id3 tag problems

[Jeff Mitchell]

Karl Vollmer wrote:
> May I suggest MPD as a good app to check, I've had the fewest problems
> with tag reading in it.
> 
> -Karl
> 
> On Mon, Dec 7, 2009 at 9:22 AM, Myriam Schweingruber
> <schweingruber at pharma-traduction.ch> wrote:
>> Hi all,
>>
>> since quite some time we get reports about id3 tag reading problems,
>> which seem to have not changed much in all the time we try to solve
>> that problem.
>>
>> Here are a few links to the various bug reports about that problem, as
>> well as some forum posts with maybe some insight for the tag
>> specialists amongst us. I skipped the forum posts in other languages
>> than English, but there are other in both German and French, and I
>> guess also in Chinese, Peter can certainly confirm:

The basic problem is that we put in this charset detector to try to
guess what encoding a file was in, and it often doesn't work and/or
screws things up even worse. In many of these cases, the people involved
already are using ID3v2.4 with UTF-8 and the charset detector just
causes corruption. This is especially a problem given that we've been
telling people to fix their tags to use the ten-year-old ID3v2.4 with
UTF-8 (and for good reason).

I put in a fix a bit ago (for the collection scanner, at least, although
I haven't for the other place the charset detector is used - I suggest
someone does this before 2.2.2) to not use it by default. It'll be in
2.2.2. That's the default for new installs; old installs will keep the
old behavior for consistency's sake and people can manually turn it off.

--Jeff

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 196 bytes
Desc: OpenPGP digital signature
Url : http://mail.kde.org/pipermail/amarok-devel/attachments/20091207/f1d99125/attachment.sig