[mythtv-users] Import and Export of Program Listing Data
robert.current at gmail.com
Thu Sep 6 20:57:25 UTC 2007
I was hoping the topic here would be a collection of ideas, proposals
for projects, and references to existing projects and discussions.
I'm quite happy and shocked that it's gone really well!
I'm going to see if I have even a slight handle on this monster…
anyone with URLs or concepts to add, please voice up! And anyone who
might be willing to outline how to make a workable solution to this
monster in a different way, I'd love to see that too.
I'll admit, I was against a couple until I learned the benefits. I
may be a bit paranoid because of all the hot debate I've seen, but
here is what sounds best to me at this point, and I would suggest
breaking each of these into separate little projects, because that way
people who want them can work on the part they want, and people who
want to attack them can attack the single parts they dislike and leave
the rest alone.
Rough Outline of Ideas Mentioned
These concepts seem clear and workable IMHO:
Files standard to be distributed would best be XMLTV, so other
projects can benefit.
If XMLTV files have xmltvid linking listing data, even xmltv files
that have non-identical lineups would provide some listing data when
imported. XMLTV files using channelid or any other identifier would
create unusable data to anyone that didn't have a completely identical
Importing xmltv data to mythconverg is very doable, almost done, just
not some pretty GUI thing.
Exporting xmltv data from mythconverg isn't done. Dumping to XML can
be done, but dumping to accurately formatted xmltv data with correct
xmltvid would have to be addressed.
These concepts seem to be going along just fine on there own:
Data sources, total can of worms, EIT, scrapers, and manually entered
data seem to get a lot of talk. Legality is the biggest can of worms.
But manually entered data seems to be a leagal use, and what the
Scripts to create quality xmltv data have been mentioned (in other
places). Pulling IMDB or other data to inhance existing data seems to
be a really wild and potentially rich way of creating good xmltv data.
But wouldn't this be best left to outside projects like Shepherd? (
These concepts seem to have no clear answer:
Distribution via torrents or USENET or a central data server was
mentioned. Torrents were dis-liked, but there is already a start
MythTV plugin for Torrents. Status of UseNet plugin to MythTV seems
unmentioned, and seems unclear to me right now. I haven't seen anyone
propose a central server, but I've seen it mentioned that if anyone
did it, it should be outside the USA for legal reasons!
How much GUI a "average user" would need to use this, totally unclear.
Manual data entry can be done slick and sweet by using a GUI frontend
for MySQL in Windows or Linux by opening the mythconverg database. A
template to pull open the right tables and joins might be cool.
That's internationally useful, totally promoting legal un-copyright
restricted data entry, and not extremely hard to conceive.
Command lines data import and export seem the easy bit, and the
commands could be quite simple. Would the "average user" even need
GUI? If someone did make it easy, would it even be allowed into the
Would there be any reason, other than file sharing and import/export
of data, to have a GUI spot in MythTV-Setup to allow cron jobs to be
scheduled? A command line might be easy; a setup script could get
around this, but... are they enough for the average user?
More information about the mythtv-users