Python grammar for tree-sitter

0.25.0 · active · verified Thu Apr 09

tree-sitter-python (version 0.25.0) provides a pre-compiled Python grammar for the tree-sitter parsing library. It enables Python applications to efficiently parse Python source code into concrete syntax trees, facilitating tasks like static analysis, code transformation, and IDE features. The library is actively maintained with regular updates following the upstream tree-sitter project.

Warnings

breaking The `tree-sitter` core library and its language grammars (like `tree-sitter-python`) use an internal ABI version. Incompatibilities can arise if the `tree-sitter` Python package and the `tree-sitter-python` grammar package are not aligned in their supported ABI versions, leading to runtime errors.
Fix: Always ensure `tree-sitter` and `tree-sitter-python` are updated to compatible versions, ideally the latest available. For example, `tree-sitter` 0.25.0 and later primarily support ABI version 15.
breaking The `tree-sitter` core Python bindings (`py-tree-sitter` package) introduced significant breaking changes in version `0.24.0`, affecting method signatures for `Parser.parse`, `Query.captures`, `Query.matches`, and deprecated `Language.query`.
Fix: Review the official `py-tree-sitter` documentation or GitHub releases for `v0.24.0` for detailed API changes. Update code to use new signatures, e.g., using `Query(language, source)` instead of `Language.query(source)`.
deprecated Version 0.24.0 of `tree-sitter-python` dropped compatibility with Python 3.9.
Fix: Upgrade your Python environment to Python 3.10 or newer.
gotcha When working with `tree-sitter` and its Python bindings, parsing very large or complex codebases, especially with certain grammar versions (e.g., specific `tree-sitter-bash` versions), has been reported to cause high memory usage or even memory leaks.
Fix: Monitor memory usage for large inputs. If excessive memory consumption is observed, check for updates to both `tree-sitter` and the specific language grammar (`tree-sitter-python`). Consider processing code in smaller chunks if possible.
gotcha While `tree-sitter-python` provides a pre-compiled grammar, building custom or third-party `tree-sitter` grammars for other languages requires a C compiler and Node.js for compilation, which can be a common setup hurdle for new users.
Fix: Ensure you have a C compiler (like GCC or Clang) and Node.js installed and accessible in your PATH if you intend to compile `tree-sitter` grammars from source.

Install

pip install tree-sitter tree-sitter-python Install with pip

Imports

Language
```
from tree_sitter import Language
```
Parser
```
from tree_sitter import Parser
```
tree_sitter_python
```
import tree_sitter_python
```

Quickstart

This quickstart demonstrates how to load the Python grammar and parse a simple Python code snippet. It initializes a `Language` object using `tree_sitter_python.language()`, creates a `Parser`, and then parses a byte string of Python code into a syntax tree. The root node's type and its first child's type are printed.

from tree_sitter import Language, Parser
import tree_sitter_python

# Initialize the Python language grammar
PY_LANGUAGE = Language(tree_sitter_python.language())

# Create a parser and configure it to use the Python language
parser = Parser(PY_LANGUAGE)

# Source code to parse
code = b"""
def greet(name):
    print(f"Hello, {name}!")

greet("World")
"""

# Parse the code
tree = parser.parse(code)

# Get the root node of the syntax tree
root_node = tree.root_node

# Print the type of the root node and its first child (for demonstration)
print(f"Root node type: {root_node.type}")
if root_node.children:
    print(f"First child type: {root_node.children[0].type}")

# Expected output might vary slightly but should show 'module' and 'function_definition'

view raw JSON →