Included Ranges

Parser.setIncludedRanges restricts lexing to byte/point ranges (editor visible regions, injections) while nodes keep their original document coordinates.

try parser.setIncludedRanges(&.{.{
    .start_byte = 120,
    .end_byte = 340,
    .start_point = .{ .row = 8, .column = 0 },
    .end_point = .{ .row = 20, .column = 5 },
}});

Semantics

• Ranges must be sorted by start_byte and non-overlapping, otherwise error.InvalidRange. An empty slice clears the restriction.
• The tokenizer skips excluded gaps silently: gaps never produce tokens and never produce errors.
• A token must lie within a single range; tokens cannot span gaps.
• Positions stay exact (points advance through skipped bytes), including in streaming mode.
• Incremental reuse and queries operate on the ranged tree normally.

Complete example

try parser.setIncludedRanges(&.{.{
    .start_byte = 0,
    .end_byte = 5,
    .start_point = .{},
    .end_point = .{ .row = 0, .column = 5 },
}});
var tree = try parser.parseString("1 + 2 + 3");
defer tree.deinit();
// root covers only the included prefix:
// root.text() == "1 + 2", endByte() == 5, hasError() == false

API used

• Parser — setIncludedRanges, includedRanges; Range.

Related guides

• Parsing Source, Custom Input