unicode

Help
kc8tbe
2004-11-06
2013-05-28
  • kc8tbe

    kc8tbe - 2004-11-06

    "Error: Your environment does not support unicode file names"

    This occurs just about whenever I try to compress recursively (e.g. 7z a -r file.7z directory). In my case, the filesystem in question is xfs. How can I get around this?

     
    • my space

      my space - 2004-11-06

      > "Error: Your environment does not support unicode file names"
      > This occurs just about whenever I try to compress recursively (e.g. 7z a -r file.7z directory).
      > In my case, the filesystem in question is xfs. How can I get around this?

      Your environment variables are not well defined to support the display
      of a name of a file or directory in your directory.
      This name is not ASCII name but a native name encoded
      for example in latin1 or en_US.utf8 ...

      Verify LC_* and LANG.
      example :
      env |grep LC_
      echo $LANG

      if none are defined, try to define only LANG with
      a better value (ls /usr/lib/locale to have the list of possible values)

      By the way, when you ask for help, give a lot more information :
      - processor : x86/sparc/powerpc ...
      - OS : linux, openbsd, cygwin, freebsd, solaris, MacOS X ...
      - language choosen to display filename : english, french, chinese ...
      - On a Unix-like system : value of "LANG" and "LC_*"

       
    • kc8tbe

      kc8tbe - 2004-11-07

      Sorry.
      Processor: amd64
      OS: Gentoo Linux

      env|grep LC_ is empty, as is echo $LANG.
      /usr/lib/locale is a directory.
      the command "locale" produces:
      LANG=
      LC_CTYPE="POSIX"
      LC_NUMERIC="POSIX"
      LC_TIME="POSIX"
      LC_COLLATE="POSIX"
      LC_MONETARY="POSIX"
      LC_MESSAGES="POSIX"
      LC_PAPER="POSIX"
      LC_NAME="POSIX"
      LC_ADDRESS="POSIX"
      LC_TELEPHONE="POSIX"
      LC_MEASUREMENT="POSIX"
      LC_IDENTIFICATION="POSIX"
      LC_ALL=

      Bear in mind, I'm not about to go converting my filesystem to unicode just so I can use p7zip :-). I was hoping for a way around this unicode reliance. p7zip actually deals with most files gracefully; its Microsoft Internet Explorer temporary files (long ugly things with lots of weird characters in them) that seem to mess it up.

       
      • my space

        my space - 2004-11-07

        7za (that is a port of 7-zip from Windows) stores the filenames in UTF16 encoding. So 7za must convert filenames :(

        To avoid unicode reliance, you can :
        - use the linux tar
        tar cf archive.tar directory
        - then 7za
        7za a archive.tar.7z archive.tar

         
    • kc8tbe

      kc8tbe - 2004-11-08

      Yeah. I've successfully filtered through tar, but

      tar -cf archive.tar directory && 7za a archive.tar.7z archive.tar

      Requires lots of scratch space, and

      tar -c directory | 7z a -si archive.tar.7z

      lacks a progress indicator. Oh well. I'm just happy it works. Thanks for all your help!

       

Log in to post a comment.