the standard and the GStreamer stuff is making this much harder work > that I thought it should be. > > Unless I am just misunderstanding. specs. Which part don't you think is similar ? Edward > > > _______________________________________________ > gstreamer-devel mailing list > [hidden email] > https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
On Fri, 2017-05-26 at 08:01 +0200, Edward Hervey wrote:
> […] > I did my best to make it a 1:1 mapping with the field naming in the > specs. Which part don't you think is similar ? > It may well be then that I am just misunderstanding – actually I hope so, as it will easier to fix that. :-) The standard presents an EIT section as follows (Table 7 of the standard): event_information_section(){ table_id section_syntax_indicator reserved_future_use reserved section_length service_id reserved version_number current_next_indicator section_number last_section_number transport_stream_id original_network_id segment_last_section_number last_table_id for(i=0;i<N;i++){ event_id start_time duration running_status free_CA_mode descriptors_loop_length for(i=0;i<N;i++){ descriptor() } } CRC_32 } The GstMpegtsEIT structure is though: struct GstMpegtsEIT { guint16 transport_stream_id; guint16 original_network_id; guint8 segment_last_section_number; guint8 last_table_id; gboolean actual_stream; gboolean present_following; GPtrArray *events; }; which doesn't seem to map to any bits of the standard's Table 7 – though it is clear that the events, a sequence of GstMpegtsEITEvent structures does somewhat map to the N events in the section: struct GstMpegtsEITEvent { guint16 event_id; GstDateTime *start_time; guint32 duration; GstMpegtsRunningStatus running_status; gboolean free_CA_mode; GPtrArray *descriptors; }; My code is getting the events, seemingly correctly (*), but it is the SID and stuff like that that give the channel numbers that I cannot see how to extract. The EIT section should have the service_id, but the GstMpegtsEIT structure does not appear to offer it. I guess I had expected the GStreamer structures to map more or less exactly to the tables in the standard, but this seems to be only partially true. Where it is true, it is easier to work things out. Where it apparently is not the case, I am at a bit of a loss as to how to proceed processing the events on the bus so as to collate the data into an in-memory representation of the EIT. (*) Though I regularly get memory errors due to all the malloc and free stuff that I really do not properly understand yet. Oh for doing this in D with it's garbage collector. -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
Hi,
On Fri, 2017-05-26 at 07:31 +0100, Russel Winder wrote: > On Fri, 2017-05-26 at 08:01 +0200, Edward Hervey wrote: > > > > […] > > I did my best to make it a 1:1 mapping with the field naming in > > the > > specs. Which part don't you think is similar ? > > > > It may well be then that I am just misunderstanding – actually I hope > so, as it will easier to fix that. :-) > > The standard presents an EIT section as follows (Table 7 of the > standard): > > event_information_section(){ > table_id > section_syntax_indicator > reserved_future_use > reserved > section_length > service_id > reserved > version_number > current_next_indicator > section_number > last_section_number > transport_stream_id > original_network_id > segment_last_section_number > last_table_id > for(i=0;i<N;i++){ > event_id > start_time > duration > running_status > free_CA_mode > descriptors_loop_length > for(i=0;i<N;i++){ > descriptor() > } > } > CRC_32 > } > > The GstMpegtsEIT structure is though: > > struct GstMpegtsEIT { > guint16 transport_stream_id; > guint16 original_network_id; > guint8 segment_last_section_number; > guint8 last_table_id; > > gboolean actual_stream; > gboolean present_following; > > GPtrArray *events; > }; > > which doesn't seem to map to any bits of the standard's Table 7 .freedesktop.org/data/doc/gstreamer/head/gst-plugins-bad-libs/html/gst- plugins-bad-libs-Base-MPEG-TS-sections.html#GstMpegtsSection-struct Just get them directly :) _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
On Fri, 2017-05-26 at 09:54 +0200, Edward Hervey wrote:
> […] > > All the standard bits are in the GstMpegtsSection : https://gstream > er > .freedesktop.org/data/doc/gstreamer/head/gst-plugins-bad- > libs/html/gst- > plugins-bad-libs-Base-MPEG-TS-sections.html#GstMpegtsSection-struct > > Just get them directly :) But that means processing all sections requiring a full-blown state machine to correctly put together all the sections for all the tables. I had hoped the GStreamer bus and the section type system would have made things easier than that. -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
Administrator
|
Re: DVB Tables
|
|
Thanks :)
Was going to point to that. Hopefully it should make the usage clearer. Edward On Fri, 2017-05-26 at 02:23 -0700, Baby Octopus wrote: > Have you looked into ts-parser to see if it gives you all the info? > https://cgit.freedesktop.org/gstreamer/gst-plugins-bad/tree/tests/exa > mples/mpegts > > ~BO > > > > -- > View this message in context: http://gstreamer-devel.966125.n4.nabble > .com/DVB-Tables-tp4683104p4683123.html > Sent from the GStreamer-devel mailing list archive at Nabble.com. > _______________________________________________ > gstreamer-devel mailing list > [hidden email] > https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
In reply to this post by Baby Octopus
On Fri, 2017-05-26 at 02:23 -0700, Baby Octopus wrote:
> Have you looked into ts-parser to see if it gives you all the info? > https://cgit.freedesktop.org/gstreamer/gst-plugins-bad/tree/tests/exa > mples/mpegts > No I hadn't looked at that, I think that will be most helpful. Thank you for pointing me to it. As well as being helpful, it also proves C is a really dreadful programming language for all this stuff. :-( -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
Re: DVB Tables
|
|
In reply to this post by Edward Hervey-4
On Fri, 2017-05-26 at 11:52 +0200, Edward Hervey wrote:
> Thanks :) > > Was going to point to that. Hopefully it should make the usage > clearer. > > Edward > Yes. And No. I pulled out the ts-parser.c file, made a meson.build so as to create an executable outside the plugins hierarchy. I ran it with a playbin using the dvb URI scheme and it dumps out all the good stuff you'd expect. And doesn't seem to fail – which is good. :-) If in my code I just look at the DVB short event a lot of the time it works but very quickly I get a memory allocation failure. (me-tv:6718): GLib-ERROR **: /build/glib2.0-B1uXKV/glib2.0-2.50.3/./glib/gmem.c:165: failed to allocate 20401094656 bytes Trace/breakpoint trap Now whilst I didn't cut and paste the ts-parser C code I cannot see any difference between it and what I have that would matter. So the implication is that each EIT section on the bus cannot be treated independently, that there is some non-local interaction happening – or that there is something really weird with the whole descriptor system. I am guessing that you really do have to fully process many different messages to avoid the problem with the short event messages. -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[hidden email] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [hidden email] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder _______________________________________________ gstreamer-devel mailing list [hidden email] https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel |
| Free forum by Nabble | Edit this page |