Crate gdck_syntax 

Lossless lexing and parsing for GDScript.

The tree this crate produces keeps every byte of the input, including whitespace, comments and blank lines. That is a deliberate constraint rather than an implementation detail: a formatter that rewrites one declaration has to leave the comments around it exactly where they were, and a linter that reports a problem has to point at a real byte offset.

Parsing never fails. Malformed input produces SyntaxKind::Error nodes and diagnostics on SyntaxTree::errors, so tools can report several problems per run and editors can work with a half-typed buffer.

Example

let tree = gdck_syntax::parse("func _ready() -> void:\n\tpass\n");
assert!(!tree.has_errors());
// The source is always recoverable from the tree.
assert_eq!(tree.text(), "func _ready() -> void:\n\tpass\n");

Structs

Checkpoint

A position in the child buffer that a node can later be opened at.

Descendants

Pre-order iterator over a subtree. See SyntaxNode::descendants.

LexResult

The result of lexing a source file.

LineCol

A 1-based line and column pair, for human-facing diagnostics.

LineIndex

Maps byte offsets to line/column positions.

NodeId

An index into a SyntaxTree’s node arena.

SyntaxError

A problem found while turning source text into a tree.

SyntaxNode

A borrowed handle to one node in a SyntaxTree.

SyntaxTree

A parsed GDScript file: the source text plus its tree and diagnostics.

TextRange

A half-open byte range into the source text.

Token

A lexed token: a kind plus the span of source it covers.

TreeBuilder

Builds a SyntaxTree as the parser walks the token stream.

Enums

Element

A child of a node: either a nested node or a leaf token.

SyntaxKind

A syntactic category. Values below SyntaxKind::FIRST_NODE are tokens.

Functions

parse

Parse GDScript source into a lossless tree.

tokenize

Tokenize source.