[mythtv] EIT updates for in-progress recordings

[Stuart Auchterlonie]

On 24/06/2010 17:14, Håkon Alstadheim wrote:
> Den 24. juni 2010 14:45, skrev Malcolm Parsons:
>> Hi,
>>
>> Last night the BBC's coverage of "Wimbledon 2010" overran due to
>> coverage of a record breaking tennis match.
>>
>> The EIT was updated several times, changing the start and end of the
>> following programme - "Today at Wimbledon".
>>
>> Both programmes were set to be recorded, and I ended up with 1
>> recording of "Wimbledon 2010" and 9 recordings of various parts of
>> what should have been "Today at Wimbledon".
>>
>> I was watching the programme live, so was able to stop the duplicate
>> recordings, but had to keep switching to a different recording when
>> the one I was watching stopped recording. It might have been easier
>> to switch to Live TV.
>>
>> I'm using MythTV 0.23-fixes.
>>
>> http://svn.mythtv.org/trac/ticket/7369 describes similar behaviour
>> with MythTV 0.21, but the ticket was closed as being a rare case.
>>
>> Is anyone else affected by this misbehaviour?
>>
>> Would it be difficult for MythTV to just make one long recording?
>>
>>
> I get this on movies from DVB-T in Norway at least once a week. Bugged
> the hell out of me until I realized what was happening. Quite often late
> night movies are scheduled for something like 22:50, and then
> rescheduled to something like 22:48. I always record a couple of minutes
> before program-start, so I end up with two recordings. The second
> recording, supposedly starting at 22:48, often starts recording LATER,
> so it misses the start. Often only one of the scheduled showings have a
> proper description. So, I end up with one showing of the full movie and
> one amputated. Which one has the proper description seems a bit random
> to me, but I have a hunch that there is a difference between long-term
> and short-term EIT data.
>
>

I wouldn't be surprised in the slightest if they send a now/next eit
data with the changed start time that doesn't have all the info.

The alternative is also possible, that they send correct now/next info
and then the 0-3 day data is incorrect.

It's also possible we just don't handle this case very well at all and
make a hash out of it...


Stuart