[Milhouse]

Ah yes, not specifying a media class basically means to cache everything (ie. all) but not songs. I'll give some thought to avoiding the error...

[Maetrik]

Is it really an error? It seems like everthing was cached correctly.

[Milhouse]

Only in the sense that it's being reported to the user, and I can/should suppress it.

[Maetrik]

Great, thank you very much for this script, it is absolutely perfect on slower devices like ATV2 or RaspberryPi. Hope that Gotham will bring some improvements that this is not necessary anymore.

You remember me blaming XBMC a few days ago for not exporting all episode thumbs? Since i ran your script everything exported flawlessly

[Milhouse]

Yeah, export is only going to export the contents of the texture cache which may in the case of low powered devices contain resized and low quality images (compared to the originals), and also as you found if you haven't yet browsed the entire media library some artwork may be missing completely.

Beware exporting... use a media manager instead!

[Maetrik]

I never used a media manager and on OS X there are not that much possibilites. Media Elch is not working with MySQL anymore, so i have to use XBMCs export system. To poor that everything was close to perfect with Eden and Frodo made it that 'challenging'.

I just export on my MacBook, so hopefully the images are not of low quality. I never heard of that before you told me.

[Milhouse]

Version 0.9.0

• Add: Summarise orphan results (r, R)
  
• Add: Notify when orphan files detected during prune (p, P)
  
• Add: Delete DDS files (when present) while pruning corresponding artwork
  
• Add: Show current version in update message
  
• Fix: Suppress "no result" error when processing sets if Movie Sets haven't been defined in the media library
  
• Fix: Handle github.com errors more gracefully
  

[CaptainKen]

Thanks to CaptainKen for helping debug his problem, which is being caused by those backslashes. It turns out that XBMC is automatically converting the erroneous backslashes into forward slashes when creating entries in Textures13.db, so reconciling the media library paths (with backslashes) against the texture cache (without backslahes) was problematic...

The script will now inspect the path and replace the slashes whenever it can't find a match, so the overhead should be fairly minimal.

Version 0.8.9

• Fix: Swap back/forward slashes on JSON artwork paths whenever a slash is used incorrectly (see here for details)
  

I sounds like you came up with a fix so I can continue to use MSAA after all. I'll just need to run "texturecache.py c sets" after running MSAA?

[Milhouse]

I sounds like you came up with a fix so I can continue to use MSAA after all. I'll just need to run "texturecache.py c sets" after running MSAA?

Yes, I added a workaround so that you can continue to use MSAA, but other scripts/addons may trip over the same flaw so really the problem needs to be fixed in MSAA.

[CaptainKen]

I sounds like you came up with a fix so I can continue to use MSAA after all. I'll just need to run "texturecache.py c sets" after running MSAA?

Yes, I added a workaround so that you can continue to use MSAA, but other scripts/addons may trip over the same flaw so really the problem needs to be fixed in MSAA.

Awesome to hear this. Thank you again for your contribution, dedication and patience in resolving this issue!

Enjoy your weekend! If you were in the Phoenix area I'd have to buy you a few beers!

[nickr]

I don't know if I am just being incredibly thick here, but my situation is this:

1. I have two client machines which I upgraded from eden to frodo today. Fresh virgin installs, no lingering db or thumbnail files/directories. Now I have no fanart, posters etc.

2. Database is on mysql on a third machine.

3. If I press "i" on a movie and choose fanart, then it gets me the fanart.

4. I thought this software would get all the missing posters etc, but it doesn't seem to.

Code:

nick@lounge:~$ python texturecache.py C movies
                                          
Caching artwork: 0 items remaining of 5 (qs: 0, qm: 0), 0 errors, 0 threads active (00.79 downloads per second, ETA: 00:00:00)

Cache pre-load activity summary for "movies":

              |    fanart   |    poster   |    thumb    |    TOTAL    
--------------+-------------+-------------+-------------+-------------
Cached        |      2      |      2      |      1      |      5      
Deleted       |      2      |      2      |      1      |      5      
Duplicate     |      -      |      -      |      1      |      1      
Error         |      -      |      -      |      -      |      0      
Ignored       |      -      |      -      |     255     |     255    
Skipped       |      -      |      -      |      -      |      0      
Undefined     |      -      |      -      |      -      |      0      
======================================================================
TOTAL         |      4      |      4      |     258     |     266    
Download Time | 00:00:01.84 | 00:00:01.89 | 00:00:01.85 | 00:00:04.85

  Threads Used: 2
   Min/Avg/Max: 0.54 / 0.79 / 1.08

       Loading: 00:00:00.63
       Parsing: 00:00:00.02
     Comparing: 00:00:00.01
   Downloading: 00:00:05.00
TOTAL RUNTIME: 00:00:05.68

The MyVideos75 database on the MySQL server seems to have a lot of fanart references, but they don't appear on my screen.

[Milhouse]

The script will only pre-load artwork that is defined in your media library (ie. what is in your MySQL database).

To "see" your media library, run this command:

Code:

python texturecache.py jd movies

and verify that all your movies are there, and they all have sane URLs for artwork.

To be honest it doesn't look as though the Eden -> Frodo upgrade went very well.

A number of your thumbnails (which may be the bulk of your old Eden movies) have been ignored - this means their artwork urls have matched one of the "cache.ignore.types" property, which defaults to "^image://video, ^image://music". This suggests the 255 ignored thumbnails are using embedded artwork (extracted from the media files) which cannot be downloaded over HTTP (so cannot be pre-loaded by this script) - you need to browse the GUI for the image extraction process to be kicked into life (unfortunately).

If you enable the log - add "@logfile=tc.log @logfile.verbose=yes" on the command line - you should be able to see what urls have been ignored and why. If it's not clear upload the log somewhere and I'll take a look.

[RockDawg]

MilhouseVH - I finally decided to just nuke my whole database as there were lots of of deprecated references including tons of paths to an old server I migrated from a while back, I re-scanned everything and used your app to restore the watched status, play counts and last played dates. Worked pretty darned good. One thing I noticed though is that there were a hand full of problems during the restore. I think most were due to mismatched names. For example the movie name was Lucky # Slevin, but because of character restrictions in file names I was forced to name it Lucky Number Slevin. I'm assuming it read the movie title from the db and then tried to find the file and couldn't?

There were a couple others it errored on where I couldn't see any naming discrepancy on them. All told I think I had around 20 movies where your utility didn't restore the watched status. Not a big deal at all, but thought I'd let you know in case you were able to fix it.

Thanks again for this wonderful utility. It is a truly useful tool!

[Milhouse]

MilhouseVH - I finally decided to just nuke my whole database as there were lots of of deprecated references including tons of paths to an old server I migrated from a while back, I re-scanned everything and used your app to restore the watched status, play counts and last played dates. Worked pretty darned good. One thing I noticed though is that there were a hand full of problems during the restore. I think most were due to mismatched names. For example the movie name was Lucky # Slevin, but because of character restrictions in file names I was forced to name it Lucky Number Slevin. I'm assuming it read the movie title from the db and then tried to find the file and couldn't?

There were a couple others it errored on where I couldn't see any naming discrepancy on them. All told I think I had around 20 movies where your utility didn't restore the watched status. Not a big deal at all, but thought I'd let you know in case you were able to fix it.

Thanks again for this wonderful utility. It is a truly useful tool!

Yes, it's using the name of the movie (not the filename) to match when restoring watched status - also the year. So if the movie name or the year changes between backup and restore, that particular item will fail to restore.

You should be able to re-run the restore - it will only attempt to restore those items that failed previously, and will ignore those that were successful - if you add "@logfile=tc.log @log.verbose=yes" on the command line and upload the log I'll try and determine the cause of the failures, hopefully it's not a coding or logic error. I'll also need to know which items failed to restore.

[Milhouse]

@RockDawg - I've just uploaded a new version with improved logging when restoring watched status, so if you do want to try the restore again (and create a log file), please use this latest version.