What is this feature of functions in Python formally known as?

Question

In Python, a function has access to all variables in-scope at the time of its creation. What is this feature called?

JavaScript has this same behavior, and we say that functions are lexical-scoped in it. Will the same apply to Python?

But I searched for this on the web.. Very few resources use the term lexical scope with Python functions. — coderboy, Nov 23 '20 at 17:05
If it's such an important concept, why do resources not use the term _lexical scope_ quite often. — coderboy, Nov 23 '20 at 17:06
I don't know what to tell you. I get a lot of hits when I google "lexically scoped closures python", among them the [PEP](https://www.python.org/dev/peps/pep-0227/) that introduced them to the language (20 years ago). — juanpa.arrivillaga, Nov 23 '20 at 17:07
*They do*. Again, I don't know what to tell you. It's common knowledge, and it is references in official documentation — juanpa.arrivillaga, Nov 23 '20 at 17:08
I saw the PEP one, but that was written in a way too complex manner. — coderboy, Nov 23 '20 at 17:13
Also, in a lot of old JS, lexically scoped closures are used to implement things that you would just use classes/modules for in Python. — juanpa.arrivillaga, Nov 23 '20 at 17:13
@coderboy You may find [this article](https://medium.com/@dannymcwaves/a-python-tutorial-to-understanding-scopes-and-closures-c6a3d3ba0937) helpful. — ekhumoro, Nov 23 '20 at 17:29

score 1 · Answer 1 · answered Nov 23 '20 at 17:09

1

In Python, a function has access to all variables in-scope at the time of its creation. What is this feature called?

A Python function actually has access to all the variable in the scope of its creation, even those defined after the function is closed.

The property that a function can access content from its definition environment is called closing over, that function defines (or colloquially is) a closure.

JavaScript has this same behavior, and we say that functions are lexical-scoped in it. Will the same apply to Python?

Lexical scoping does apply to function, but it mostly has to do with the chain taken to resolve names: lexical scoping means the names are resolved at the point of definition, dynamic scoping is the other big one and means the names are resolved as the point of use.

Lexical or dynamic scoping applies to all scopes, so in most languages it will also (or only) apply to blocks (e.g. Python only has function-level scoping, C only has block-level scoping, Go has both).

answered Nov 23 '20 at 17:09

Masklinn

34,759
3
38
57

I believe you are talking about early vs late binding. Dynamic scoping would work something like this: `def foo(): print(x)` then def `bar(): x = 12; foo()` would print `12`. i.e. the resolution of free variables are determined *by call site* not by the *static/lexical* site of where they are defined. – juanpa.arrivillaga Nov 23 '20 at 17:15
@juanpa.arrivillaga I don't get what code you mean to show here.. Please be more elaborate... – coderboy Nov 23 '20 at 17:20
@coderboy it's supposed to demonstrate the difference between dynamic and static/lexical scope. If you run the code above it will give you a `NameError` in Python, but it would print `12` if python had dynamic scope – juanpa.arrivillaga Nov 23 '20 at 17:24
@juanpa.arrivillaga That's right! Sure Python would throw an error. I guess most languages use this lexical-scoping thing, in contrast to dynamic-scoping. – coderboy Nov 23 '20 at 17:26
@coderboy yes, it's not common in modern languages. The only language I use that has it would probably be bash. Perl and Common Lisp allow you to choose dynamic or lexical scoping. – juanpa.arrivillaga Nov 23 '20 at 17:29

Wasif · Answer 2 · 2020-11-23T17:15:00.607

0

Yeah, python has lexical scoping, but environments are mutable. Such as if x=1 and f() returns x but before calling if x is 10 then return value is 10.

In Python, just as in SML, or modern Lisp, the body of a function is evaluated in the environment where it was defined. So, all three languages are lexically scoped.

But In Python, environments are mutable. That is, you can assign a new value to an existing variable, and that mutates the environment the variable is part of. Any functions defined within that environment will be evaluated in that environment—which means they will see the new value of the variable.

From: Does Python scoping rule fits the definition of lexical scoping?

edited Nov 23 '20 at 17:15

answered Nov 23 '20 at 17:10

Wasif

14,755
3
14
34

2

Please don't copy [other people's answers](https://stackoverflow.com/a/51604511/984421) without giving proper attribution. – ekhumoro Nov 23 '20 at 17:13
@ekhumoro added attribution, i was just editing, but accidentaly closed all tabs in browser and my workstation was auto-scheduled to restart – Wasif Nov 23 '20 at 17:15
1

This should probably be marked as a duplicate, not simply re-posted here. – juanpa.arrivillaga Nov 23 '20 at 17:16
Is the other name for lexically-scoping, static-scoping? – coderboy Nov 23 '20 at 17:19
@coderboy yes that is a synonym. Although, I think that is avoided because "static" means something else in class-based OOP sense. – juanpa.arrivillaga Nov 23 '20 at 17:21

What is this feature of functions in Python formally known as?

2 Answers2