unable to search japanese encoded filename

[Newk]

i'm unable to download files with filenames that are encoded by what i think is Shift-JIS character encoding, wich is Japanese.

The an example of files i'm trying to download are here: slsk://tommyFFF@tommyfff/j:/mp3s/japan/sharpnel sound/# dj sharpnel/02 - [srpc-0002] high speed music team sharpnel - double dutch/

note that only daelstorm, SeeSchloss and Cravings are granted to access this users shared files.

[lbponey]

If I understand correctly, the problem is that Shift-JIS is not in the list of available encodings in museeq.

So the fix would be to have the complete list of iconv supported encodings in museeq's codeccombo (ie the combobox available for example in shares browser).

'iconv -l' is a command that returns this list... maybe this would be used in museeq's codeccombo constructor.

Don't know about other clients though.

Can you try the following: edit your museekd config file (usually ~/.museekd/config.xml, SAVE IT before in case you mess it) and add "<key id="tommyFFF">SHIFT-JIS</key>" inside the <domain id="encoding.users"></domain>. It should look something like this:

<?xml version="1.0" encoding="UTF-8"?>
<museekd>
  ...some <domain></domain>...
  <domain id="encoding.users">
    <key id="tommyFFF">SHIFT-JIS</key>
    ... maybe some other <key> elements
  </domain>
  ...some <domain></domain>...
</museekd>

Once you've done that, (re)start museekd, browse tommyFFF shares and retry to download the files. And of course tell me if it works.

[SeeSchloss]

Cravings pointed me to this ticket while I was thinking about adding support for korean and chinese encodings, so, I tested all that, and added the following charsets : Big5 (traditional chinese), EUC-CN (simplified chinese), EUC-KR (korean), EUC-JP (japanese but mostly used by unix systems) and Shift-JIS (japanese, used under windows).

They probably need to be added to murmur and mucous too ? But not this night :)

I don't think adding all the encodings iconv supports is such a good idea... the list would be huge, and most of them probably aren't even used at all in our context (like UTF-16 and UTF-32, which ARE included in the list and which won't ever be encountered by museeq).

Other than that, a related issue is the ability to search files using such encodings. Right now searches are made using the default network encoding, but it means that you cannot search for files in another encoding, which can be a problem. Maybe adding an encoding combobox in the search tab would be a good idea.

[lbponey]

Changing the title... Don't know exactly if it's possible with the current soulseek protocol.