So, I've read the chat logs earlier and Tim Wessman said some things about weird things that could happen in functions/plots
[08:21:15]<tw_hpcalc>i am quite excited to get more people pounding on it with the user programs, but also nervous
[08:21:24]<tw_hpcalc> as user programs are one of the hardest things to allow
[08:21:34]<tw_hpcalc> so many ways people can do things you never anticipated
[08:23:25]<tw_hpcalc> so here's a good one for example
[08:23:35]<tw_hpcalc> you can make a user program, and graph the thing
[08:23:58]<tw_hpcalc> put the function call inside your equation, and have that create user dialogs....
[08:24:10]<tw_hpcalc> what's going to happen there? :-|
[08:24:21]<tw_hpcalc> just crazy stuff like that
So far... no luck in finding anything weird in that (well, at least on the computer software, but at this level it shouldn't matter)
Indeed I can call a function I defined earlier from a plot (within F1(X) for example), and as long as the program RETURNs something, it can be used.
However... that stops working as soon as there is some I/O commands in there. (Then, the graph (and table) says the values are undefined)
I have a question, though, if TW ever reads this : a function(X) returning TICKS() and then plotted will only produce one point ... Any idea why ?
And... weird behavious finally found : With a single-screen (plot) : I can't autoscale because of this ^ (it says it can't because there is one point). However, when looking at a split view, with table, it then produces more values (it's funny because when you click, or do whatever, it updates the values
. And then, *sometimes* you can use autoscale., don't ask me why
Looks like this :
(Also, you still can click and move around even if that popup text is displayed)
Anyway, any other cool ideas of crazy things to try ?
PS : What about syntax-coloring in the program editor ... ? Would be darn cool/useful.