ast.walk

Learn to traverse the tree with ast.walk.

What is ast.walk?

ast.walk(tree) is a generator that yields every node in an AST in order. Depth or breadth does not matter; if you just want to see every node it is ideal.

For example for node in ast.walk(tree): lets you iterate all nodes and filter a specific type with isinstance(node, ast.Name). This is enough for simple counters and analysis tools. For more controlled traversal you use NodeVisitor; but for a quick scan ast.walk is the shortest path.

Counting node types

Counting nodes of a specific type with ast.walk is a common task. You initialize a counter, check the type with isinstance inside the loop and increment on match.

For example finding how many Name nodes a piece of code has is useful to measure variable usage. Similarly the count of ast.Call shows how often functions are called. These metrics are the foundation of code-quality tools. You report the result with a print.

Exercise Count nodes with ast.walk

Import the module with import ast. Build the tree with ast.parse("a = 1\nb = 2\nc = 3"). Walk the tree with ast.walk; count Name nodes with isinstance(node, ast.Name). Initialize a counter and increment on each match. Print the result.

Show example solution

import ast

tree = ast.parse("a = 1\nb = 2\nc = 3")
count = 0
for node in ast.walk(tree):
    if isinstance(node, ast.Name):
        count += 1
print(count)

Test yourself

Match ast.walk concepts

ast.walk

Walks all nodes
for node in

Loop variable
isinstance

Type filter
tree

Tree root
What does ast.walk return?
```
import ast
tree = ast.parse("x = 1")
for n in ast.walk(tree):
    print(type(n).__name__)
```
- A list
- A generator
- A single node
- A dict
Show answer

A generator — answer: a generator. ast.walk yields one node at a time.

Complete the ast.walk call

for node in ____(tree):
    print(type(node).__name__)

Show answer

for node in ast.walk(tree):
    print(type(node).__name__)

Next lesson: NodeVisitor

We Are Enhancing Your Experience

Cookie Preferences

Strictly Necessary Cookies

Analytical Cookies