[Octave-bug-tracker] [bug #59597] bogus result returned by mexCallMATLAB

[Rik]

Follow-up Comment #14, bug #59597 (project octave):

@jwe: Unfortunately my patch did not fix the bug, but do you understand
whether it is okay to use octave::feval rather than interp.feval?  It seems
like the patch might be a good thing to apply anyways.  For reference, here is
the entire mexCallMATLAB function from mex.cc.


int
mexCallMATLAB (int nargout, mxArray *argout[], int nargin,
               mxArray *argin[], const char *fname)
{
  octave_value_list args;

  // FIXME: do we need unwind protect to clean up args?  Off hand, I
  // would say that this problem is endemic to Octave and we will
  // continue to have memory leaks after Ctrl-C until proper exception
  // handling is implemented.

  // FIXME: Proper exception handling has been implemented (Jan. 2016).
  //        Can this code be re-factored?
  args.resize (nargin);

  for (int i = 0; i < nargin; i++)
    args(i) = mxArray::as_octave_value (argin[i]);

  octave::interpreter& interp = octave::__get_interpreter__
("mexCallMATLAB");

  bool execution_error = false;

  octave_value_list retval;

  try
    {
      retval = octave::feval (fname, args, nargout);
      // My proposed change below
      //retval = interp.feval (fname, args, nargout);
    }
  catch (const octave::execution_exception&)
    {
      if (mex_context->trap_feval_error)
        {
          // FIXME: is there a way to indicate what error occurred?
          // Should the error message be displayed here?  Do we need to
          // save the exception info for lasterror?

          interp.recover_from_exception ();

          execution_error = true;
        }
      else
        {
          args.resize (0);
          retval.resize (0);

          throw;
        }
    }

  int num_to_copy = retval.length ();

  if (nargout < retval.length ())
    num_to_copy = nargout;

  for (int i = 0; i < num_to_copy; i++)
    {
      // FIXME: it would be nice to avoid copying the value here,
      // but there is no way to steal memory from a matrix, never mind
      // that matrix memory is allocated by new[] and mxArray memory
      // is allocated by malloc().
      argout[i] = mex_context->make_value (retval(i));
    }

  while (num_to_copy < nargout)
    argout[num_to_copy++] = nullptr;

  return execution_error ? 1 : 0;
}




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?59597>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/