[mythtv-users] "New Episodes Only" filtering for recording rules

Nick Morrott knowledgejunkie at gmail.com
Tue Jun 14 01:29:52 UTC 2011

On 13 June 2011 22:51, Martin Moores <moores.martin at gmail.com> wrote:
> On 13 June 2011 20:23, Jeffrey Hahn <jmhahn at gmail.com> wrote:
>> I've seen a similar issue with recording Merlin on Syfy in the US. The
>> original airdates were not the same as the date they were premiering
>> here, so it was considered a rerun on the schedule. I made a custom
>> power search to handle it:
>> channel.channum = '829' AND program.title = 'Merlin' AND
>> datediff(now(),program.originalairdate) < 365
>> You'd want to change the channel number and title, of course, and it
>> assumes you will get the show within 1 year of the original airdate.
>> With duplicate detection set to "Subtitle and Description" and no
>> filter, I've never had a false positive and never missed a new
>> episode.
>> -Jeff
> I am using xmltv data, which has been much better than EIT so far.
> The first example of this is CSI: High And Low, which is on Five USA on
> Wednesday at 12:40, this episode is from 2002, but not marked as a repeat
> and shows an original airdate of 2011!
> I did think about setting the rule to just record from channel 5, but then
> if for whatever reason that showing was missed, but it was repeated on
> another channel, myth wouldn't schedule it on another channel, unless I
> mis-understood how that works.
> I guess duplicate detection doesn't work here, as I have only ever recorded
> new showings of CSI, never all the re-runs, so myth wouldn't pick these up
> as previously recorded etc.  It does work well with picking the new showing
> each week (my rules are find and record one showing each week, apart from
> the odd time where the new episode is missed in its normal timeslot and a
> re-run is recorded instead, never got to the bottom of that).  I guess if
> these re-runs are not marked as repeats or generic, then myth would not know
> not to record them, so it is only as good as the data it gets.
> Thanks for that rule Jeff, will give that a try

Note that the Radio Times XMLTV data does not provide original airdate
information in the same way that Schedules Direct does in the US and,
as you have realised, the scheduler is at the mercy of the source data
flagging episodes as premieres/repeats.

I have a couple of ideas for XMLTV/uk_rt going forwards that may help
to further improve listings data, including:

i) marking descriptions as generic where they are not specific to a
particular episode - this could allow MythTV and other PVRs to treat
episodes as generic in the same way that the Schedules Direct listings

ii) marking repeats by tracking the current season of popular shows
and automatically flagging older episodes from previous series as
repeats (as determined from a comparison with a given episode number)
if that flag is not already specified (this would help in your current
CSI example)


Nick Morrott

MythTV Official wiki: http://mythtv.org/wiki/
MythTV users list archive: http://www.gossamer-threads.com/lists/mythtv/users

"An investment in knowledge always pays the best interest." - Benjamin Franklin

More information about the mythtv-users mailing list