Description: EncodingConverter is a plugin for Amarok that offers a conversion from MP3 ID3v1 tags with non-latin1 data to unicode with ID3v2.
Important: This script is in alpha status. As it makes changes to ID3 tag information please make a backup of relevant files. Furthermore only a limited number of languages is supported though more can be easily added. Please submit encodings and their general language or language family name (e.g. ISO 8859-1 = west european) to the author.
ID3v1 doesn't specify in which encoding the meta data is given. Many people believe that only latin1 (ISO 8859-1) should be specified, though a lot of music is tagged in the encoding used locally and a lot of music players will interpret ID3v1 meta data as being in latin1 or in a previously specified encoding set for all system wide files.
With mixing music of different languages from different sources, ID3v1 tags will be mixed with different encodings, so that a default encoding can't work for all files. ID3v2 does allow Unicode, so an easy solution is to just convert ID3v1 tags to ID3v2. This script will offer the user a method to select a supported encoding and will even guess the encoding as to try to offer the right encoding as the first match.
Usage:
Right clicking on the file and choosing "EncodingConverter" and "Convert to Unicode" will open a window that shows all relevant tag fields and an encoding chooser. The encoding guesser will try to find a best match for the given content but if the tag fields do show wrong content and weired characters the current encoding is wrong and another one needs to be chosen.
The fields actually to be converted can be chosen by ticking the checkboxes. By default only fields will be selected for conversion where no ID3v2 information exists, or the ID3v1 and ID3v2 entries are the same. On conversion the ID3v2 tag will be overwritten, the ID3v1 tag stays untouched.
See README for more information.Last changelog:
2007.11.26 (0.1): initial release, alpha
2007.12.10 (0.2): new version of ngram.py, includes language/encoding pairs even if Textcat not available updated documentation
There was a problem, that amarok doesn't represent changes in file tags until full collection rescan. I've written small patch, that solves problem.
Quote:
Hi,
I know about this problem, but as this shouldn't be found too often I am just to lazy to change this script, which for me isn't more then a quick hack (though I wish to keep the encoding guesser clean).
What you can do is try to copy the ID3V2 tag to ID3V1 and then use this script. This isn't the best solution as ID3V1 tags have less space so it only works for short tags. Try to make a backup of your mp3 though first.
I'm sorry that there isn't any more elegant solution available here. Well, unless you don't mind changing some stuff in the script, which I would warmly welcome.
It seems a nice utility but although it installs succefully it fails when I press run:
The script 'EncodingConverter' exited with error code: 1
Traceback (most recent call last):
File "/home/manos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 69, in <module>
id3v2.FrameFactory.instance().setDefaultTextEncoding(tagpy.StringType.UTF8)
AttributeError: 'id3v2_FrameFactory' object has no attribute 'setDefaultTextEncoding'
the same was happening with 0.1
If you need any information let me know..
I'm using ubuntu 7.10 (with kde libraries - not the whole kde installed)
You have to upgrade to the newest version of tagpy (or as it's named in debian: python-tagpy).
I'm running 0.93-1 here and the function that's missing in your version has been added just lately.
You can alternatively try to disable the one line
id3v2.FrameFactory.instance().setDefaultTextEncoding(tagpy.StringType.UTF8)
that's quite at the beginning of the whole script. This might render some conversions unuseful though.
yes that worked! very nice indeed. Although I don't understand why the converted tags doesn't show up correctly in amarok. But I dont this it has to do with your script. Probably a configuration option.
Nice work!
If you convert a tag in the background Amarok then doesn't know about it. I don't see a simple way of telling Amarok to reload the song into the playlist.
But if you open the Amarok "metadata" dialog where you can see the tag info you should be at least able to see the changed data immediately.
If not and you have removed the line as I proposed as an alternate (though bad) solution, then the tags haven't been converted properly.
This script could be made nicer, with some more simple options, but I hope it is useable for the beginning.
I have same error. I've tried to install python-tagpy 0.93 from dedian and ubuntu hardy repositories. Script runs, but I don't see configuration dialog or converting one.
I'v tried to comment this line in script (python-tagpy 0.91 from ubuntu gusty) with same result.
In console I get this:
kdecore (KAction): WARNING: KAction::insertKAccel( kaccel = 0x80ef60 ): KAccel o bject already contains an action name "play_pause"
QLayout "unnamed" added to QVBox "unnamed", which already has a layout
kdecore (KAction): WARNING: KAction::insertKAccel( kaccel = 0x80ef60 ): KAccel o bject already contains an action name "play_pause"
QLayout: Adding KToolBar/mainToolBar (child of QVBox/unnamed) to layout for Play listWindow/PlaylistWindow
QObject::connect: Incompatible sender/receiver arguments
StarManager::ratingsColorsChanged() --> ContextBrowser::ratingOrScoreOrL abelsChanged(const QString&)
PS Kubuntu 7.10 amd64
The output you give doesn't seem to be connected to the script.
Please try to give more detail. For example you can check the debug output in /tmp/encodingconverter.debug if you have debugging turned on (default).
Furthermore you can see the script output in Amarok if you go to the script settings and right click on the script's entry.
Traceback (most recent call last):
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 528, in <module>
main( sys.argv )
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 524, in main
app = EncodingConverter(sys.argv)
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 269, in __init__
self.readSettings()
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 295, in readSettings
language_order=language_order)
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encoding.py", line 80, in __init__
language_order=language_order)
File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/ngram.py", line 135, in __init__
raise ValueError("no language files found")
ValueError: no language files found
[encodingconverter] Started.
[encodingconverter] reading settings
[encodingconverter] No config file found, using defaults.
[encodingconverter] config read settings: textcat_LM_path='/usr/share/libtextcat/LM', language_encoding_pref='', keep log file: True
You don't have Textcat installed, so the script tells you it can't find the files. Can you try to upgrade to the new version of the script? I asume you are still running the first version (0.1). The second version will not force you to install Textcat.
Once running, you can see a new entry in the context menu when clicking on a playlist entry.
If you wan't quicker support, please write me an email.
Below is the current table of language encodings. These encodings are supported by this script. To add an encoding for your language please add a comment.
Adding your language will mean: a) making its encoding show up in the selector, b) make textcat guess your language.
As 0 people already made a comment your feedback is appreciated. Does this script work for you?
Ratings & Comments
15 Comments
There was a problem, that amarok doesn't represent changes in file tags until full collection rescan. I've written small patch, that solves problem. Quote:
437a438,439
>
> os.system("touch '"+os.path.dirname(file)+"'");
448a451,452
> os.popen("dcop amarok collection scanCollectionChanges")
>
It should be applied to encodingconverter.py file. Sorry, if something is bad, but i'm newby in python and in patch making.
Hello, how can I convert ID3V2 non UTF-8 Tags to UTF-8?
Hi, I know about this problem, but as this shouldn't be found too often I am just to lazy to change this script, which for me isn't more then a quick hack (though I wish to keep the encoding guesser clean). What you can do is try to copy the ID3V2 tag to ID3V1 and then use this script. This isn't the best solution as ID3V1 tags have less space so it only works for short tags. Try to make a backup of your mp3 though first. I'm sorry that there isn't any more elegant solution available here. Well, unless you don't mind changing some stuff in the script, which I would warmly welcome.
Hello there! What `bout russian CP-1251? Tnx
You can find this under the entry "cyrillic (windows1251)". If you have it installed textcat should be able to guess this encoding.
It seems a nice utility but although it installs succefully it fails when I press run: The script 'EncodingConverter' exited with error code: 1 Traceback (most recent call last): File "/home/manos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 69, in <module> id3v2.FrameFactory.instance().setDefaultTextEncoding(tagpy.StringType.UTF8) AttributeError: 'id3v2_FrameFactory' object has no attribute 'setDefaultTextEncoding' the same was happening with 0.1 If you need any information let me know.. I'm using ubuntu 7.10 (with kde libraries - not the whole kde installed)
You have to upgrade to the newest version of tagpy (or as it's named in debian: python-tagpy). I'm running 0.93-1 here and the function that's missing in your version has been added just lately. You can alternatively try to disable the one line id3v2.FrameFactory.instance().setDefaultTextEncoding(tagpy.StringType.UTF8) that's quite at the beginning of the whole script. This might render some conversions unuseful though.
yes that worked! very nice indeed. Although I don't understand why the converted tags doesn't show up correctly in amarok. But I dont this it has to do with your script. Probably a configuration option. Nice work!
If you convert a tag in the background Amarok then doesn't know about it. I don't see a simple way of telling Amarok to reload the song into the playlist. But if you open the Amarok "metadata" dialog where you can see the tag info you should be at least able to see the changed data immediately. If not and you have removed the line as I proposed as an alternate (though bad) solution, then the tags haven't been converted properly. This script could be made nicer, with some more simple options, but I hope it is useable for the beginning.
I have same error. I've tried to install python-tagpy 0.93 from dedian and ubuntu hardy repositories. Script runs, but I don't see configuration dialog or converting one. I'v tried to comment this line in script (python-tagpy 0.91 from ubuntu gusty) with same result. In console I get this: kdecore (KAction): WARNING: KAction::insertKAccel( kaccel = 0x80ef60 ): KAccel o bject already contains an action name "play_pause" QLayout "unnamed" added to QVBox "unnamed", which already has a layout kdecore (KAction): WARNING: KAction::insertKAccel( kaccel = 0x80ef60 ): KAccel o bject already contains an action name "play_pause" QLayout: Adding KToolBar/mainToolBar (child of QVBox/unnamed) to layout for Play listWindow/PlaylistWindow QObject::connect: Incompatible sender/receiver arguments StarManager::ratingsColorsChanged() --> ContextBrowser::ratingOrScoreOrL abelsChanged(const QString&) PS Kubuntu 7.10 amd64
The output you give doesn't seem to be connected to the script. Please try to give more detail. For example you can check the debug output in /tmp/encodingconverter.debug if you have debugging turned on (default). Furthermore you can see the script output in Amarok if you go to the script settings and right click on the script's entry.
/tmp/encodingconverter.debug the script output
Traceback (most recent call last): File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 528, in <module> main( sys.argv ) File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 524, in main app = EncodingConverter(sys.argv) File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 269, in __init__ self.readSettings() File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encodingconverter.py", line 295, in readSettings language_order=language_order) File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/encoding.py", line 80, in __init__ language_order=language_order) File "/home/drchaos/.kde/share/apps/amarok/scripts/encodingconverter/ngram.py", line 135, in __init__ raise ValueError("no language files found") ValueError: no language files found [encodingconverter] Started. [encodingconverter] reading settings [encodingconverter] No config file found, using defaults. [encodingconverter] config read settings: textcat_LM_path='/usr/share/libtextcat/LM', language_encoding_pref='', keep log file: TrueYou don't have Textcat installed, so the script tells you it can't find the files. Can you try to upgrade to the new version of the script? I asume you are still running the first version (0.1). The second version will not force you to install Textcat. Once running, you can see a new entry in the context menu when clicking on a playlist entry. If you wan't quicker support, please write me an email.
Sorry, that should read: If you want quicker support, please write me an email.
Below is the current table of language encodings. These encodings are supported by this script. To add an encoding for your language please add a comment.
defaultEncodings = {'windows1250': ('slovak', ), 'windows1256': ('arabic', ), 'cp874': ('thai', ), 'gb2312': ('chinese', ), 'big5': ('chinese', ), 'gbk': ('chinese', ), 'euc_jp': ('japanese', ), 'shift_jis': ('japanese', ), 'euc_kr': ('korean', ), 'iso8859_1': ('english', 'german', 'french', 'africaans', 'spanish', 'danish', 'swahili', 'finnish', 'portuguese', 'norwegian', 'swedish', 'rumantsch', 'catalan', 'basque', 'latin'), 'iso8859_2': ('bosnian', 'polish'), 'iso8859_6': ('arabic', ), 'iso88597': ('greek', ), 'iso8859_8': ('hebrew', )} # 'tscii': ('tamil', ) # seems not to be supported in pythonAdding your language will mean: a) making its encoding show up in the selector, b) make textcat guess your language. As 0 people already made a comment your feedback is appreciated. Does this script work for you?