On Sun, May 16, 2010 at 2:51 AM, Eugeniy Mikhailov <[email protected]> wrote: > Hi, > > I think I see a bug in the parcellfun.m function of the general-1.2.0 > package. > > The following output produces an error with general-1.2.0 while it was > fine with general-1.1.4 > ========================================================== > octave:8> x{1}=pi/2; > octave:9> x{2}=pi; > octave:10> parcellfun(2,'sin',x) > error: common_size: only makes sense if nargin >= 2 > error: called from: > error: /usr/share/octave/3.2.4/m/general/common_size.m at line 49, column 5 > error: /home/evmik/octave/general-1.2.0/parcellfun.m at line 97, column 21 > ========================================================== > > However if after installation of general-1.2.0 I replace parcellfun.m > from general-1.1.4 everything works as expected. > > The only relevant differences of the parcellfun.m are shown below > (truncated output of: diff -u parcellfun.m.1.1.4 parcellfun.m.1.2.0) > ========================================================== > if (length (args) == 0) > print_usage (); > - elseif (length (args) > 1 && ! size_equal (args{:})) > - error ("arguments size must match"); > + else > + [err, args{:}] = common_size (args{:}); > + if (err) > + error ("parcellfun: arguments size must match"); > + endif > endif > ========================================================== > > I am a bit confused about 'common_size' usage. Why is it needed at all? >
To allow auto-expansion of singleton cells, like Octave's cellfun does. This bug has already been fixed in SVN. I'm requesting the release of 1.2.1 now. thanks -- RNDr. Jaroslav Hajek, PhD computing expert & GNU Octave developer Aeronautical Research and Test Institute (VZLU) Prague, Czech Republic url: www.highegg.matfyz.cz ------------------------------------------------------------------------------ _______________________________________________ Octave-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/octave-dev
