oggflac 1.1.1 crashes slimserver

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Stephen Rush

    oggflac 1.1.1 crashes slimserver

    I have tried this with the official 5.3.1 release as well as the 10-15
    nightly.
    I have encoded using new flac 1.1.1 to Ogg FLAC to try it out.
    I have also downloaded the latest flac.exe (1.1.1) and oggdec.exe (1.1.2)
    and I still get the same errors.
    After adding the .ogg files encoded with flac 1.1.1 and rescan, the server
    eventually displays 'page not found' message. The following message is
    displayed in the Debug window:
    SlimServer has started!
    Wrong vorbis header type, giving up. at
    /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 197.
    Use of uninitialized value in seek at /PerlApp/Ogg/Vorbis/Header/PurePerl.pm
    line 224.
    No comment header? at /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 295.
    Can't use an undefined value as an ARRAY reference at
    /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 69.

    My initial thought was to maybe tweak the convert.conf for ogg>wav to use
    flac instead of oggdec
    Ideas?

  • kdf
    NOT a Slim Devices Employee
    • Apr 2005
    • 9493

    #2
    oggflac 1.1.1 crashes slimserver

    Quoting Stephen Rush <stephen_rush (AT) hotmail (DOT) com>:

    > I have tried this with the official 5.3.1 release as well as the 10-15
    > nightly.
    > I have encoded using new flac 1.1.1 to Ogg FLAC to try it out.
    > I have also downloaded the latest flac.exe (1.1.1) and oggdec.exe (1.1.2)
    > and I still get the same errors.
    > After adding the .ogg files encoded with flac 1.1.1 and rescan, the server
    > eventually displays 'page not found' message. The following message is
    > displayed in the Debug window:
    > SlimServer has started!
    > Wrong vorbis header type, giving up. at
    > /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 197.
    > Use of uninitialized value in seek at /PerlApp/Ogg/Vorbis/Header/PurePerl.pm
    > line 224.
    > No comment header? at /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 295.
    > Can't use an undefined value as an ARRAY reference at
    > /PerlApp/Ogg/Vorbis/Header/PurePerl.pm line 69.
    >
    > My initial thought was to maybe tweak the convert.conf for ogg>wav to use
    > flac instead of oggdec
    > Ideas?


    the server will think that any .ogg file is an Ogg file, not an OggFLAC file.
    Likely the headers are arranged differently, and the current CPAN module to
    pars Ogg headers clealry doesn't support OggFLAC.

    -kdf

    Comment

    • Dan Sully

      #3
      oggflac 1.1.1 crashes slimserver

      * kdf <slim-mail (AT) deane-freeman (DOT) com> shaped the electrons to say...

      >> My initial thought was to maybe tweak the convert.conf for ogg>wav to use
      >> flac instead of oggdec
      >> Ideas?

      >
      >the server will think that any .ogg file is an Ogg file, not an OggFLAC file.
      >Likely the headers are arranged differently, and the current CPAN module to
      >pars Ogg headers clealry doesn't support OggFLAC.


      That is correct - OggFLAC is rarely used, and not even promoted by the FLAC author.

      If you haven't gone to far already, try to make them plain FLAC files with Vorbis tags.

      Thanks.

      -D
      --
      "Hey, careful, man, there's a beverage here!"

      Comment

      • michael

        #4
        container formats (was: oggflac 1.1.1 crashes slimserver)

        kdf <slim-mail (AT) deane-freeman (DOT) com> writes:
        > the server will think that any .ogg file is an Ogg file, not an OggFLAC file.


        As an enhancement idea, it should be noted that ogg is a container
        format, and even if ogg-flac never catches on, there may be other
        things in an ogg container than just vorbis streams. For that matter,
        quicktime is also a container format, and could one day present
        similar issues. The parsing of flac files with embedded cuesheets
        contains an ugly hack around what is basically the same problem. We
        should really think about moving away from the flawed assumption that
        a file name extension uniquely identifies the files content type.

        Thoughts?

        -michael

        Comment

        Working...