Violation of bounds

[Peter230655]

opty can easily give the violations of the eoms and the instance constraints.
To my knowledge there is no method to check whether the bounds are kept,
If I am right, would it make sense to have such a method in Problem?

[moorepants]

If an optimal solution is found, doesn't than mean the bounds are met?

[Peter230655]

I do not know. The instance constraints also have errors.
And if a solution is not found, maybe it is useful to know, which bounds were violited how much?

[moorepants]

You could easily plot dotted lines on the trajectory plots showing the bounds.

[Peter230655]

That would be easy!
Should I try or not worth it?

[Peter230655]

Nice idea, would not add additional plots.
Worth doing?

[Peter230655]

I think, it would only make sense to draw these lines if min(x(t) <= a or max(x(t) >= b, where bounds = {x: (a, b)}.
Otherwise, if a << x(t) << b, x(t) would appear as a featureless line it the plot.

Is it worth doing this?
I can do it -and like to do such things which I can do- but is there a danger, that opty get 'overloaded'?

[moorepants]

You are right that you may not want these lines in the plot, as the solution could be very small and the bounds very large, so the plot will no longer show useful information. It can be an option to show such lines.

[Peter230655]

...but this would add an optional kwarg.
If this is o.k., I can do so, otherwise I'd only draw the bounds if, say, x(t) <= a

[moorepants]

You continue to believe that I am against optional kwargs just because it is an optional kwarg, even though I have tried to explain that is not true multiple times.

[Peter230655]

I admit it's still a bit in the back of my mind - and yes, you told me otherwise!
I will do it with a kwarg. Would show_bounds be a good name?
Default would be False?

[Peter230655]

Taken care of with PR #419