On Sun, Mar 30, 2008 at 8:16 PM, Neal Holtz <[EMAIL PROTECTED]> wrote: > > > > On Mar 30, 10:22 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > > > > On Sun, Mar 30, 2008 at 7:09 PM, Neal Holtz <[EMAIL PROTECTED]> wrote: > > > > > Hi, > > > > > I've been playing a lot with interact recently -- I *really* like it. > > > While attempting to develop a step-at-a-time example of Newton-Raphson, > > > a couple of things came to mind. > > > > > 1. What do you think about a control that saves state between > > > interactions, but does not give you any visible control widgets? > > > I know you can save state using globals, but if this works without > > > too much trouble, it might be cleaner. In fact, I have an > implementation > > > now that seems to work, that allows things like: > > > > > @interact > > > def _( a = save_state(i=0), b = ['Go'] ): > > > a.i += 1 > > > print 'i=', a.i > > > > > 'a' does not appear as an interactive widget. > > > Every time you press 'Go' you get the next number. > > > > > Only minor testing do far, so I don't know about all the > repercusions... > > > > > 2. What about a similar kind of thing, but gives you access to > > > all the controls, so perhaps your function could change > > > default values, reposition sliders, etc. Perhaps something > > > like > > > > > @interact > > > def _ ( f = sin(x), range=(0..20), controls=interact_controls() ): > > > ... > > > controls['range'].set_value(...) > > > > > The value of controls could be a dictionary indexed by variable > name ... > > > > > If this has any worth, I could attempt a proof-of-concept in a > couple of days. > > > > I think both of these are good ideas if they can be implemented > > robustly. The second will be harder to implement but be *extremely* > > useful in practice, I think. Probably somehow whenever the controls > > are changed via the second thing, we'll have to somehow know to > > redraw the entire output cell (including the input controls) in order to > > set the values of the controls appropriately. When I was writing > > > funtool:http://wiki.sagemath.org/interact#head-16e9c0180afb4f04813e76ab5685ef... > > > > > I could have done something much more powerful using something > > like your suggestion 2. > > > > William > > > > -- William > > Ah yes -- I just took a crack at #2 and now realize its trickier than > first thought. > The controls are rendered to HTML only once, right, and not each time > the function is invoked.
Yes, that's correct. > Oh well, learning my way around here. I'll > continue > worrying about it for a few days ... Good. It was a (very surprisingly) huge amount of hard work to implement the current interact functionality. It's not trivial to add more. But it's well worth it. > My compliments on the code -- it was pretty easy to find my way > around. Cool. Thanks. > > > -- William Stein Associate Professor of Mathematics University of Washington http://wstein.org --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-support URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
