[mythtv] Same old mind reading crud.
schachte at csse.unimelb.edu.au
Fri Apr 13 04:38:34 UTC 2007
Bruce Markey wrote:
> There is no magic solution. If the content of a
> program is expected to extend outside the listed times, then
> the scheduler needs to be told with endoffset that the highest
> priority shows must record to their conclusion. If you expect
> to require extra time to capture content for the majority of
> your rules, DefaultEndOffset can be used.
There's no magic mind-reading solution, but we could *tell* Myth when it's OK
to drop padding and when it's not, and have myth do that. Just as we give
priorities to programs and let Myth do the actual scheduling accordingly.
If each program had a "Drop the padding if necessary" option in the scheduling
options screen (right after Start Early and End Late), then if Myth finds it
can't schedule a program whose hard padding is permitted to be dropped, it
could try again without the padding. OK, it's more complicated than that;
you'd want to keep as much of the padding as possible, and you'd want to allow
two programs of the same priority both of whose padding can be dropped to share
the available time for padding. But the point is that it could optionally drop
padding without mind reading if it had just a little bit of extra info.
Peter Schachte Let's face it, conceptual purity is not where
schachte at cs.mu.OZ.AU Prolog's at.
www.cs.mu.oz.au/~schachte/ -- Richard O'Keefe
Phone: +61 3 8344 1338
More information about the mythtv-dev