[mythtv-users] Schedules Direct TBA Listings

[Robert Kulagowski]

On Thu, Jul 17, 2014 at 11:53 AM, Gary Buhrmaster
<gary.buhrmaster at gmail.com> wrote:
> On Thu, Jul 17, 2014 at 4:32 PM, Robert Kulagowski <rkulagow at gmail.com> wrote:
> ...
>> Please contact me at rkulagow at gmail.com if you'd like to participate
>> in the beta.
>
> While it has been some time since I looked, and the
> beta was an interesting proof of concept, it had a fatal
> flaw (putting my DBA hat on) where it overloaded the
> mythconverg database for an application that should
> have been independent of MythTV itself.  Has the
> more recent betas moved to a standalone grabber
> database (perhaps using a standalone sqlite db if you
> did not want to require a full mysql stack)?  If not,
> what is the schedule for doing so?

The raw JSON for schedules and programs is downloaded and put into
their own tables in mythconverg. The JSON is then processed and the
appropriate fields in channels, videosource, etc are updated, then a
reschedule is sent to mythbackend.

The JSON contains more data than can be accommodated in the existing
schema, so it's stored in case someone wants to implement support for
the extra fields. (Programs have recommendations "If you like 'this',
you may also want to schedule 'foo', 'bar', 'baz'")

The grabber can be used in "grabber only" mode as a starting point for
other developers who just want to download the JSON data and process
it themselves; not everyone runs MythTV. I don't have plans for sqlite
at this point because it's pretty much a MythTV-specific proof of
concept.

I don't plan on doing much more client-side at this point because I'm
working on the next server API which will include support for images
and other metadata. You'll be able to query a particular programID and
get back images for that program; no more trying to do metadata
lookups based on guessing from the title / subtitle.