[mythtv-users] Myth Recording Extender (MRE) v1.0.0 Released
Michael T. Dean
mtdean at thirdcontact.com
Fri Sep 1 02:40:27 UTC 2006
On 08/31/06 22:07, Ross Campbell wrote:
>You know, as I was just reading replies, it occured to me that
>tracking overtime sporting events and last minute scheduling changes
>is really something the good folks at Zap2It labs should do (or are
I think they only "publish" the data they are given by the stations. I
don't think they watch all the channels for schedule discrepancies, etc.
Now, if you could convince stations to provide up-to-the-minute
schedule-change information to TMS so they could publish it to allow
people to more easily record the shows on that station so they could
>Only problem is... we don't want to overload their servers with more
>frequent updates... so I got to thinking about all the ugly wget stuff
>that flies by when I manually run mythfilldatabase, and how so much of
>that data is existing data...
>Why couldn't Zap2It put their XML files into some directory structure
>that accounts for area code and provider and store that data in
>subversion? Then we could pull ONLY THE CHANGES with a svn update,
>which is presumably *MUCH* nicer on their servers than wget of 14 days
>of schedule data.
>Perhaps nice enough that MythTV could check more frequently for
Customized (per-user) lineups. If they weren't customized, DD usage
would require significantly more bandwidth. For example, downloading
listings for the 13 channels I get on DISH network is significantly less
bandwidth than downloading listings for the 890 stations TMS lists for DISH.
TMS explicitly chose the customized lineups--knowing it would require
significantly more server resources than standard lineups from which the
clients could customize listings--to save bandwidth. They also asked
developers of clients to use the bandwidth effectively.
Therefore, MythTV downloads only +1 (tomorrow) and +maxDays (where
maxDays is either a default decided by the grabber maintainer or the
value the user specifies to mythfilldatabase with --max-days). In
theory, the data from +maxDays should be new, so only the data for +1 is
existing data (possibly with a few changes). The only time there would
be significantly more existing data downloaded than new data is when the
user irresponsibly species the --refresh-all option to mythfilldatabase
in MythFillDatabaseArgs or runs the grabber more than once per day.
More information about the mythtv-users