[mythtvnz] Using the Freeview EPG guide in MythTV

[Steven Ellis]

On Mon, June 8, 2015 2:07 pm, Steven Ellis wrote:
>
> On Sat, March 7, 2015 12:48 pm, David Moore wrote:
>> On 07/03/15 11:45, Paulgir wrote:
>>>
>>> I see the Freeview DVB-T EPG now has 8 days of data included.
>>> Up until now I have been using http://epg.org.nz data.
>>> If I want to use the over-the-air guide,can I just check the "use on
>>> air eit" boxes in Myth Backend set up and disable the running of
>>> tv_grab_nz-py ? or is there other set up necessary?
>>>
>>> Paul
>>>
>> Interesting. If I feel the urge I will investigate further. But I am
>> also somewhat disturbed by related things I have just read regarding
>> what TVNZ and the BBC are doing. Here is a quote from the "Metadata"
>> section at http://en.wikipedia.org/wiki/Freeview_%28New_Zealand%29.
>>
>> "During the third week of December 2014, TVNZ tested using the same
>> Huffman look-up tables the BBC implemented to force viewers to use
>> approved Freeview receivers that restrict HD recording and viewing. The
>> Huffman tables are being used to compress the EIT text used for
>> terrestrial schedule event names and descriptions. From March 2015, TVNZ
>> began compressing the EIT schedule again. Compressing the EIT text in
>> the schedule would not achieve the same receiver use given the
>> terrestrial EIT only has limited programme details. Receivers that don't
>> use the BBC huffman tables will either display no details or display
>> garbage characters."
>>
>> If TVNZ is really compressing the EIT using Huffman coding then the
>> MythTV EIT decoding code must be pretty clever to decompress it without
>> access to the Huffman lookup table(s). (Unless the Huffman table is
>> embedded in the EIT data in some standard way maybe?)
>>
>> After reading about how the BBC got OFCOM approval to implement "content
>> protection" for their HD service and some of the interweb comments about
>> how this could affect MythTV users I am nervous about TVNZ's plans. As
>> the paragraph above from Wikipedia says you can't decompress data that
>> has been compressed with Huffman coding unless you know the Huffman
>> decoding tree/lookup table. If TVNZ (and others?) follow the BBC path
>> and compress the EPG and require users to get a licenced copy of the
>> Huffman table then we'll lose our relatively simple access to the EPG
>> over the air.
>>
>> Also it could be the thin end of the wedge heading towards encryption of
>> "premium" content although I don't see why this would be allowed on
>> Freeview. But given what OFCOM allowed the BBC to do you never know what
>> warped reasoning might allow Freeview content to be encrypted.
>>
>> Hopefully my fears are completely ungrounded and we can continue on our
>> merry, open source way with our EPG and recording working fine.
>
> If they are using the same table as FreeSat this has already been
> implemented in MythTV
>
> https://code.mythtv.org/trac/ticket/6498#no1
>
> It would be interesting to see of epgsnoop could be updated to include the
> huffman tables.
>
> Looks like another EIT grabber already has the huffman support merged from
> MythTV.
>  - https://github.com/ubaldus/tv_grab_dvb_plus
>

Having  play against DTT with tv_grab_dvb_plus


$ ./src/tv_grab_dvb_plus -a 2 -d DEBUG -o /tmp/dtt.xml
tv_grab_dvb_plus DEBUG    midnight today="20150626000000 +1200"
tv_grab_dvb_plus DEBUG   start of period="20150626000000 +1200"
tv_grab_dvb_plus DEBUG     end of period="20160625000000 +1200"
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus ERROR   data or length is wrong for pid 20 (0x14) tableid
112 (0x70), len 8. skipping packet.
tv_grab_dvb_plus DEBUG   timeout occurred
tv_grab_dvb_plus DEBUG   freesat huffman average expansion ratio: 3.000000

Looks like the huffman decode works.

Wonder if this can be merged into epgsnoop?

--------------------------------------------
Steven Ellis - Technical Director
OpenMedia Limited - The Home of myPVR
email   - steven at openmedia.co.nz
website - http://www.openmedia.co.nz