[mythtv-users] HD Homerun Prime Scripting

[David Engel]

On Wed, Oct 19, 2011 at 02:42:29PM -0400, Eric Sharkey wrote:
> On Wed, Oct 19, 2011 at 1:02 PM, Gary Buhrmaster
> <gary.buhrmaster at gmail.com> wrote:
> > On Wed, Oct 19, 2011 at 06:37, Eric Sharkey <eric at lisaneric.org> wrote:
> > ...
> >> You could separate this out by splitting the channel table in two.
> >> Have chanid, callsign, xmltvid, virtual channel number, etc. in one
> >> table, and make the video sources table be just the sourceid, chanid,
> >> and tuning information for that that channel on that source.  That
> >> would simplify maintenance of the qam channel list considerably.
> >
> > Does your proposed redesign handle the cases
> > where you have multiple providers and slightly
> > differing content from each?
> 
> Mostly.
> 
> > Extreme US example:
> > One is Comcast, where they rebroadcast the OTA
> > NBC channel as channel 2, but your outside antenna
> > receives it as channel 3.1, and your AT&T U-verse
> > box makes it available as channel 4.
> 
> That part of the scenario is already covered by the above proposal.
> (I admit that there are a lot of details I've waved my hands over, but
> this would still work.)
> 
> > Your Dish
> > receiver gets that local channel on 1003, but needs
> > a different content source identifier xmltvid because
> > they occasionally preempt that channel to broadcast
> > something of more importance.
> 
> I'm not sure how to handle that one.  How do we handle that today?
> Imperfectly I'd imagine.
> 
> > Your satellite
> > gets NBC from the national feed, so when you
> > watch the Today show, you do not get the local
> > weather (but you did get more of Willard Scott).
> 
> I'd call that a different channel, then.  I get NBC out of New York
> and I also get NBC out of Philadelphia.  Only the New York channel is
> WNBC.  NBC out of Philly is WCAU.  They're different channels with 98%
> identical content.
> 
> > simplifying the
> > channel tables to not provide the necessary
> > flexibility is not the way to go.
> 
> I wouldn't dream of it.

Guys, no one disagrees that the current schema is less than ideal.
FWIW, the program and related tables have huge duplication of data too
when programs are repeated many times.  All of the devs have much more
pressing things to work on, however, so unless you're willing to back
up your redesign with code, you're just tilting at windmills.

David
-- 
David Engel
david at istwok.net