MythTV Dropping Support for Data Direct in MythTV 31

Forum rules
Get Help using the Schedules Direct service. NOTE: application issues are probably better resolved via the application's support methods. If you post here, at least include your application name!

MythTV Dropping Support for Data Direct in MythTV 31

Postby gtb » Wed Jan 09, 2019 11:20 am

Was posted on the MythTV developers list. Archive can be seen
at: http://lists.mythtv.org/pipermail/mytht ... 77489.html

---

FYI, MythTV will be dropping support for Schedules Direct's legacy,
Data Direct, guide data service in version 31. Going forward, MythTV
will only support Schedules Direct's newer, JSON-based, guide data
service via XMLTV. Currently, there are two XMLTV grabbers
(tv_grab_zz_sdjson and tv_grab_zz_sdjson_sqlite) which support
Schedules Direct's service

We expect this change to be made in the master branch of MythTV
shortly after the release of version 30. That means probably in the
next few weeks. If you are tracking MythTV's master branch and are
still using Data Direct for your guide data, now is the time to switch
to XMLTV. Instructions for setting up XMLTV for use with Schedules
Direct can be found in the MythTV Wiki at
https://www.mythtv.org/wiki/XMLTV#Sched ... rect_Setup.

The goal of this change is to simplify the logic in mythfilldatabase
and pave the way for channel management and possibly other
improvements.

David
gtb
 
Posts: 82
Joined: Thu Oct 02, 2014 2:07 pm

Re: MythTV Dropping Support for Data Direct in MythTV 31

Postby rmeden » Tue Jan 22, 2019 10:26 pm

You're not *REQUIRED* to use the JSON service... you could use tv_grab_na_dd to get XMLTV format from the DataDirect service (although I'm not sure why you'd want to).
rmeden
SD Board Member
 
Posts: 1518
Joined: Tue Aug 14, 2007 2:31 pm
Location: Cedar Hill, TX

Re: MythTV Dropping Support for Data Direct in MythTV 31

Postby gtb » Thu Feb 21, 2019 2:03 pm

rmeden wrote:You're not *REQUIRED* to use the JSON service... you could use tv_grab_na_dd to get XMLTV format from the DataDirect service (although I'm not sure why you'd want to).

You could always post on the MythTV (dev) list to point out to David that additional option.

I think(*) the point David was trying to make was that the dedicated zap2it labs EPG ingestion (from which all the DD code derives) is disappearing(**) in that project. For external sources, it is now XMLTV full steam ahead (well, there is still embedded EIT, but that is a different thing to tackle). There are clearly many XMLTV format guide sources available (not all of them even US centric, as you are aware).



(*) That is a guess, I can't know exactly the entire thought process, I simply thought re-posting here was a good place where it might get noticed (and in a year or two or three when some future MythTV versions is released and someone upgrades and then complains here about MythTV no longer working with Schedules Direct one can point to this notice).

(**) I have looked at that code. I have even provided some patches to the larger EPG code area. Due to the history of the MythTV project, and the work over the years for other sources, the DD code adds some complexity to some code paths that would require some hoop jumping to add some features that are desirable(***). And now that (arguably) the SDJSON XMLTV grabbers produce roughly equivalent guide quality(****), and someone (David) seems to be planning to work in that code area, the cost to continue it has apparently reached its terminus.

(***) I once had (what I thought would be) a useful approach for a completely different issue, but the existing (legacy) DD code, which had its feet in certain code paths, made it not viable (well, without a bunch of special case checks, which would never be accepted as a patch). Once the DD code is removed I might be able to go back and see if that stash can be re-based (if the fix is even still useful, which it might not be given the time that has elapsed). David alluded to his own set of enhancements in his posting. As with much else on that project, the payoff may not happen for quite some time ("free time" to work on that project seems to ebb and flow), but the direction has been stated.

(****) I would claim better quality for most use cases, but since I am clearly biased, I will let others make that evaluation.
gtb
 
Posts: 82
Joined: Thu Oct 02, 2014 2:07 pm


Return to Support

Who is online

Users browsing this forum: No registered users and 10 guests

cron