☰

Creating parsers

Developing Tree-sitter grammars can have a difficult learning curve, but once you get the hang of it, it can be fun and even zen-like. This document will help you to get started and to develop a useful mental model.

Getting Started

Dependencies

In order to develop a Tree-sitter parser, there are two dependencies that you need to install:

Installation

To create a Tree-sitter parser, you need to use the tree-sitter CLI. You can install the CLI in a few different ways:

Project Setup

The preferred convention is to name the parser repository “tree-sitter-“ followed by the name of the language.

mkdir tree-sitter-${YOUR_LANGUAGE_NAME}
cd tree-sitter-${YOUR_LANGUAGE_NAME}

You can use the tree-sitter CLI tool to set up your project, and allows your parser to be used from multiple languages.

# This will prompt you for input
tree-sitter init

Once you have installed the CLI and run through the init command’s prompts, a file called grammar.js should exist with the following contents:

/// <reference types="tree-sitter-cli/dsl" />
// @ts-check

module.exports = grammar({
  name: 'YOUR_LANGUAGE_NAME',

  rules: {
    // TODO: add the actual grammar rules
    source_file: $ => 'hello'
  }
});

Now, run the following command:

tree-sitter generate

This will generate the C code required to parse this trivial language, as well as a few files that are needed to compile and load this native parser as a Node.js module.

You can test this parser by creating a source file with the contents “hello” and parsing it:

echo 'hello' > example-file
tree-sitter parse example-file

Alternatively, in Windows PowerShell:

"hello" | Out-File example-file -Encoding utf8
tree-sitter parse example-file

This should print the following:

(source_file [0, 0] - [1, 0])

You now have a working parser.

Finally, look back at the triple-slash and @ts-check comments in grammar.js; these tell your editor to provide documentation and type information as you edit your grammar. For these to work, you must download Tree-sitter’s TypeScript API from npm into a node_modules directory in your project:

npm install

Tool Overview

Let’s go over all of the functionality of the tree-sitter command line tool.

Command: init

The first command you will likely run is the init command. This command sets up an empty repository with everything you need to get going with a grammar repository. It only has one optional argument, --update, which will update outdated generated files, if needed.

The main file of interest for users to configure is tree-sitter.json, which tells the CLI information about your grammar, such as the queries.

Structure of tree-sitter.json

The grammars field

This field is an array of objects, you typically only need one object in this array, unless your repo has multiple grammars (e.g. like Typescript and TSX)

Basics

These keys specify basic information about the parser:

Language Detection

These keys help to decide whether the language applies to a given file:

Query Paths

These keys specify relative paths from the directory containing tree-sitter.json to the files that control syntax highlighting:

The behaviors of these three files are described in the next section.

The metadata field

This field contains information that tree-sitter will use to populate relevant bindings’ files, especially their versions. A future bump-version and publish subcommand will leverage this version information as well. Typically, this will all be set up when you run tree-sitter init, but you are welcome to update it as you see fit.

The bindings field

This field controls what bindings are generated when the init command is run. Each key is a language name, and the value is a boolean.

Command: version

The version command prints the version of the tree-sitter CLI tool that you have installed.

tree-sitter version 1.0.0

The only argument is the version itself, which is the first positional argument. This will update the version in several files, if they exist:

As a grammar author, you should keep the version of your grammar in sync across different bindings. However, doing so manually is error-prone and tedious, so this command takes care of the burden.

Command: generate

The most important command you’ll use is tree-sitter generate. This command reads the grammar.js file in your current working directory and creates a file called src/parser.c, which implements the parser. After making changes to your grammar, just run tree-sitter generate again.

The first time you run tree-sitter generate, it will also generate a few other files for bindings for the following languages:

C/C++

Go

Node

Python

Rust

Swift

If there is an ambiguity or local ambiguity in your grammar, Tree-sitter will detect it during parser generation, and it will exit with a Unresolved conflict error message. See below for more information on these errors.

Command: build

The build command compiles your parser into a dynamically-loadable library, either as a shared object (.so, .dylib, or .dll) or as a WASM module.

You can change the compiler executable via the CC environment variable and add extra flags via CFLAGS. For macOS or iOS, you can set MACOSX_DEPLOYMENT_TARGET or IPHONEOS_DEPLOYMENT_TARGET respectively to define the minimum supported version.

You can specify whether to compile it as a wasm module with the --wasm/-w flag, and you can opt to use docker or podman to supply emscripten with the --docker/-d flag. This removes the need to install emscripten on your machine locally.

You can specify where to output the shared object file (native or WASM) with the --output/-o flag, which accepts either an absolute path or relative path. Note that if you don’t supply this flag, the CLI will attempt to figure out what the language name is based on the parent directory (so building in tree-sitter-javascript will resolve to javascript) to use for the output file. If it can’t figure it out, it will default to parser, thus generating parser.so or parser.wasm in the current working directory.

Lastly, you can also specify a path to the actual grammar directory, in case you are not currently in one. This is done by providing a path as the first positional argument.

Example:

tree-sitter build --wasm --output ./build/parser.wasm tree-sitter-javascript

Notice how the tree-sitter-javascript argument is the first positional argument.

Command: test

The tree-sitter test command allows you to easily test that your parser is working correctly.

For each rule that you add to the grammar, you should first create a test that describes how the syntax trees should look when parsing that rule. These tests are written using specially-formatted text files in the test/corpus/ directory within your parser’s root folder.

For example, you might have a file called test/corpus/statements.txt that contains a series of entries like this:

==================
Return statements
==================

func x() int {
  return 1;
}

---

(source_file
  (function_definition
    (identifier)
    (parameter_list)
    (primitive_type)
    (block
      (return_statement (number)))))

(source_file
  (function_definition
    name: (identifier)
    parameters: (parameter_list)
    result: (primitive_type)
    body: (block
      (return_statement (number)))))

==================|||
Basic module
==================|||

---- MODULE Test ----
increment(n) == n + 1
====

---|||

(source_file
  (module (identifier)
    (operator (identifier)
      (parameter_list (identifier))
      (plus (identifier_ref) (number)))))

These tests are important. They serve as the parser’s API documentation, and they can be run every time you change the grammar to verify that everything still parses correctly.

By default, the tree-sitter test command runs all of the tests in your test/corpus/ folder. To run a particular test, you can use the -f flag:

tree-sitter test -f 'Return statements'

The recommendation is to be comprehensive in adding tests. If it’s a visible node, add it to a test file in your test/corpus directory. It’s typically a good idea to test all of the permutations of each language construct. This increases test coverage, but doubly acquaints readers with a way to examine expected outputs and understand the “edges” of a language.

Attributes

Tests can be annotated with a few attributes. Attributes must be put in the header, below the test name, and start with a :. A couple of attributes also take in a parameter, which require the use of parenthesis.

Note: If you’d like to supply in multiple parameters, e.g. to run tests on multiple platforms or to test multiple languages, you can repeat the attribute on a new line.

The following attributes are available:

Examples using attributes:

=========================
Test that will be skipped
:skip
=========================

int main() {}

-------------------------

====================================
Test that will run on Linux or macOS

:platform(linux)
:platform(macos)
====================================

int main() {}

------------------------------------

========================================================================
Test that expects an error, and will fail fast if there's no parse error
:fail-fast
:error
========================================================================

int main ( {}

------------------------------------------------------------------------

=================================================
Test that will parse with both Typescript and TSX
:language(typescript)
:language(tsx)
=================================================

console.log('Hello, world!');

-------------------------------------------------

Automatic Compilation

You might notice that the first time you run tree-sitter test after regenerating your parser, it takes some extra time. This is because Tree-sitter automatically compiles your C code into a dynamically-loadable library. It recompiles your parser as-needed whenever you update it by re-running tree-sitter generate.

Syntax Highlighting Tests

The tree-sitter test command will also run any syntax highlighting tests in the test/highlight folder, if it exists. For more information about syntax highlighting tests, see the syntax highlighting page.

Command: parse

You can run your parser on an arbitrary file using tree-sitter parse. This will print the resulting the syntax tree, including nodes’ ranges and field names, like this:

(source_file [0, 0] - [3, 0]
  (function_declaration [0, 0] - [2, 1]
    name: (identifier [0, 5] - [0, 9])
    parameters: (parameter_list [0, 9] - [0, 11])
    result: (type_identifier [0, 12] - [0, 15])
    body: (block [0, 16] - [2, 1]
      (return_statement [1, 2] - [1, 10]
        (expression_list [1, 9] - [1, 10]
          (int_literal [1, 9] - [1, 10]))))))

You can pass any number of file paths and glob patterns to tree-sitter parse, and it will parse all of the given files. The command will exit with a non-zero status code if any parse errors occurred. Passing the --cst flag will output a pretty-printed CST instead of the normal S-expression representation. You can also prevent the syntax trees from being printed using the --quiet flag. Additionally, the --stat flag prints out aggregated parse success/failure information for all processed files. This makes tree-sitter parse usable as a secondary testing strategy: you can check that a large number of files parse without error:

tree-sitter parse 'examples/**/*.go' --quiet --stat

Command: highlight

You can run syntax highlighting on an arbitrary file using tree-sitter highlight. This can either output colors directly to your terminal using ansi escape codes, or produce HTML (if the --html flag is passed). For more information, see the syntax highlighting page.

The Grammar DSL

The following is a complete list of built-in functions you can use in your grammar.js to define rules. Use-cases for some of these functions will be explained in more detail in later sections.

In addition to the name and rules fields, grammars have a few other optional public fields that influence the behavior of the parser.

Writing the Grammar

Writing a grammar requires creativity. There are an infinite number of CFGs (context-free grammars) that can be used to describe any given language. In order to produce a good Tree-sitter parser, you need to create a grammar with two important properties:

  1. An intuitive structure - Tree-sitter’s output is a concrete syntax tree; each node in the tree corresponds directly to a terminal or non-terminal symbol in the grammar. So in order to produce an easy-to-analyze tree, there should be a direct correspondence between the symbols in your grammar and the recognizable constructs in the language. This might seem obvious, but it is very different from the way that context-free grammars are often written in contexts like language specifications or Yacc/Bison parsers.

  2. A close adherence to LR(1) - Tree-sitter is based on the GLR parsing algorithm. This means that while it can handle any context-free grammar, it works most efficiently with a class of context-free grammars called LR(1) Grammars. In this respect, Tree-sitter’s grammars are similar to (but less restrictive than) Yacc and Bison grammars, but different from ANTLR grammars, Parsing Expression Grammars, or the ambiguous grammars commonly used in language specifications.

It’s unlikely that you’ll be able to satisfy these two properties just by translating an existing context-free grammar directly into Tree-sitter’s grammar format. There are a few kinds of adjustments that are often required. The following sections will explain these adjustments in more depth.

The First Few Rules

It’s usually a good idea to find a formal specification for the language you’re trying to parse. This specification will most likely contain a context-free grammar. As you read through the rules of this CFG, you will probably discover a complex and cyclic graph of relationships. It might be unclear how you should navigate this graph as you define your grammar.

Although languages have very different constructs, their constructs can often be categorized in to similar groups like Declarations, Definitions, Statements, Expressions, Types, and Patterns. In writing your grammar, a good first step is to create just enough structure to include all of these basic groups of symbols. For a language like Go, you might start with something like this:

{
  // ...

  rules: {
    source_file: $ => repeat($._definition),

    _definition: $ => choice(
      $.function_definition
      // TODO: other kinds of definitions
    ),

    function_definition: $ => seq(
      'func',
      $.identifier,
      $.parameter_list,
      $._type,
      $.block
    ),

    parameter_list: $ => seq(
      '(',
       // TODO: parameters
      ')'
    ),

    _type: $ => choice(
      'bool'
      // TODO: other kinds of types
    ),

    block: $ => seq(
      '{',
      repeat($._statement),
      '}'
    ),

    _statement: $ => choice(
      $.return_statement
      // TODO: other kinds of statements
    ),

    return_statement: $ => seq(
      'return',
      $._expression,
      ';'
    ),

    _expression: $ => choice(
      $.identifier,
      $.number
      // TODO: other kinds of expressions
    ),

    identifier: $ => /[a-z]+/,

    number: $ => /\d+/
  }
}

Some of the details of this grammar will be explained in more depth later on, but if you focus on the TODO comments, you can see that the overall strategy is breadth-first. Notably, this initial skeleton does not need to directly match an exact subset of the context-free grammar in the language specification. It just needs to touch on the major groupings of rules in as simple and obvious a way as possible.

With this structure in place, you can now freely decide what part of the grammar to flesh out next. For example, you might decide to start with types. One-by-one, you could define the rules for writing basic types and composing them into more complex types:

{
  // ...

  _type: $ => choice(
    $.primitive_type,
    $.array_type,
    $.pointer_type
  ),

  primitive_type: $ => choice(
    'bool',
    'int'
  ),

  array_type: $ => seq(
    '[',
    ']',
    $._type
  ),

  pointer_type: $ => seq(
    '*',
    $._type
  )
}

After developing the type sublanguage a bit further, you might decide to switch to working on statements or expressions instead. It’s often useful to check your progress by trying to parse some real code using tree-sitter parse.

And remember to add tests for each rule in your test/corpus folder!

Structuring Rules Well

Imagine that you were just starting work on the Tree-sitter JavaScript parser. Naively, you might try to directly mirror the structure of the ECMAScript Language Spec. To illustrate the problem with this approach, consider the following line of code:

return x + y;

According to the specification, this line is a ReturnStatement, the fragment x + y is an AdditiveExpression, and x and y are both IdentifierReferences. The relationship between these constructs is captured by a complex series of production rules:

ReturnStatement          ->  'return' Expression
Expression               ->  AssignmentExpression
AssignmentExpression     ->  ConditionalExpression
ConditionalExpression    ->  LogicalORExpression
LogicalORExpression      ->  LogicalANDExpression
LogicalANDExpression     ->  BitwiseORExpression
BitwiseORExpression      ->  BitwiseXORExpression
BitwiseXORExpression     ->  BitwiseANDExpression
BitwiseANDExpression     ->  EqualityExpression
EqualityExpression       ->  RelationalExpression
RelationalExpression     ->  ShiftExpression
ShiftExpression          ->  AdditiveExpression
AdditiveExpression       ->  MultiplicativeExpression
MultiplicativeExpression ->  ExponentiationExpression
ExponentiationExpression ->  UnaryExpression
UnaryExpression          ->  UpdateExpression
UpdateExpression         ->  LeftHandSideExpression
LeftHandSideExpression   ->  NewExpression
NewExpression            ->  MemberExpression
MemberExpression         ->  PrimaryExpression
PrimaryExpression        ->  IdentifierReference

The language spec encodes the twenty different precedence levels of JavaScript expressions using twenty levels of indirection between IdentifierReference and Expression. If we were to create a concrete syntax tree representing this statement according to the language spec, it would have twenty levels of nesting, and it would contain nodes with names like BitwiseXORExpression, which are unrelated to the actual code.

Using Precedence

To produce a readable syntax tree, we’d like to model JavaScript expressions using a much flatter structure like this:

{
  // ...

  _expression: $ => choice(
    $.identifier,
    $.unary_expression,
    $.binary_expression,
    // ...
  ),

  unary_expression: $ => choice(
    seq('-', $._expression),
    seq('!', $._expression),
    // ...
  ),

  binary_expression: $ => choice(
    seq($._expression, '*', $._expression),
    seq($._expression, '+', $._expression),
    // ...
  ),
}

Of course, this flat structure is highly ambiguous. If we try to generate a parser, Tree-sitter gives us an error message:

Error: Unresolved conflict for symbol sequence:

  '-'  _expression  •  '*'  …

Possible interpretations:

  1:  '-'  (binary_expression  _expression  •  '*'  _expression)
  2:  (unary_expression  '-'  _expression)  •  '*'  …

Possible resolutions:

  1:  Specify a higher precedence in `binary_expression` than in the other rules.
  2:  Specify a higher precedence in `unary_expression` than in the other rules.
  3:  Specify a left or right associativity in `unary_expression`
  4:  Add a conflict for these rules: `binary_expression` `unary_expression`

Note: The • character in the error message indicates where exactly during parsing the conflict occurs, or in other words, where the parser is encountering ambiguity.

For an expression like -a * b, it’s not clear whether the - operator applies to the a * b or just to the a. This is where the prec function described above comes into play. By wrapping a rule with prec, we can indicate that certain sequence of symbols should bind to each other more tightly than others. For example, the '-', $._expression sequence in unary_expression should bind more tightly than the $._expression, '+', $._expression sequence in binary_expression:

{
  // ...

  unary_expression: $ => prec(2, choice(
    seq('-', $._expression),
    seq('!', $._expression),
    // ...
  ))
}

Using Associativity

Applying a higher precedence in unary_expression fixes that conflict, but there is still another conflict:

Error: Unresolved conflict for symbol sequence:

  _expression  '*'  _expression  •  '*'  …

Possible interpretations:

  1:  _expression  '*'  (binary_expression  _expression  •  '*'  _expression)
  2:  (binary_expression  _expression  '*'  _expression)  •  '*'  …

Possible resolutions:

  1:  Specify a left or right associativity in `binary_expression`
  2:  Add a conflict for these rules: `binary_expression`

For an expression like a * b * c, it’s not clear whether we mean a * (b * c) or (a * b) * c. This is where prec.left and prec.right come into use. We want to select the second interpretation, so we use prec.left.

{
  // ...

  binary_expression: $ => choice(
    prec.left(2, seq($._expression, '*', $._expression)),
    prec.left(1, seq($._expression, '+', $._expression)),
    // ...
  ),
}

Hiding Rules

You may have noticed in the above examples that some of the grammar rule name like _expression and _type began with an underscore. Starting a rule’s name with an underscore causes the rule to be hidden in the syntax tree. This is useful for rules like _expression in the grammars above, which always just wrap a single child node. If these nodes were not hidden, they would add substantial depth and noise to the syntax tree without making it any easier to understand.

Using Fields

Often, it’s easier to analyze a syntax node if you can refer to its children by name instead of by their position in an ordered list. Tree-sitter grammars support this using the field function. This function allows you to assign unique names to some or all of a node’s children:

function_definition: $ => seq(
  'func',
  field('name', $.identifier),
  field('parameters', $.parameter_list),
  field('return_type', $._type),
  field('body', $.block)
)

Adding fields like this allows you to retrieve nodes using the field APIs.

Lexical Analysis

Tree-sitter’s parsing process is divided into two phases: parsing (which is described above) and lexing - the process of grouping individual characters into the language’s fundamental tokens. There are a few important things to know about how Tree-sitter’s lexing works.

Conflicting Tokens

Grammars often contain multiple tokens that can match the same characters. For example, a grammar might contain the tokens ("if" and /[a-z]+/). Tree-sitter differentiates between these conflicting tokens in a few ways.

  1. Context-aware Lexing - Tree-sitter performs lexing on-demand, during the parsing process. At any given position in a source document, the lexer only tries to recognize tokens that are valid at that position in the document.

  2. Lexical Precedence - When the precedence functions described above are used within the token function, the given explicit precedence values serve as instructions to the lexer. If there are two valid tokens that match the characters at a given position in the document, Tree-sitter will select the one with the higher precedence.

  3. Match Length - If multiple valid tokens with the same precedence match the characters at a given position in a document, Tree-sitter will select the token that matches the longest sequence of characters.

  4. Match Specificity - If there are two valid tokens with the same precedence and which both match the same number of characters, Tree-sitter will prefer a token that is specified in the grammar as a String over a token specified as a RegExp.

  5. Rule Order - If none of the above criteria can be used to select one token over another, Tree-sitter will prefer the token that appears earlier in the grammar.

If there is an external scanner it may have an additional impact over regular tokens defined in the grammar.

Lexical Precedence vs. Parse Precedence

One common mistake involves not distinguishing lexical precedence from parse precedence. Parse precedence determines which rule is chosen to interpret a given sequence of tokens. Lexical precedence determines which token is chosen to interpret at a given position of text and it is a lower-level operation that is done first. The above list fully captures Tree-sitter’s lexical precedence rules, and you will probably refer back to this section of the documentation more often than any other. Most of the time when you really get stuck, you’re dealing with a lexical precedence problem. Pay particular attention to the difference in meaning between using prec inside of the token function versus outside of it. The lexical precedence syntax is token(prec(N, ...)).

Keywords

Many languages have a set of keyword tokens (e.g. if, for, return), as well as a more general token (e.g. identifier) that matches any word, including many of the keyword strings. For example, JavaScript has a keyword instanceof, which is used as a binary operator, like this:

if (a instanceof Something) b();

The following, however, is not valid JavaScript:

if (a instanceofSomething) b();

A keyword like instanceof cannot be followed immediately by another letter, because then it would be tokenized as an identifier, even though an identifier is not valid at that position. Because Tree-sitter uses context-aware lexing, as described above, it would not normally impose this restriction. By default, Tree-sitter would recognize instanceofSomething as two separate tokens: the instanceof keyword followed by an identifier.

Keyword Extraction

Fortunately, Tree-sitter has a feature that allows you to fix this, so that you can match the behavior of other standard parsers: the word token. If you specify a word token in your grammar, Tree-sitter will find the set of keyword tokens that match strings also matched by the word token. Then, during lexing, instead of matching each of these keywords individually, Tree-sitter will match the keywords via a two-step process where it first matches the word token.

For example, suppose we added identifier as the word token in our JavaScript grammar:

grammar({
  name: 'javascript',

  word: $ => $.identifier,

  rules: {
    _expression: $ => choice(
      $.identifier,
      $.unary_expression,
      $.binary_expression
      // ...
    ),

    binary_expression: $ => choice(
      prec.left(1, seq($._expression, 'instanceof', $._expression))
      // ...
    ),

    unary_expression: $ => choice(
      prec.left(2, seq('typeof', $._expression))
      // ...
    ),

    identifier: $ => /[a-z_]+/
  }
});

Tree-sitter would identify typeof and instanceof as keywords. Then, when parsing the invalid code above, rather than scanning for the instanceof token individually, it would scan for an identifier first, and find instanceofSomething. It would then correctly recognize the code as invalid.

Aside from improving error detection, keyword extraction also has performance benefits. It allows Tree-sitter to generate a smaller, simpler lexing function, which means that the parser will compile much more quickly.

External Scanners

Many languages have some tokens whose structure is impossible or inconvenient to describe with a regular expression. Some examples:

Tree-sitter allows you to handle these kinds of tokens using external scanners. An external scanner is a set of C functions that you, the grammar author, can write by hand in order to add custom logic for recognizing certain tokens.

To use an external scanner, there are a few steps. First, add an externals section to your grammar. This section should list the names of all of your external tokens. These names can then be used elsewhere in your grammar.

grammar({
  name: 'my_language',

  externals: $ => [
    $.indent,
    $.dedent,
    $.newline
  ],

  // ...
});

Then, add another C source file to your project. Currently, its path must be src/scanner.c for the CLI to recognize it. Be sure to add this file to the sources section of your binding.gyp file so that it will be included when your project is compiled by Node.js and uncomment the appropriate block in your bindings/rust/build.rs file so that it will be included in your Rust crate.

In this new source file, define an enum type containing the names of all of your external tokens. The ordering of this enum must match the order in your grammar’s externals array; the actual names do not matter.

#include "tree_sitter/parser.h"
#include "tree_sitter/alloc.h"
#include "tree_sitter/array.h"

enum TokenType {
  INDENT,
  DEDENT,
  NEWLINE
}

Finally, you must define five functions with specific names, based on your language’s name and five actions: create, destroy, serialize, deserialize, and scan.

Create

void *tree_sitter_my_language_external_scanner_create(void) {
  // ...
}

This function should create your scanner object. It will only be called once anytime your language is set on a parser. Often, you will want to allocate memory on the heap and return a pointer to it. If your external scanner doesn’t need to maintain any state, it’s ok to return NULL.

Destroy

void tree_sitter_my_language_external_scanner_destroy(void *payload) {
  // ...
}

This function should free any memory used by your scanner. It is called once when a parser is deleted or assigned a different language. It receives as an argument the same pointer that was returned from the create function. If your create function didn’t allocate any memory, this function can be a noop.

Serialize

unsigned tree_sitter_my_language_external_scanner_serialize(
  void *payload,
  char *buffer
) {
  // ...
}

This function should copy the complete state of your scanner into a given byte buffer, and return the number of bytes written. The function is called every time the external scanner successfully recognizes a token. It receives a pointer to your scanner and a pointer to a buffer. The maximum number of bytes that you can write is given by the TREE_SITTER_SERIALIZATION_BUFFER_SIZE constant, defined in the tree_sitter/parser.h header file.

The data that this function writes will ultimately be stored in the syntax tree so that the scanner can be restored to the right state when handling edits or ambiguities. For your parser to work correctly, the serialize function must store its entire state, and deserialize must restore the entire state. For good performance, you should design your scanner so that its state can be serialized as quickly and compactly as possible.

Deserialize

void tree_sitter_my_language_external_scanner_deserialize(
  void *payload,
  const char *buffer,
  unsigned length
) {
  // ...
}

This function should restore the state of your scanner based the bytes that were previously written by the serialize function. It is called with a pointer to your scanner, a pointer to the buffer of bytes, and the number of bytes that should be read. It is good practice to explicitly erase your scanner state variables at the start of this function, before restoring their values from the byte buffer.

Scan

bool tree_sitter_my_language_external_scanner_scan(
  void *payload,
  TSLexer *lexer,
  const bool *valid_symbols
) {
  // ...
}

This function is responsible for recognizing external tokens. It should return true if a token was recognized, and false otherwise. It is called with a “lexer” struct with the following fields:

The third argument to the scan function is an array of booleans that indicates which of external tokens are currently expected by the parser. You should only look for a given token if it is valid according to this array. At the same time, you cannot backtrack, so you may need to combine certain pieces of logic.

if (valid_symbols[INDENT] || valid_symbols[DEDENT]) {

  // ... logic that is common to both `INDENT` and `DEDENT`

  if (valid_symbols[INDENT]) {

    // ... logic that is specific to `INDENT`

    lexer->result_symbol = INDENT;
    return true;
  }
}

External Scanner Helpers

Allocator

Instead of using libc’s malloc, calloc, realloc, and free, you should use the versions prefixed with ts_ from tree_sitter/alloc.h. These macros can allow a potential consumer to override the default allocator with their own implementation, but by default will use the libc functions.

As a consumer of the tree-sitter core library as well as any parser libraries that might use allocations, you can enable overriding the default allocator and have it use the same one as the library allocator, of which you can set with ts_set_allocator. To enable this overriding in scanners, you must compile them with the TREE_SITTER_REUSE_ALLOCATOR macro defined, and tree-sitter the library must be linked into your final app dynamically, since it needs to resolve the internal functions at runtime. If you are compiling an executable binary that uses the core library, but want to load parsers dynamically at runtime, then you will have to use a special linker flag on Unix. For non-Darwin systems, that would be --dynamic-list and for Darwin systems, that would be -exported_symbols_list. The CLI does exactly this, so you can use it as a reference (check out cli/build.rs).

For example, assuming you wanted to allocate 100 bytes for your scanner, you’d do so like the following example:

#include "tree_sitter/parser.h"
#include "tree_sitter/alloc.h"

// ...

void *tree_sitter_my_language_external_scanner_create(void) {
  return ts_calloc(100, 1); // or ts_malloc(100)
}

// ...
Arrays

If you need to use array-like types in your scanner, such as tracking a stack of indentations or tags, you should use the array macros from tree_sitter/array.h.

There are quite a few of them provided for you, but here’s how you could get started tracking some . Check out the header itself for more detailed documentation.

NOTE: Do not use any of the array functions or macros that are prefixed with an underscore and have comments saying that it is not what you are looking for. These are internal functions used as helpers by other macros that are public. They are not meant to be used directly, nor are they what you want.

#include "tree_sitter/parser.h"
#include "tree_sitter/array.h"

enum TokenType {
  INDENT,
  DEDENT,
  NEWLINE,
  STRING,
}

// Create the array in your create function

void *tree_sitter_my_language_external_scanner_create(void) {
  return ts_calloc(1, sizeof(Array(int)));

  // or if you want to zero out the memory yourself

  Array(int) *stack = ts_malloc(sizeof(Array(int)));
  array_init(&stack);
  return stack;
}

bool tree_sitter_my_language_external_scanner_scan(
  void *payload,
  TSLexer *lexer,
  const bool *valid_symbols
) {
  Array(int) *stack = payload;
  if (valid_symbols[INDENT]) {
    array_push(stack, lexer->get_column(lexer));
    lexer->result_symbol = INDENT;
    return true;
  }
  if (valid_symbols[DEDENT]) {
    array_pop(stack); // this returns the popped element by value, but we don't need it
    lexer->result_symbol = DEDENT;
    return true;
  }

  // we can also use an array on the stack to keep track of a string

  Array(char) next_string = array_new();

  if (valid_symbols[STRING] && lexer->lookahead == '"') {
    lexer->advance(lexer, false);
    while (lexer->lookahead != '"' && lexer->lookahead != '\n' && !lexer->eof(lexer)) {
      array_push(&next_string, lexer->lookahead);
      lexer->advance(lexer, false);
    }

    // assume we have some arbitrary constraint of not having more than 100 characters in a string
    if (lexer->lookahead == '"' && next_string.size <= 100) {
      lexer->advance(lexer, false);
      lexer->result_symbol = STRING;
      return true;
    }
  }

  return false;
}

Other External Scanner Details

If a token in the externals array is valid at a given position in the parse, the external scanner will be called first before anything else is done. This means the external scanner functions as a powerful override of Tree-sitter’s lexing behavior, and can be used to solve problems that can’t be cracked with ordinary lexical, parse, or dynamic precedence.

If a syntax error is encountered during regular parsing, Tree-sitter’s first action during error recovery will be to call the external scanner’s scan function with all tokens marked valid. The scanner should detect this case and handle it appropriately. One simple method of detection is to add an unused token to the end of the externals array, for example externals: $ => [$.token1, $.token2, $.error_sentinel], then check whether that token is marked valid to determine whether Tree-sitter is in error correction mode.

If you put terminal keywords in the externals array, for example externals: $ => ['if', 'then', 'else'], then any time those terminals are present in the grammar they will be tokenized by the external scanner. It is similar to writing externals: [$.if_keyword, $.then_keyword, $.else_keyword] then using alias($.if_keyword, 'if') in the grammar.

If in the externals array use literal keywords then lexing works in two steps, the external scanner will be called first and if it sets a resulting token and returns true then the token considered as recognized and Tree-sitter moves to a next token. But the external scanner may return false and in this case Tree-sitter fallbacks to the internal lexing mechanism.

In case of some keywords defined in the externals array in a rule referencing form like $.if_keyword and there is no additional definition of that rule in the grammar rules, e.g., if_keyword: $ => 'if' then fallback to the internal lexer isn’t possible because Tree-sitter doesn’t know the actual keyword and it’s fully the external scanner resposibilty to recognize such tokens.

External scanners are a common cause of infinite loops. Be very careful when emitting zero-width tokens from your external scanner, and if you consume characters in a loop be sure use the eof function to check whether you are at the end of the file.