Charset Detector screwups

[Jeff Mitchell]

Ian Monroe wrote:
> On Wed, Oct 28, 2009 at 12:08 AM, Peter Zhou <peterzhoulei at gmail.com> wrote:
>> I don't think the id3 version has anything to do with the tag encodings.
>> Many non-UTF-8 tags are using id3v2
> 
> I agree with this point. The spec of id3v1 just supports western, but
> that didn't stop folks from sticking whatever the hell they wanted
> into it. Ditto for id3v2.

Well, right.

But you can be certain that in ID3v1 UTF-8 isn't valid, so you don't
have to worry about the charset detector screwing up already-valid
UTF-8, unlike with ID3v2.4.

Honestly, this should be able to be configured off. There's no reason
why people that *do* have valid, UTF-8 ID3v2.4 tags should ever have
their tags fubared instead of read in (and with Picard and such, it's
getting a lot easier to have valid tags).

--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/20091029/878482b9/attachment.sig