[mythtv-users] another scheduling strangeness/question

[Michael T. Dean]

  On 08/30/2010 05:52 PM, Brian J. Murrell wrote:
> So perhaps this is an issue that they didn't match nearly two weeks ago
> when that data was initially pulled from SD but was brought into
> alignment with each other at some time between when myth fetched it for
> me and now.  IIUC, Myth wouldn't have "refetched" that data between the
> first time it gets it and the day before it's supposed to be used,
> correct?  If so that might explain it.
>
> But that would mean that nearly 14 days ago when this data was initially
> fetched, it was (perhaps only for a short period of time) not in sync
> between the digital and analog channels.  Not having that data corrected
> for nearly two weeks does yield this kinds of myth scheduling mistakes
> and they are mistakes that can cost a recording if interim scheduling
> moves a show around and it's not discovered until the day before it was
> supposed to be recorded.
>
> Perhaps if these scheduling "sync" problems are only a problem for the
> showings that are near the end of the ~14 day window, each day, not only
> should the listings for "tomorrow" be refreshed but the listings for the
> final 2-3 days should be refreshed to catch up with some of these short
> term mis-syncs?

Far more likely, it was a generic episode when it was retrieved for both 
channels at +13 days (since they have the same xmltvid, it /must/ have 
been identical).  Then, on +12 days, mythfilldatabase re-fetched the 
data for only one of your sources because of its "do we have a 
substantial amount of 'No Data' when averaged against all channels in 
the source" logic.  The data had been updated by +12, so the source with 
the channel with current data got updated and the other didn't.

Mike