On 12 Nov 2011, at 14:17, Michele Martone wrote: > On 20111108@15:41, Michele Martone wrote: >> On 20111108@15:12, Carlo de Falco wrote: >>> ... >>> There is an example of the correct way to create a new octave matrix class >>> in the package "triangular": >>> http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/extra/triangular/ >> OK I'll check it out to see if there is something I've missed in >> "sparsersb". >> ... > > Hi Carlo, > > I've looked at the "triangular" example. > So, from the .texi and .cc files by myself, I've learned: > > - the need for using mlock() in the oct-file (commit'ed the change) > - that the 'triangular' example does not fit completely to sparsersb: > class octave_triangular_matrix extends class octave_matrix, while > class octave_sparse_rsb_matrix extends class octave_sparse_matrix
I see, there are still a lot of similarities however. > - subsref, subsasgn, dotref are unfinished in sparsersb > - the 'print(std::ostream&...' method is incomplete: is this method > the one ensuring the cool output paging capability in Octave ? that is what is invoked when you type: >> a = uppertri (randn (3,3)); >> a or >> disp(a) and yes, the output of that command is piped through the pager if >> page_screen_output () ans = 1 > - assignments, conversions: these are important but a bit obscure to > me now; however there is no hurry with them I would say, as they > are not in the "critical" features list one would expect from > sparsersb ; you are right, I think to get started is the most important thing. It's useful anyway to have a full list of things that would be need to be done eventually, and also to make sure that informative error messages are given when user tries to access unimplemented features. > > Since I would prefer to follow a principle of caution, I would not > implement the mentioned missing features, until I get positive responses > (so, octave-dev's peple help is critical here) about current sparsersb's > stability, error reporting, robustness for installation and (shared > memory parallel, optimized) core operations (e.g.: matrix creation, > multiplication, triangular solution). as I mentioned in the previous mail I'm installing your package right now to start having a look, anyway it seems your contribution has already attracted quite some interest. > m. c. ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev
