@Szabolcs thanks for the tips!

@Nasser did you ever have auto save enabled?

@Szabolcs FYI, the FE just hanged again. Same steps. I stop the local kernel, following by evaluating an open notebook. This keep happening like once or twice a day it seems for me. Always same steps.

@CATrevillian what is auto save enabled? No, I never knew about it. I always make sure I save the notebook myself after any minor changes I make, since I do not trust Mathematica not to crash or hang ending up losing lots of work, like what just happened now. FE hanged.

Hello everyone

I would very much like some help with something I know absolutely nothing about. How does mathematica deals with wrappers so that e.g. a Sparse Matrix is shown with a little graphics. In the case of interest I want to be able to wrap something around an object which is like an InterpolatingFunction in mathematica? Thanks for any pointer to anything? Cheers

@chris mathematica.stackexchange.com/q/77658/5478

@Kuba thanks !!

@Kuba would you know how to grab the summary code used by Interpolation?

@Szabolcs I haven't heard about it. Do you have a link perhaps?

Hey! Is it possible to use custom ScalingFunctions inside RegionPlot3D?

I am not able to make it work

@Kuba Great! do you want to write a quick answer here (which I could edit and complement): mathematica.stackexchange.com/questions/218764/… or should I do it?

@chris I am sorry but I don't have time today to write properly :) Feel free to proceed.

sure: thanks again, I just wanted to be fair to your input.

oops

ok I found it

@chris Copy As > Input Text I think but I am not sure

Ok gtg, have a nice day.

plain text seems to do the trick

thanks again

@IstvánZachar Just this but both the syntax and the package name have changed since then.

@IstvánZachar Try this to get started, or look up the included (but very sparse) documentation of the paclets:

There is currently a limit on how many lints are shown, so play with SeveritiyExclusions and ConfidenceLevel to filter only the more important ones.

@jose Are you aware of the bug when part of the dataset is cut off on the right (even after scrolling)?

I have to say copying our community ads each year and having them upvoted by our rather small number of core users to reach the 5 upvote barrier is cumbersome...

@Szabolcs Yes, we are aware of this problem. Many thanks again for reporting these issues.

@Silvia @Silvia have you reported this?

So the usual culprit here is that inverse trig functions are tricky and have multiple valid branches.

ArcCos[k] can basically really represent (-ArcCos[k] + 2 Pi anInteger) or (ArcCos[k] + 2 Pi anInteger) where anInteger is an Integer.

But I'm still not able to resolve the output it's giving when I assume that.

If you'd like I'd be glad to report it

Does anyone know where I may find a WL binding to the StackExchange API (if it currently exists?)

@user21 Here on MSE, there is a recurrent schema we must take care of :

A OP ask an underspecified problem (I mean that there are many solutions) or ask a problem with very different possible interpretations. All this without being aware of the heavy ambiguity.

Furthermore he wants the problem to be solved with FEM (simply because it is the standard approach, I think)

Often, FEM can only solve some of the solutions, not all.

So we propose of course a solution that FEM can solve.

The problem is that there is a risk that this solution is not the one the OP expected (typically : infinite domain, NeumannPeriodic + Dirichlet Periodic problems).

I' m afraid that the following conclusion comes up : Mathematica can' t do it.

In fact the reality is that FEM (not specially Mma) can't (straitforwardly) do it, or Mathematica can do it, but with other methods.

This explain my motivation when I try analyse deeply OP' s questions. (it isn't a pernickety disorder)

Example for this question in the chatroom Electrostatic

I have tried to ask the relevant questions from the very beginning of the question, but I didn' t get response, problably because my question were too advanced.

After that Alex Trounev ask exactly the most important of these questions (are the object conductors?).

Not even a response for him.

So I deleted my complicated question, observe what has happened, and explain the situation afterward (but it's not simple).

This is not a claim. It is just an explanation of my motivations.

Concerning Neumann periodic boundary : I try to understand enough things about FEM to see if it is possible to modify the matrice (s) : for example in 1D, the stiffness matrice should become periodic. The idea is the same in 2D, but the indices are a mess. One thing seems OK : I think there's no nees to change its dimensions.

But I can't promise any interesting result (and above all a timing, consider timing = +Infinity)

typo in my text : nees -> need

No promises that any of it is up to date