[Imgcif-l] Possible bug in cbf_byte_offset.c

[Herbert J. Bernstein]

Dear Nick,

   You are absolutely right.  I am also having trouble working up a test
case.  It needs an image with a very large negative delta.   I'll try
to force something.

   In the meantime, I have updated the sourceforge cbflib svn with the
fix in CBFlib_bleeding_edge.  After I have a test case, I'll propagate
the change more widely.

   I am copying this to the imgCIF list to warn others in the meantime.

   Thank you very much.  Well done.

   Regards,
     Herbert




At 11:06 AM -0800 2/17/10, Nicholas K. Sauter wrote:
>Herbert,
>
>Please see line 1166 in src/cbf_byte_offset.c.
>
>Based on the logic of the algorithm I think this line should be
>
>if (delta[0] & 0x8000) {
>
>i.e., there is an extra "0" in the hex value in the CVS code.
>That being said, I don't think this line has an effect on any of my 
>test cases, so I'm at a loss about how to test this out.
>
>Nick


-- 
=====================================================
  Herbert J. Bernstein, Professor of Computer Science
    Dowling College, Kramer Science Center, KSC 121
         Idle Hour Blvd, Oakdale, NY, 11769

                  +1-631-244-3035
                  yaya at dowling.edu
=====================================================