Key Features

• Parses Python source code into an Abstract Syntax Tree
• Generates ASTs directly from compiled code with the ast.PyCF_ONLY_AST flag
• Provides tools for traversing and modifying ASTs
• Includes base classes for visiting and transforming nodes
• Preserves and updates source-location metadata (line and column information)
• Can compile ASTs back into executable code
• Can convert ASTs back into Python source code (but not original formatting or comments)

Frequently Used Classes and Functions

Examples

Modify an AST to change variable names:

Compile and execute an AST:

Common Use Cases

The most common tasks for ast include:

• Analyzing code for static analysis tools (such as linters and type checkers)
• Extracting structure for documentation and dependency analysis (such as imports)
• Refactoring code programmatically (such as renaming symbols or rewriting APIs)
• Instrumenting code automatically (such as inserting logging or timing)
• Generating code from templates or other inputs
• Building code formatters, transpilers, and domain-specific languages

Real-World Example

Suppose you want to programmatically rename all occurrences of a variable in a script:

In this example, the ast module was used to rename a variable in the source code and execute the modified code. Note the use of ast.fix_missing_locations() to ensure the modified AST has proper line number information before compilation.

For additional information on related topics, take a look at the following resources:

• Python eval(): Evaluate Expressions Dynamically (Tutorial)
• Python's exec(): Execute Dynamically Generated Code (Tutorial)
• Evaluate Expressions Dynamically With Python eval() (Course)