Re: 3.2 status report

[John W. Eaton]

On  6-Aug-2008, Jaroslav Hajek wrote:

| On Wed, Aug 6, 2008 at 6:05 PM, John W. Eaton <address@hidden> wrote:
| > On  6-Aug-2008, Jaroslav Hajek wrote:
| >
| > | The reason for this change is that Matlab's choice makes the structure
| > | visually useless: if you dump it on the screen, the relevant options
| > | are hindered by a flood of empty fields which you're completely not
| > | interested in (and sometimes they don't even make sense). I really
| > | don't understand why the Matlab people made this choice.
| > | Thus, there is no simple way to inspect what options a structure
| > | actually contains.
| >
| > Other than the clutter, I'm not sure I understand what real problem
| > this change is trying to solve.  Can you give an example?
| 
| There's no other problem, just the cluttering. But I think it is a real 
problem.

OK, I don't see that it hurts (I guess someone will tell us if they
depend on this behavior) so if you'll submit a changeset for just this
change with a ChangeLog entry, I'll apply it in my archive.

I've checked in changes so that the current fsolve function accepts
optimset and also produces a more compatible value of INFO.
Currently, the only optimset option that has any effect for fsolve is
"tolx".  The values of INFO produced by Minpack don't clearly map to
the values produced by Matlab's fsolve, so I made my best guess about
what the translation should be.

These changes are enough for me for 3.2, but I'd certainly consider
other improvements.

Thanks,

jwe