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

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

def tokenize_rs(self, sql: str) -> List[Token]:
1360    def tokenize_rs(self, sql: str) -> t.List[Token]:
1361        if not self._RS_TOKENIZER:
1362            raise SqlglotError("Rust tokenizer is not available")
1363
1364        try:
1365            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1366            for token in tokens:
1367                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1368            return tokens
1369        except Exception as e:
1370            raise TokenError(str(e))
size
sql
tokens