core/text/regex/virtual_machine

regex_vm

Types

Jump :: struct #packed {
	opcode: Opcode,
	target: i16,
}

Source

Machine :: struct {
	// Program state
	memory:            string,
	class_data:        []Rune_Class_Data,
	code:              Program,
	// Thread state
	top_thread:        int,
	threads:           [^]Thread,
	next_threads:      [^]Thread,
	// The busy map is used to merge threads based on their program counters.
	busy_map:          []u64,
	// Global state
	string_pointer:    int,
	current_rune:      rune,
	current_rune_size: int,
	next_rune:         rune,
	next_rune_size:    int,
	last_rune:         rune,
}

Source

Opcode :: enum u8 {
	// | [ operands ]
	Match                       = 0, // |
	Match_And_Exit              = 1, // |
	Byte                        = 2, // | u8
	Rune                        = 3, // | i32
	Rune_Class                  = 4, // | u8
	Rune_Class_Negated          = 5, // | u8
	Wildcard                    = 6, // |
	Jump                        = 7, // | u16
	Split                       = 8, // | u16, u16
	Save                        = 9, // | u8
	Assert_Start                = 10, // |
	Assert_Start_Multiline      = 11, // |
	Assert_End                  = 12, // |
	Assert_Word_Boundary        = 13, // |
	Assert_Non_Word_Boundary    = 14, // |
	Multiline_Open              = 15, // |
	Multiline_Close             = 16, // |
	Wait_For_Byte               = 17, // | u8
	Wait_For_Rune               = 18, // | i32
	Wait_For_Rune_Class         = 19, // | u8
	Wait_For_Rune_Class_Negated = 20, // | u8
	Match_All_And_Escape        = 21, // |
}

Source

Opcode_Iterator :: struct {
	code: Program,
	pc:   int,
}

Source

Made available under Odin's license.

	List of contributors:
		Feoramund: Initial implementation.

Program :: []OpcodeSource

Rune_Class_Data :: struct {
	runes:  []rune,
	ranges: []Rune_Class_Range,
}

Source

NOTE: This structure differs intentionally from the one in regex/parser, as this data doesn't need to be a dynamic array once it hits the VM.

Save :: struct #packed {
	opcode:  Opcode,
	operand: Opcode,
}

Source

Split :: struct #packed {
	opcode: Opcode,
	left:   i16,
	right:  i16,
}

Source

Thread :: struct {
	pc:    int,
	saved: ^[2 * common.MAX_CAPTURE_GROUPS]int,
}

Source

Wait_For_Byte :: struct #packed {
	opcode:  Opcode,
	operand: Opcode,
}

Source

Wait_For_Rune :: struct #packed {
	opcode:  Opcode,
	operand: rune,
}

Source

Wait_For_Rune_Class :: struct #packed {
	opcode:  Opcode,
	operand: Opcode,
}

Source

Wait_For_Rune_Class_Negated :: struct #packed {
	opcode:  Opcode,
	operand: Opcode,
}

Source

Constants

Rune_Class_Range :: parser.Rune_Class_RangeSource

Procedures

add_thread :: proc(vm: ^Machine, saved: ^[2 * common.MAX_CAPTURE_GROUPS]int, pc: int)Source

check_busy_map :: proc(vm: ^Machine, pc: int) -> (bool)Source

create :: proc(code: Program, str: string, allocator: mem.Allocator = context.allocator) -> (vm: Machine)Source

destroy :: proc(vm: Machine, allocator: mem.Allocator = context.allocator)Source

is_word_class :: proc(r: rune) -> (bool)Source

@MetaCharacter NOTE: This must be kept in sync with the compiler & tokenizer.

iterate_opcodes :: proc(iter: ^Opcode_Iterator) -> (opcode: Opcode, pc: int, ok: bool)Source

opcode_count :: proc(code: Program) -> (opcodes: int)Source

opcode_to_name :: proc(opcode: Opcode) -> (str: string)Source

run :: proc(vm: ^Machine, UNICODE_MODE: bool) -> (saved: ^[2 * common.MAX_CAPTURE_GROUPS]int, ok: bool)Source

set_busy_map :: proc(vm: ^Machine, pc: int) -> (bool)Source

regex_vm

Types

Constants

Procedures

Find anything

Settings