Basics

What you'll learn

• Parsing a string, reading the root node, and detecting errors.

When to use this

First contact with the library, smoke tests, and language experiments.

Complete example

examples/basic_parse.zig:

var parser = treesitter.Parser.init(gpa);
defer parser.deinit();
try parser.setLanguage(treesitter.expression_language);

var tree = try parser.parseString("1 + 2 * 3");
defer tree.deinit();

const root = tree.rootNode();
std.debug.print("root: {s} [{d}, {d}]\n", .{ root.nodeType(), root.startByte(), root.endByte() });
std.debug.print("has error: {}\n", .{tree.hasError()});
std.debug.print("text: {s}\n", .{root.text()});

Running the example

zig build run-basic_parse

Expected output

root: program [0, 9]
has error: false
text: 1 + 2 * 3

How it works

parseString lexes and reduces 1 + 2 * 3 into a program node spanning the whole input. hasError() reads the root's aggregated flag — false here because every token had a valid table action. text() slices the tree's owned source copy.

API used

• Parser, Tree, Node

Memory ownership

Parser owns scratch; tree owns pool and source; handles borrow. Three deinit calls release everything.

Related guides

• Your First Parser, Error Recovery for the has_error=true side.