From: nma@12000.org   
      
   On 7/16/2015 1:22 AM, Christopher Creutzig wrote:   
      
   >   
   > In MuPAD, the function is only defined (aka implemented) for integer   
   > order. That may or may not be a weakness, but it can hardly be “wrong.”   
      
   The "wrong" is then in how Mupad handles this. If Mupad does not   
   support non-integer order, then it should simply return the   
   integral unevaluated (i.e it could not do it) which I think   
   is the normal practice when a CAS can not do something.   
      
   My small complaint was on the message saying that the "order   
   must be integer".  May be this should be in the documentation   
   only, but it should not give an error and abort.   
      
   > There's a lot of code available, for example, for Bessel functions of   
   > integer or half-integer order, which is not “wrong” either.   
   >   
   >> But to be able to run Mupad, I have to have the same number of   
   >> tests for each CAS. I'd like to remove these 6 integrals   
   >   
   > You could always use some wrapper function in your input files that   
   > avoids the error and will fail to integrate:   
   >   
   > polyLog := (n,x) -> if testtype(n,DOM_INT) then   
   >                        polylog(n,x)   
   >                      else   
   >                        procname(args())   
   >                      end_if:   
   >   
   >   
      
   Thanks. Will look into this. The test files are large   
   and contain all the integrals in a long list or lists.   
      
   So if the loading fails, all the integrals can not be read, since   
   they are all read in one read statement (they are all in one   
   list of lists).  But will look into it for next version of   
   my build program to try to handle such case differently, for   
   now, this affects these 6 integrals out of 51,000, so   
   not a big problem)   
      
   --Nasser   
   > HTH,   
   >   
   > Christopher   
   >   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)