From john.maguire on April 28, 2010 07:58:16

Does taglib support this?

Labels: -Type-Defect -Priority-Medium Type-Enhancement Priority-Low

From 2sabio on April 28, 2010 08:52:56

"ID3v1 should in theory always contain ISO-8859-1 (Latin1) data. In practice it does
not. TagLib by default only supports ISO-8859-1 data in ID3v1 tags.

However by subclassing this class and reimplementing parse() and render() and setting
your reimplementation as the default with ID3v1::Tag::setStringHandler() you can
define how you would like these transformations to be done." http://developer.kde.org/~wheeler/taglib/api/classTagLib_1_1ID3v1_1_1StringHandler.html

From john.maguire on April 28, 2010 09:26:57

Sounds like you want to write a patch ;-)

From 2sabio on April 28, 2010 09:55:00

I just typed "taglib id3v1 encoding" in Google :)

And even though I'm a developer indeed I've never developed something for Linux (it's
java+web and a bit of Python what I work with).

From kadaimx on April 28, 2010 10:23:58

Yes, having a way to detect the correct encoding of the id3 v1 might be very very
useful, specially with those files that were tagged on systems (usually windows xp
and before) that do not used unicode.

Now, there was a script for amarok that did something similar detecting the correct
encoding and converting it.

The name is encodingconverter.

From 2sabio on April 29, 2010 02:38:20

As a second thought.
To avoid having an UI option for ID3v1 encoding you might consider implementing
encoding auto-detection.

In that case you might find these pages interesting: http://www.mozilla.org/projects/intl/detectorsrc.html http://www.mozilla.org/projects/intl/UniversalCharsetDetection.html

From john.maguire on May 10, 2010 06:32:05

Can someone upload some files with non Latin1 ID3v1 tags?

From 2sabio on May 10, 2010 06:42:04

This one has both ID3v1 and ID3v2 in cp1251 - http://drop.io/gng5kwf (the title should read "Выдыхай")

Let me know if you need more/other files

From john.maguire on May 11, 2010 04:41:38

So that file is very broken :-)

It has cp1251 in the ID3v1 tag which is wrong but I can fix. The cp1251 in the ID3v
2 tag though is just broken, there's no reason for that not to be UTF-8 and taglib,
for sensible reasons, picks the ID3v2 tag first.

From john.maguire on May 11, 2010 05:04:07

This issue was closed by revision r893 .

Status: Fixed

From 2sabio on May 11, 2010 05:20:44

Yes, ID3v2 in this case is out of question. :)
I've chosen this file only because it was the smallest of those I have at hand.

Glad to hear the problem is solved. :)

From Taras.Puchko on January 18, 2011 07:20:14

Clementine 0.6 cannot detect cp1251 in ID3v2.3.0 of the following file (and many others): http://www.sofiarotaru.com/download_music/06-track6.mp3 It would be great if where was an option in a configuration file that would force a particular encoding for such cases.

From davidsansome on January 18, 2011 10:51:31

Taras: can you open a new issue for that so it doesn't get lost? Test cases like that are really useful.