Re: 3.0.1 release?

[Michael Goffioul]

On Mon, Apr 7, 2008 at 9:02 PM, John W. Eaton <address@hidden> wrote:
>  The second test release is here:
>
>   http://jbrwww.che.wisc.edu/home/jwe/octave-3.0.1b.tar.bz2
>
>  I think I've fixed all of the problems reported specifically about the
>  3.0.1a test release.  Since this release should fix bugs and not
>  introduce them, I'd like to have some people try it before releasing
>  and announcing it.

The package compiles and runs fine under Win32/MSVC. The test suite
reports the same problem as before: see for instance
http://www.nabble.com/fntests-errors-on-mingw-Octave-1%3Adispatch.cc-to14574608.html#a14582630

Among "known" failures, the 2 most problematic ones are:

========================

  ***** assert(bitcmp(A,Amax),bitor(bitshift(1,Amax-1),bitshift(1,Amax-2)));
!!!!! test failed
error: assert (bitcmp (A, Amax),bitor (bitshift (1, Amax - 1),
bitshift (1, Amax - 2))) expected
6.7554e+015
but got
1.1259e+016
values do not match
shared variables {
  Amax =  53
  Bmax = 9.0072e+015
  A = 2.2518e+015
}

========================

  ***** test
 charset = setstr (0:127);

 result = zeros (1, 128);

 result (33:127) = 1;

 assert(all (isprint (charset) == result));
!!!!! test failed
error: assert (all (isprint (charset) == result)) failed

========================

For the first one, I have no clue. The second one is due to the
fact that under Win32: isprint(setstr(9)) returns 1, while the
test expects 0.

Michael.