Home of the Squeezebox™ & Transporter® network music players.
Page 2 of 2 FirstFirst 12
Results 11 to 19 of 19
  1. #11
    Junior Member
    Join Date
    Jan 2013
    Posts
    17
    Quote Originally Posted by mherger View Post
    > Now I found the root cause:
    > Source encoding = CP1251 (WINDOWS-1251)
    > Displayed in ISO-8859-15


    BTW: what's your Windows' language setting (see Settings/Information in
    LMS)?

    --

    Michael
    LMS is installed on a NAS ).
    Operating System: Linux -EN - UTF8 (see the very first attachment)

  2. #12
    Babelfish's Best Boy mherger's Avatar
    Join Date
    Apr 2005
    Location
    Switzerland
    Posts
    19,682

    Artist name & song title played by radiostations are not in Cyrillic

    > Perhaps including a call to Mozilla Universal Character Set Detector
    > would bring back info about which charset is being used.
    > http://search.cpan.org/~jgmyers/Enco...01/Detector.pm


    It does indeed. What makes me hesitate is the fact that we already use
    it, but only _after_ some of our own tests. And I would assume there was
    some good reason why we don't rely on EDD alone.

    --

    Michael

  3. #13
    Babelfish's Best Boy mherger's Avatar
    Join Date
    Apr 2005
    Location
    Switzerland
    Posts
    19,682
    Quote Originally Posted by tals View Post
    LMS is installed on a NAS ).
    Operating System: Linux -EN - UTF8 (see the very first attachment)
    Hmm... I guess you compiled that Perl yourself? It's using non-standard (for LMS) module versions eg. for SQLite. Would you have Encode::Detect::Detector? Or could you compile it?

    Code:
    perl -MEncode::Detect::Detector -e'print $Encode::Detect::Detector::VERSION'
    You could then apply the following patch to Slim::Utils::Unicode: https://gist.github.com/mherger/0a96...b3b50b749e0a32

    You can try that patch anyway. It would silently fail if aforementioned module was missing.
    Michael

    http://www.herger.net/slim-plugins - MusicArtistInfo, MusicInfoSCR

  4. #14
    Junior Member
    Join Date
    Jan 2013
    Posts
    17
    Quote Originally Posted by mherger View Post
    Hmm... I guess you compiled that Perl yourself? It's using non-standard (for LMS) module versions eg. for SQLite. Would you have Encode:etect:etector? Or could you compile it?

    Code:
    perl -MEncode::Detect::Detector -e'print $Encode::Detect::Detector::VERSION'
    You could then apply the following patch to Slim::Utils::Unicode: https://gist.github.com/mherger/0a96...b3b50b749e0a32

    You can try that patch anyway. It would silently fail if aforementioned module was missing.
    Well, yes. My platform is not a standard one where an out-of-the-box installation of LMS is possible, so I compiled the packages from tarball ).
    Code:
    perl -MEncode::Detect::Detector -e'print $Encode::Detect::Detector::VERSION'
    Code:
    1.00
    Thanks for the patch! Already applied.
    Let me test it and come back with a feedback.

    ---
    Sergey

  5. #15
    Junior Member
    Join Date
    Jan 2013
    Posts
    17
    Now it works as expected!)
    Thanks a lot!

    ---
    Sergey

  6. #16
    Babelfish's Best Boy mherger's Avatar
    Join Date
    Apr 2005
    Location
    Switzerland
    Posts
    19,682

    Artist name & song title played by radiostations are not in Cyrillic

    > Now it works as expected!)

    Cool! Please let me know if you see any negative side-effect. I'm not
    sure yet whether I want to include this by default or not, as I'd rather
    not want to break things for non-Cyrillic users...

    --

    Michael

  7. #17
    Junior Member
    Join Date
    Jan 2013
    Posts
    17
    Quote Originally Posted by mherger View Post
    > Now it works as expected!)

    Cool! Please let me know if you see any negative side-effect. I'm not
    sure yet whether I want to include this by default or not, as I'd rather
    not want to break things for non-Cyrillic users...

    --

    Michael
    Sure.
    Hope no side-effects will occur. However, I'll make a screenshot if any.

    --
    Sergey

  8. #18
    Junior Member
    Join Date
    Jan 2013
    Posts
    17
    Hi Michael!

    Yesterday I updated 'Encode :: Detect' to 1.01 and performed a smoke test playing different radio stations available from the list.
    Here are the results:
    1) Very seldom Cyrillic text displayed as mentioned in my first comment. However, as soon as a new song appears the text is displayed correctly. [See attachment]

    Name:  seldom.jpg
Views: 47
Size:  48.9 KB

    2) German/Czech/Slovakian/French etc. - OK
    3) Greek - OK
    4) Turkish - some characters displayed as <?>
    5) Chinese / Arabic and similar - displayed as []

    Hope it helps.

    ---
    Sergey

  9. #19
    Junior Member
    Join Date
    Mar 2017
    Posts
    8
    Quote Originally Posted by mherger View Post
    > Now it works as expected!)

    Cool! Please let me know if you see any negative side-effect. I'm not
    sure yet whether I want to include this by default or not, as I'd rather
    not want to break things for non-Cyrillic users...

    --

    Michael
    Hi Michael,

    I was happy to find your patch. It proved to be very helpful for making cp1251 encoded Cyrillic titles readable. Still the auto-detection is not 100% reliable and occasionally fails. In my case 10-20% of the song titles remain unreadable. The patch also somehow does not affect any cp1251 encoded in-stream station names. I wonder whether further improvement could be possible? Probably forcing specific encoding for a selected stream.

    Thanks,
    Sergey

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •