[mythtv-users] SchedulesDirect Changes

[f-myth-users at media.mit.edu]

    > Date: Thu, 18 Sep 2014 17:00:28 -0700
    > From: Jay Foster <jayf0ster at roadrunner.com>

    > It would be helpful to see support for older MythTV versions, or at 
    > least some documentation and recommendations about what will or will not 
    > work with specific older versions of MythTV (0.21-fixes, 0.22-fixes, 
    > 0.23, 0.24, 0.25, 0.26, etc.).  I don't think everyone is running the 
    > latest bleeding edge version of MythTV, or should be forced to do so 
    > just to get listings from SD.

This is why I asked about Robert Eden's comment (in the forum
announcement) that a service is also being developed to mimic the
DataDirect feed.  (I'm hesitant to migrate this thread to the -dev
list or ask him directly, since that feels a bit pushy, and I don't
even know if he's the developer who might be working on this or if
it's someone else, if anyone.  And asking about schedule for a
software project is always fraught.  :)  I suspect that there may be a
number of users who installed myth and treat it as an appliance---not
necessarily upgrading at every point (because that can always break
device support and have other bad effects), and who are probably not
on any of the lists.  We may see a number of them suddenly appearing
out of the woodwork when their listings data suddenly stops, and
forced upgrades after noticing lack of listings but before they run
out is going to be a fun time for them.

Perhaps at least SchedulesDirect should consider sending a note to all
subscribers warning them that this may happen.  Those who aren't on
the lists and don't happen to visit SD (e.g., to renew) in the next
month are going to be caught unawares.  That's why an XML DD shim,
if possible, would help a lot---they may be able to switch URLs
(f necessary) and continue to get listings as a stopgap.

[Plus, not all users of the DD XML are Myth.  They may not be able
to upgrade in time, or maybe ever.]

Meanwhile, since I have no idea if a DD XML shim will actually happen,
I'm going to continue to experiment with running the JSON grabber on a
separate host and then figure out if it will insert listings into an
old database on an old host.  One concern:  UTC vs local time.
Sufficiently old myths predate the conversion to UTC.  I don't know
if that's going to be an issue here, though presumably I can convert
locally if it is by patching the grabber.