From: Etienne G. <et...@is...> - 2003-02-11 08:34:57
|
On Sun, Feb 09, 2003 at 11:18:32PM -0500, Paul Kienzle wrote: # Since the current octave-forge release doesn't work with the # new octave, I believe we need an early release. Please # post your updates. # # I've tested against 2.1.44 on Windows, but not 2.1.43 or # linux. Anyone know whether we are still 2.1.43 compatible? # Or 2.1.40 for that matter. Can't tell for sure because I upgraded to 2.1.44 from 2.1.43 (always linux) last week; until then, all went fine. # I've put together some release notes in the file # RELEASE-NOTES. I'm trying a new format --- please # let me know if there is enough detail to be useful. # Etienne, Alois --- please give a brief summary of what you # have done to VRML, NaN and TSA and post the results. Just committed. When is the release? I have received a suggested change to bfgs.m and I am trying to improve the behavior of this function. Cheers, Etienne # Things not included in this release: # # * Testing!!! We have to start doing better tests --- some # of the recent octave changes break script files. E.g., # global now defines an initial value of empty if none is # assigned, which breaks all the colormap functions. # * Dispatch: now that we have a dispatch mechanism, it # would be nice to make sparse/symbolic look more like # builtin types by allowing users to use, e.g., find instead # of spfind. # * Commands: octave has mark_as_command so that # functions such as grid and edit can be turned into # commands which don't need parenthesis and quotes # around their arguments. # # Comments? # # Paul Kienzle # pki...@us... -- Etienne Grossmann ------ http://www.isr.ist.utl.pt/~etienne |