Hi,

Upgrading to 2.1.1 suddenly made some songs with Japanese tags appear as question marks and some songs with a Japanese file name not appear at all in the collection.

I'm pretty sure the tags are properly encoded but as a test I tried to rewrite some using easy tag and the UTF-8 setting wherever possible. No success.
I'm using the en_US.utf8 locale and I'm confident the file names are UTF8 encoded.
The weird thing about this is that it affects only some songs, even among the same album so that leads me to think amarok is the culprit rather my files (usually when I'm renaming/tagging an entire album I use the same setting for all songs)
So I've got a mix of albums only half-imported with tags that show up properly, full albums with tags showing up as question marks and full albums working flawlessly.
I've also noticed that in some cases the "edit metadata" window the file path at the bottom shows a messed up file name.
FIY the files work perfectly in audacious as well as in songbird.
I tried to delete the .kde4/share/apps/amarok folder and reimport the collection: same result.

Downgrading to 2.1 fixes both problems.

Before filling a bug report I thought I'd post here to see if other people had the same issues or knew a way to fix the problem.

I'm using gentoo, KDE 4.2.4, files are mp3s (didn't test with FLAC, my bad) and the engine is Xine.

Hello,

I have exactly the same problem on Kubuntu 9.04. I indeed think this is a serious bug. Unfortunately even after downgrading to 2.1.0 my collection is still screwed. After examining with id3tool is appears that Amarok changed the tags to question marks within the files.

I already reported the bug: https://bugs.launchpad.net/ubuntu/+sour ... bug/395956

Hi,

Same here :(:(

Ubuntu 9.04
Unsupported updates (Jaunty backports) checked
Software source for KDE stuff: http://ppa.launchpad.net/kubuntu-ppa/backports/ubuntu

Checking the unsupported updates (Jaunty backports) caused Amarok to upgrade in version 2.1.1.
After the upgrade, more than half of the non latin strings are replaced with question marks. I cannot understand the pattern, the change seems random. All files in the collection have been processed with EasyTAG before importing and they tags were UTF8. The question marks show in the collection view and in the  "Edit track details" context menu shows question marks too.

In my case this only an issue for the amarok database and not the ID3 tags themselves. I examined the tags with EasyTAG and they are perfectly OK.
But I opened the amarok database with mysql and all the non latin characters are indeed replaced by question marks in the database.

Rescanning the collection did not help.
Using the perl MP3::Mplib to set the UTF8 header on the tag as suggested by the amarok FAQ, had no effect too.

I am switching back to the launchpad ppa maintained package (v 2.1) and probably never tick the unsupported updates again

I managed to solve the problem with the question marks by reverting back to 2.1 from the launchpad kubuntu ppa (http://ppa.launchpad.net/kubuntu-ppa/backports) and rescanning the collection.

But I created another problem: I found at least one track that was in my playlist during rescanning now is wrongly reported under a different album of a different artist. Maybe this is a bug and collection should never be rescanned with tracks in the playlist.

After repetitive scans the track now shows under the wrong artist/album. This is not as worrying as the fact that I am not sure how many other errors there could be.

I've music tagged with any sort of japanese, cyrillic, german chars. All tagged with EasyTag on a UTF system. Never got any question marks or placeholders in my tags. On the other hand (I'm no friend of *ubuntu btw.) you're using backports, running into trouble and put the blame on Amarok. And yes, I know, that 2.0.2 is broken.
I've a homebrew A2 (2.2svn) running and can't reproduce most of the errors.

m0nk

@Dieter
As a Gentoo I assume the package is compiled from source the same way that version 2.1 is on my system so it makes me think there really is a problem somewhere with 2.1.1.
Besides the 2.1 ebuild and the 2.1.1 one don't differ:
http://sources.gentoo.org/viewcvs.py/ge ... iew=markup
http://sources.gentoo.org/viewcvs.py/ge ... iew=markup

Since I'm not familiar with Amarok's source I can't take part in the hunt for the bug but looking at the changes summary that states:
# Fixed non-UTF8 MPEG tag wrongly shown.
makes me think there might be a possibility that some work on UTF8 tag handling introduced some kind of regression somewhere.
Just my 2 cents, I really don't have a clue.

btw, just like you I'm using Easytag with UTF all other the place (system locale, easytag setting etc.).


It seems there are quite a few Ubuntu users with this problem but since I'm not using that distro I don't think it's a distro specific problem.

Sorry julien, my last post was addressed to ich.
There could be problems with non-latin UTF-8 chars in 2.1.1.
But it should be solved in 2.2.

m0nk