On Sat, Dec 28, 2019 at 3:40 PM David Mertz <me...@gnosis.cx> wrote: > What about Decimal snan? >
That version tries to call .is_nan() first, so works fine with the Decimal snan. But as Nark pointed out, having an s an raise maybe the right thing to do anyway— it should not be used to indicate a missing value anyway. I’m pretty sure the issue with numpy array scalars is that it’s not calling __float__. Which does make think that maybe all __float__ implementations should return a float NaN when appropriate. -CHB > > On Sat, Dec 28, 2019, 5:53 PM Christopher Barker <python...@gmail.com> > wrote: > >> On Sat, Dec 28, 2019 at 2:41 AM Antoine Pitrou <solip...@pitrou.net> >> wrote: >> >>> +1 for a .is_nan() method on suitable types. That's the most natural >>> and elegant solution, IMHO. Tricks like "x == x" are nice when you >>> *know* that x is a float or Decimal, but not in the general case. >>> >> >> agreed -- while it may work in almost all cases, what it is really >> checking is whether an object compares to itself, which is not question >> being asked. >> >> I suppose we could do something like: >> >> def is_nan(num): >> try: >> return num.is_nan() >> except AttributeError: >> if isinstance(num, Number): >> return not (num == num) >> else: >> return False >> >> Running it on my test code, it works for everything I thought to test >> except numpy arrays of size 1. >> >> -CHB >> >> >> -- >> Christopher Barker, PhD >> >> Python Language Consulting >> - Teaching >> - Scientific Software Development >> - Desktop GUI and Web Development >> - wxPython, numpy, scipy, Cython >> -- >> Christopher Barker, PhD >> >> Python Language Consulting >> - Teaching >> - Scientific Software Development >> - Desktop GUI and Web Development >> - wxPython, numpy, scipy, Cython >> > _______________________________________________ >> Python-ideas mailing list -- python-ideas@python.org >> To unsubscribe send an email to python-ideas-le...@python.org >> https://mail.python.org/mailman3/lists/python-ideas.python.org/ >> Message archived at >> https://mail.python.org/archives/list/python-ideas@python.org/message/WYRLHSVZTPE756XXGHDXW2HK624DPW5P/ > > >> Code of Conduct: http://python.org/psf/codeofconduct/ >> > -- Christopher Barker, PhD Python Language Consulting - Teaching - Scientific Software Development - Desktop GUI and Web Development - wxPython, numpy, scipy, Cython
_______________________________________________ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-le...@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/XA2WSUXRAERUX2U4WZJ6FVNUO5JHHWYX/ Code of Conduct: http://python.org/psf/codeofconduct/
