Le jeudi 19 avril 2012 à 19:05 +0200, Joachim Durchholz a écrit : > Am 19.04.2012 15:25, schrieb Ronan Lamy: > > I'm referring to sympy.utilities, which does import basically all of > > sympy.core. It's actually not as bad as I thought, but still seems quite > > dangerous for something as basic as a deprecation warning. > > What dangers do you mean? > > The ones I can think of are performance and namespace pollution. > > Performance isn't a problem, because at least SymPy core will be pulled > in by any SymPy program anyway. > True.
> Namespace pollution is a non-issue with explicit imports of the > from m import f > form, which the code is already doing. > Since the import is inside a function, even an import * wouldn't matter, since it only pollutes the local namespace of the function. > Am I overlooking something? > The danger I mentioned is that it increases the risk of circular imports. Getting an ImportError because of a warning is no fun, but that will happen to us if this gets in as is. More broadly, warnings tend to appear where you least expect them, so fiddling with the mechanism is likely to increase the probability of them having nasty consequences. -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com. To unsubscribe from this group, send email to sympy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
