Tree-sitter HTML Grammar

0.23.2 · active · verified Sun Apr 12

tree-sitter-html provides the HTML grammar for the Tree-sitter parsing system. It enables parsing HTML code into a concrete syntax tree, facilitating static analysis, syntax highlighting, and code transformation. The library is actively developed with regular updates, currently at version 0.23.2.

Warnings

gotcha The `tree-sitter` core library is a mandatory dependency and must be installed alongside `tree-sitter-html` for the Python bindings to function.
Fix: Ensure both `tree-sitter` and `tree-sitter-html` are installed: `pip install tree-sitter tree-sitter-html`.
gotcha Grammar updates can lead to changes in the generated Abstract Syntax Tree (AST), potentially affecting existing queries or code analysis logic. Common issues include unexpected handling of whitespace, attribute values, or void elements.
Fix: Thoroughly test parsing results with your specific HTML structures after updating the grammar. Refer to GitHub issues for known parsing quirks.
gotcha When `tree-sitter-html` is used for parsing HTML embedded within other languages (e.g., PHP, JavaScript in `script` tags, CSS in `style` tags), ensure that the corresponding `tree-sitter` grammars (e.g., `tree-sitter-javascript`, `tree-sitter-css`) are also installed for complete and accurate highlighting/parsing of injected languages.
Fix: Install all relevant language grammars for files containing multiple embedded languages: `pip install tree-sitter-html tree-sitter-javascript tree-sitter-css`.
deprecated Older examples might show `tree_sitter.Language.build_library()` for loading grammars. This method is deprecated for pre-compiled Python wheels (like `tree-sitter-html`).
Fix: Instead of `Language.build_library()`, directly pass the grammar's `language()` function to `tree_sitter.Language()`: `HTML_LANGUAGE = tree_sitter.Language(tree_sitter_html.language())`.

Install

pip install tree-sitter tree-sitter-html Install with pip

Imports

Language, Parser
```
from tree_sitter import Language, Parser
import tree_sitter_html
```
`tree_sitter_html` provides the grammar object via `tree_sitter_html.language()`, while `Language` and `Parser` classes come from the core `tree_sitter` library.

Quickstart

This quickstart demonstrates how to initialize a `tree-sitter` parser with the `tree-sitter-html` grammar and parse a basic HTML string into a syntax tree, then print its S-expression representation.

import tree_sitter
import tree_sitter_html

# Load the HTML language grammar
HTML_LANGUAGE = tree_sitter.Language(tree_sitter_html.language())

# Initialize the parser and set the language
parser = tree_sitter.Parser()
parser.set_language(HTML_LANGUAGE)

# Sample HTML code (must be bytes for tree-sitter)
html_code = b"<!DOCTYPE html>\n<html><body><h1>Hello, Tree-sitter!</h1></body></html>"

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

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

# Print a S-expression representation of the tree
print(root_node.sexp())

view raw JSON →