Edit on GitHub

sqlglot.tokens

   1from __future__ import annotations
   2
   3import os
   4import typing as t
   5from enum import auto
   6
   7from sqlglot.errors import SqlglotError, TokenError
   8from sqlglot.helper import AutoName
   9from sqlglot.trie import TrieResult, in_trie, new_trie
  10
  11if t.TYPE_CHECKING:
  12    from sqlglot.dialects.dialect import DialectType
  13
  14
  15try:
  16    from sqlglotrs import (  # type: ignore
  17        Tokenizer as RsTokenizer,
  18        TokenizerDialectSettings as RsTokenizerDialectSettings,
  19        TokenizerSettings as RsTokenizerSettings,
  20        TokenTypeSettings as RsTokenTypeSettings,
  21    )
  22
  23    USE_RS_TOKENIZER = os.environ.get("SQLGLOTRS_TOKENIZER", "1") == "1"
  24except ImportError:
  25    USE_RS_TOKENIZER = False
  26
  27
  28class TokenType(AutoName):
  29    L_PAREN = auto()
  30    R_PAREN = auto()
  31    L_BRACKET = auto()
  32    R_BRACKET = auto()
  33    L_BRACE = auto()
  34    R_BRACE = auto()
  35    COMMA = auto()
  36    DOT = auto()
  37    DASH = auto()
  38    PLUS = auto()
  39    COLON = auto()
  40    DCOLON = auto()
  41    DQMARK = auto()
  42    SEMICOLON = auto()
  43    STAR = auto()
  44    BACKSLASH = auto()
  45    SLASH = auto()
  46    LT = auto()
  47    LTE = auto()
  48    GT = auto()
  49    GTE = auto()
  50    NOT = auto()
  51    EQ = auto()
  52    NEQ = auto()
  53    NULLSAFE_EQ = auto()
  54    COLON_EQ = auto()
  55    AND = auto()
  56    OR = auto()
  57    AMP = auto()
  58    DPIPE = auto()
  59    PIPE = auto()
  60    PIPE_SLASH = auto()
  61    DPIPE_SLASH = auto()
  62    CARET = auto()
  63    TILDA = auto()
  64    ARROW = auto()
  65    DARROW = auto()
  66    FARROW = auto()
  67    HASH = auto()
  68    HASH_ARROW = auto()
  69    DHASH_ARROW = auto()
  70    LR_ARROW = auto()
  71    DAT = auto()
  72    LT_AT = auto()
  73    AT_GT = auto()
  74    DOLLAR = auto()
  75    PARAMETER = auto()
  76    SESSION_PARAMETER = auto()
  77    DAMP = auto()
  78    XOR = auto()
  79    DSTAR = auto()
  80
  81    BLOCK_START = auto()
  82    BLOCK_END = auto()
  83
  84    SPACE = auto()
  85    BREAK = auto()
  86
  87    STRING = auto()
  88    NUMBER = auto()
  89    IDENTIFIER = auto()
  90    DATABASE = auto()
  91    COLUMN = auto()
  92    COLUMN_DEF = auto()
  93    SCHEMA = auto()
  94    TABLE = auto()
  95    VAR = auto()
  96    BIT_STRING = auto()
  97    HEX_STRING = auto()
  98    BYTE_STRING = auto()
  99    NATIONAL_STRING = auto()
 100    RAW_STRING = auto()
 101    HEREDOC_STRING = auto()
 102    UNICODE_STRING = auto()
 103
 104    # types
 105    BIT = auto()
 106    BOOLEAN = auto()
 107    TINYINT = auto()
 108    UTINYINT = auto()
 109    SMALLINT = auto()
 110    USMALLINT = auto()
 111    MEDIUMINT = auto()
 112    UMEDIUMINT = auto()
 113    INT = auto()
 114    UINT = auto()
 115    BIGINT = auto()
 116    UBIGINT = auto()
 117    INT128 = auto()
 118    UINT128 = auto()
 119    INT256 = auto()
 120    UINT256 = auto()
 121    FLOAT = auto()
 122    DOUBLE = auto()
 123    DECIMAL = auto()
 124    UDECIMAL = auto()
 125    BIGDECIMAL = auto()
 126    CHAR = auto()
 127    NCHAR = auto()
 128    VARCHAR = auto()
 129    NVARCHAR = auto()
 130    BPCHAR = auto()
 131    TEXT = auto()
 132    MEDIUMTEXT = auto()
 133    LONGTEXT = auto()
 134    MEDIUMBLOB = auto()
 135    LONGBLOB = auto()
 136    TINYBLOB = auto()
 137    TINYTEXT = auto()
 138    NAME = auto()
 139    BINARY = auto()
 140    VARBINARY = auto()
 141    JSON = auto()
 142    JSONB = auto()
 143    TIME = auto()
 144    TIMETZ = auto()
 145    TIMESTAMP = auto()
 146    TIMESTAMPTZ = auto()
 147    TIMESTAMPLTZ = auto()
 148    TIMESTAMP_S = auto()
 149    TIMESTAMP_MS = auto()
 150    TIMESTAMP_NS = auto()
 151    DATETIME = auto()
 152    DATETIME64 = auto()
 153    DATE = auto()
 154    DATE32 = auto()
 155    INT4RANGE = auto()
 156    INT4MULTIRANGE = auto()
 157    INT8RANGE = auto()
 158    INT8MULTIRANGE = auto()
 159    NUMRANGE = auto()
 160    NUMMULTIRANGE = auto()
 161    TSRANGE = auto()
 162    TSMULTIRANGE = auto()
 163    TSTZRANGE = auto()
 164    TSTZMULTIRANGE = auto()
 165    DATERANGE = auto()
 166    DATEMULTIRANGE = auto()
 167    UUID = auto()
 168    GEOGRAPHY = auto()
 169    NULLABLE = auto()
 170    GEOMETRY = auto()
 171    HLLSKETCH = auto()
 172    HSTORE = auto()
 173    SUPER = auto()
 174    SERIAL = auto()
 175    SMALLSERIAL = auto()
 176    BIGSERIAL = auto()
 177    XML = auto()
 178    YEAR = auto()
 179    UNIQUEIDENTIFIER = auto()
 180    USERDEFINED = auto()
 181    MONEY = auto()
 182    SMALLMONEY = auto()
 183    ROWVERSION = auto()
 184    IMAGE = auto()
 185    VARIANT = auto()
 186    OBJECT = auto()
 187    INET = auto()
 188    IPADDRESS = auto()
 189    IPPREFIX = auto()
 190    IPV4 = auto()
 191    IPV6 = auto()
 192    ENUM = auto()
 193    ENUM8 = auto()
 194    ENUM16 = auto()
 195    FIXEDSTRING = auto()
 196    LOWCARDINALITY = auto()
 197    NESTED = auto()
 198    AGGREGATEFUNCTION = auto()
 199    SIMPLEAGGREGATEFUNCTION = auto()
 200    UNKNOWN = auto()
 201
 202    # keywords
 203    ALIAS = auto()
 204    ALTER = auto()
 205    ALWAYS = auto()
 206    ALL = auto()
 207    ANTI = auto()
 208    ANY = auto()
 209    APPLY = auto()
 210    ARRAY = auto()
 211    ASC = auto()
 212    ASOF = auto()
 213    AUTO_INCREMENT = auto()
 214    BEGIN = auto()
 215    BETWEEN = auto()
 216    CACHE = auto()
 217    CASE = auto()
 218    CHARACTER_SET = auto()
 219    CLUSTER_BY = auto()
 220    COLLATE = auto()
 221    COMMAND = auto()
 222    COMMENT = auto()
 223    COMMIT = auto()
 224    CONNECT_BY = auto()
 225    CONSTRAINT = auto()
 226    CREATE = auto()
 227    CROSS = auto()
 228    CUBE = auto()
 229    CURRENT_DATE = auto()
 230    CURRENT_DATETIME = auto()
 231    CURRENT_TIME = auto()
 232    CURRENT_TIMESTAMP = auto()
 233    CURRENT_USER = auto()
 234    DEFAULT = auto()
 235    DELETE = auto()
 236    DESC = auto()
 237    DESCRIBE = auto()
 238    DICTIONARY = auto()
 239    DISTINCT = auto()
 240    DISTRIBUTE_BY = auto()
 241    DIV = auto()
 242    DROP = auto()
 243    ELSE = auto()
 244    END = auto()
 245    ESCAPE = auto()
 246    EXCEPT = auto()
 247    EXECUTE = auto()
 248    EXISTS = auto()
 249    FALSE = auto()
 250    FETCH = auto()
 251    FILTER = auto()
 252    FINAL = auto()
 253    FIRST = auto()
 254    FOR = auto()
 255    FORCE = auto()
 256    FOREIGN_KEY = auto()
 257    FORMAT = auto()
 258    FROM = auto()
 259    FULL = auto()
 260    FUNCTION = auto()
 261    GLOB = auto()
 262    GLOBAL = auto()
 263    GROUP_BY = auto()
 264    GROUPING_SETS = auto()
 265    HAVING = auto()
 266    HINT = auto()
 267    IGNORE = auto()
 268    ILIKE = auto()
 269    ILIKE_ANY = auto()
 270    IN = auto()
 271    INDEX = auto()
 272    INNER = auto()
 273    INSERT = auto()
 274    INTERSECT = auto()
 275    INTERVAL = auto()
 276    INTO = auto()
 277    INTRODUCER = auto()
 278    IRLIKE = auto()
 279    IS = auto()
 280    ISNULL = auto()
 281    JOIN = auto()
 282    JOIN_MARKER = auto()
 283    KEEP = auto()
 284    KILL = auto()
 285    LANGUAGE = auto()
 286    LATERAL = auto()
 287    LEFT = auto()
 288    LIKE = auto()
 289    LIKE_ANY = auto()
 290    LIMIT = auto()
 291    LOAD = auto()
 292    LOCK = auto()
 293    MAP = auto()
 294    MATCH_RECOGNIZE = auto()
 295    MEMBER_OF = auto()
 296    MERGE = auto()
 297    MOD = auto()
 298    MODEL = auto()
 299    NATURAL = auto()
 300    NEXT = auto()
 301    NOTNULL = auto()
 302    NULL = auto()
 303    OBJECT_IDENTIFIER = auto()
 304    OFFSET = auto()
 305    ON = auto()
 306    ONLY = auto()
 307    OPERATOR = auto()
 308    ORDER_BY = auto()
 309    ORDER_SIBLINGS_BY = auto()
 310    ORDERED = auto()
 311    ORDINALITY = auto()
 312    OUTER = auto()
 313    OVER = auto()
 314    OVERLAPS = auto()
 315    OVERWRITE = auto()
 316    PARTITION = auto()
 317    PARTITION_BY = auto()
 318    PERCENT = auto()
 319    PIVOT = auto()
 320    PLACEHOLDER = auto()
 321    POSITIONAL = auto()
 322    PRAGMA = auto()
 323    PREWHERE = auto()
 324    PRIMARY_KEY = auto()
 325    PROCEDURE = auto()
 326    PROPERTIES = auto()
 327    PSEUDO_TYPE = auto()
 328    QUALIFY = auto()
 329    QUOTE = auto()
 330    RANGE = auto()
 331    RECURSIVE = auto()
 332    REFRESH = auto()
 333    REPLACE = auto()
 334    RETURNING = auto()
 335    REFERENCES = auto()
 336    RIGHT = auto()
 337    RLIKE = auto()
 338    ROLLBACK = auto()
 339    ROLLUP = auto()
 340    ROW = auto()
 341    ROWS = auto()
 342    SELECT = auto()
 343    SEMI = auto()
 344    SEPARATOR = auto()
 345    SEQUENCE = auto()
 346    SERDE_PROPERTIES = auto()
 347    SET = auto()
 348    SETTINGS = auto()
 349    SHOW = auto()
 350    SIMILAR_TO = auto()
 351    SOME = auto()
 352    SORT_BY = auto()
 353    START_WITH = auto()
 354    STORAGE_INTEGRATION = auto()
 355    STRUCT = auto()
 356    TABLE_SAMPLE = auto()
 357    TEMPORARY = auto()
 358    TOP = auto()
 359    THEN = auto()
 360    TRUE = auto()
 361    TRUNCATE = auto()
 362    UNCACHE = auto()
 363    UNION = auto()
 364    UNNEST = auto()
 365    UNPIVOT = auto()
 366    UPDATE = auto()
 367    USE = auto()
 368    USING = auto()
 369    VALUES = auto()
 370    VIEW = auto()
 371    VOLATILE = auto()
 372    WHEN = auto()
 373    WHERE = auto()
 374    WINDOW = auto()
 375    WITH = auto()
 376    UNIQUE = auto()
 377    VERSION_SNAPSHOT = auto()
 378    TIMESTAMP_SNAPSHOT = auto()
 379    OPTION = auto()
 380
 381
 382_ALL_TOKEN_TYPES = list(TokenType)
 383_TOKEN_TYPE_TO_INDEX = {token_type: i for i, token_type in enumerate(_ALL_TOKEN_TYPES)}
 384
 385
 386class Token:
 387    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
 388
 389    @classmethod
 390    def number(cls, number: int) -> Token:
 391        """Returns a NUMBER token with `number` as its text."""
 392        return cls(TokenType.NUMBER, str(number))
 393
 394    @classmethod
 395    def string(cls, string: str) -> Token:
 396        """Returns a STRING token with `string` as its text."""
 397        return cls(TokenType.STRING, string)
 398
 399    @classmethod
 400    def identifier(cls, identifier: str) -> Token:
 401        """Returns an IDENTIFIER token with `identifier` as its text."""
 402        return cls(TokenType.IDENTIFIER, identifier)
 403
 404    @classmethod
 405    def var(cls, var: str) -> Token:
 406        """Returns an VAR token with `var` as its text."""
 407        return cls(TokenType.VAR, var)
 408
 409    def __init__(
 410        self,
 411        token_type: TokenType,
 412        text: str,
 413        line: int = 1,
 414        col: int = 1,
 415        start: int = 0,
 416        end: int = 0,
 417        comments: t.Optional[t.List[str]] = None,
 418    ) -> None:
 419        """Token initializer.
 420
 421        Args:
 422            token_type: The TokenType Enum.
 423            text: The text of the token.
 424            line: The line that the token ends on.
 425            col: The column that the token ends on.
 426            start: The start index of the token.
 427            end: The ending index of the token.
 428            comments: The comments to attach to the token.
 429        """
 430        self.token_type = token_type
 431        self.text = text
 432        self.line = line
 433        self.col = col
 434        self.start = start
 435        self.end = end
 436        self.comments = [] if comments is None else comments
 437
 438    def __repr__(self) -> str:
 439        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
 440        return f"<Token {attributes}>"
 441
 442
 443class _Tokenizer(type):
 444    def __new__(cls, clsname, bases, attrs):
 445        klass = super().__new__(cls, clsname, bases, attrs)
 446
 447        def _convert_quotes(arr: t.List[str | t.Tuple[str, str]]) -> t.Dict[str, str]:
 448            return dict(
 449                (item, item) if isinstance(item, str) else (item[0], item[1]) for item in arr
 450            )
 451
 452        def _quotes_to_format(
 453            token_type: TokenType, arr: t.List[str | t.Tuple[str, str]]
 454        ) -> t.Dict[str, t.Tuple[str, TokenType]]:
 455            return {k: (v, token_type) for k, v in _convert_quotes(arr).items()}
 456
 457        klass._QUOTES = _convert_quotes(klass.QUOTES)
 458        klass._IDENTIFIERS = _convert_quotes(klass.IDENTIFIERS)
 459
 460        klass._FORMAT_STRINGS = {
 461            **{
 462                p + s: (e, TokenType.NATIONAL_STRING)
 463                for s, e in klass._QUOTES.items()
 464                for p in ("n", "N")
 465            },
 466            **_quotes_to_format(TokenType.BIT_STRING, klass.BIT_STRINGS),
 467            **_quotes_to_format(TokenType.BYTE_STRING, klass.BYTE_STRINGS),
 468            **_quotes_to_format(TokenType.HEX_STRING, klass.HEX_STRINGS),
 469            **_quotes_to_format(TokenType.RAW_STRING, klass.RAW_STRINGS),
 470            **_quotes_to_format(TokenType.HEREDOC_STRING, klass.HEREDOC_STRINGS),
 471            **_quotes_to_format(TokenType.UNICODE_STRING, klass.UNICODE_STRINGS),
 472        }
 473
 474        klass._STRING_ESCAPES = set(klass.STRING_ESCAPES)
 475        klass._IDENTIFIER_ESCAPES = set(klass.IDENTIFIER_ESCAPES)
 476        klass._COMMENTS = {
 477            **dict(
 478                (comment, None) if isinstance(comment, str) else (comment[0], comment[1])
 479                for comment in klass.COMMENTS
 480            ),
 481            "{#": "#}",  # Ensure Jinja comments are tokenized correctly in all dialects
 482        }
 483
 484        klass._KEYWORD_TRIE = new_trie(
 485            key.upper()
 486            for key in (
 487                *klass.KEYWORDS,
 488                *klass._COMMENTS,
 489                *klass._QUOTES,
 490                *klass._FORMAT_STRINGS,
 491            )
 492            if " " in key or any(single in key for single in klass.SINGLE_TOKENS)
 493        )
 494
 495        if USE_RS_TOKENIZER:
 496            settings = RsTokenizerSettings(
 497                white_space={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.WHITE_SPACE.items()},
 498                single_tokens={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.SINGLE_TOKENS.items()},
 499                keywords={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.KEYWORDS.items()},
 500                numeric_literals=klass.NUMERIC_LITERALS,
 501                identifiers=klass._IDENTIFIERS,
 502                identifier_escapes=klass._IDENTIFIER_ESCAPES,
 503                string_escapes=klass._STRING_ESCAPES,
 504                quotes=klass._QUOTES,
 505                format_strings={
 506                    k: (v1, _TOKEN_TYPE_TO_INDEX[v2])
 507                    for k, (v1, v2) in klass._FORMAT_STRINGS.items()
 508                },
 509                has_bit_strings=bool(klass.BIT_STRINGS),
 510                has_hex_strings=bool(klass.HEX_STRINGS),
 511                comments=klass._COMMENTS,
 512                var_single_tokens=klass.VAR_SINGLE_TOKENS,
 513                commands={_TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMANDS},
 514                command_prefix_tokens={
 515                    _TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMAND_PREFIX_TOKENS
 516                },
 517                heredoc_tag_is_identifier=klass.HEREDOC_TAG_IS_IDENTIFIER,
 518            )
 519            token_types = RsTokenTypeSettings(
 520                bit_string=_TOKEN_TYPE_TO_INDEX[TokenType.BIT_STRING],
 521                break_=_TOKEN_TYPE_TO_INDEX[TokenType.BREAK],
 522                dcolon=_TOKEN_TYPE_TO_INDEX[TokenType.DCOLON],
 523                heredoc_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEREDOC_STRING],
 524                hex_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEX_STRING],
 525                identifier=_TOKEN_TYPE_TO_INDEX[TokenType.IDENTIFIER],
 526                number=_TOKEN_TYPE_TO_INDEX[TokenType.NUMBER],
 527                parameter=_TOKEN_TYPE_TO_INDEX[TokenType.PARAMETER],
 528                semicolon=_TOKEN_TYPE_TO_INDEX[TokenType.SEMICOLON],
 529                string=_TOKEN_TYPE_TO_INDEX[TokenType.STRING],
 530                var=_TOKEN_TYPE_TO_INDEX[TokenType.VAR],
 531                heredoc_string_alternative=_TOKEN_TYPE_TO_INDEX[klass.HEREDOC_STRING_ALTERNATIVE],
 532            )
 533            klass._RS_TOKENIZER = RsTokenizer(settings, token_types)
 534        else:
 535            klass._RS_TOKENIZER = None
 536
 537        return klass
 538
 539
 540class Tokenizer(metaclass=_Tokenizer):
 541    SINGLE_TOKENS = {
 542        "(": TokenType.L_PAREN,
 543        ")": TokenType.R_PAREN,
 544        "[": TokenType.L_BRACKET,
 545        "]": TokenType.R_BRACKET,
 546        "{": TokenType.L_BRACE,
 547        "}": TokenType.R_BRACE,
 548        "&": TokenType.AMP,
 549        "^": TokenType.CARET,
 550        ":": TokenType.COLON,
 551        ",": TokenType.COMMA,
 552        ".": TokenType.DOT,
 553        "-": TokenType.DASH,
 554        "=": TokenType.EQ,
 555        ">": TokenType.GT,
 556        "<": TokenType.LT,
 557        "%": TokenType.MOD,
 558        "!": TokenType.NOT,
 559        "|": TokenType.PIPE,
 560        "+": TokenType.PLUS,
 561        ";": TokenType.SEMICOLON,
 562        "/": TokenType.SLASH,
 563        "\\": TokenType.BACKSLASH,
 564        "*": TokenType.STAR,
 565        "~": TokenType.TILDA,
 566        "?": TokenType.PLACEHOLDER,
 567        "@": TokenType.PARAMETER,
 568        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 569        "'": TokenType.QUOTE,
 570        "`": TokenType.IDENTIFIER,
 571        '"': TokenType.IDENTIFIER,
 572        "#": TokenType.HASH,
 573    }
 574
 575    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 576    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 577    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 578    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 579    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 580    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 581    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 582    IDENTIFIER_ESCAPES = ['"']
 583    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 584    STRING_ESCAPES = ["'"]
 585    VAR_SINGLE_TOKENS: t.Set[str] = set()
 586
 587    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 588    HEREDOC_TAG_IS_IDENTIFIER = False
 589
 590    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 591    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 592
 593    # Autofilled
 594    _COMMENTS: t.Dict[str, str] = {}
 595    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 596    _IDENTIFIERS: t.Dict[str, str] = {}
 597    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 598    _QUOTES: t.Dict[str, str] = {}
 599    _STRING_ESCAPES: t.Set[str] = set()
 600    _KEYWORD_TRIE: t.Dict = {}
 601    _RS_TOKENIZER: t.Optional[t.Any] = None
 602
 603    KEYWORDS: t.Dict[str, TokenType] = {
 604        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 605        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 606        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 607        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 608        "/*+": TokenType.HINT,
 609        "==": TokenType.EQ,
 610        "::": TokenType.DCOLON,
 611        "||": TokenType.DPIPE,
 612        ">=": TokenType.GTE,
 613        "<=": TokenType.LTE,
 614        "<>": TokenType.NEQ,
 615        "!=": TokenType.NEQ,
 616        ":=": TokenType.COLON_EQ,
 617        "<=>": TokenType.NULLSAFE_EQ,
 618        "->": TokenType.ARROW,
 619        "->>": TokenType.DARROW,
 620        "=>": TokenType.FARROW,
 621        "#>": TokenType.HASH_ARROW,
 622        "#>>": TokenType.DHASH_ARROW,
 623        "<->": TokenType.LR_ARROW,
 624        "&&": TokenType.DAMP,
 625        "??": TokenType.DQMARK,
 626        "ALL": TokenType.ALL,
 627        "ALWAYS": TokenType.ALWAYS,
 628        "AND": TokenType.AND,
 629        "ANTI": TokenType.ANTI,
 630        "ANY": TokenType.ANY,
 631        "ASC": TokenType.ASC,
 632        "AS": TokenType.ALIAS,
 633        "ASOF": TokenType.ASOF,
 634        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 635        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 636        "BEGIN": TokenType.BEGIN,
 637        "BETWEEN": TokenType.BETWEEN,
 638        "CACHE": TokenType.CACHE,
 639        "UNCACHE": TokenType.UNCACHE,
 640        "CASE": TokenType.CASE,
 641        "CHARACTER SET": TokenType.CHARACTER_SET,
 642        "CLUSTER BY": TokenType.CLUSTER_BY,
 643        "COLLATE": TokenType.COLLATE,
 644        "COLUMN": TokenType.COLUMN,
 645        "COMMIT": TokenType.COMMIT,
 646        "CONNECT BY": TokenType.CONNECT_BY,
 647        "CONSTRAINT": TokenType.CONSTRAINT,
 648        "CREATE": TokenType.CREATE,
 649        "CROSS": TokenType.CROSS,
 650        "CUBE": TokenType.CUBE,
 651        "CURRENT_DATE": TokenType.CURRENT_DATE,
 652        "CURRENT_TIME": TokenType.CURRENT_TIME,
 653        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 654        "CURRENT_USER": TokenType.CURRENT_USER,
 655        "DATABASE": TokenType.DATABASE,
 656        "DEFAULT": TokenType.DEFAULT,
 657        "DELETE": TokenType.DELETE,
 658        "DESC": TokenType.DESC,
 659        "DESCRIBE": TokenType.DESCRIBE,
 660        "DISTINCT": TokenType.DISTINCT,
 661        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 662        "DIV": TokenType.DIV,
 663        "DROP": TokenType.DROP,
 664        "ELSE": TokenType.ELSE,
 665        "END": TokenType.END,
 666        "ENUM": TokenType.ENUM,
 667        "ESCAPE": TokenType.ESCAPE,
 668        "EXCEPT": TokenType.EXCEPT,
 669        "EXECUTE": TokenType.EXECUTE,
 670        "EXISTS": TokenType.EXISTS,
 671        "FALSE": TokenType.FALSE,
 672        "FETCH": TokenType.FETCH,
 673        "FILTER": TokenType.FILTER,
 674        "FIRST": TokenType.FIRST,
 675        "FULL": TokenType.FULL,
 676        "FUNCTION": TokenType.FUNCTION,
 677        "FOR": TokenType.FOR,
 678        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 679        "FORMAT": TokenType.FORMAT,
 680        "FROM": TokenType.FROM,
 681        "GEOGRAPHY": TokenType.GEOGRAPHY,
 682        "GEOMETRY": TokenType.GEOMETRY,
 683        "GLOB": TokenType.GLOB,
 684        "GROUP BY": TokenType.GROUP_BY,
 685        "GROUPING SETS": TokenType.GROUPING_SETS,
 686        "HAVING": TokenType.HAVING,
 687        "ILIKE": TokenType.ILIKE,
 688        "IN": TokenType.IN,
 689        "INDEX": TokenType.INDEX,
 690        "INET": TokenType.INET,
 691        "INNER": TokenType.INNER,
 692        "INSERT": TokenType.INSERT,
 693        "INTERVAL": TokenType.INTERVAL,
 694        "INTERSECT": TokenType.INTERSECT,
 695        "INTO": TokenType.INTO,
 696        "IS": TokenType.IS,
 697        "ISNULL": TokenType.ISNULL,
 698        "JOIN": TokenType.JOIN,
 699        "KEEP": TokenType.KEEP,
 700        "KILL": TokenType.KILL,
 701        "LATERAL": TokenType.LATERAL,
 702        "LEFT": TokenType.LEFT,
 703        "LIKE": TokenType.LIKE,
 704        "LIMIT": TokenType.LIMIT,
 705        "LOAD": TokenType.LOAD,
 706        "LOCK": TokenType.LOCK,
 707        "MERGE": TokenType.MERGE,
 708        "NATURAL": TokenType.NATURAL,
 709        "NEXT": TokenType.NEXT,
 710        "NOT": TokenType.NOT,
 711        "NOTNULL": TokenType.NOTNULL,
 712        "NULL": TokenType.NULL,
 713        "OBJECT": TokenType.OBJECT,
 714        "OFFSET": TokenType.OFFSET,
 715        "ON": TokenType.ON,
 716        "OR": TokenType.OR,
 717        "XOR": TokenType.XOR,
 718        "ORDER BY": TokenType.ORDER_BY,
 719        "ORDINALITY": TokenType.ORDINALITY,
 720        "OUTER": TokenType.OUTER,
 721        "OVER": TokenType.OVER,
 722        "OVERLAPS": TokenType.OVERLAPS,
 723        "OVERWRITE": TokenType.OVERWRITE,
 724        "PARTITION": TokenType.PARTITION,
 725        "PARTITION BY": TokenType.PARTITION_BY,
 726        "PARTITIONED BY": TokenType.PARTITION_BY,
 727        "PARTITIONED_BY": TokenType.PARTITION_BY,
 728        "PERCENT": TokenType.PERCENT,
 729        "PIVOT": TokenType.PIVOT,
 730        "PRAGMA": TokenType.PRAGMA,
 731        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 732        "PROCEDURE": TokenType.PROCEDURE,
 733        "QUALIFY": TokenType.QUALIFY,
 734        "RANGE": TokenType.RANGE,
 735        "RECURSIVE": TokenType.RECURSIVE,
 736        "REGEXP": TokenType.RLIKE,
 737        "REPLACE": TokenType.REPLACE,
 738        "RETURNING": TokenType.RETURNING,
 739        "REFERENCES": TokenType.REFERENCES,
 740        "RIGHT": TokenType.RIGHT,
 741        "RLIKE": TokenType.RLIKE,
 742        "ROLLBACK": TokenType.ROLLBACK,
 743        "ROLLUP": TokenType.ROLLUP,
 744        "ROW": TokenType.ROW,
 745        "ROWS": TokenType.ROWS,
 746        "SCHEMA": TokenType.SCHEMA,
 747        "SELECT": TokenType.SELECT,
 748        "SEMI": TokenType.SEMI,
 749        "SET": TokenType.SET,
 750        "SETTINGS": TokenType.SETTINGS,
 751        "SHOW": TokenType.SHOW,
 752        "SIMILAR TO": TokenType.SIMILAR_TO,
 753        "SOME": TokenType.SOME,
 754        "SORT BY": TokenType.SORT_BY,
 755        "START WITH": TokenType.START_WITH,
 756        "TABLE": TokenType.TABLE,
 757        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 758        "TEMP": TokenType.TEMPORARY,
 759        "TEMPORARY": TokenType.TEMPORARY,
 760        "THEN": TokenType.THEN,
 761        "TRUE": TokenType.TRUE,
 762        "TRUNCATE": TokenType.TRUNCATE,
 763        "UNION": TokenType.UNION,
 764        "UNKNOWN": TokenType.UNKNOWN,
 765        "UNNEST": TokenType.UNNEST,
 766        "UNPIVOT": TokenType.UNPIVOT,
 767        "UPDATE": TokenType.UPDATE,
 768        "USE": TokenType.USE,
 769        "USING": TokenType.USING,
 770        "UUID": TokenType.UUID,
 771        "VALUES": TokenType.VALUES,
 772        "VIEW": TokenType.VIEW,
 773        "VOLATILE": TokenType.VOLATILE,
 774        "WHEN": TokenType.WHEN,
 775        "WHERE": TokenType.WHERE,
 776        "WINDOW": TokenType.WINDOW,
 777        "WITH": TokenType.WITH,
 778        "APPLY": TokenType.APPLY,
 779        "ARRAY": TokenType.ARRAY,
 780        "BIT": TokenType.BIT,
 781        "BOOL": TokenType.BOOLEAN,
 782        "BOOLEAN": TokenType.BOOLEAN,
 783        "BYTE": TokenType.TINYINT,
 784        "MEDIUMINT": TokenType.MEDIUMINT,
 785        "INT1": TokenType.TINYINT,
 786        "TINYINT": TokenType.TINYINT,
 787        "INT16": TokenType.SMALLINT,
 788        "SHORT": TokenType.SMALLINT,
 789        "SMALLINT": TokenType.SMALLINT,
 790        "INT128": TokenType.INT128,
 791        "HUGEINT": TokenType.INT128,
 792        "INT2": TokenType.SMALLINT,
 793        "INTEGER": TokenType.INT,
 794        "INT": TokenType.INT,
 795        "INT4": TokenType.INT,
 796        "INT32": TokenType.INT,
 797        "INT64": TokenType.BIGINT,
 798        "LONG": TokenType.BIGINT,
 799        "BIGINT": TokenType.BIGINT,
 800        "INT8": TokenType.TINYINT,
 801        "DEC": TokenType.DECIMAL,
 802        "DECIMAL": TokenType.DECIMAL,
 803        "BIGDECIMAL": TokenType.BIGDECIMAL,
 804        "BIGNUMERIC": TokenType.BIGDECIMAL,
 805        "MAP": TokenType.MAP,
 806        "NULLABLE": TokenType.NULLABLE,
 807        "NUMBER": TokenType.DECIMAL,
 808        "NUMERIC": TokenType.DECIMAL,
 809        "FIXED": TokenType.DECIMAL,
 810        "REAL": TokenType.FLOAT,
 811        "FLOAT": TokenType.FLOAT,
 812        "FLOAT4": TokenType.FLOAT,
 813        "FLOAT8": TokenType.DOUBLE,
 814        "DOUBLE": TokenType.DOUBLE,
 815        "DOUBLE PRECISION": TokenType.DOUBLE,
 816        "JSON": TokenType.JSON,
 817        "CHAR": TokenType.CHAR,
 818        "CHARACTER": TokenType.CHAR,
 819        "NCHAR": TokenType.NCHAR,
 820        "VARCHAR": TokenType.VARCHAR,
 821        "VARCHAR2": TokenType.VARCHAR,
 822        "NVARCHAR": TokenType.NVARCHAR,
 823        "NVARCHAR2": TokenType.NVARCHAR,
 824        "BPCHAR": TokenType.BPCHAR,
 825        "STR": TokenType.TEXT,
 826        "STRING": TokenType.TEXT,
 827        "TEXT": TokenType.TEXT,
 828        "LONGTEXT": TokenType.LONGTEXT,
 829        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 830        "TINYTEXT": TokenType.TINYTEXT,
 831        "CLOB": TokenType.TEXT,
 832        "LONGVARCHAR": TokenType.TEXT,
 833        "BINARY": TokenType.BINARY,
 834        "BLOB": TokenType.VARBINARY,
 835        "LONGBLOB": TokenType.LONGBLOB,
 836        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 837        "TINYBLOB": TokenType.TINYBLOB,
 838        "BYTEA": TokenType.VARBINARY,
 839        "VARBINARY": TokenType.VARBINARY,
 840        "TIME": TokenType.TIME,
 841        "TIMETZ": TokenType.TIMETZ,
 842        "TIMESTAMP": TokenType.TIMESTAMP,
 843        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 844        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 845        "DATE": TokenType.DATE,
 846        "DATETIME": TokenType.DATETIME,
 847        "INT4RANGE": TokenType.INT4RANGE,
 848        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 849        "INT8RANGE": TokenType.INT8RANGE,
 850        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 851        "NUMRANGE": TokenType.NUMRANGE,
 852        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 853        "TSRANGE": TokenType.TSRANGE,
 854        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 855        "TSTZRANGE": TokenType.TSTZRANGE,
 856        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 857        "DATERANGE": TokenType.DATERANGE,
 858        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 859        "UNIQUE": TokenType.UNIQUE,
 860        "STRUCT": TokenType.STRUCT,
 861        "SEQUENCE": TokenType.SEQUENCE,
 862        "VARIANT": TokenType.VARIANT,
 863        "ALTER": TokenType.ALTER,
 864        "ANALYZE": TokenType.COMMAND,
 865        "CALL": TokenType.COMMAND,
 866        "COMMENT": TokenType.COMMENT,
 867        "COPY": TokenType.COMMAND,
 868        "EXPLAIN": TokenType.COMMAND,
 869        "GRANT": TokenType.COMMAND,
 870        "OPTIMIZE": TokenType.COMMAND,
 871        "PREPARE": TokenType.COMMAND,
 872        "VACUUM": TokenType.COMMAND,
 873        "USER-DEFINED": TokenType.USERDEFINED,
 874        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 875        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 876    }
 877
 878    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 879        " ": TokenType.SPACE,
 880        "\t": TokenType.SPACE,
 881        "\n": TokenType.BREAK,
 882        "\r": TokenType.BREAK,
 883    }
 884
 885    COMMANDS = {
 886        TokenType.COMMAND,
 887        TokenType.EXECUTE,
 888        TokenType.FETCH,
 889        TokenType.SHOW,
 890    }
 891
 892    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 893
 894    # Handle numeric literals like in hive (3L = BIGINT)
 895    NUMERIC_LITERALS: t.Dict[str, str] = {}
 896
 897    COMMENTS = ["--", ("/*", "*/")]
 898
 899    __slots__ = (
 900        "sql",
 901        "size",
 902        "tokens",
 903        "dialect",
 904        "_start",
 905        "_current",
 906        "_line",
 907        "_col",
 908        "_comments",
 909        "_char",
 910        "_end",
 911        "_peek",
 912        "_prev_token_line",
 913        "_rs_dialect_settings",
 914    )
 915
 916    def __init__(self, dialect: DialectType = None) -> None:
 917        from sqlglot.dialects import Dialect
 918
 919        self.dialect = Dialect.get_or_raise(dialect)
 920
 921        if USE_RS_TOKENIZER:
 922            self._rs_dialect_settings = RsTokenizerDialectSettings(
 923                escape_sequences=self.dialect.ESCAPE_SEQUENCES,
 924                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 925            )
 926
 927        self.reset()
 928
 929    def reset(self) -> None:
 930        self.sql = ""
 931        self.size = 0
 932        self.tokens: t.List[Token] = []
 933        self._start = 0
 934        self._current = 0
 935        self._line = 1
 936        self._col = 0
 937        self._comments: t.List[str] = []
 938
 939        self._char = ""
 940        self._end = False
 941        self._peek = ""
 942        self._prev_token_line = -1
 943
 944    def tokenize(self, sql: str) -> t.List[Token]:
 945        """Returns a list of tokens corresponding to the SQL string `sql`."""
 946        if USE_RS_TOKENIZER:
 947            return self.tokenize_rs(sql)
 948
 949        self.reset()
 950        self.sql = sql
 951        self.size = len(sql)
 952
 953        try:
 954            self._scan()
 955        except Exception as e:
 956            start = max(self._current - 50, 0)
 957            end = min(self._current + 50, self.size - 1)
 958            context = self.sql[start:end]
 959            raise TokenError(f"Error tokenizing '{context}'") from e
 960
 961        return self.tokens
 962
 963    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
 964        while self.size and not self._end:
 965            current = self._current
 966
 967            # Skip spaces here rather than iteratively calling advance() for performance reasons
 968            while current < self.size:
 969                char = self.sql[current]
 970
 971                if char.isspace() and (char == " " or char == "\t"):
 972                    current += 1
 973                else:
 974                    break
 975
 976            offset = current - self._current if current > self._current else 1
 977
 978            self._start = current
 979            self._advance(offset)
 980
 981            if not self._char.isspace():
 982                if self._char.isdigit():
 983                    self._scan_number()
 984                elif self._char in self._IDENTIFIERS:
 985                    self._scan_identifier(self._IDENTIFIERS[self._char])
 986                else:
 987                    self._scan_keywords()
 988
 989            if until and until():
 990                break
 991
 992        if self.tokens and self._comments:
 993            self.tokens[-1].comments.extend(self._comments)
 994
 995    def _chars(self, size: int) -> str:
 996        if size == 1:
 997            return self._char
 998
 999        start = self._current - 1
1000        end = start + size
1001
1002        return self.sql[start:end] if end <= self.size else ""
1003
1004    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1005        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1006            # Ensures we don't count an extra line if we get a \r\n line break sequence
1007            if not (self._char == "\r" and self._peek == "\n"):
1008                self._col = 1
1009                self._line += 1
1010        else:
1011            self._col += i
1012
1013        self._current += i
1014        self._end = self._current >= self.size
1015        self._char = self.sql[self._current - 1]
1016        self._peek = "" if self._end else self.sql[self._current]
1017
1018        if alnum and self._char.isalnum():
1019            # Here we use local variables instead of attributes for better performance
1020            _col = self._col
1021            _current = self._current
1022            _end = self._end
1023            _peek = self._peek
1024
1025            while _peek.isalnum():
1026                _col += 1
1027                _current += 1
1028                _end = _current >= self.size
1029                _peek = "" if _end else self.sql[_current]
1030
1031            self._col = _col
1032            self._current = _current
1033            self._end = _end
1034            self._peek = _peek
1035            self._char = self.sql[_current - 1]
1036
1037    @property
1038    def _text(self) -> str:
1039        return self.sql[self._start : self._current]
1040
1041    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1042        self._prev_token_line = self._line
1043
1044        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1045            self.tokens[-1].comments.extend(self._comments)
1046            self._comments = []
1047
1048        self.tokens.append(
1049            Token(
1050                token_type,
1051                text=self._text if text is None else text,
1052                line=self._line,
1053                col=self._col,
1054                start=self._start,
1055                end=self._current - 1,
1056                comments=self._comments,
1057            )
1058        )
1059        self._comments = []
1060
1061        # If we have either a semicolon or a begin token before the command's token, we'll parse
1062        # whatever follows the command's token as a string
1063        if (
1064            token_type in self.COMMANDS
1065            and self._peek != ";"
1066            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1067        ):
1068            start = self._current
1069            tokens = len(self.tokens)
1070            self._scan(lambda: self._peek == ";")
1071            self.tokens = self.tokens[:tokens]
1072            text = self.sql[start : self._current].strip()
1073            if text:
1074                self._add(TokenType.STRING, text)
1075
1076    def _scan_keywords(self) -> None:
1077        size = 0
1078        word = None
1079        chars = self._text
1080        char = chars
1081        prev_space = False
1082        skip = False
1083        trie = self._KEYWORD_TRIE
1084        single_token = char in self.SINGLE_TOKENS
1085
1086        while chars:
1087            if skip:
1088                result = TrieResult.PREFIX
1089            else:
1090                result, trie = in_trie(trie, char.upper())
1091
1092            if result == TrieResult.FAILED:
1093                break
1094            if result == TrieResult.EXISTS:
1095                word = chars
1096
1097            end = self._current + size
1098            size += 1
1099
1100            if end < self.size:
1101                char = self.sql[end]
1102                single_token = single_token or char in self.SINGLE_TOKENS
1103                is_space = char.isspace()
1104
1105                if not is_space or not prev_space:
1106                    if is_space:
1107                        char = " "
1108                    chars += char
1109                    prev_space = is_space
1110                    skip = False
1111                else:
1112                    skip = True
1113            else:
1114                char = ""
1115                break
1116
1117        if word:
1118            if self._scan_string(word):
1119                return
1120            if self._scan_comment(word):
1121                return
1122            if prev_space or single_token or not char:
1123                self._advance(size - 1)
1124                word = word.upper()
1125                self._add(self.KEYWORDS[word], text=word)
1126                return
1127
1128        if self._char in self.SINGLE_TOKENS:
1129            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1130            return
1131
1132        self._scan_var()
1133
1134    def _scan_comment(self, comment_start: str) -> bool:
1135        if comment_start not in self._COMMENTS:
1136            return False
1137
1138        comment_start_line = self._line
1139        comment_start_size = len(comment_start)
1140        comment_end = self._COMMENTS[comment_start]
1141
1142        if comment_end:
1143            # Skip the comment's start delimiter
1144            self._advance(comment_start_size)
1145
1146            comment_end_size = len(comment_end)
1147            while not self._end and self._chars(comment_end_size) != comment_end:
1148                self._advance(alnum=True)
1149
1150            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1151            self._advance(comment_end_size - 1)
1152        else:
1153            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1154                self._advance(alnum=True)
1155            self._comments.append(self._text[comment_start_size:])
1156
1157        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1158        # Multiple consecutive comments are preserved by appending them to the current comments list.
1159        if comment_start_line == self._prev_token_line:
1160            self.tokens[-1].comments.extend(self._comments)
1161            self._comments = []
1162            self._prev_token_line = self._line
1163
1164        return True
1165
1166    def _scan_number(self) -> None:
1167        if self._char == "0":
1168            peek = self._peek.upper()
1169            if peek == "B":
1170                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1171            elif peek == "X":
1172                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1173
1174        decimal = False
1175        scientific = 0
1176
1177        while True:
1178            if self._peek.isdigit():
1179                self._advance()
1180            elif self._peek == "." and not decimal:
1181                decimal = True
1182                self._advance()
1183            elif self._peek in ("-", "+") and scientific == 1:
1184                scientific += 1
1185                self._advance()
1186            elif self._peek.upper() == "E" and not scientific:
1187                scientific += 1
1188                self._advance()
1189            elif self._peek.isidentifier():
1190                number_text = self._text
1191                literal = ""
1192
1193                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1194                    literal += self._peek
1195                    self._advance()
1196
1197                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1198
1199                if token_type:
1200                    self._add(TokenType.NUMBER, number_text)
1201                    self._add(TokenType.DCOLON, "::")
1202                    return self._add(token_type, literal)
1203                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1204                    return self._add(TokenType.VAR)
1205
1206                self._advance(-len(literal))
1207                return self._add(TokenType.NUMBER, number_text)
1208            else:
1209                return self._add(TokenType.NUMBER)
1210
1211    def _scan_bits(self) -> None:
1212        self._advance()
1213        value = self._extract_value()
1214        try:
1215            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1216            int(value, 2)
1217            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1218        except ValueError:
1219            self._add(TokenType.IDENTIFIER)
1220
1221    def _scan_hex(self) -> None:
1222        self._advance()
1223        value = self._extract_value()
1224        try:
1225            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1226            int(value, 16)
1227            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1228        except ValueError:
1229            self._add(TokenType.IDENTIFIER)
1230
1231    def _extract_value(self) -> str:
1232        while True:
1233            char = self._peek.strip()
1234            if char and char not in self.SINGLE_TOKENS:
1235                self._advance(alnum=True)
1236            else:
1237                break
1238
1239        return self._text
1240
1241    def _scan_string(self, start: str) -> bool:
1242        base = None
1243        token_type = TokenType.STRING
1244
1245        if start in self._QUOTES:
1246            end = self._QUOTES[start]
1247        elif start in self._FORMAT_STRINGS:
1248            end, token_type = self._FORMAT_STRINGS[start]
1249
1250            if token_type == TokenType.HEX_STRING:
1251                base = 16
1252            elif token_type == TokenType.BIT_STRING:
1253                base = 2
1254            elif token_type == TokenType.HEREDOC_STRING:
1255                if (
1256                    self.HEREDOC_TAG_IS_IDENTIFIER
1257                    and not self._peek.isidentifier()
1258                    and not self._peek == end
1259                ):
1260                    if self.HEREDOC_STRING_ALTERNATIVE != token_type.VAR:
1261                        self._add(self.HEREDOC_STRING_ALTERNATIVE)
1262                    else:
1263                        self._scan_var()
1264
1265                    return True
1266
1267                self._advance()
1268
1269                if self._char == end:
1270                    tag = ""
1271                else:
1272                    tag = self._extract_string(
1273                        end, raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER
1274                    )
1275
1276                if self._end and tag and self.HEREDOC_TAG_IS_IDENTIFIER:
1277                    self._advance(-len(tag))
1278                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1279                    return True
1280
1281                end = f"{start}{tag}{end}"
1282        else:
1283            return False
1284
1285        self._advance(len(start))
1286        text = self._extract_string(end)
1287
1288        if base:
1289            try:
1290                int(text, base)
1291            except Exception:
1292                raise TokenError(
1293                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1294                )
1295
1296        self._add(token_type, text)
1297        return True
1298
1299    def _scan_identifier(self, identifier_end: str) -> None:
1300        self._advance()
1301        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1302        self._add(TokenType.IDENTIFIER, text)
1303
1304    def _scan_var(self) -> None:
1305        while True:
1306            char = self._peek.strip()
1307            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1308                self._advance(alnum=True)
1309            else:
1310                break
1311
1312        self._add(
1313            TokenType.VAR
1314            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1315            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1316        )
1317
1318    def _extract_string(
1319        self, delimiter: str, escapes: t.Optional[t.Set[str]] = None, raise_unmatched: bool = True
1320    ) -> str:
1321        text = ""
1322        delim_size = len(delimiter)
1323        escapes = self._STRING_ESCAPES if escapes is None else escapes
1324
1325        while True:
1326            if (
1327                self._char in escapes
1328                and (self._peek == delimiter or self._peek in escapes)
1329                and (self._char not in self._QUOTES or self._char == self._peek)
1330            ):
1331                if self._peek == delimiter:
1332                    text += self._peek
1333                else:
1334                    text += self._char + self._peek
1335
1336                if self._current + 1 < self.size:
1337                    self._advance(2)
1338                else:
1339                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1340            else:
1341                if self._chars(delim_size) == delimiter:
1342                    if delim_size > 1:
1343                        self._advance(delim_size - 1)
1344                    break
1345
1346                if self._end:
1347                    if not raise_unmatched:
1348                        return text + self._char
1349
1350                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1351
1352                if (
1353                    self.dialect.ESCAPE_SEQUENCES
1354                    and self._peek
1355                    and self._char in self.STRING_ESCAPES
1356                ):
1357                    escaped_sequence = self.dialect.ESCAPE_SEQUENCES.get(self._char + self._peek)
1358                    if escaped_sequence:
1359                        self._advance(2)
1360                        text += escaped_sequence
1361                        continue
1362
1363                current = self._current - 1
1364                self._advance(alnum=True)
1365                text += self.sql[current : self._current - 1]
1366
1367        return text
1368
1369    def tokenize_rs(self, sql: str) -> t.List[Token]:
1370        if not self._RS_TOKENIZER:
1371            raise SqlglotError("Rust tokenizer is not available")
1372
1373        try:
1374            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1375            for token in tokens:
1376                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1377            return tokens
1378        except Exception as e:
1379            raise TokenError(str(e))
class TokenType(sqlglot.helper.AutoName):
 29class TokenType(AutoName):
 30    L_PAREN = auto()
 31    R_PAREN = auto()
 32    L_BRACKET = auto()
 33    R_BRACKET = auto()
 34    L_BRACE = auto()
 35    R_BRACE = auto()
 36    COMMA = auto()
 37    DOT = auto()
 38    DASH = auto()
 39    PLUS = auto()
 40    COLON = auto()
 41    DCOLON = auto()
 42    DQMARK = auto()
 43    SEMICOLON = auto()
 44    STAR = auto()
 45    BACKSLASH = auto()
 46    SLASH = auto()
 47    LT = auto()
 48    LTE = auto()
 49    GT = auto()
 50    GTE = auto()
 51    NOT = auto()
 52    EQ = auto()
 53    NEQ = auto()
 54    NULLSAFE_EQ = auto()
 55    COLON_EQ = auto()
 56    AND = auto()
 57    OR = auto()
 58    AMP = auto()
 59    DPIPE = auto()
 60    PIPE = auto()
 61    PIPE_SLASH = auto()
 62    DPIPE_SLASH = auto()
 63    CARET = auto()
 64    TILDA = auto()
 65    ARROW = auto()
 66    DARROW = auto()
 67    FARROW = auto()
 68    HASH = auto()
 69    HASH_ARROW = auto()
 70    DHASH_ARROW = auto()
 71    LR_ARROW = auto()
 72    DAT = auto()
 73    LT_AT = auto()
 74    AT_GT = auto()
 75    DOLLAR = auto()
 76    PARAMETER = auto()
 77    SESSION_PARAMETER = auto()
 78    DAMP = auto()
 79    XOR = auto()
 80    DSTAR = auto()
 81
 82    BLOCK_START = auto()
 83    BLOCK_END = auto()
 84
 85    SPACE = auto()
 86    BREAK = auto()
 87
 88    STRING = auto()
 89    NUMBER = auto()
 90    IDENTIFIER = auto()
 91    DATABASE = auto()
 92    COLUMN = auto()
 93    COLUMN_DEF = auto()
 94    SCHEMA = auto()
 95    TABLE = auto()
 96    VAR = auto()
 97    BIT_STRING = auto()
 98    HEX_STRING = auto()
 99    BYTE_STRING = auto()
100    NATIONAL_STRING = auto()
101    RAW_STRING = auto()
102    HEREDOC_STRING = auto()
103    UNICODE_STRING = auto()
104
105    # types
106    BIT = auto()
107    BOOLEAN = auto()
108    TINYINT = auto()
109    UTINYINT = auto()
110    SMALLINT = auto()
111    USMALLINT = auto()
112    MEDIUMINT = auto()
113    UMEDIUMINT = auto()
114    INT = auto()
115    UINT = auto()
116    BIGINT = auto()
117    UBIGINT = auto()
118    INT128 = auto()
119    UINT128 = auto()
120    INT256 = auto()
121    UINT256 = auto()
122    FLOAT = auto()
123    DOUBLE = auto()
124    DECIMAL = auto()
125    UDECIMAL = auto()
126    BIGDECIMAL = auto()
127    CHAR = auto()
128    NCHAR = auto()
129    VARCHAR = auto()
130    NVARCHAR = auto()
131    BPCHAR = auto()
132    TEXT = auto()
133    MEDIUMTEXT = auto()
134    LONGTEXT = auto()
135    MEDIUMBLOB = auto()
136    LONGBLOB = auto()
137    TINYBLOB = auto()
138    TINYTEXT = auto()
139    NAME = auto()
140    BINARY = auto()
141    VARBINARY = auto()
142    JSON = auto()
143    JSONB = auto()
144    TIME = auto()
145    TIMETZ = auto()
146    TIMESTAMP = auto()
147    TIMESTAMPTZ = auto()
148    TIMESTAMPLTZ = auto()
149    TIMESTAMP_S = auto()
150    TIMESTAMP_MS = auto()
151    TIMESTAMP_NS = auto()
152    DATETIME = auto()
153    DATETIME64 = auto()
154    DATE = auto()
155    DATE32 = auto()
156    INT4RANGE = auto()
157    INT4MULTIRANGE = auto()
158    INT8RANGE = auto()
159    INT8MULTIRANGE = auto()
160    NUMRANGE = auto()
161    NUMMULTIRANGE = auto()
162    TSRANGE = auto()
163    TSMULTIRANGE = auto()
164    TSTZRANGE = auto()
165    TSTZMULTIRANGE = auto()
166    DATERANGE = auto()
167    DATEMULTIRANGE = auto()
168    UUID = auto()
169    GEOGRAPHY = auto()
170    NULLABLE = auto()
171    GEOMETRY = auto()
172    HLLSKETCH = auto()
173    HSTORE = auto()
174    SUPER = auto()
175    SERIAL = auto()
176    SMALLSERIAL = auto()
177    BIGSERIAL = auto()
178    XML = auto()
179    YEAR = auto()
180    UNIQUEIDENTIFIER = auto()
181    USERDEFINED = auto()
182    MONEY = auto()
183    SMALLMONEY = auto()
184    ROWVERSION = auto()
185    IMAGE = auto()
186    VARIANT = auto()
187    OBJECT = auto()
188    INET = auto()
189    IPADDRESS = auto()
190    IPPREFIX = auto()
191    IPV4 = auto()
192    IPV6 = auto()
193    ENUM = auto()
194    ENUM8 = auto()
195    ENUM16 = auto()
196    FIXEDSTRING = auto()
197    LOWCARDINALITY = auto()
198    NESTED = auto()
199    AGGREGATEFUNCTION = auto()
200    SIMPLEAGGREGATEFUNCTION = auto()
201    UNKNOWN = auto()
202
203    # keywords
204    ALIAS = auto()
205    ALTER = auto()
206    ALWAYS = auto()
207    ALL = auto()
208    ANTI = auto()
209    ANY = auto()
210    APPLY = auto()
211    ARRAY = auto()
212    ASC = auto()
213    ASOF = auto()
214    AUTO_INCREMENT = auto()
215    BEGIN = auto()
216    BETWEEN = auto()
217    CACHE = auto()
218    CASE = auto()
219    CHARACTER_SET = auto()
220    CLUSTER_BY = auto()
221    COLLATE = auto()
222    COMMAND = auto()
223    COMMENT = auto()
224    COMMIT = auto()
225    CONNECT_BY = auto()
226    CONSTRAINT = auto()
227    CREATE = auto()
228    CROSS = auto()
229    CUBE = auto()
230    CURRENT_DATE = auto()
231    CURRENT_DATETIME = auto()
232    CURRENT_TIME = auto()
233    CURRENT_TIMESTAMP = auto()
234    CURRENT_USER = auto()
235    DEFAULT = auto()
236    DELETE = auto()
237    DESC = auto()
238    DESCRIBE = auto()
239    DICTIONARY = auto()
240    DISTINCT = auto()
241    DISTRIBUTE_BY = auto()
242    DIV = auto()
243    DROP = auto()
244    ELSE = auto()
245    END = auto()
246    ESCAPE = auto()
247    EXCEPT = auto()
248    EXECUTE = auto()
249    EXISTS = auto()
250    FALSE = auto()
251    FETCH = auto()
252    FILTER = auto()
253    FINAL = auto()
254    FIRST = auto()
255    FOR = auto()
256    FORCE = auto()
257    FOREIGN_KEY = auto()
258    FORMAT = auto()
259    FROM = auto()
260    FULL = auto()
261    FUNCTION = auto()
262    GLOB = auto()
263    GLOBAL = auto()
264    GROUP_BY = auto()
265    GROUPING_SETS = auto()
266    HAVING = auto()
267    HINT = auto()
268    IGNORE = auto()
269    ILIKE = auto()
270    ILIKE_ANY = auto()
271    IN = auto()
272    INDEX = auto()
273    INNER = auto()
274    INSERT = auto()
275    INTERSECT = auto()
276    INTERVAL = auto()
277    INTO = auto()
278    INTRODUCER = auto()
279    IRLIKE = auto()
280    IS = auto()
281    ISNULL = auto()
282    JOIN = auto()
283    JOIN_MARKER = auto()
284    KEEP = auto()
285    KILL = auto()
286    LANGUAGE = auto()
287    LATERAL = auto()
288    LEFT = auto()
289    LIKE = auto()
290    LIKE_ANY = auto()
291    LIMIT = auto()
292    LOAD = auto()
293    LOCK = auto()
294    MAP = auto()
295    MATCH_RECOGNIZE = auto()
296    MEMBER_OF = auto()
297    MERGE = auto()
298    MOD = auto()
299    MODEL = auto()
300    NATURAL = auto()
301    NEXT = auto()
302    NOTNULL = auto()
303    NULL = auto()
304    OBJECT_IDENTIFIER = auto()
305    OFFSET = auto()
306    ON = auto()
307    ONLY = auto()
308    OPERATOR = auto()
309    ORDER_BY = auto()
310    ORDER_SIBLINGS_BY = auto()
311    ORDERED = auto()
312    ORDINALITY = auto()
313    OUTER = auto()
314    OVER = auto()
315    OVERLAPS = auto()
316    OVERWRITE = auto()
317    PARTITION = auto()
318    PARTITION_BY = auto()
319    PERCENT = auto()
320    PIVOT = auto()
321    PLACEHOLDER = auto()
322    POSITIONAL = auto()
323    PRAGMA = auto()
324    PREWHERE = auto()
325    PRIMARY_KEY = auto()
326    PROCEDURE = auto()
327    PROPERTIES = auto()
328    PSEUDO_TYPE = auto()
329    QUALIFY = auto()
330    QUOTE = auto()
331    RANGE = auto()
332    RECURSIVE = auto()
333    REFRESH = auto()
334    REPLACE = auto()
335    RETURNING = auto()
336    REFERENCES = auto()
337    RIGHT = auto()
338    RLIKE = auto()
339    ROLLBACK = auto()
340    ROLLUP = auto()
341    ROW = auto()
342    ROWS = auto()
343    SELECT = auto()
344    SEMI = auto()
345    SEPARATOR = auto()
346    SEQUENCE = auto()
347    SERDE_PROPERTIES = auto()
348    SET = auto()
349    SETTINGS = auto()
350    SHOW = auto()
351    SIMILAR_TO = auto()
352    SOME = auto()
353    SORT_BY = auto()
354    START_WITH = auto()
355    STORAGE_INTEGRATION = auto()
356    STRUCT = auto()
357    TABLE_SAMPLE = auto()
358    TEMPORARY = auto()
359    TOP = auto()
360    THEN = auto()
361    TRUE = auto()
362    TRUNCATE = auto()
363    UNCACHE = auto()
364    UNION = auto()
365    UNNEST = auto()
366    UNPIVOT = auto()
367    UPDATE = auto()
368    USE = auto()
369    USING = auto()
370    VALUES = auto()
371    VIEW = auto()
372    VOLATILE = auto()
373    WHEN = auto()
374    WHERE = auto()
375    WINDOW = auto()
376    WITH = auto()
377    UNIQUE = auto()
378    VERSION_SNAPSHOT = auto()
379    TIMESTAMP_SNAPSHOT = auto()
380    OPTION = auto()

An enumeration.

L_PAREN = <TokenType.L_PAREN: 'L_PAREN'>
R_PAREN = <TokenType.R_PAREN: 'R_PAREN'>
L_BRACKET = <TokenType.L_BRACKET: 'L_BRACKET'>
R_BRACKET = <TokenType.R_BRACKET: 'R_BRACKET'>
L_BRACE = <TokenType.L_BRACE: 'L_BRACE'>
R_BRACE = <TokenType.R_BRACE: 'R_BRACE'>
COMMA = <TokenType.COMMA: 'COMMA'>
DOT = <TokenType.DOT: 'DOT'>
DASH = <TokenType.DASH: 'DASH'>
PLUS = <TokenType.PLUS: 'PLUS'>
COLON = <TokenType.COLON: 'COLON'>
DCOLON = <TokenType.DCOLON: 'DCOLON'>
DQMARK = <TokenType.DQMARK: 'DQMARK'>
SEMICOLON = <TokenType.SEMICOLON: 'SEMICOLON'>
STAR = <TokenType.STAR: 'STAR'>
BACKSLASH = <TokenType.BACKSLASH: 'BACKSLASH'>
SLASH = <TokenType.SLASH: 'SLASH'>
LT = <TokenType.LT: 'LT'>
LTE = <TokenType.LTE: 'LTE'>
GT = <TokenType.GT: 'GT'>
GTE = <TokenType.GTE: 'GTE'>
NOT = <TokenType.NOT: 'NOT'>
EQ = <TokenType.EQ: 'EQ'>
NEQ = <TokenType.NEQ: 'NEQ'>
NULLSAFE_EQ = <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>
COLON_EQ = <TokenType.COLON_EQ: 'COLON_EQ'>
AND = <TokenType.AND: 'AND'>
OR = <TokenType.OR: 'OR'>
AMP = <TokenType.AMP: 'AMP'>
DPIPE = <TokenType.DPIPE: 'DPIPE'>
PIPE = <TokenType.PIPE: 'PIPE'>
PIPE_SLASH = <TokenType.PIPE_SLASH: 'PIPE_SLASH'>
DPIPE_SLASH = <TokenType.DPIPE_SLASH: 'DPIPE_SLASH'>
CARET = <TokenType.CARET: 'CARET'>
TILDA = <TokenType.TILDA: 'TILDA'>
ARROW = <TokenType.ARROW: 'ARROW'>
DARROW = <TokenType.DARROW: 'DARROW'>
FARROW = <TokenType.FARROW: 'FARROW'>
HASH = <TokenType.HASH: 'HASH'>
HASH_ARROW = <TokenType.HASH_ARROW: 'HASH_ARROW'>
DHASH_ARROW = <TokenType.DHASH_ARROW: 'DHASH_ARROW'>
LR_ARROW = <TokenType.LR_ARROW: 'LR_ARROW'>
DAT = <TokenType.DAT: 'DAT'>
LT_AT = <TokenType.LT_AT: 'LT_AT'>
AT_GT = <TokenType.AT_GT: 'AT_GT'>
DOLLAR = <TokenType.DOLLAR: 'DOLLAR'>
PARAMETER = <TokenType.PARAMETER: 'PARAMETER'>
SESSION_PARAMETER = <TokenType.SESSION_PARAMETER: 'SESSION_PARAMETER'>
DAMP = <TokenType.DAMP: 'DAMP'>
XOR = <TokenType.XOR: 'XOR'>
DSTAR = <TokenType.DSTAR: 'DSTAR'>
BLOCK_START = <TokenType.BLOCK_START: 'BLOCK_START'>
BLOCK_END = <TokenType.BLOCK_END: 'BLOCK_END'>
SPACE = <TokenType.SPACE: 'SPACE'>
BREAK = <TokenType.BREAK: 'BREAK'>
STRING = <TokenType.STRING: 'STRING'>
NUMBER = <TokenType.NUMBER: 'NUMBER'>
IDENTIFIER = <TokenType.IDENTIFIER: 'IDENTIFIER'>
DATABASE = <TokenType.DATABASE: 'DATABASE'>
COLUMN = <TokenType.COLUMN: 'COLUMN'>
COLUMN_DEF = <TokenType.COLUMN_DEF: 'COLUMN_DEF'>
SCHEMA = <TokenType.SCHEMA: 'SCHEMA'>
TABLE = <TokenType.TABLE: 'TABLE'>
VAR = <TokenType.VAR: 'VAR'>
BIT_STRING = <TokenType.BIT_STRING: 'BIT_STRING'>
HEX_STRING = <TokenType.HEX_STRING: 'HEX_STRING'>
BYTE_STRING = <TokenType.BYTE_STRING: 'BYTE_STRING'>
NATIONAL_STRING = <TokenType.NATIONAL_STRING: 'NATIONAL_STRING'>
RAW_STRING = <TokenType.RAW_STRING: 'RAW_STRING'>
HEREDOC_STRING = <TokenType.HEREDOC_STRING: 'HEREDOC_STRING'>
UNICODE_STRING = <TokenType.UNICODE_STRING: 'UNICODE_STRING'>
BIT = <TokenType.BIT: 'BIT'>
BOOLEAN = <TokenType.BOOLEAN: 'BOOLEAN'>
TINYINT = <TokenType.TINYINT: 'TINYINT'>
UTINYINT = <TokenType.UTINYINT: 'UTINYINT'>
SMALLINT = <TokenType.SMALLINT: 'SMALLINT'>
USMALLINT = <TokenType.USMALLINT: 'USMALLINT'>
MEDIUMINT = <TokenType.MEDIUMINT: 'MEDIUMINT'>
UMEDIUMINT = <TokenType.UMEDIUMINT: 'UMEDIUMINT'>
INT = <TokenType.INT: 'INT'>
UINT = <TokenType.UINT: 'UINT'>
BIGINT = <TokenType.BIGINT: 'BIGINT'>
UBIGINT = <TokenType.UBIGINT: 'UBIGINT'>
INT128 = <TokenType.INT128: 'INT128'>
UINT128 = <TokenType.UINT128: 'UINT128'>
INT256 = <TokenType.INT256: 'INT256'>
UINT256 = <TokenType.UINT256: 'UINT256'>
FLOAT = <TokenType.FLOAT: 'FLOAT'>
DOUBLE = <TokenType.DOUBLE: 'DOUBLE'>
DECIMAL = <TokenType.DECIMAL: 'DECIMAL'>
UDECIMAL = <TokenType.UDECIMAL: 'UDECIMAL'>
BIGDECIMAL = <TokenType.BIGDECIMAL: 'BIGDECIMAL'>
CHAR = <TokenType.CHAR: 'CHAR'>
NCHAR = <TokenType.NCHAR: 'NCHAR'>
VARCHAR = <TokenType.VARCHAR: 'VARCHAR'>
NVARCHAR = <TokenType.NVARCHAR: 'NVARCHAR'>
BPCHAR = <TokenType.BPCHAR: 'BPCHAR'>
TEXT = <TokenType.TEXT: 'TEXT'>
MEDIUMTEXT = <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>
LONGTEXT = <TokenType.LONGTEXT: 'LONGTEXT'>
MEDIUMBLOB = <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>
LONGBLOB = <TokenType.LONGBLOB: 'LONGBLOB'>
TINYBLOB = <TokenType.TINYBLOB: 'TINYBLOB'>
TINYTEXT = <TokenType.TINYTEXT: 'TINYTEXT'>
NAME = <TokenType.NAME: 'NAME'>
BINARY = <TokenType.BINARY: 'BINARY'>
VARBINARY = <TokenType.VARBINARY: 'VARBINARY'>
JSON = <TokenType.JSON: 'JSON'>
JSONB = <TokenType.JSONB: 'JSONB'>
TIME = <TokenType.TIME: 'TIME'>
TIMETZ = <TokenType.TIMETZ: 'TIMETZ'>
TIMESTAMP = <TokenType.TIMESTAMP: 'TIMESTAMP'>
TIMESTAMPTZ = <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>
TIMESTAMPLTZ = <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>
TIMESTAMP_S = <TokenType.TIMESTAMP_S: 'TIMESTAMP_S'>
TIMESTAMP_MS = <TokenType.TIMESTAMP_MS: 'TIMESTAMP_MS'>
TIMESTAMP_NS = <TokenType.TIMESTAMP_NS: 'TIMESTAMP_NS'>
DATETIME = <TokenType.DATETIME: 'DATETIME'>
DATETIME64 = <TokenType.DATETIME64: 'DATETIME64'>
DATE = <TokenType.DATE: 'DATE'>
DATE32 = <TokenType.DATE32: 'DATE32'>
INT4RANGE = <TokenType.INT4RANGE: 'INT4RANGE'>
INT4MULTIRANGE = <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>
INT8RANGE = <TokenType.INT8RANGE: 'INT8RANGE'>
INT8MULTIRANGE = <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>
NUMRANGE = <TokenType.NUMRANGE: 'NUMRANGE'>
NUMMULTIRANGE = <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>
TSRANGE = <TokenType.TSRANGE: 'TSRANGE'>
TSMULTIRANGE = <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>
TSTZRANGE = <TokenType.TSTZRANGE: 'TSTZRANGE'>
TSTZMULTIRANGE = <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>
DATERANGE = <TokenType.DATERANGE: 'DATERANGE'>
DATEMULTIRANGE = <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>
UUID = <TokenType.UUID: 'UUID'>
GEOGRAPHY = <TokenType.GEOGRAPHY: 'GEOGRAPHY'>
NULLABLE = <TokenType.NULLABLE: 'NULLABLE'>
GEOMETRY = <TokenType.GEOMETRY: 'GEOMETRY'>
HLLSKETCH = <TokenType.HLLSKETCH: 'HLLSKETCH'>
HSTORE = <TokenType.HSTORE: 'HSTORE'>
SUPER = <TokenType.SUPER: 'SUPER'>
SERIAL = <TokenType.SERIAL: 'SERIAL'>
SMALLSERIAL = <TokenType.SMALLSERIAL: 'SMALLSERIAL'>
BIGSERIAL = <TokenType.BIGSERIAL: 'BIGSERIAL'>
XML = <TokenType.XML: 'XML'>
YEAR = <TokenType.YEAR: 'YEAR'>
UNIQUEIDENTIFIER = <TokenType.UNIQUEIDENTIFIER: 'UNIQUEIDENTIFIER'>
USERDEFINED = <TokenType.USERDEFINED: 'USERDEFINED'>
MONEY = <TokenType.MONEY: 'MONEY'>
SMALLMONEY = <TokenType.SMALLMONEY: 'SMALLMONEY'>
ROWVERSION = <TokenType.ROWVERSION: 'ROWVERSION'>
IMAGE = <TokenType.IMAGE: 'IMAGE'>
VARIANT = <TokenType.VARIANT: 'VARIANT'>
OBJECT = <TokenType.OBJECT: 'OBJECT'>
INET = <TokenType.INET: 'INET'>
IPADDRESS = <TokenType.IPADDRESS: 'IPADDRESS'>
IPPREFIX = <TokenType.IPPREFIX: 'IPPREFIX'>
IPV4 = <TokenType.IPV4: 'IPV4'>
IPV6 = <TokenType.IPV6: 'IPV6'>
ENUM = <TokenType.ENUM: 'ENUM'>
ENUM8 = <TokenType.ENUM8: 'ENUM8'>
ENUM16 = <TokenType.ENUM16: 'ENUM16'>
FIXEDSTRING = <TokenType.FIXEDSTRING: 'FIXEDSTRING'>
LOWCARDINALITY = <TokenType.LOWCARDINALITY: 'LOWCARDINALITY'>
NESTED = <TokenType.NESTED: 'NESTED'>
AGGREGATEFUNCTION = <TokenType.AGGREGATEFUNCTION: 'AGGREGATEFUNCTION'>
SIMPLEAGGREGATEFUNCTION = <TokenType.SIMPLEAGGREGATEFUNCTION: 'SIMPLEAGGREGATEFUNCTION'>
UNKNOWN = <TokenType.UNKNOWN: 'UNKNOWN'>
ALIAS = <TokenType.ALIAS: 'ALIAS'>
ALTER = <TokenType.ALTER: 'ALTER'>
ALWAYS = <TokenType.ALWAYS: 'ALWAYS'>
ALL = <TokenType.ALL: 'ALL'>
ANTI = <TokenType.ANTI: 'ANTI'>
ANY = <TokenType.ANY: 'ANY'>
APPLY = <TokenType.APPLY: 'APPLY'>
ARRAY = <TokenType.ARRAY: 'ARRAY'>
ASC = <TokenType.ASC: 'ASC'>
ASOF = <TokenType.ASOF: 'ASOF'>
AUTO_INCREMENT = <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>
BEGIN = <TokenType.BEGIN: 'BEGIN'>
BETWEEN = <TokenType.BETWEEN: 'BETWEEN'>
CACHE = <TokenType.CACHE: 'CACHE'>
CASE = <TokenType.CASE: 'CASE'>
CHARACTER_SET = <TokenType.CHARACTER_SET: 'CHARACTER_SET'>
CLUSTER_BY = <TokenType.CLUSTER_BY: 'CLUSTER_BY'>
COLLATE = <TokenType.COLLATE: 'COLLATE'>
COMMAND = <TokenType.COMMAND: 'COMMAND'>
COMMENT = <TokenType.COMMENT: 'COMMENT'>
COMMIT = <TokenType.COMMIT: 'COMMIT'>
CONNECT_BY = <TokenType.CONNECT_BY: 'CONNECT_BY'>
CONSTRAINT = <TokenType.CONSTRAINT: 'CONSTRAINT'>
CREATE = <TokenType.CREATE: 'CREATE'>
CROSS = <TokenType.CROSS: 'CROSS'>
CUBE = <TokenType.CUBE: 'CUBE'>
CURRENT_DATE = <TokenType.CURRENT_DATE: 'CURRENT_DATE'>
CURRENT_DATETIME = <TokenType.CURRENT_DATETIME: 'CURRENT_DATETIME'>
CURRENT_TIME = <TokenType.CURRENT_TIME: 'CURRENT_TIME'>
CURRENT_TIMESTAMP = <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>
CURRENT_USER = <TokenType.CURRENT_USER: 'CURRENT_USER'>
DEFAULT = <TokenType.DEFAULT: 'DEFAULT'>
DELETE = <TokenType.DELETE: 'DELETE'>
DESC = <TokenType.DESC: 'DESC'>
DESCRIBE = <TokenType.DESCRIBE: 'DESCRIBE'>
DICTIONARY = <TokenType.DICTIONARY: 'DICTIONARY'>
DISTINCT = <TokenType.DISTINCT: 'DISTINCT'>
DISTRIBUTE_BY = <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>
DIV = <TokenType.DIV: 'DIV'>
DROP = <TokenType.DROP: 'DROP'>
ELSE = <TokenType.ELSE: 'ELSE'>
END = <TokenType.END: 'END'>
ESCAPE = <TokenType.ESCAPE: 'ESCAPE'>
EXCEPT = <TokenType.EXCEPT: 'EXCEPT'>
EXECUTE = <TokenType.EXECUTE: 'EXECUTE'>
EXISTS = <TokenType.EXISTS: 'EXISTS'>
FALSE = <TokenType.FALSE: 'FALSE'>
FETCH = <TokenType.FETCH: 'FETCH'>
FILTER = <TokenType.FILTER: 'FILTER'>
FINAL = <TokenType.FINAL: 'FINAL'>
FIRST = <TokenType.FIRST: 'FIRST'>
FOR = <TokenType.FOR: 'FOR'>
FORCE = <TokenType.FORCE: 'FORCE'>
FOREIGN_KEY = <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>
FORMAT = <TokenType.FORMAT: 'FORMAT'>
FROM = <TokenType.FROM: 'FROM'>
FULL = <TokenType.FULL: 'FULL'>
FUNCTION = <TokenType.FUNCTION: 'FUNCTION'>
GLOB = <TokenType.GLOB: 'GLOB'>
GLOBAL = <TokenType.GLOBAL: 'GLOBAL'>
GROUP_BY = <TokenType.GROUP_BY: 'GROUP_BY'>
GROUPING_SETS = <TokenType.GROUPING_SETS: 'GROUPING_SETS'>
HAVING = <TokenType.HAVING: 'HAVING'>
HINT = <TokenType.HINT: 'HINT'>
IGNORE = <TokenType.IGNORE: 'IGNORE'>
ILIKE = <TokenType.ILIKE: 'ILIKE'>
ILIKE_ANY = <TokenType.ILIKE_ANY: 'ILIKE_ANY'>
IN = <TokenType.IN: 'IN'>
INDEX = <TokenType.INDEX: 'INDEX'>
INNER = <TokenType.INNER: 'INNER'>
INSERT = <TokenType.INSERT: 'INSERT'>
INTERSECT = <TokenType.INTERSECT: 'INTERSECT'>
INTERVAL = <TokenType.INTERVAL: 'INTERVAL'>
INTO = <TokenType.INTO: 'INTO'>
INTRODUCER = <TokenType.INTRODUCER: 'INTRODUCER'>
IRLIKE = <TokenType.IRLIKE: 'IRLIKE'>
IS = <TokenType.IS: 'IS'>
ISNULL = <TokenType.ISNULL: 'ISNULL'>
JOIN = <TokenType.JOIN: 'JOIN'>
JOIN_MARKER = <TokenType.JOIN_MARKER: 'JOIN_MARKER'>
KEEP = <TokenType.KEEP: 'KEEP'>
KILL = <TokenType.KILL: 'KILL'>
LANGUAGE = <TokenType.LANGUAGE: 'LANGUAGE'>
LATERAL = <TokenType.LATERAL: 'LATERAL'>
LEFT = <TokenType.LEFT: 'LEFT'>
LIKE = <TokenType.LIKE: 'LIKE'>
LIKE_ANY = <TokenType.LIKE_ANY: 'LIKE_ANY'>
LIMIT = <TokenType.LIMIT: 'LIMIT'>
LOAD = <TokenType.LOAD: 'LOAD'>
LOCK = <TokenType.LOCK: 'LOCK'>
MAP = <TokenType.MAP: 'MAP'>
MATCH_RECOGNIZE = <TokenType.MATCH_RECOGNIZE: 'MATCH_RECOGNIZE'>
MEMBER_OF = <TokenType.MEMBER_OF: 'MEMBER_OF'>
MERGE = <TokenType.MERGE: 'MERGE'>
MOD = <TokenType.MOD: 'MOD'>
MODEL = <TokenType.MODEL: 'MODEL'>
NATURAL = <TokenType.NATURAL: 'NATURAL'>
NEXT = <TokenType.NEXT: 'NEXT'>
NOTNULL = <TokenType.NOTNULL: 'NOTNULL'>
NULL = <TokenType.NULL: 'NULL'>
OBJECT_IDENTIFIER = <TokenType.OBJECT_IDENTIFIER: 'OBJECT_IDENTIFIER'>
OFFSET = <TokenType.OFFSET: 'OFFSET'>
ON = <TokenType.ON: 'ON'>
ONLY = <TokenType.ONLY: 'ONLY'>
OPERATOR = <TokenType.OPERATOR: 'OPERATOR'>
ORDER_BY = <TokenType.ORDER_BY: 'ORDER_BY'>
ORDER_SIBLINGS_BY = <TokenType.ORDER_SIBLINGS_BY: 'ORDER_SIBLINGS_BY'>
ORDERED = <TokenType.ORDERED: 'ORDERED'>
ORDINALITY = <TokenType.ORDINALITY: 'ORDINALITY'>
OUTER = <TokenType.OUTER: 'OUTER'>
OVER = <TokenType.OVER: 'OVER'>
OVERLAPS = <TokenType.OVERLAPS: 'OVERLAPS'>
OVERWRITE = <TokenType.OVERWRITE: 'OVERWRITE'>
PARTITION = <TokenType.PARTITION: 'PARTITION'>
PARTITION_BY = <TokenType.PARTITION_BY: 'PARTITION_BY'>
PERCENT = <TokenType.PERCENT: 'PERCENT'>
PIVOT = <TokenType.PIVOT: 'PIVOT'>
PLACEHOLDER = <TokenType.PLACEHOLDER: 'PLACEHOLDER'>
POSITIONAL = <TokenType.POSITIONAL: 'POSITIONAL'>
PRAGMA = <TokenType.PRAGMA: 'PRAGMA'>
PREWHERE = <TokenType.PREWHERE: 'PREWHERE'>
PRIMARY_KEY = <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>
PROCEDURE = <TokenType.PROCEDURE: 'PROCEDURE'>
PROPERTIES = <TokenType.PROPERTIES: 'PROPERTIES'>
PSEUDO_TYPE = <TokenType.PSEUDO_TYPE: 'PSEUDO_TYPE'>
QUALIFY = <TokenType.QUALIFY: 'QUALIFY'>
QUOTE = <TokenType.QUOTE: 'QUOTE'>
RANGE = <TokenType.RANGE: 'RANGE'>
RECURSIVE = <TokenType.RECURSIVE: 'RECURSIVE'>
REFRESH = <TokenType.REFRESH: 'REFRESH'>
REPLACE = <TokenType.REPLACE: 'REPLACE'>
RETURNING = <TokenType.RETURNING: 'RETURNING'>
REFERENCES = <TokenType.REFERENCES: 'REFERENCES'>
RIGHT = <TokenType.RIGHT: 'RIGHT'>
RLIKE = <TokenType.RLIKE: 'RLIKE'>
ROLLBACK = <TokenType.ROLLBACK: 'ROLLBACK'>
ROLLUP = <TokenType.ROLLUP: 'ROLLUP'>
ROW = <TokenType.ROW: 'ROW'>
ROWS = <TokenType.ROWS: 'ROWS'>
SELECT = <TokenType.SELECT: 'SELECT'>
SEMI = <TokenType.SEMI: 'SEMI'>
SEPARATOR = <TokenType.SEPARATOR: 'SEPARATOR'>
SEQUENCE = <TokenType.SEQUENCE: 'SEQUENCE'>
SERDE_PROPERTIES = <TokenType.SERDE_PROPERTIES: 'SERDE_PROPERTIES'>
SET = <TokenType.SET: 'SET'>
SETTINGS = <TokenType.SETTINGS: 'SETTINGS'>
SHOW = <TokenType.SHOW: 'SHOW'>
SIMILAR_TO = <TokenType.SIMILAR_TO: 'SIMILAR_TO'>
SOME = <TokenType.SOME: 'SOME'>
SORT_BY = <TokenType.SORT_BY: 'SORT_BY'>
START_WITH = <TokenType.START_WITH: 'START_WITH'>
STORAGE_INTEGRATION = <TokenType.STORAGE_INTEGRATION: 'STORAGE_INTEGRATION'>
STRUCT = <TokenType.STRUCT: 'STRUCT'>
TABLE_SAMPLE = <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>
TEMPORARY = <TokenType.TEMPORARY: 'TEMPORARY'>
TOP = <TokenType.TOP: 'TOP'>
THEN = <TokenType.THEN: 'THEN'>
TRUE = <TokenType.TRUE: 'TRUE'>
TRUNCATE = <TokenType.TRUNCATE: 'TRUNCATE'>
UNCACHE = <TokenType.UNCACHE: 'UNCACHE'>
UNION = <TokenType.UNION: 'UNION'>
UNNEST = <TokenType.UNNEST: 'UNNEST'>
UNPIVOT = <TokenType.UNPIVOT: 'UNPIVOT'>
UPDATE = <TokenType.UPDATE: 'UPDATE'>
USE = <TokenType.USE: 'USE'>
USING = <TokenType.USING: 'USING'>
VALUES = <TokenType.VALUES: 'VALUES'>
VIEW = <TokenType.VIEW: 'VIEW'>
VOLATILE = <TokenType.VOLATILE: 'VOLATILE'>
WHEN = <TokenType.WHEN: 'WHEN'>
WHERE = <TokenType.WHERE: 'WHERE'>
WINDOW = <TokenType.WINDOW: 'WINDOW'>
WITH = <TokenType.WITH: 'WITH'>
UNIQUE = <TokenType.UNIQUE: 'UNIQUE'>
VERSION_SNAPSHOT = <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>
TIMESTAMP_SNAPSHOT = <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>
OPTION = <TokenType.OPTION: 'OPTION'>
Inherited Members
enum.Enum
name
value
class Token:
387class Token:
388    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
389
390    @classmethod
391    def number(cls, number: int) -> Token:
392        """Returns a NUMBER token with `number` as its text."""
393        return cls(TokenType.NUMBER, str(number))
394
395    @classmethod
396    def string(cls, string: str) -> Token:
397        """Returns a STRING token with `string` as its text."""
398        return cls(TokenType.STRING, string)
399
400    @classmethod
401    def identifier(cls, identifier: str) -> Token:
402        """Returns an IDENTIFIER token with `identifier` as its text."""
403        return cls(TokenType.IDENTIFIER, identifier)
404
405    @classmethod
406    def var(cls, var: str) -> Token:
407        """Returns an VAR token with `var` as its text."""
408        return cls(TokenType.VAR, var)
409
410    def __init__(
411        self,
412        token_type: TokenType,
413        text: str,
414        line: int = 1,
415        col: int = 1,
416        start: int = 0,
417        end: int = 0,
418        comments: t.Optional[t.List[str]] = None,
419    ) -> None:
420        """Token initializer.
421
422        Args:
423            token_type: The TokenType Enum.
424            text: The text of the token.
425            line: The line that the token ends on.
426            col: The column that the token ends on.
427            start: The start index of the token.
428            end: The ending index of the token.
429            comments: The comments to attach to the token.
430        """
431        self.token_type = token_type
432        self.text = text
433        self.line = line
434        self.col = col
435        self.start = start
436        self.end = end
437        self.comments = [] if comments is None else comments
438
439    def __repr__(self) -> str:
440        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
441        return f"<Token {attributes}>"
Token( token_type: TokenType, text: str, line: int = 1, col: int = 1, start: int = 0, end: int = 0, comments: Optional[List[str]] = None)
410    def __init__(
411        self,
412        token_type: TokenType,
413        text: str,
414        line: int = 1,
415        col: int = 1,
416        start: int = 0,
417        end: int = 0,
418        comments: t.Optional[t.List[str]] = None,
419    ) -> None:
420        """Token initializer.
421
422        Args:
423            token_type: The TokenType Enum.
424            text: The text of the token.
425            line: The line that the token ends on.
426            col: The column that the token ends on.
427            start: The start index of the token.
428            end: The ending index of the token.
429            comments: The comments to attach to the token.
430        """
431        self.token_type = token_type
432        self.text = text
433        self.line = line
434        self.col = col
435        self.start = start
436        self.end = end
437        self.comments = [] if comments is None else comments

Token initializer.

Arguments:
  • token_type: The TokenType Enum.
  • text: The text of the token.
  • line: The line that the token ends on.
  • col: The column that the token ends on.
  • start: The start index of the token.
  • end: The ending index of the token.
  • comments: The comments to attach to the token.
@classmethod
def number(cls, number: int) -> Token:
390    @classmethod
391    def number(cls, number: int) -> Token:
392        """Returns a NUMBER token with `number` as its text."""
393        return cls(TokenType.NUMBER, str(number))

Returns a NUMBER token with number as its text.

@classmethod
def string(cls, string: str) -> Token:
395    @classmethod
396    def string(cls, string: str) -> Token:
397        """Returns a STRING token with `string` as its text."""
398        return cls(TokenType.STRING, string)

Returns a STRING token with string as its text.

@classmethod
def identifier(cls, identifier: str) -> Token:
400    @classmethod
401    def identifier(cls, identifier: str) -> Token:
402        """Returns an IDENTIFIER token with `identifier` as its text."""
403        return cls(TokenType.IDENTIFIER, identifier)

Returns an IDENTIFIER token with identifier as its text.

@classmethod
def var(cls, var: str) -> Token:
405    @classmethod
406    def var(cls, var: str) -> Token:
407        """Returns an VAR token with `var` as its text."""
408        return cls(TokenType.VAR, var)

Returns an VAR token with var as its text.

token_type
text
line
col
start
end
comments
class Tokenizer:
 541class Tokenizer(metaclass=_Tokenizer):
 542    SINGLE_TOKENS = {
 543        "(": TokenType.L_PAREN,
 544        ")": TokenType.R_PAREN,
 545        "[": TokenType.L_BRACKET,
 546        "]": TokenType.R_BRACKET,
 547        "{": TokenType.L_BRACE,
 548        "}": TokenType.R_BRACE,
 549        "&": TokenType.AMP,
 550        "^": TokenType.CARET,
 551        ":": TokenType.COLON,
 552        ",": TokenType.COMMA,
 553        ".": TokenType.DOT,
 554        "-": TokenType.DASH,
 555        "=": TokenType.EQ,
 556        ">": TokenType.GT,
 557        "<": TokenType.LT,
 558        "%": TokenType.MOD,
 559        "!": TokenType.NOT,
 560        "|": TokenType.PIPE,
 561        "+": TokenType.PLUS,
 562        ";": TokenType.SEMICOLON,
 563        "/": TokenType.SLASH,
 564        "\\": TokenType.BACKSLASH,
 565        "*": TokenType.STAR,
 566        "~": TokenType.TILDA,
 567        "?": TokenType.PLACEHOLDER,
 568        "@": TokenType.PARAMETER,
 569        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 570        "'": TokenType.QUOTE,
 571        "`": TokenType.IDENTIFIER,
 572        '"': TokenType.IDENTIFIER,
 573        "#": TokenType.HASH,
 574    }
 575
 576    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 577    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 578    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 579    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 580    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 581    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 582    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 583    IDENTIFIER_ESCAPES = ['"']
 584    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 585    STRING_ESCAPES = ["'"]
 586    VAR_SINGLE_TOKENS: t.Set[str] = set()
 587
 588    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 589    HEREDOC_TAG_IS_IDENTIFIER = False
 590
 591    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 592    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 593
 594    # Autofilled
 595    _COMMENTS: t.Dict[str, str] = {}
 596    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 597    _IDENTIFIERS: t.Dict[str, str] = {}
 598    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 599    _QUOTES: t.Dict[str, str] = {}
 600    _STRING_ESCAPES: t.Set[str] = set()
 601    _KEYWORD_TRIE: t.Dict = {}
 602    _RS_TOKENIZER: t.Optional[t.Any] = None
 603
 604    KEYWORDS: t.Dict[str, TokenType] = {
 605        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 606        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 607        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 608        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 609        "/*+": TokenType.HINT,
 610        "==": TokenType.EQ,
 611        "::": TokenType.DCOLON,
 612        "||": TokenType.DPIPE,
 613        ">=": TokenType.GTE,
 614        "<=": TokenType.LTE,
 615        "<>": TokenType.NEQ,
 616        "!=": TokenType.NEQ,
 617        ":=": TokenType.COLON_EQ,
 618        "<=>": TokenType.NULLSAFE_EQ,
 619        "->": TokenType.ARROW,
 620        "->>": TokenType.DARROW,
 621        "=>": TokenType.FARROW,
 622        "#>": TokenType.HASH_ARROW,
 623        "#>>": TokenType.DHASH_ARROW,
 624        "<->": TokenType.LR_ARROW,
 625        "&&": TokenType.DAMP,
 626        "??": TokenType.DQMARK,
 627        "ALL": TokenType.ALL,
 628        "ALWAYS": TokenType.ALWAYS,
 629        "AND": TokenType.AND,
 630        "ANTI": TokenType.ANTI,
 631        "ANY": TokenType.ANY,
 632        "ASC": TokenType.ASC,
 633        "AS": TokenType.ALIAS,
 634        "ASOF": TokenType.ASOF,
 635        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 636        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 637        "BEGIN": TokenType.BEGIN,
 638        "BETWEEN": TokenType.BETWEEN,
 639        "CACHE": TokenType.CACHE,
 640        "UNCACHE": TokenType.UNCACHE,
 641        "CASE": TokenType.CASE,
 642        "CHARACTER SET": TokenType.CHARACTER_SET,
 643        "CLUSTER BY": TokenType.CLUSTER_BY,
 644        "COLLATE": TokenType.COLLATE,
 645        "COLUMN": TokenType.COLUMN,
 646        "COMMIT": TokenType.COMMIT,
 647        "CONNECT BY": TokenType.CONNECT_BY,
 648        "CONSTRAINT": TokenType.CONSTRAINT,
 649        "CREATE": TokenType.CREATE,
 650        "CROSS": TokenType.CROSS,
 651        "CUBE": TokenType.CUBE,
 652        "CURRENT_DATE": TokenType.CURRENT_DATE,
 653        "CURRENT_TIME": TokenType.CURRENT_TIME,
 654        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 655        "CURRENT_USER": TokenType.CURRENT_USER,
 656        "DATABASE": TokenType.DATABASE,
 657        "DEFAULT": TokenType.DEFAULT,
 658        "DELETE": TokenType.DELETE,
 659        "DESC": TokenType.DESC,
 660        "DESCRIBE": TokenType.DESCRIBE,
 661        "DISTINCT": TokenType.DISTINCT,
 662        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 663        "DIV": TokenType.DIV,
 664        "DROP": TokenType.DROP,
 665        "ELSE": TokenType.ELSE,
 666        "END": TokenType.END,
 667        "ENUM": TokenType.ENUM,
 668        "ESCAPE": TokenType.ESCAPE,
 669        "EXCEPT": TokenType.EXCEPT,
 670        "EXECUTE": TokenType.EXECUTE,
 671        "EXISTS": TokenType.EXISTS,
 672        "FALSE": TokenType.FALSE,
 673        "FETCH": TokenType.FETCH,
 674        "FILTER": TokenType.FILTER,
 675        "FIRST": TokenType.FIRST,
 676        "FULL": TokenType.FULL,
 677        "FUNCTION": TokenType.FUNCTION,
 678        "FOR": TokenType.FOR,
 679        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 680        "FORMAT": TokenType.FORMAT,
 681        "FROM": TokenType.FROM,
 682        "GEOGRAPHY": TokenType.GEOGRAPHY,
 683        "GEOMETRY": TokenType.GEOMETRY,
 684        "GLOB": TokenType.GLOB,
 685        "GROUP BY": TokenType.GROUP_BY,
 686        "GROUPING SETS": TokenType.GROUPING_SETS,
 687        "HAVING": TokenType.HAVING,
 688        "ILIKE": TokenType.ILIKE,
 689        "IN": TokenType.IN,
 690        "INDEX": TokenType.INDEX,
 691        "INET": TokenType.INET,
 692        "INNER": TokenType.INNER,
 693        "INSERT": TokenType.INSERT,
 694        "INTERVAL": TokenType.INTERVAL,
 695        "INTERSECT": TokenType.INTERSECT,
 696        "INTO": TokenType.INTO,
 697        "IS": TokenType.IS,
 698        "ISNULL": TokenType.ISNULL,
 699        "JOIN": TokenType.JOIN,
 700        "KEEP": TokenType.KEEP,
 701        "KILL": TokenType.KILL,
 702        "LATERAL": TokenType.LATERAL,
 703        "LEFT": TokenType.LEFT,
 704        "LIKE": TokenType.LIKE,
 705        "LIMIT": TokenType.LIMIT,
 706        "LOAD": TokenType.LOAD,
 707        "LOCK": TokenType.LOCK,
 708        "MERGE": TokenType.MERGE,
 709        "NATURAL": TokenType.NATURAL,
 710        "NEXT": TokenType.NEXT,
 711        "NOT": TokenType.NOT,
 712        "NOTNULL": TokenType.NOTNULL,
 713        "NULL": TokenType.NULL,
 714        "OBJECT": TokenType.OBJECT,
 715        "OFFSET": TokenType.OFFSET,
 716        "ON": TokenType.ON,
 717        "OR": TokenType.OR,
 718        "XOR": TokenType.XOR,
 719        "ORDER BY": TokenType.ORDER_BY,
 720        "ORDINALITY": TokenType.ORDINALITY,
 721        "OUTER": TokenType.OUTER,
 722        "OVER": TokenType.OVER,
 723        "OVERLAPS": TokenType.OVERLAPS,
 724        "OVERWRITE": TokenType.OVERWRITE,
 725        "PARTITION": TokenType.PARTITION,
 726        "PARTITION BY": TokenType.PARTITION_BY,
 727        "PARTITIONED BY": TokenType.PARTITION_BY,
 728        "PARTITIONED_BY": TokenType.PARTITION_BY,
 729        "PERCENT": TokenType.PERCENT,
 730        "PIVOT": TokenType.PIVOT,
 731        "PRAGMA": TokenType.PRAGMA,
 732        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 733        "PROCEDURE": TokenType.PROCEDURE,
 734        "QUALIFY": TokenType.QUALIFY,
 735        "RANGE": TokenType.RANGE,
 736        "RECURSIVE": TokenType.RECURSIVE,
 737        "REGEXP": TokenType.RLIKE,
 738        "REPLACE": TokenType.REPLACE,
 739        "RETURNING": TokenType.RETURNING,
 740        "REFERENCES": TokenType.REFERENCES,
 741        "RIGHT": TokenType.RIGHT,
 742        "RLIKE": TokenType.RLIKE,
 743        "ROLLBACK": TokenType.ROLLBACK,
 744        "ROLLUP": TokenType.ROLLUP,
 745        "ROW": TokenType.ROW,
 746        "ROWS": TokenType.ROWS,
 747        "SCHEMA": TokenType.SCHEMA,
 748        "SELECT": TokenType.SELECT,
 749        "SEMI": TokenType.SEMI,
 750        "SET": TokenType.SET,
 751        "SETTINGS": TokenType.SETTINGS,
 752        "SHOW": TokenType.SHOW,
 753        "SIMILAR TO": TokenType.SIMILAR_TO,
 754        "SOME": TokenType.SOME,
 755        "SORT BY": TokenType.SORT_BY,
 756        "START WITH": TokenType.START_WITH,
 757        "TABLE": TokenType.TABLE,
 758        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 759        "TEMP": TokenType.TEMPORARY,
 760        "TEMPORARY": TokenType.TEMPORARY,
 761        "THEN": TokenType.THEN,
 762        "TRUE": TokenType.TRUE,
 763        "TRUNCATE": TokenType.TRUNCATE,
 764        "UNION": TokenType.UNION,
 765        "UNKNOWN": TokenType.UNKNOWN,
 766        "UNNEST": TokenType.UNNEST,
 767        "UNPIVOT": TokenType.UNPIVOT,
 768        "UPDATE": TokenType.UPDATE,
 769        "USE": TokenType.USE,
 770        "USING": TokenType.USING,
 771        "UUID": TokenType.UUID,
 772        "VALUES": TokenType.VALUES,
 773        "VIEW": TokenType.VIEW,
 774        "VOLATILE": TokenType.VOLATILE,
 775        "WHEN": TokenType.WHEN,
 776        "WHERE": TokenType.WHERE,
 777        "WINDOW": TokenType.WINDOW,
 778        "WITH": TokenType.WITH,
 779        "APPLY": TokenType.APPLY,
 780        "ARRAY": TokenType.ARRAY,
 781        "BIT": TokenType.BIT,
 782        "BOOL": TokenType.BOOLEAN,
 783        "BOOLEAN": TokenType.BOOLEAN,
 784        "BYTE": TokenType.TINYINT,
 785        "MEDIUMINT": TokenType.MEDIUMINT,
 786        "INT1": TokenType.TINYINT,
 787        "TINYINT": TokenType.TINYINT,
 788        "INT16": TokenType.SMALLINT,
 789        "SHORT": TokenType.SMALLINT,
 790        "SMALLINT": TokenType.SMALLINT,
 791        "INT128": TokenType.INT128,
 792        "HUGEINT": TokenType.INT128,
 793        "INT2": TokenType.SMALLINT,
 794        "INTEGER": TokenType.INT,
 795        "INT": TokenType.INT,
 796        "INT4": TokenType.INT,
 797        "INT32": TokenType.INT,
 798        "INT64": TokenType.BIGINT,
 799        "LONG": TokenType.BIGINT,
 800        "BIGINT": TokenType.BIGINT,
 801        "INT8": TokenType.TINYINT,
 802        "DEC": TokenType.DECIMAL,
 803        "DECIMAL": TokenType.DECIMAL,
 804        "BIGDECIMAL": TokenType.BIGDECIMAL,
 805        "BIGNUMERIC": TokenType.BIGDECIMAL,
 806        "MAP": TokenType.MAP,
 807        "NULLABLE": TokenType.NULLABLE,
 808        "NUMBER": TokenType.DECIMAL,
 809        "NUMERIC": TokenType.DECIMAL,
 810        "FIXED": TokenType.DECIMAL,
 811        "REAL": TokenType.FLOAT,
 812        "FLOAT": TokenType.FLOAT,
 813        "FLOAT4": TokenType.FLOAT,
 814        "FLOAT8": TokenType.DOUBLE,
 815        "DOUBLE": TokenType.DOUBLE,
 816        "DOUBLE PRECISION": TokenType.DOUBLE,
 817        "JSON": TokenType.JSON,
 818        "CHAR": TokenType.CHAR,
 819        "CHARACTER": TokenType.CHAR,
 820        "NCHAR": TokenType.NCHAR,
 821        "VARCHAR": TokenType.VARCHAR,
 822        "VARCHAR2": TokenType.VARCHAR,
 823        "NVARCHAR": TokenType.NVARCHAR,
 824        "NVARCHAR2": TokenType.NVARCHAR,
 825        "BPCHAR": TokenType.BPCHAR,
 826        "STR": TokenType.TEXT,
 827        "STRING": TokenType.TEXT,
 828        "TEXT": TokenType.TEXT,
 829        "LONGTEXT": TokenType.LONGTEXT,
 830        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 831        "TINYTEXT": TokenType.TINYTEXT,
 832        "CLOB": TokenType.TEXT,
 833        "LONGVARCHAR": TokenType.TEXT,
 834        "BINARY": TokenType.BINARY,
 835        "BLOB": TokenType.VARBINARY,
 836        "LONGBLOB": TokenType.LONGBLOB,
 837        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 838        "TINYBLOB": TokenType.TINYBLOB,
 839        "BYTEA": TokenType.VARBINARY,
 840        "VARBINARY": TokenType.VARBINARY,
 841        "TIME": TokenType.TIME,
 842        "TIMETZ": TokenType.TIMETZ,
 843        "TIMESTAMP": TokenType.TIMESTAMP,
 844        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 845        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 846        "DATE": TokenType.DATE,
 847        "DATETIME": TokenType.DATETIME,
 848        "INT4RANGE": TokenType.INT4RANGE,
 849        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 850        "INT8RANGE": TokenType.INT8RANGE,
 851        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 852        "NUMRANGE": TokenType.NUMRANGE,
 853        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 854        "TSRANGE": TokenType.TSRANGE,
 855        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 856        "TSTZRANGE": TokenType.TSTZRANGE,
 857        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 858        "DATERANGE": TokenType.DATERANGE,
 859        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 860        "UNIQUE": TokenType.UNIQUE,
 861        "STRUCT": TokenType.STRUCT,
 862        "SEQUENCE": TokenType.SEQUENCE,
 863        "VARIANT": TokenType.VARIANT,
 864        "ALTER": TokenType.ALTER,
 865        "ANALYZE": TokenType.COMMAND,
 866        "CALL": TokenType.COMMAND,
 867        "COMMENT": TokenType.COMMENT,
 868        "COPY": TokenType.COMMAND,
 869        "EXPLAIN": TokenType.COMMAND,
 870        "GRANT": TokenType.COMMAND,
 871        "OPTIMIZE": TokenType.COMMAND,
 872        "PREPARE": TokenType.COMMAND,
 873        "VACUUM": TokenType.COMMAND,
 874        "USER-DEFINED": TokenType.USERDEFINED,
 875        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 876        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 877    }
 878
 879    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 880        " ": TokenType.SPACE,
 881        "\t": TokenType.SPACE,
 882        "\n": TokenType.BREAK,
 883        "\r": TokenType.BREAK,
 884    }
 885
 886    COMMANDS = {
 887        TokenType.COMMAND,
 888        TokenType.EXECUTE,
 889        TokenType.FETCH,
 890        TokenType.SHOW,
 891    }
 892
 893    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 894
 895    # Handle numeric literals like in hive (3L = BIGINT)
 896    NUMERIC_LITERALS: t.Dict[str, str] = {}
 897
 898    COMMENTS = ["--", ("/*", "*/")]
 899
 900    __slots__ = (
 901        "sql",
 902        "size",
 903        "tokens",
 904        "dialect",
 905        "_start",
 906        "_current",
 907        "_line",
 908        "_col",
 909        "_comments",
 910        "_char",
 911        "_end",
 912        "_peek",
 913        "_prev_token_line",
 914        "_rs_dialect_settings",
 915    )
 916
 917    def __init__(self, dialect: DialectType = None) -> None:
 918        from sqlglot.dialects import Dialect
 919
 920        self.dialect = Dialect.get_or_raise(dialect)
 921
 922        if USE_RS_TOKENIZER:
 923            self._rs_dialect_settings = RsTokenizerDialectSettings(
 924                escape_sequences=self.dialect.ESCAPE_SEQUENCES,
 925                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 926            )
 927
 928        self.reset()
 929
 930    def reset(self) -> None:
 931        self.sql = ""
 932        self.size = 0
 933        self.tokens: t.List[Token] = []
 934        self._start = 0
 935        self._current = 0
 936        self._line = 1
 937        self._col = 0
 938        self._comments: t.List[str] = []
 939
 940        self._char = ""
 941        self._end = False
 942        self._peek = ""
 943        self._prev_token_line = -1
 944
 945    def tokenize(self, sql: str) -> t.List[Token]:
 946        """Returns a list of tokens corresponding to the SQL string `sql`."""
 947        if USE_RS_TOKENIZER:
 948            return self.tokenize_rs(sql)
 949
 950        self.reset()
 951        self.sql = sql
 952        self.size = len(sql)
 953
 954        try:
 955            self._scan()
 956        except Exception as e:
 957            start = max(self._current - 50, 0)
 958            end = min(self._current + 50, self.size - 1)
 959            context = self.sql[start:end]
 960            raise TokenError(f"Error tokenizing '{context}'") from e
 961
 962        return self.tokens
 963
 964    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
 965        while self.size and not self._end:
 966            current = self._current
 967
 968            # Skip spaces here rather than iteratively calling advance() for performance reasons
 969            while current < self.size:
 970                char = self.sql[current]
 971
 972                if char.isspace() and (char == " " or char == "\t"):
 973                    current += 1
 974                else:
 975                    break
 976
 977            offset = current - self._current if current > self._current else 1
 978
 979            self._start = current
 980            self._advance(offset)
 981
 982            if not self._char.isspace():
 983                if self._char.isdigit():
 984                    self._scan_number()
 985                elif self._char in self._IDENTIFIERS:
 986                    self._scan_identifier(self._IDENTIFIERS[self._char])
 987                else:
 988                    self._scan_keywords()
 989
 990            if until and until():
 991                break
 992
 993        if self.tokens and self._comments:
 994            self.tokens[-1].comments.extend(self._comments)
 995
 996    def _chars(self, size: int) -> str:
 997        if size == 1:
 998            return self._char
 999
1000        start = self._current - 1
1001        end = start + size
1002
1003        return self.sql[start:end] if end <= self.size else ""
1004
1005    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1006        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1007            # Ensures we don't count an extra line if we get a \r\n line break sequence
1008            if not (self._char == "\r" and self._peek == "\n"):
1009                self._col = 1
1010                self._line += 1
1011        else:
1012            self._col += i
1013
1014        self._current += i
1015        self._end = self._current >= self.size
1016        self._char = self.sql[self._current - 1]
1017        self._peek = "" if self._end else self.sql[self._current]
1018
1019        if alnum and self._char.isalnum():
1020            # Here we use local variables instead of attributes for better performance
1021            _col = self._col
1022            _current = self._current
1023            _end = self._end
1024            _peek = self._peek
1025
1026            while _peek.isalnum():
1027                _col += 1
1028                _current += 1
1029                _end = _current >= self.size
1030                _peek = "" if _end else self.sql[_current]
1031
1032            self._col = _col
1033            self._current = _current
1034            self._end = _end
1035            self._peek = _peek
1036            self._char = self.sql[_current - 1]
1037
1038    @property
1039    def _text(self) -> str:
1040        return self.sql[self._start : self._current]
1041
1042    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1043        self._prev_token_line = self._line
1044
1045        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1046            self.tokens[-1].comments.extend(self._comments)
1047            self._comments = []
1048
1049        self.tokens.append(
1050            Token(
1051                token_type,
1052                text=self._text if text is None else text,
1053                line=self._line,
1054                col=self._col,
1055                start=self._start,
1056                end=self._current - 1,
1057                comments=self._comments,
1058            )
1059        )
1060        self._comments = []
1061
1062        # If we have either a semicolon or a begin token before the command's token, we'll parse
1063        # whatever follows the command's token as a string
1064        if (
1065            token_type in self.COMMANDS
1066            and self._peek != ";"
1067            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1068        ):
1069            start = self._current
1070            tokens = len(self.tokens)
1071            self._scan(lambda: self._peek == ";")
1072            self.tokens = self.tokens[:tokens]
1073            text = self.sql[start : self._current].strip()
1074            if text:
1075                self._add(TokenType.STRING, text)
1076
1077    def _scan_keywords(self) -> None:
1078        size = 0
1079        word = None
1080        chars = self._text
1081        char = chars
1082        prev_space = False
1083        skip = False
1084        trie = self._KEYWORD_TRIE
1085        single_token = char in self.SINGLE_TOKENS
1086
1087        while chars:
1088            if skip:
1089                result = TrieResult.PREFIX
1090            else:
1091                result, trie = in_trie(trie, char.upper())
1092
1093            if result == TrieResult.FAILED:
1094                break
1095            if result == TrieResult.EXISTS:
1096                word = chars
1097
1098            end = self._current + size
1099            size += 1
1100
1101            if end < self.size:
1102                char = self.sql[end]
1103                single_token = single_token or char in self.SINGLE_TOKENS
1104                is_space = char.isspace()
1105
1106                if not is_space or not prev_space:
1107                    if is_space:
1108                        char = " "
1109                    chars += char
1110                    prev_space = is_space
1111                    skip = False
1112                else:
1113                    skip = True
1114            else:
1115                char = ""
1116                break
1117
1118        if word:
1119            if self._scan_string(word):
1120                return
1121            if self._scan_comment(word):
1122                return
1123            if prev_space or single_token or not char:
1124                self._advance(size - 1)
1125                word = word.upper()
1126                self._add(self.KEYWORDS[word], text=word)
1127                return
1128
1129        if self._char in self.SINGLE_TOKENS:
1130            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1131            return
1132
1133        self._scan_var()
1134
1135    def _scan_comment(self, comment_start: str) -> bool:
1136        if comment_start not in self._COMMENTS:
1137            return False
1138
1139        comment_start_line = self._line
1140        comment_start_size = len(comment_start)
1141        comment_end = self._COMMENTS[comment_start]
1142
1143        if comment_end:
1144            # Skip the comment's start delimiter
1145            self._advance(comment_start_size)
1146
1147            comment_end_size = len(comment_end)
1148            while not self._end and self._chars(comment_end_size) != comment_end:
1149                self._advance(alnum=True)
1150
1151            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1152            self._advance(comment_end_size - 1)
1153        else:
1154            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1155                self._advance(alnum=True)
1156            self._comments.append(self._text[comment_start_size:])
1157
1158        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1159        # Multiple consecutive comments are preserved by appending them to the current comments list.
1160        if comment_start_line == self._prev_token_line:
1161            self.tokens[-1].comments.extend(self._comments)
1162            self._comments = []
1163            self._prev_token_line = self._line
1164
1165        return True
1166
1167    def _scan_number(self) -> None:
1168        if self._char == "0":
1169            peek = self._peek.upper()
1170            if peek == "B":
1171                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1172            elif peek == "X":
1173                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1174
1175        decimal = False
1176        scientific = 0
1177
1178        while True:
1179            if self._peek.isdigit():
1180                self._advance()
1181            elif self._peek == "." and not decimal:
1182                decimal = True
1183                self._advance()
1184            elif self._peek in ("-", "+") and scientific == 1:
1185                scientific += 1
1186                self._advance()
1187            elif self._peek.upper() == "E" and not scientific:
1188                scientific += 1
1189                self._advance()
1190            elif self._peek.isidentifier():
1191                number_text = self._text
1192                literal = ""
1193
1194                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1195                    literal += self._peek
1196                    self._advance()
1197
1198                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1199
1200                if token_type:
1201                    self._add(TokenType.NUMBER, number_text)
1202                    self._add(TokenType.DCOLON, "::")
1203                    return self._add(token_type, literal)
1204                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1205                    return self._add(TokenType.VAR)
1206
1207                self._advance(-len(literal))
1208                return self._add(TokenType.NUMBER, number_text)
1209            else:
1210                return self._add(TokenType.NUMBER)
1211
1212    def _scan_bits(self) -> None:
1213        self._advance()
1214        value = self._extract_value()
1215        try:
1216            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1217            int(value, 2)
1218            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1219        except ValueError:
1220            self._add(TokenType.IDENTIFIER)
1221
1222    def _scan_hex(self) -> None:
1223        self._advance()
1224        value = self._extract_value()
1225        try:
1226            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1227            int(value, 16)
1228            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1229        except ValueError:
1230            self._add(TokenType.IDENTIFIER)
1231
1232    def _extract_value(self) -> str:
1233        while True:
1234            char = self._peek.strip()
1235            if char and char not in self.SINGLE_TOKENS:
1236                self._advance(alnum=True)
1237            else:
1238                break
1239
1240        return self._text
1241
1242    def _scan_string(self, start: str) -> bool:
1243        base = None
1244        token_type = TokenType.STRING
1245
1246        if start in self._QUOTES:
1247            end = self._QUOTES[start]
1248        elif start in self._FORMAT_STRINGS:
1249            end, token_type = self._FORMAT_STRINGS[start]
1250
1251            if token_type == TokenType.HEX_STRING:
1252                base = 16
1253            elif token_type == TokenType.BIT_STRING:
1254                base = 2
1255            elif token_type == TokenType.HEREDOC_STRING:
1256                if (
1257                    self.HEREDOC_TAG_IS_IDENTIFIER
1258                    and not self._peek.isidentifier()
1259                    and not self._peek == end
1260                ):
1261                    if self.HEREDOC_STRING_ALTERNATIVE != token_type.VAR:
1262                        self._add(self.HEREDOC_STRING_ALTERNATIVE)
1263                    else:
1264                        self._scan_var()
1265
1266                    return True
1267
1268                self._advance()
1269
1270                if self._char == end:
1271                    tag = ""
1272                else:
1273                    tag = self._extract_string(
1274                        end, raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER
1275                    )
1276
1277                if self._end and tag and self.HEREDOC_TAG_IS_IDENTIFIER:
1278                    self._advance(-len(tag))
1279                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1280                    return True
1281
1282                end = f"{start}{tag}{end}"
1283        else:
1284            return False
1285
1286        self._advance(len(start))
1287        text = self._extract_string(end)
1288
1289        if base:
1290            try:
1291                int(text, base)
1292            except Exception:
1293                raise TokenError(
1294                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1295                )
1296
1297        self._add(token_type, text)
1298        return True
1299
1300    def _scan_identifier(self, identifier_end: str) -> None:
1301        self._advance()
1302        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1303        self._add(TokenType.IDENTIFIER, text)
1304
1305    def _scan_var(self) -> None:
1306        while True:
1307            char = self._peek.strip()
1308            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1309                self._advance(alnum=True)
1310            else:
1311                break
1312
1313        self._add(
1314            TokenType.VAR
1315            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1316            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1317        )
1318
1319    def _extract_string(
1320        self, delimiter: str, escapes: t.Optional[t.Set[str]] = None, raise_unmatched: bool = True
1321    ) -> str:
1322        text = ""
1323        delim_size = len(delimiter)
1324        escapes = self._STRING_ESCAPES if escapes is None else escapes
1325
1326        while True:
1327            if (
1328                self._char in escapes
1329                and (self._peek == delimiter or self._peek in escapes)
1330                and (self._char not in self._QUOTES or self._char == self._peek)
1331            ):
1332                if self._peek == delimiter:
1333                    text += self._peek
1334                else:
1335                    text += self._char + self._peek
1336
1337                if self._current + 1 < self.size:
1338                    self._advance(2)
1339                else:
1340                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1341            else:
1342                if self._chars(delim_size) == delimiter:
1343                    if delim_size > 1:
1344                        self._advance(delim_size - 1)
1345                    break
1346
1347                if self._end:
1348                    if not raise_unmatched:
1349                        return text + self._char
1350
1351                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1352
1353                if (
1354                    self.dialect.ESCAPE_SEQUENCES
1355                    and self._peek
1356                    and self._char in self.STRING_ESCAPES
1357                ):
1358                    escaped_sequence = self.dialect.ESCAPE_SEQUENCES.get(self._char + self._peek)
1359                    if escaped_sequence:
1360                        self._advance(2)
1361                        text += escaped_sequence
1362                        continue
1363
1364                current = self._current - 1
1365                self._advance(alnum=True)
1366                text += self.sql[current : self._current - 1]
1367
1368        return text
1369
1370    def tokenize_rs(self, sql: str) -> t.List[Token]:
1371        if not self._RS_TOKENIZER:
1372            raise SqlglotError("Rust tokenizer is not available")
1373
1374        try:
1375            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1376            for token in tokens:
1377                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1378            return tokens
1379        except Exception as e:
1380            raise TokenError(str(e))
Tokenizer( dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None)
917    def __init__(self, dialect: DialectType = None) -> None:
918        from sqlglot.dialects import Dialect
919
920        self.dialect = Dialect.get_or_raise(dialect)
921
922        if USE_RS_TOKENIZER:
923            self._rs_dialect_settings = RsTokenizerDialectSettings(
924                escape_sequences=self.dialect.ESCAPE_SEQUENCES,
925                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
926            )
927
928        self.reset()
SINGLE_TOKENS = {'(': <TokenType.L_PAREN: 'L_PAREN'>, ')': <TokenType.R_PAREN: 'R_PAREN'>, '[': <TokenType.L_BRACKET: 'L_BRACKET'>, ']': <TokenType.R_BRACKET: 'R_BRACKET'>, '{': <TokenType.L_BRACE: 'L_BRACE'>, '}': <TokenType.R_BRACE: 'R_BRACE'>, '&': <TokenType.AMP: 'AMP'>, '^': <TokenType.CARET: 'CARET'>, ':': <TokenType.COLON: 'COLON'>, ',': <TokenType.COMMA: 'COMMA'>, '.': <TokenType.DOT: 'DOT'>, '-': <TokenType.DASH: 'DASH'>, '=': <TokenType.EQ: 'EQ'>, '>': <TokenType.GT: 'GT'>, '<': <TokenType.LT: 'LT'>, '%': <TokenType.MOD: 'MOD'>, '!': <TokenType.NOT: 'NOT'>, '|': <TokenType.PIPE: 'PIPE'>, '+': <TokenType.PLUS: 'PLUS'>, ';': <TokenType.SEMICOLON: 'SEMICOLON'>, '/': <TokenType.SLASH: 'SLASH'>, '\\': <TokenType.BACKSLASH: 'BACKSLASH'>, '*': <TokenType.STAR: 'STAR'>, '~': <TokenType.TILDA: 'TILDA'>, '?': <TokenType.PLACEHOLDER: 'PLACEHOLDER'>, '@': <TokenType.PARAMETER: 'PARAMETER'>, "'": <TokenType.QUOTE: 'QUOTE'>, '`': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '"': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '#': <TokenType.HASH: 'HASH'>}
BIT_STRINGS: List[Union[str, Tuple[str, str]]] = []
BYTE_STRINGS: List[Union[str, Tuple[str, str]]] = []
HEX_STRINGS: List[Union[str, Tuple[str, str]]] = []
RAW_STRINGS: List[Union[str, Tuple[str, str]]] = []
HEREDOC_STRINGS: List[Union[str, Tuple[str, str]]] = []
UNICODE_STRINGS: List[Union[str, Tuple[str, str]]] = []
IDENTIFIERS: List[Union[str, Tuple[str, str]]] = ['"']
IDENTIFIER_ESCAPES = ['"']
QUOTES: List[Union[str, Tuple[str, str]]] = ["'"]
STRING_ESCAPES = ["'"]
VAR_SINGLE_TOKENS: Set[str] = set()
HEREDOC_TAG_IS_IDENTIFIER = False
HEREDOC_STRING_ALTERNATIVE = <TokenType.VAR: 'VAR'>
KEYWORDS: Dict[str, TokenType] = {'{%': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%-': <TokenType.BLOCK_START: 'BLOCK_START'>, '%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '+%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '{{+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{{-': <TokenType.BLOCK_START: 'BLOCK_START'>, '+}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '/*+': <TokenType.HINT: 'HINT'>, '==': <TokenType.EQ: 'EQ'>, '::': <TokenType.DCOLON: 'DCOLON'>, '||': <TokenType.DPIPE: 'DPIPE'>, '>=': <TokenType.GTE: 'GTE'>, '<=': <TokenType.LTE: 'LTE'>, '<>': <TokenType.NEQ: 'NEQ'>, '!=': <TokenType.NEQ: 'NEQ'>, ':=': <TokenType.COLON_EQ: 'COLON_EQ'>, '<=>': <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>, '->': <TokenType.ARROW: 'ARROW'>, '->>': <TokenType.DARROW: 'DARROW'>, '=>': <TokenType.FARROW: 'FARROW'>, '#>': <TokenType.HASH_ARROW: 'HASH_ARROW'>, '#>>': <TokenType.DHASH_ARROW: 'DHASH_ARROW'>, '<->': <TokenType.LR_ARROW: 'LR_ARROW'>, '&&': <TokenType.DAMP: 'DAMP'>, '??': <TokenType.DQMARK: 'DQMARK'>, 'ALL': <TokenType.ALL: 'ALL'>, 'ALWAYS': <TokenType.ALWAYS: 'ALWAYS'>, 'AND': <TokenType.AND: 'AND'>, 'ANTI': <TokenType.ANTI: 'ANTI'>, 'ANY': <TokenType.ANY: 'ANY'>, 'ASC': <TokenType.ASC: 'ASC'>, 'AS': <TokenType.ALIAS: 'ALIAS'>, 'ASOF': <TokenType.ASOF: 'ASOF'>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'BEGIN': <TokenType.BEGIN: 'BEGIN'>, 'BETWEEN': <TokenType.BETWEEN: 'BETWEEN'>, 'CACHE': <TokenType.CACHE: 'CACHE'>, 'UNCACHE': <TokenType.UNCACHE: 'UNCACHE'>, 'CASE': <TokenType.CASE: 'CASE'>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 'CHARACTER_SET'>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 'CLUSTER_BY'>, 'COLLATE': <TokenType.COLLATE: 'COLLATE'>, 'COLUMN': <TokenType.COLUMN: 'COLUMN'>, 'COMMIT': <TokenType.COMMIT: 'COMMIT'>, 'CONNECT BY': <TokenType.CONNECT_BY: 'CONNECT_BY'>, 'CONSTRAINT': <TokenType.CONSTRAINT: 'CONSTRAINT'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ENUM': <TokenType.ENUM: 'ENUM'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'TRUNCATE': <TokenType.TRUNCATE: 'TRUNCATE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, 'INT1': <TokenType.TINYINT: 'TINYINT'>, 'TINYINT': <TokenType.TINYINT: 'TINYINT'>, 'INT16': <TokenType.SMALLINT: 'SMALLINT'>, 'SHORT': <TokenType.SMALLINT: 'SMALLINT'>, 'SMALLINT': <TokenType.SMALLINT: 'SMALLINT'>, 'INT128': <TokenType.INT128: 'INT128'>, 'HUGEINT': <TokenType.INT128: 'INT128'>, 'INT2': <TokenType.SMALLINT: 'SMALLINT'>, 'INTEGER': <TokenType.INT: 'INT'>, 'INT': <TokenType.INT: 'INT'>, 'INT4': <TokenType.INT: 'INT'>, 'INT32': <TokenType.INT: 'INT'>, 'INT64': <TokenType.BIGINT: 'BIGINT'>, 'LONG': <TokenType.BIGINT: 'BIGINT'>, 'BIGINT': <TokenType.BIGINT: 'BIGINT'>, 'INT8': <TokenType.TINYINT: 'TINYINT'>, 'DEC': <TokenType.DECIMAL: 'DECIMAL'>, 'DECIMAL': <TokenType.DECIMAL: 'DECIMAL'>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'MAP': <TokenType.MAP: 'MAP'>, 'NULLABLE': <TokenType.NULLABLE: 'NULLABLE'>, 'NUMBER': <TokenType.DECIMAL: 'DECIMAL'>, 'NUMERIC': <TokenType.DECIMAL: 'DECIMAL'>, 'FIXED': <TokenType.DECIMAL: 'DECIMAL'>, 'REAL': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT4': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT8': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE PRECISION': <TokenType.DOUBLE: 'DOUBLE'>, 'JSON': <TokenType.JSON: 'JSON'>, 'CHAR': <TokenType.CHAR: 'CHAR'>, 'CHARACTER': <TokenType.CHAR: 'CHAR'>, 'NCHAR': <TokenType.NCHAR: 'NCHAR'>, 'VARCHAR': <TokenType.VARCHAR: 'VARCHAR'>, 'VARCHAR2': <TokenType.VARCHAR: 'VARCHAR'>, 'NVARCHAR': <TokenType.NVARCHAR: 'NVARCHAR'>, 'NVARCHAR2': <TokenType.NVARCHAR: 'NVARCHAR'>, 'BPCHAR': <TokenType.BPCHAR: 'BPCHAR'>, 'STR': <TokenType.TEXT: 'TEXT'>, 'STRING': <TokenType.TEXT: 'TEXT'>, 'TEXT': <TokenType.TEXT: 'TEXT'>, 'LONGTEXT': <TokenType.LONGTEXT: 'LONGTEXT'>, 'MEDIUMTEXT': <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, 'TINYTEXT': <TokenType.TINYTEXT: 'TINYTEXT'>, 'CLOB': <TokenType.TEXT: 'TEXT'>, 'LONGVARCHAR': <TokenType.TEXT: 'TEXT'>, 'BINARY': <TokenType.BINARY: 'BINARY'>, 'BLOB': <TokenType.VARBINARY: 'VARBINARY'>, 'LONGBLOB': <TokenType.LONGBLOB: 'LONGBLOB'>, 'MEDIUMBLOB': <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, 'TINYBLOB': <TokenType.TINYBLOB: 'TINYBLOB'>, 'BYTEA': <TokenType.VARBINARY: 'VARBINARY'>, 'VARBINARY': <TokenType.VARBINARY: 'VARBINARY'>, 'TIME': <TokenType.TIME: 'TIME'>, 'TIMETZ': <TokenType.TIMETZ: 'TIMETZ'>, 'TIMESTAMP': <TokenType.TIMESTAMP: 'TIMESTAMP'>, 'TIMESTAMPTZ': <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, 'TIMESTAMPLTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'DATE': <TokenType.DATE: 'DATE'>, 'DATETIME': <TokenType.DATETIME: 'DATETIME'>, 'INT4RANGE': <TokenType.INT4RANGE: 'INT4RANGE'>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, 'INT8RANGE': <TokenType.INT8RANGE: 'INT8RANGE'>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, 'NUMRANGE': <TokenType.NUMRANGE: 'NUMRANGE'>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, 'TSRANGE': <TokenType.TSRANGE: 'TSRANGE'>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, 'TSTZRANGE': <TokenType.TSTZRANGE: 'TSTZRANGE'>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, 'DATERANGE': <TokenType.DATERANGE: 'DATERANGE'>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, 'UNIQUE': <TokenType.UNIQUE: 'UNIQUE'>, 'STRUCT': <TokenType.STRUCT: 'STRUCT'>, 'SEQUENCE': <TokenType.SEQUENCE: 'SEQUENCE'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'COPY': <TokenType.COMMAND: 'COMMAND'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>}
WHITE_SPACE: Dict[Optional[str], TokenType] = {' ': <TokenType.SPACE: 'SPACE'>, '\t': <TokenType.SPACE: 'SPACE'>, '\n': <TokenType.BREAK: 'BREAK'>, '\r': <TokenType.BREAK: 'BREAK'>}
COMMANDS = {<TokenType.EXECUTE: 'EXECUTE'>, <TokenType.SHOW: 'SHOW'>, <TokenType.FETCH: 'FETCH'>, <TokenType.COMMAND: 'COMMAND'>}
COMMAND_PREFIX_TOKENS = {<TokenType.SEMICOLON: 'SEMICOLON'>, <TokenType.BEGIN: 'BEGIN'>}
NUMERIC_LITERALS: Dict[str, str] = {}
COMMENTS = ['--', ('/*', '*/')]
dialect
def reset(self) -> None:
930    def reset(self) -> None:
931        self.sql = ""
932        self.size = 0
933        self.tokens: t.List[Token] = []
934        self._start = 0
935        self._current = 0
936        self._line = 1
937        self._col = 0
938        self._comments: t.List[str] = []
939
940        self._char = ""
941        self._end = False
942        self._peek = ""
943        self._prev_token_line = -1
def tokenize(self, sql: str) -> List[Token]:
945    def tokenize(self, sql: str) -> t.List[Token]:
946        """Returns a list of tokens corresponding to the SQL string `sql`."""
947        if USE_RS_TOKENIZER:
948            return self.tokenize_rs(sql)
949
950        self.reset()
951        self.sql = sql
952        self.size = len(sql)
953
954        try:
955            self._scan()
956        except Exception as e:
957            start = max(self._current - 50, 0)
958            end = min(self._current + 50, self.size - 1)
959            context = self.sql[start:end]
960            raise TokenError(f"Error tokenizing '{context}'") from e
961
962        return self.tokens

Returns a list of tokens corresponding to the SQL string sql.

def tokenize_rs(self, sql: str) -> List[Token]:
1370    def tokenize_rs(self, sql: str) -> t.List[Token]:
1371        if not self._RS_TOKENIZER:
1372            raise SqlglotError("Rust tokenizer is not available")
1373
1374        try:
1375            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1376            for token in tokens:
1377                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1378            return tokens
1379        except Exception as e:
1380            raise TokenError(str(e))
size
sql
tokens