Kodi Community Forum

I don't know, I am not a core developer. Best would be to ask in the feature requests forum.

(2012-10-27, 17:00)olympia Wrote: [ -> ]

(2012-10-27, 14:47)`Black Wrote: [ -> ]Why is album cover scraping from fanart.tv etc. preferred over local folder.jpg? Would be nice if that could be fixed.

How this comes down to the scraper?

Was this resolved? Just starting to scan my albums with the Universal Album Scraper for the first time, was worried it might start downloading different album cover images or even replace my folder.jpg files? I have the cover images embedded in to the mp3 files and also have a hi-res folder.jpg in each albums folder.

Thanks

---

Have selected query info for all albums. The scraper starts scanning and goes through all the albums A-Z. However some albums still show no information. Could just be that there is no information for that album. But on some albums if I go into album information on them and select reload then it does show an album destription etc, so why didnt the query all albums bring that info down in the first place?

Many thanks

First of all, thanks for a great scraper! It works flawlessly for most of my albums, but I'm having problems with some albums. For one example see below for the intresting parts from the debug log:

03:13:21 T:139994579154688 DEBUG: FindAlbum: Searching for 'Bembeya Jazz National - Bembeya' using Universal Album Scraper scraper (path: '/home/<user>/.xbmc/addons/metadata.album.universal', content: 'albums', version: '1.7.2')
03:13:21 T:139994579154688 DEBUG: scraper: CreateAlbumSearchUrl returned <url>http://search.musicbrainz.org/ws/2/release/?fmt=xml&query=release:Bembeya%20AND%20artist:Bembeya%20Jazz%20National</url>
03:13:21 T:139994579154688 DEBUG: CurlFile::Open(0x7f53081bb750) http://search.musicbrainz.org/ws/2/relea...20National
03:13:22 T:139994579154688 DEBUG: scraper: GetAlbumSearchResults returned <results></results>

When I look at the url beeing used by "CurlFile::Open" it looks like I should be finding one hit but it doesn't, does anyone know what could be the problem?

I looked into the problem and the issue was that the release was lacking the language tag on musicbrainz, since we dont use what's in that tag it shouldn't be any problem to make it optional. I tested the following and it worked for me:

changed:
&lt;language&gt;[^&lt;]*&lt;/language&gt;&lt;script&gt;[^&lt;]*&lt;/script&gt
To:
(?:&lt;language&gt;[^&lt;]*&lt;/language&gt*(?:&lt;script&gt;[^&lt;]*&lt;/script&gt*

For the four regex:es inside the GetAlbumSearchResults in albumuniversal.xml

Is it possible to get this included in the scraper so I dont have to patch it manually? If it is, patch on: http://www.xbmclogs.com/show.php?id=6734

Right, make sense, good catch. Just pushed to Frodo repo. Cheers!

Hi,

I ripped the Ben Howard: Empty Kingdom CD yesterday.
I updated my library, and the album is under the "unkwnown" artist!
I checked and tried to manualy updated informations.
The album is scrapped, with all titles, but there is no "artist" field.

Is it a scrapper bug, or maybe audiodb or another lacks information?
How can I debug it?
Thanks for the great work ^^

We need a dubog log to see what the issue is. Odds are high it is not the scraper.

ok, I'll post it asap

I uploaded the lgo file here

The scrapper finds the album artist.
I don't undestand where is the problem.

What file types do you have? MP3? FLAC? Are you sure all involved files are tagged correctly? The artist name XBMC is reading out from the tags and not using the one scraped by the scraper.

Look at line number 773-775 in your log, there is no artist read out from the tags there.

It's a FLAC album.
I checked tags with latest mp3Tag version, and they are correct.
So the scrapper is not the guilty one but my tags, right?

...or the tag reader of XBMC...
I would play with ID3 versions... Not sure what you set there.

Hi,

I've been using this scraper for a while now, great great great! I've a question though, it's not scraper related, so I hope people don't mind me posting here, but .... I really like having all the artwork just so, and part of my music collection includes singles (and in 2 cases free CDs from newspapers). Now, I know theaudiodb.com does not support singles, so if I were to want to add public (online) artwork, are there any alternatives?

e.g.

Singles:
McFly - All About You / You've Got a Friend (it's the wife's, I swear! )
http://musicbrainz.org/release-group/1a4...38/details

Richard Ashcroft - A Song for the Lovers
http://musicbrainz.org/release-group/52a...be/details

Newspaper freebie:
Jimi Hendrix - Albert Hall from the Sunday Times
http://musicbrainz.org/release-group/ed5...f5/details

Question 2, for the less than well known artists am I right to assume a local metadata is the only option?

e.g.
Paul Finley, is not on MusicBrainz

Any help would be appreciated.

@n1md4
1.
if i'm correct you can artwork to fanart.tv for singles too. If it's not listed on the artist page use the release-group ID to add it.

2.
possibly or add the less well known artists to musicbrainz/theaudiodb and add their information.
adding to musicbrainz allows you to upload artwork to fanart.tv (where a large portion of the artwork from theaudiodb comes from).

This is a great scraper but one main issue I have is that it skips non-english text for example Pyotr Ilyich Tchaikovsky. In MusicBrainz it is the top result when searching for Tchaikovsky, (there is another english entry but it has no detail or links at all). Would it be possible for the scraper to look against the Sort Name which is always in English and, as I'm assuming it cannot display the russian characters at least display the bracketed text e.g. [Russian romantic composer] from the name field so it can be correctly linked in xbmc? Rainbow (South Korean band) is another example where it won't show in the manual refresh list.