<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    On 9.12.2014 20:33, Tristan Matthews wrote:<br>
    <blockquote
cite="mid:CAN8HRDk5dYqAtraPsZV2GtN2=22rUEhQinS9hVVayLfUNgt5mw@mail.gmail.com"
      type="cite">
      <div dir="ltr">On Tue, Dec 9, 2014 at 1:31 PM, Janne Hyvärinen <span
          dir="ltr">&lt;<a moz-do-not-send="true"
            href="mailto:cse@sci.fi" target="_blank">cse@sci.fi</a>&gt;</span>
        wrote:<br>
        <div class="gmail_extra">
          <div class="gmail_quote">
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex"><span
                class="">On 25.11.2014 12:14, Miroslav Lichvar wrote:<br>
                &gt; I think the case with non-zero partition order may
                need to be fixed<br>
                &gt; too. For example, with partition order of 1,
                predictor order of 16 and<br>
                &gt; blocksize of 4, the function would return true and
                blocksize-order in<br>
                &gt; the caller would still underflow.<br>
                &gt;<br>
                &gt; --- a/src/libFLAC/stream_decoder.c<br>
                &gt; +++ b/src/libFLAC/stream_decoder.c<br>
                &gt; @@ -2744,7 +2744,7 @@ FLAC__bool
                read_residual_partitioned_rice_(FLAC__StreamDecoder
                *decoder, unsigne<br>
                &gt;                  if(partition_samples &lt;
                predictor_order) {<br>
                &gt;                         
                send_error_to_client_(decoder,
                FLAC__STREAM_DECODER_ERROR_STATUS_LOST_SYNC);<br>
                &gt;                         
                decoder-&gt;protected_-&gt;state =
                FLAC__STREAM_DECODER_SEARCH_FOR_FRAME_SYNC;<br>
                &gt; -                       return true;<br>
                &gt; +                       return false;<br>
                &gt;                  }<br>
                &gt;          }<br>
                &gt;<br>
                &gt; Thoughts?<br>
                <br>
              </span>This patch breaks seeking in some perfectly valid
              files. So far I have<br>
              received one sample full CD image from a foobar2000 user
              where a track<br>
              is rendered inaccessible because of this. Re-encoding the
              file with FLAC<br>
              1.2.1 - 1.3.1 with identical settings doesn't remove the
              seeking problem.<br>
              Either this patch needs to go or it needs to be altered to
              not prevent<br>
              seek sync.<br>
            </blockquote>
            <div><br>
            </div>
            <div>Can you share samples?<br>
            </div>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
    It's a 470 MB copyrighted music album. I could but I don't think
    it's legal.<br>
  </body>
</html>