lazy nsolve

[Krastanov]

A way to get the function "a -> solution for x of f(x,a)=0" working for well behaved f(x,a)

A proposal was implemented in PR 968 but Ronan raised some objections. Discussion and arguments from both sides are present on the pull request page #968

[Krastanov]

solving issue #3276 may be a more idiomatic approach. It is not the same, however I will mark it as blocked-on

[asmeurer]

What you really want, I think, is a SymPy object that encapsulates an arbitrary black box callable on numeric arguments, but also holds unevaluated symbolic arguments. This is already doable with Function and .eval, but I guess you want something a little more streamlined for that purpose. Maybe you could allow to pass a second argument to Function when using it to create an anonymous function that would be called on numerical arguments. You would probably need to add nargs as well. Ronan, do you see any issues with that proposal?

[asmeurer]

We have moved issues to GitHub https://github.com/sympy/sympy/issues .

**Labels:** Restrict-AddIssueComment-Commit  
**Blockedon:** -sympy:3276 sympy:3276  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3259#c3
Original author: https://code.google.com/u/asmeurer@gmail.com/

[asmeurer]

We have moved issues to GitHub https://github.com/sympy/sympy/issues .

Original comment: http://code.google.com/p/sympy/issues/detail?id=3259#c4
Original author: https://code.google.com/u/asmeurer@gmail.com/

[oscarbenjamin]

Related to #4678