AF1 wrote:
> Are you already using the lastest version of DPL3, 3.6.9 or 3.6.10
> (which should be available later tonight)?
> I'm asking because you said that new album*s* (plural) = 150 tracks were
> added. Your SQLite statement (like the built-in dynamic playlists)
> should only add 1 album at a time (unless no of tracks in album < min.
> number of unplayed tracks as explained).
> After updating to the latest DPL3 version you could clear your client
> playlist and restart your dynamic playlist. In my tests with the
> built-in album playlists no single tracks were added.
I definitely am using 3.6.10 now and the behavior I saw last night
seemed exactly right. I didn't see it add any odd tracks and what's out
there right now waiting for me to go to bed tonight is the remains of a
large single album. So it looks like it's back to working the way it did
before, which is also perfect for what I need. Thank you!
AF1 wrote:
>
> Switching from the old DPL to DPL3 is no reason to nuke your LMS
> installation or delete all of your databse tables. Apart from a
> temporary database table (to prevent repeating tracks already played in
> a session) DPL3 has no database tables.
I can no longer remember the exact details, but when the new DPL3 came
along I was getting the impression that already-played albums were
playing again. Pre-retirement, my primary job function was in a billing
audit center writing code to catch weird little exceptions. So I looked
into the old DPL tables as well as the regular LMS ones and discovered
discrepancies between last played dates and whether the track had been
played at all. I have no idea how it happened, but I figured that making
my query check both last played and the play count would ultimately get
me where I want to go.
AF1 wrote:
>
> Depending on the kind of the inconsistencies in your database, dynamic
> playlists may show different results for you than for users without
> these inconsistencies. That's not a catastrophe but not good either. A
> rescan (clear) can wipe/clear the library.db and repopulate it with the
> correct current data. The most important values in the -persist.db- are
> ratings, play counts, date last played and date added. Those are not
> affected by a rescan. Maybe get rid of the inconsistencies there, do a
> restart and then a rescan (clear). As always, make a copy of library.db
> and persist.db before you start working on them. Good luck.
For now, I'm pretty happy with how it's working. At some point in the
distant future I'll finally run out of stuff to play, at which point
I'll finally do a full rescan, which will then pick up all the stuff
I've added since the last rescan.
To give you an idea of how desperately nerdy I am about this, I get a
report emailed to me every morning at 8:30:
Code:
--------------------
This script is: /mnt/WD8TBNAS03/Documents/ShellScripts/TrackStat-new.py
1448 Unplayed Albums
Unplayed Tracks: 8,090
Played Tracks: 95,712
Oldest and most recently played tracks:
2017-02-03 21:40:22
file:///mnt/WD6TBNAS01/Squeezebox%20Music/NotDTS/Hair/Hair%20Broadway%20Deluxe/D1-T01%20-%20Aquarius.flac
2022-09-25 05:02:48
file:///mnt/WD6TBNAS01/Squeezebox%20Music/NotDTS/Various%20Artists/The%20Complete%20Motown%20Singles%20Collection%20Volume%2006/D1-T06-This%20Old%20Heart%20Of%20Mine%20(Is%20Weak%20For%20You)%20(Alternate%20Mix).flac
ID|Album Title|Total Tracks|Unplayed Tracks
194|(What's the Story) Morning Glory? [DSD stereo]|12|10
2649|1 [2015 remix]|27|25
4335|1,039/Smoothed Out Slappy Hours|19|16
4161|10|16|11
6829|12 X 5 [2002 DSD remaster]|12|3
7481|12 x 12 Original Remixes|12|11
7465|12x12 Original Remixes|12|7
2653|1967-1970 [2010 reissue]|28|25
7851|1968-05-09 - Fillmore Soundboard [sector boundary errors]|10|9
8777|1969: Velvet Underground Live|19|15
6161|1970-12-22 - Disgusting, Hardly Music|8|7
5292|1971-1979 Demos|28|25
6226|1972-12-09 - In a Neutral Land [Revision 1]|14|13
6796|1972-12-09 - Orchestral Tommy Live at the Rainbow|26|20
6237|1973-06-29 - Vaulting Ambition|16|15
7369|1974-03-20 - Live at the Record Plant|10|8
--------------------
That last list goes on forever, but you get the idea. Every morning I
compare the current report to the prior to see how many albums and
tracks are left to go.
You get a pretty good idea of how out of whack the data is just from the
little bit above. You can see almost everything is showing as partly
played, which probably means that somehow or other some of the tracks
lost their play status. It's very, very unlikely that so many albums
were aborted before finishing, but I have no idea how it all got so
jumbled.
------------------------------------------------------------------------
atrocity's Profile: http://forums.slimdevices.com/member.php?userid=16009
View this thread: http://forums.slimdevices.com/showthread.php?t=115501
_______________________________________________
plugins mailing list
plugins@lists.slimdevices.com
http://lists.slimdevices.com/mailman/listinfo/plugins
