[mythtv-users] An companion script for jamu.py

[Epinephrine Junky]

> If you are using 0.24, there is no scan.  Batch metadata processing
> triggers when a scan is manually performed (if configured as such), or
when
> manually run from the MythVideo menu. 

You lost me there.  "there is no scan... processing triggers when a scan
is performed"
How should MythVideo know when new files have appeared on the drive?

> Once started, it will run until
> complete, or until the user exits MythVideo.  Each new video will be
> processed exactly once.  Once processed, the video is tagged as
processed
> in the database and will not run through batch processing again. 
Metadata
> downloading can be repeated manually per video through the MythVideo
menu,
> or with the 'w' hotkey.  This is designed intentionally to keep down
> unnecessary load on the metadata providers.

My goal is not to increase load on the metadata providers. I want to
understand how to set Myth to periodically check the drive for new/moved
video files.  If all the files showed up automatically and I just had to
hit 'w' to grab the metadata that would be acceptable but the files aren't
showing up.

> If you are using 0.25

I'm using 0.24.1

>> If I "scan for changes" from the frontend once then around 20% of my
>> previously available videos disappear entirely from my mythvideo
> listings
>> (they are still on the drive though) and then I "scan for changes"
again
>> and most, but not all, of the time it finds all videos including the
new
>> ones.
> 
> When you scan for changes, MythVideo pulls a list of files from each
> folder defined in the Videos storage group in each accessible backend. 
Any
> new files without a matching path in the database get hashed.  Any
previous
> files that are now found missing are compared to these new hashes to
detect
> moved files.  Remaining new files are added to the database, and
remaining
> missing files are removed.  The only reason I can think of that content
> would be removed one scan and added back the next would be if your video
> content were accessed over some unstable NFS or SMB mount.

Reasonable guess but the files are on a local 1.5TB drive on the combined
BE/FE. Any other ideas?  Should I try to get a log when this happens?
 
> If you have content available through Miro, check out MiroBridge (
> http://www.mythtv.org/wiki/MiroBridge ).  It will automatically
synchronize
> the content subscribed to in Miro with either recordings or MythVideo,
user
> selected.  Individual services such as this are better handled through
> individual applications such as MiroBridge, or through individual
grabber
> scripts for MythNetvision, that can be tuned to handle the quirks of
each
> service.

The fundamental flaw with mirobridge is that I have to shut down Miro for
it to work.  It's not clear to me how to use mirobridge effectively.  Start
Miro, download content, close Miro, watch content?  What if something else
is downloading in Miro when I want to watch the content that already
finished?  To say nothing of the hassle of getting LIRC to control Miro
start&exit.  My system has no keyboard or mouse so I start Miro in the
background when the system starts and only shut it down if I'm debugging
something.

I also haven't been able to use MythNetvision because it seems to rely on
mouse interaction.  By the time I grab my laptop to vnc into the box to
control MythNetvision I might as well watch the content on my laptop.  I
much prefer the MythTV UI that is blissfully keyboard free (and doesn't
heat up my lap).  :)