[mythtv] trying to understand mythfilldatabase
Michael T. Dean
mtdean at thirdcontact.com
Wed Feb 6 17:52:48 UTC 2008
On 02/06/2008 09:07 AM, Simon Kenyon wrote:
> why does the status say:
> Last mythfilldatabase run started on 2008-02-06 03:14 and ended on
> 2008-02-06 03:16. mythfilldatabase ran, but did not insert any new data
> into the Guide for 4 of sources. This can indicate a potential grabber
> There's guide data until 2008-02-12 06:00 (6 days).
> before and after the run and compares the two. this say if new data was
> added at the end, but says nothing about modified data.
> this can give the wrong impression.
Which is why it says, "This *can* indicate a *potential* grabber
failure," (emphasis added).
> in libs/libmythtv/programdata.cpp, ProgramData::handlePrograms()
> calculates the updated and unchanged values. could they be passed back
> to mythfilldatabase?
The theory is that if you're updating your data only one time per day
and your source (i.e. Schedules Direct or whomever) updates their data
once per day, you will always see data going farther into the future
than you did before the mythfilldatabase run. It's a very quick sanity
That being said, I would have absolutely no problems with removing the,
"This can indicate a potential grabber failure," warning as it tends to
confuse more people than it helps--especially since the warning should
only really apply when the user is first configuring
mythfilldatabase/XMLTV (or, for those rare cases when an XMLTV grabber
script is obsoleted/broken). In theory, the "did not insert any new
data" part of the message should be equally telling for an experienced
XMLTV user to notice updating XMLTV may be necessary.
I'd like to come up with a better wording for the "did not insert any
new data" part, but I can't really think of one... "Changed data but
did not insert new data..." "Did not insert data beyond the end of the
existing data..." I don't really know how to say it in a user-friendly way.
Oh, and Simon, if you're using EIT, I'd bet you've been bitten by the
EIT-inserted-program-information-for-2037 issue (or some other
ludicrously-far-in-the-future date) that's caused by bad data. That
(specific date) would cause every run of mythfilldatabase to give you
the indicated warning for another almost 30 years... And, if you're
using a combination of EIT and XMLTV, it's quite possible you may always
see the warning depending on how far into the future each goes.
More information about the mythtv-dev