What's the best way to minimize the need of run time checks?

Juan Pablo Romero Méndez jpablo.romero at gmail.com
Wed Aug 10 13:43:02 EDT 2016


2016-08-09 23:16 GMT-07:00 Gregory Ewing <greg.ewing at canterbury.ac.nz>:

> Juan Pablo Romero Méndez wrote:
>
> This is interesting. You are Ok having runtime errors?
>>
>
> You're going to have runtime errors in any case, whether
> they come from code you've put there yourself to check
> types, or from somewhere deeper down.
>


You are correct. What I was trying to understand is if there is a much more
relaxed attitude towards runtime errors in the Python community (even
embracing it) than others* (and of course, how to handle those errors).

(In certain communities people favor a design where exceptions / runtime
errors only happen at program boundaries)



>
> The only difference is that checks you make yourself
> *might* catch errors slightly sooner, and *might* be able
> to provide better diagnostics.
>
> However, experience has shown that, the vast majority of
> the time, type errors in Python are caught pretty much
> immediately, and the stack trace provides more than
> enough information to pin down the source of the problem.
> So, putting in manual type checks is hardly ever worth
> the effort, and can even be counterproductive, since it
> interferes with duck typing.
>
> Occasionally there will be a situation where a type
> error results in a corrupted data structure that leads
> to problems later. But those cases are relatively rare
> and best dealt with as they come up.
>
> --
> Greg
> --
> https://mail.python.org/mailman/listinfo/python-list
>



More information about the Python-list mailing list