Check for the type of arguments

[Joshua Landau]

On 17 August 2013 13:34, Fernando Saldanha <fsaldan1 at gmail.com> wrote:
> I am new to Python, with some experience in Java, C++ and R.
>
> Writing in other languages I usually check the type and values of function arguments. In the Python code examples I have seen this is rarely done.
>
> Questions:
>
> 1) Is this because it would be "unpythonic" or just because the examples are not really production code?

Unpythonic. Python duck-types so we tend to take things as long as the
seem like they work. This is really helpful if you want to provide a
custom type or data object to a function (or anything really). If it
looks like a duck and quacks like a duck... you can probably make a
duck sandwich.

> 2) If I still want to check the type of my arguments, do I
>
> a) use type() or is instance() to check for type?

You'd want to travel down this stack, choosing the first reasonable one:

• Don't check at all

• Check that it can do what you need it to do, such as by calling
"iter(input)" to check that it's iterable.

• Check using an ABC (http://docs.python.org/3/library/abc.html) with isinstance

• Check that it has the methods you need using hasattr

• Check using "isinstance(...)" against a type

• Check using "type(...) is"

The choices higher up are better than the choices lower down.

> b) use assert (I guess not),

"assert" is for things that *can't* be wrong (yet still sometimes
are). Don't normally assert user input, I'd say.

>raise a ValueError,

Sounds right.

>or sys.exit()?

No. You should never be throwing a SystemExit except at top-level.

> (I noticed that raising a ValueError does not stop execution when I am running the Interactive Interpreter under PTVS, which I find inconvenient, but it does stop execution when running the code non-interactively.)